LEGENDS, TRADITIONS AND HISTORY IN MEDIEVAL ENGLAND
This page intentionally left blank
LEGENDS, TRADITIONS AND HISTORY IN MEDIEVAL ENGLAND
ANTONIA GRANSDEN
THE HAMBLEDON PRESS LONDON
AND
RIO GRANDE
Published by The Hambledon Press 1992 102 Gloucester Avenue, London NW1 8HX (U.K.) P.O. Box 162, Rio Grande, Ohio 45672 (U.S.A.) ISBN 1 852850167 © Antonia Gransden 1992 A description of this book is available from the British Library and from the Library of Congress Printed on acid-free paper and bound in Great Britain at the University Press, Cambridge
Contents
Acknowledgements
vii
Preface
ix
List of Illustrations
xi
1
Bede's Reputation as an Historian in Medieval England
2
Traditionalism and Continuity during the Last Century of Anglo-Saxon Monasticism
31
3
Legends and Traditions concerning the Origins of the Abbey of Bury St Edmunds
81
4
Cultural Transition at Worcester in the Anglo-Norman Period
107
5
Prologues in the Historiography of Twelfth-Century England
125
6
The Growth of the Glastonbury Traditions and Legends in the Twelfth Century
153
7
Realistic Observation in Twelfth-Century England
175
8
The Chronicles of Medieval England and Scotland
199
9
The Cronica Buriensis and the Abbey of St Benet ofHulme
239
10
The Continuations of the Flores Historiarum from 1265 to 1327
245
11
The Alleged Rape by Edward III of the Countess ofSalisbury
267
12
A Fourteenth-Century Chronicle from the Grey Friars at Lynn
279
1
vi
Legends, Traditions and History in Medieval England
13
The Date and Authorship ofjohn of Glastonbury's Cronica sive Antiquitates Glastoniensis Ecclesie
289
14
Antiquarian Studies in Fifteenth-Century England
299
Additional Notes to Chapters 6, 8 and 10
329
Index
335 360
Index of Manuscripts
Acknowledgements
The articles reprinted here appeared first in the following places and are reprinted by kind permission of the original publishers. 1. Journal of Ecclesiastical History, 32 (1981), 397-425. 2. Journal of'EcclesiasticalHistory,40(1989), 159-207. 3. English Historical Review, c (1985), 1 -24. 4. Medieval Art and Architecture at Worcester Cathedral British Archaeological Association, Conference Trans actionsfor theyear 1975,1 (1978), 1-14. 5. England in the Twelfth- Century, Proceedings of the 1988 Harlaxton Symposium, edited by Daniel Williams (Woodbridge, 1990), pp. 55-81. 6. Journal of Ecclesiastical History, 27 (1976), 337-58. 7. Speculum, xvu( 1972),29-51. 8. Journal of Medieval History, xvi (1990), 129-50, ibid., xvii (1991), 217-43. 9. Bulletin of the Institute of Historical Research, xxxvi (1963), 77-82. 10. Mediaeval Studies, xxxvi (1974), 472-92. 11. English Historical Review, Ixxxvii (1972), 333-44. 12. English Historical Review, Ixxii (1957), 270-8. 13. English Historical Review, xcv (1980), 358-63. 14. Antiquaries Journal, Ix (1980), 75-97.
This page intentionally left blank
Preface
The value to the historian of narrative sources as quarries of facts has always been recognized. However, there is a growing consciousness that, unless these sources are properly understood, they can neither be correctly interpreted nor their historical worth assessed. Used uncritically, it can be argued, they provide less reliable factual information than charters, public records and other kinds of official documents. This is true, but an unwary historian can also be led astray by, for example, forged charters and political bias in public records. In any case, narrative sources have more to offer than just solid facts, as historians increasingly appreciate. They provide clues to what the author and, by implication, some of his contemporaries, thought about events and personalities in their times. They also give some idea about how the author saw the past. For instance, they raise the puzzling question of the extent to which the author understood the difference between true history and legend. My interest in the narrative sources of English history was first aroused by Professor V.H. Galbraith, who was an excellent example of a scholar who successfully used narrative, particularly chronicle, evidence in conjunction with that of the public records. The enthusiasm, knowledge and perception with which he treated the chronicles of Richard IPs reign in his seminars at Oxford made him an inspiring and invigorating teacher. I have also been especially indebted to two other eminent medievalists, Sir Goronwy Edwards and Dom David Knowles. Sir Goronwy, my supervisor for a London Ph.D., combined erudition with good sense and good humour — and a sense of humour — an invaluable combination in a supervisor. Later, Dom David's belief that my plan to write a history of historical writing in medieval England was feasible, and that such a book should be very useful, gave me confidence and encouragement, but without his sustained interest and help I might not have persisted. Both he and Sir Goronwy read chapters of the first volume of my Historical Writing in England in draft and always had valuable comments and suggestions to make. The articles reprinted in the present book have been written in the last thirty-five years. They have been chosen because they all concern the study in medieval England of the past, medieval attitudes to history and
x
Legends, Traditions and History in Medieval England
medieval modes of historiographic expression. Most of them discuss in detail topics briefly alluded to in one or other of the two volumes of my Historical Writing. But Chapter 1, on Bede's reputation in medieval England, Chapter 5, on prologues to twelfth-century historical works, and nearly all Chapter 8, on the writing of chronicles, are independent of Historical Writing. Chapter 14, on fifteenth-century antiquarianisrn, deals thematically and more systematically with material which is scattered, though more detailed, in Historical Writing, ii. It has not been practicable to revise the articles thoroughly nor to bring them completely up to date. However, many corrections have been made, and to some extent bibiliographical references have been up-dated and new evidence taken into account. Wherever possible revisions are incorporated in the preexisting text; if there was not enough room they have been put in a note at the end of the article; in three instances it has been necessary to put a revisory note at the end of the volume (see pp. 329-35, Additional Notes to Chapters 6, 8 and 10). Debts to scholars who have helped me revise articles are acknowledged in the footnotes where appropriate. But here I must especially thank Dr Martin Brett for checking through Chapter 4, and Professor John Greatrex for telling me about, and sending me copies of, entries in the precentors' accounts of Worcester cathedral which throw light on the composing of a chronicle (cf. Article 8 and pp. 330-32). I am also deeply indebted to Mr Martin Sheppard of the Hambledon Press for his patient and sympathetic handling of the book, and to my daughter, Deborah Shields, for kindly and efficiently helping me with the index. Antonia Gransden Cambridge January 1992
List of Illustrations
1 David harping, from St Wulfstan's portiforium. (Cambridge, Corpus Ckrist{College, MS391 Jo. 24v)
105
Between pp. 119-120 2 The visions of Henry I, from the chronicle of John of Worcester: Henry is beset by peasants and by knights. (Oxford, Corpus Christi College, MS 157, p. 382) 3 The visions of Henry I, from the chronicle of John of Worcester: Henry is beset by bishops; his stormy crossing from Normandy. (Oxford, Corpus Christi College, MS 157, p. 383) Between pp. 184-185 4 'Benedict of Peterborough' (rota of William II of Sicily). (London, British Library, MS Cotton Vitellius E xvii,fo. 28) 5 A cripple from Gerald of Wales, Topographia Hibernica, dist. iii, cap. xxxv. (London, British Library, MS Royal 13 B viii,fo. 30v) 6 Irish hornblowers from Gerald of Wales, Topographia Hibernica, dist. iii, cap. xxxiv. (London, British Library, MS Royal 13 B viiijo. 30) 1 Plan of waterworks of Christ Church, Canterbury, c. 1165. (Cambridge, Trinity College, MSR.17.lJos. 284v-285) Between pp. 193-194 8 A crane, grus, from Gerald of Wales, Topographia Hibernica, dist. i, cap. xiv. (London, British Library, MS Royal 13 B viii,fo. 9) 9 A swan from a bestiary of c. 1200. (Oxford, Bodleian Library, MS Ashmole 1511, fo. 71)
xii
Legends, Traditions and History in Medieval England
10 Pigeons from a bestiary of c. 1120-30. (Oxford, Bodleian Library, MS Laud Misc. 247, fo. 161v) 11 Hedgehogs from a bestiary of c. 1200. (Cambridge, University Library, MS Ii.4.26,fo. 28v] 12 A bramble from an early twelfth-century herbal. (Oxford, Bodleian Library, MS Bodley 130, fo. 26) 13 A milk thistle from an early-twelfth century herbal. (Oxford, Bodleian Library, MS Bodley 130, fo. 37) Between pp. 227-228 14 The Chronicle of Bury St Edmunds, end of the annal for 1279 and beginning of that for 1280. (London, College of Arms, MS Arundel, 30, fo. 167v) 15 Chronicle of Bury St Edmunds, the end of last annal, for 1265, of the first recension, with the passage alleging that Simon de Montfort's corpse worked miracles erased. (London, British Library, MS Julius A l,fo. 43v) 16 The Chronicle of Bury St Edmund: annals from 1260 to 1263 with spaces between the annals for 1260 and 1262. (London, College of Arms, MS Arundel 30, fo. 150) 17 The Chronicle of Bury St Edmunds: the end of the annal for 1282. (Bury St Edmunds, Moyses Hall Museum, fo. 103v) 18 The Chronicle of Melrose Abbey: annals from 1208 to 1210, with spaces between the annals. (London, British Library, MS Faustina B ix,fo. 28v) 19 Scotichronicon by Walter Bower, Book XVI, the end of chapter 15 and the beginning of chapter 16 (1428-30). (Cambridge, Corpus Christi College, MS 171,fo. 343) 20 The Chronicle of Bury St Edmunds: part of the annal for 1291. (London, College of Arms, MS Arundel 30, fo. 179) 21 The Chronicle of Bury St Edmunds: part of the (penultimate) annal for 1295. (London, College of Arms, MS Arundel 30, fo. 190) 22 The Westminster Chronicle: annal for 1390, including a notice of the consecration of Brother Alexander ((Bache)) as bishop of St Asaph. (Cambridge, Corpus Christi College, MS 197a, p. 197)
List of Illustrations
xiii
23 The Chronicle of Bury St Edmunds: a marginal addition to the annal for 772. (London, British Library, MS Cotton Julius A I Jo. 23v) 24 The Chronicle of Winchcombe Priory: annals from 1074 to 1088, with numerous additions. (London, British Library, MS Cotton Faustina B.I.,fo. 13) 25 The Chronicle of Bury St Edmunds: part of the annal for 1294. The succession of William de Hotham to the archbishopric of Dublin has been added in the margin. (London, College of Arms, MS Arundel 30, f o . 188v) 26 The Chronicle of St Benet of Hulme, commonly known as 'John de Oxenedes': part of the annal for 1282. (London, British Library, MS Cotton Nero D II, fo. 235v) 27 The St Albans copy of Ralph Diceto's Ymagines Historiarum: the annal for 1198 'edited' for the scribe making an abbreviated version for Dunstable Priory. (London, British Library, MS Royal 13 E VI, fo. 134v) Between pp. 247-248 28 The 'Merton' Flores Historiarum: the coronation of Henry III. (Windsor, Eton College, MS 123, fo. 194) 29 The 'Merton' Flores Historiarum'. the coronation of Edward I (who holds the rod with the dove). (Windsor, Eton College, MS 123, fo. 237) Between pp. 308-309 30 Thomas Elmham's facsimile of the (forged) privilege of St Augustine to St Augustine's, Canterbury. (Cambridge, Trinity Hall, MSI Jo. 2) 31 Thomas Elmham's picture of the seal of an abbey dedicated to St Stephen. (Cambridge, Trinity Hall, MS I Jo. 24) 32 Thomas Elmham's plan of the high altar and sanctuary at St Augustine's, Canterbury. (Cambridge, Trinity Hall, MS I Jo. 77) 33 Thomas Elmham's map of Thanet. (Cambridge, Trinity Hall, MS I,fo. 42v)
xiv
Legends, Traditions and History in Medieval England
34 Arthgallus, a legendary British earl of Warwick, from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976) 35 Ufa, from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976) 36 The seal, 1353-54, of Thomas Beauchamp, earl of Warwick (1329-62). (London, British Library, seal no. xliii. 18) 37 Thurkill, from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976) 38 William de Newburgh, earl of Warwick (d. 1184), from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976) 39 Waleran de Newburgh (d. ?1203), from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976) 40 Thomas Beauchamp, earl of Warwick (1370-1401), from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976) 41 Richard Beauchamp, earl of Warwick (1403-39), from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976) 42 Two pages from William Worcester's Itinerary. The page on the right shows a cross section of the mouldings of the south porch of St Stephen's church, Bristol. (Cambridge, Corpus Christi College, MS 210Jo. 129)
To Jonathan
This page intentionally left blank
1 Bede's Reputation as an Historian in Medieval England1
T
he Hand-List of Bede Manuscripts by M. L. W. Laistner and H. H. King is not definitive, since it is based partly on the evidence of catalogues and other printed material and not on the manuscripts themselves. However, it provides a useful rough guide. It lists approximately 160 surviving manuscripts of the Historia Ecclesiastica. Laistner constructed a table showing the percentage of the total number copied in each century: 2 BEDE Century copied eighth 8c eighth/ninth ninth tenth eleventh twelfth
MANUSCRIPTS
Percentage 3'4 7-0
5-o 9-0 31-0
Century copied
Percentage
thirteenth fourteenth fifteenth sixteenth doubtful
10-0 17-0 12-0 1-2 4.4
Laistner estimated that at least 70 per cent of the extant manuscripts of the Historia Ecclesiastica were copied in England.3 The pattern of survival can be seen, therefore, as throwing some light on the relative popularity of the Historia Ecclesiastica in one century and another. The most remarkable feature is the large percentage of manuscripts written in the twelfth century. It is also noteworthy that the Historia Ecclesiastica maintained its popularity in the late Middle Ages; in fact, a fairly high percentage of the copies were made in the fourteenth century. 1
This article is based on a paper delivered at the British Library in September 1980, in a series of lectures on the Benedictines in Britain. 2 M. L. W. Laistner, with the collaboration of H. H. King, A Hand-List of Bede Manuscripts, Ithaca, New York 1943, 7. The edition of the Historia Ecclesiastica cited below is Venerabilis Baedae Historiam Ecclesiasticam . ., ed. C. Plummer, Oxford 1896, 2 vols. s Laistner, op. cit., 8.
2
Legends, Traditions and History in Medieval England
Evidence suggesting why the Historia Ecclesiastica was more popular in some centuries than in others can be sought in contemporary sources. The opinions the chroniclers express about Bede, aftd the use they make of the Historia Ecclesiastica, are particularly valuable, but references in other types of sources, for example, official documents, can also be revealing. In such a study some account must be taken of Bede's minor historical works, the two universal histories in De Temporibus* and De Temporum Rationed the prose Life of St Cuthbert6 and the metrical one,7 and the Lives of the Abbots of Wearmouth. 8 Nor can Bede's reputation as an historian be wholly dissociated from his reputation as a saint and Father of the Church. The Anglo-Norman eulogies cited below make it plain that Bede's historical works were considered merely as one element in his total achievement. Moreover, his venerability and learning in the scriptures added weight to his authority on historical matters.9 The chronicles and other sources show that the Historia Ecclesiastica was used in the Middle Ages for various reasons. For instance, it could provide spiritual inspiration for a religious revival, since it was the main source of information about the Northumbrian renaissance, the golden age of the Anglo-Saxon Church, of which Bede was the principal luminary. Religious leaders often wanted their movement to emulate the achievements of that period and, therefore, consulted the Historia Ecclesiastica to discover details. Indeed, the Historia Ecclesiastica seems on occasion to have provided them with an initial incentive. The chroniclers themselves used Bede's historical works for information, in order to establish the continuous sequence of English history. Occasionally, when a chronicler found Bede's narrative conflicted with someone else's account of the same period, he might prefer Bede's authority as being of greater weight. Some chroniclers had a tendentious purpose. The Northumbrian period provided the English Church with a glorious past. Therefore, a chronicler might extol Bede himself arid use his works to paint a flattering picture of the Anglo-Saxon Church in the 4 Printed in M.G.H., Auctorum Antiquissimorum, xm, Chronica Minorum saec. iv—vii, iii, ed. T. Mommscn, Berlin 1898, 247-317. 5 Printed in ibid., iii. 247-327. 6 Printed in Two Lives of St Cuthbert, ed., with an English translation, B. Colgrave, Cambridge 1940. 7 Printed as 'Bedas metrische Vita Sancti Cuthberti', ed. W. Jaager, Palaestra, cxcviii. Leipzig 1935. 8 Printed in Plummer, Venerabilis Baedae, i. 364-87. 9 Professor Laistner constructed a table which demonstrates the percentage of the surviving manuscripts of Bede's biblical commentaries belonging to each century, similar to his table for the Historia Ecclesiastica', Laistner, Hand-List, 4. The pattern resembles that for the Historia Ecclesiastica, in so far as there is a high percentage belonging to the twelfth century. However, as most of the extant manuscripts of the commentaries are of continental provenance, his table does not indicate their popularity in England as accurately as does that for the Historia Ecclesiastica.
Bede's Reputation as an Historian in Medieval England
3
seventh century. Above all, Bede's Historia Ecclesiastica was an authority. Ancient origins, whether of the English Church as a whole or of individual monasteries, attested by Bede had especial prestige. But the Historia Ecclesiastica did more than increase fame; it could also be appealed to in specific disputes as a storehouse of precedents. It could be cited by 'official' chroniclers presenting the case of a party to a dispute and by those actually involved in the proceedings. The evidence supplied by the chroniclers and others concerning Bede's reputation as an historian in medieval England will be treated chronologically. This arrangement makes possible correlation with Laistner's table. However, it is not practicable to divide the material century by century, as Laistner did the manuscripts. This is because evidence is scarce or totally lacking in some periods, though plentiful in others. Moreover, when abundant, it tends to have distinctive characteristics which link it with one or more of the motives for interest in Bede outlined above. In the divisions and subdivisions adopted below, such characteristics have been taken into account. I. The Anglo-Saxon Period
Bede's influence as an historian was established soon after his death. He played a vital role in the genesis of annalistic historiography among the Anglo-Saxons. The chronological epitome which he had appended to the Historia Ecclesiastica became the stock from which contemporary annals grew. He himself had probably added annals from 731 to 734, the so-called Appendix, and shortly afterwards a continuation from 735 to 766 was written in Northumbria. 10 Another series of Northumbrian annals, incorporating the Appendix and continuing until 802, was added to the epitome in the late eighth and early ninth centuries.11 The importance of the Historia Ecclesiastica was recognised in King Alfred's reign. It was translated into the vernacular,12 possibly at the king's instigation as part of his programme to raise the standard of education and Christian observance.13 Moreover, Alfred needed to 10 The epitome is printed in Plummer, Venerabilis Baedae, i. 352-6, and the continuations to 734 and 766 in ibid., i. 361-3. Cf. ibid., ii. 345 note. 11 See Chronica Magistri Rogeri de Houedene, ed. W. Stubbs (Rolls Series (hereafter cited as R.S.), 1868/71, 4 vols), xxviii-xxx, and P. H. Blair, 'Some observations on the Historia Regum attributed to Symeon of Durham', Celt and Saxon, ed. N. K. Chadwick, Cambridge 1963,86-99. 12 Printed as The Old English Version of Bede's Ecclesiastical History of the English People, ed. T. Miller (Early English Text Society (hereafter cited as E.E.T.S.), o.s., xcv-xcvi, cx-cxi, 1890, 1898, 2 pts). Six manuscripts of the Old English Bede survive, all of the ninth and tenth centuries; Laistner, Hand-List, 111-12. 13 Professor Whitelock, after a careful examination of the Old English Bede, concluded: 'That the work was undertaken at Alfred's instigation remains a probability'; D. Whitelock, 'The Old English Bede', Proceedings of the British Academy, xlviii (1962), 77.
4
Legends, Traditions and History in Medieval England
ifnprove the prestige of his dynasty and the morale of his subjects; an obvious expedient was to make accessible the history of the early achievements of the English people—as narrated by Bede. As mentioned above, Bede can be regarded as the father of the early Northumbrian annals. These annals were, of course, in Latin, but Bede can claim almost equal importance in the inception of vernacular annals in the south of England. The form of the Anglo-Saxon Chronicle, which King Alfred himself may have commissioned,14 was partly determined by the epitome of the Historia Ecclesiastica, and both the epitome and the main text supplied some of the information. 15 In addition, at about the same time Bede made an important contribution to the hagiographical tradition, which had been fairly quiescent since the eighth century. The Old English martyrology was probably written under King Alfred. Much of its material was borrowed from the Historia Ecclesiastica; and some derived from Bede's Lives of St Cuthbert, and from the Lives of the Abbots.16 Bede's Historia Ecclesiastica and Lives of St Cuthbert were influential during the monastic revival in the last half of the tenth century. But their influence was on the actual course of the movement rather than on historical literature. The Historia Ecclesiastica provided the reformers, SS Dunstan, Ethelwold and Oswald, with a spiritual tradition native to England. Apparently they saw themselves to some extent as successors to the seventh-century religious leaders. The council at Winchester, summoned by King Edgar to draw up the Regularis Concordia, which imposed
14
Plummer believed that King Alfred commissioned the Anglo-Saxon Chronicle; see Two of the Saxon Chronicles Parallel, ed. C. Plummer and J. Earle, Oxford 1892-9, 2 vols, 11. civ-cv). A case can be made for regarding it as an official history; see A. Gransden, Historical Writing in England c. 55o-c. ijoo, London 1974, pp. 34-5. However, Professor Whitelock takes a cautious view; The Anglo-Saxon Chronicle, ed. D. Whitelock, with D. C. Douglas and S. I. Tucker, London 1961, xxii-xxiii. 15 See Plummer and Earle, op. cit;, n. xli and n. 2, Ixviii-lxix, cxiii and n. i. It is noteworthy that in the southern versions of the Anglo-Saxon Chronicle (A, B and C) nearly all the annals derived from the Historia Ecclesiastica are from the epitome, while in the northern version, that lying behind D and E, most such annals are from the main text. (The northern version also used the Northumbrian annals; see ibid., n. Ixviii-lxxi.) The complex relationship of the Anglo-Saxon Chronicle to Bede is fully discussed byj. Bately, 'Bede and the Anglo-Saxon Chronicle', Saints, Scholars and Heroes, ed. M. H. King and W. M. Stevens, Hill Monastic MS Library, Minnesota 1979, 233-44. 16 For the use of Bede in the Old English martyrology see An Old English Martyrology, ed. G. Herzfield (E.E.T.S., o.s., cxvi, 1900), xxxiii/xxxiv, xxxvi/xlii passim, and C. E. Fell, 'Edward King and Martyr and the Anglo-Saxon hagiographical tradition', Ethelred the Unready: Papers from the Millenary Conference, ed. D. Hill (British Archaeological Reports, British series, lix, 1978), 2, 3. For the date of the Old English Martyrology, and its probable association with King Alfred's reforms, see C. Sisam, 'An early fragment of the Old English Martyrology', The Review of English Studies, N.S., iv (1953), 217. It is probably based on a lost ninth-century Latin original; ibid., 212-13, Herzfeld, op. cit., xxxii, xxxvi, and F. Liebermann, 'Zum Old English Martyrology', Archiv fur das Studium der neueren Sprachen, cv (1900), 86 et seq.
Bede's Reputation as an Historian in Medieval England
5
some unity on the monasteries, had a precedent in the synod of Whitby summoned by King Oswy to heal the division in the English Church and fully described by Bede. The Regularis Concordia, indeed, has an allusion to the Histona Ecclesiastica.11 Two of the monasteries founded by St Ethelwold during the revival, Peterborough and Ely, were in fact re-foundations, for there were monasteries on those sites in the seventh century, as Bede records (Histona Ecclesiastica, iv. 6, 19). And St Oswald apparently tried to refound yet another 'Bedan' monastery, Ripon. 18 Both St Ethelwold and St Oswald venerated the early saints, whose acts Bede described. St Ethelwold included a solemn blessing for the feast of St ^thelthryth of Ely, and a full page miniature of her, in his Benedictional. (St Cuthbert figures in the miniature of the Confessors.)19 And St Oswald searched Northumbria for relics of the ancient saints.20 In so far as Bede's Histona Ecclesiastica and Lives of St Cuthbert had a literary influence during the revival, it was on hagiography. St Ethelwold's pupil, yElfric, used the Histona Ecclesiastica to provide the acts of SS Alban, ^Ethelthryth and Oswald for his Lives of Saints.21 And possibly it was owing partly to the example of Bede's Lives of St Cuthbert that biographies were produced of the three tenth-century religious leaders, SS Dunstan, Ethelwold and Oswald.22 Indeed, the Life of St Oswald contains evidence of Bede's influence. The author compares a miracle performed by St Oswald, the rescue of monks in a sinking boat, with one of St Cuthbert's—which is described in Bede's metrical Life.23 And he has an allusion to the Historia Ecclesiastica; he comments that Foldbriht, abbot of Pershore, ignored the advice given by Bede in the Historia Ecclesiastica (iii. 25), that it is better to speak well than ill of those whom you do not know, and proceeds to relate, perhaps with Drythelm in mind (Historia
11 Regularis Concordia, ed., with an English translation, T. Symons, Nelson's Medieval Texts 1953, 2 and n. 2. Cf. pp. xlv n. i, 3 and n.c. 18 The Historians of the Church of York, ed.J. Raine (R.S., 1879-94, 3 vols), i. 462. 19 See The Benedictional of Saint /Ethelwold, ed., in facsimile, G. F. Warner and H. A. Wilson (Roxburghe Club), Oxford 1910, xv-xvi, 37, and fos i, gov. 20 Raine, op cit., i. 462. 21 Aelfric's Lives of Saints, ed. W. W. Skeat (E.E.T.S., o.s., Ixxvi, Ixxxii, xciv, cxiv, 18811900, 2 vols), i. 414-41 (nos xix, xx); ii. 124-43 (no. xxvi); cf. ibid., ii. xlviii. For /Elfric's use of Bede for his Lives see J. H. Ott, Uber die Quellen der Heiligenleben in Aelfrics Lives of Saints, i (Inaugural-Dissertation), Halle 1892, 44-7 (for SS Alban and /Ethelthryth), Ruth Waterhouse, '^Ifric's use of discourse in some saints' lives', Anglo-Saxon England, ¥ ( 1 9 7 6 ) , 83/91 and nn. passim, and Fell, 'Edward King and Martyr', 2, 3. 22 The earliest Lives of SS Dunstan, Ethelwold and Oswald are printed respectively in: Memorials of St Dunstan, ed. W. Stubbs (R.S., 1874), 3-52; Chronicon Monasterii de Abingdon, ed.J. Stevenson (R.S., 1858, 2 vols), ii. 255-66; and Raine, op. cit., i. 399-475. 23 Raine, op cit., i. 448. Cf. Bede's metrical life chap, ix (ed. Jaager, 'Bedas metrische Vita', 77-8).
6
Legends, Traditions and History in Medieval England
Ecclesiastic a, v. 12), how Foldbriht died and came alive again, having had a vision of the afterworld.24 However, Bede's influence as an historian on the tenth-century monastic revival was limited. Scholars were at least as interested in his computistical works as in his historical ones. Byrhtferth based his Manual25 and /Elfric his De Temporibus Anni26 on the De Temporum Ratione, the De Temporibus and the De Natura Rerum. Nor did the revival rely exclusively, or even mainly, for inspiration on the tradition of the seventh-century Anglo-Saxon Church. No serious attempt was made to revive monasticism in the North, and the English characteristics of the revival are less conspicuous than its continental ones; its strongest influences came from abroad. Meanwhile, it seems that Bede's cult as a saint developed throughout the Anglo-Saxon period. His relics at Jarrow were reputed to work miracles at least by the late eighth century.27 Then, in the mid-eleventh century, Alfred (or Alured), son of Westou, sacrist of Durham, thought it worth his while to steal Bede's relics from Jarrow and re-bury them in Durham cathedral, next to St Cuthbert. 28 Bede's reputation is also witnessed by the fact that one of the first acts of Wulfstan as bishop of Worcester, in 1062, was to dedicate a church to him: 'It is most fitting', wrote Wulfstan's biographer, Coleman (sometime between 1095 and 1113), 'that his first dedication should have been to him who is first in fame as a man of letters among the English people.'29
24 Raine, op. cit., i. 439-41. Foldbriht was abbot of Pershore c. 970-88; M. D. Knowles, C. N. L. Brooke and V. C. M. London, The Heads of Religious Houses, England and Wales 940—1216, Cambridge 1972, 58. For later examples of visions resembling Drythelm's experienced by inhabitants of the north of England, see p. 8 n. 36 below. 25 Printed as Byrhtferth's Manual, ed. S. J. Crawford (E.E.T.S, o.s., clxxvii, 1929; only one volume, comprising text with footnotes, published). For the use of Bede in the Manual see Crawford's footnotes and H. Henel, 'Byrhtferth's Preface: the Epilogue of his Manual?', Speculum, xviii (1943), 290. Byrhtferth has a eulogy on Bede, especially praising his contribution to computistics, in his 'epilogue', which was written to accompany a copy of Bede's computistical works, or perhaps as an epilogue to the Manual', see the printed text in G. F. Forsey, 'Byrhtferth's Preface', Speculum, iii (1928), 516-19; cf. Henel, op cit., 288-302, and M. Lapidge, 'The hermeneutic style in tenth-century Anglo-Latin literature', Anglo-Saxon England, iv (1975), 90 and n. 3. 26 Printed as Aelfric's De Temporibus Anni, ed. H. Henel (E.E.T.S., o.s., ccxiii, 1942). For /Elfric's use of Bede therein see ibid., liii-lvi. 27 Alcuin, De Pontificibus et Sanctis Ecclesiae Eboracensis Carmen in Raine, Historians o f . . . York, i. 388. 2 *Symeonis Monachi Opera Omnia, ed. T. Arnold (R.S., 1882, 1885, 2 vols), i. 88-9. Cf. H. S. Offler, 'The date of Durham (Carmen de Situ DunelmiY, Jnl English and Germanic Philology, Ixi (1962), 592-3. 29 The Vita Wulfstani of William of Malmesbury, ed. R. R. Darlington (Camden Society, 3rd ser., xl, 1928), 20.
Bede's Reputation as an Historian in Medieval England
7
II. c. io66-c. 1136 Perhaps at no time in medieval England were Bede's historical works more intensively studied than in the Anglo-Norman period. There were two main reasons for this. i. Bede as the inspiration for the monasticisation of the North in the late eleventh century Aldwin, a monk of Winchcombe, was moved by the Historia Ecclesiastica and the Lives of the Abbots of Wearmouth to try to revive religious life in the region formerly ancient Northumbria. A near contemporary, the chronicler Symeon of Durham, recorded that Aldwin 'learnt from the Historia Anglorum [i.e. the Historia Ecclesiastica} that the province of Northumbria was once crowded with communities of monks and packed with multitudes of saints'.30 Accordingly, in 1073 or 1074, Aldwin went to Evesham, where he was joined by two monks, Elfwy and Reinfrid, and with them travelled north. The group went to Walcher, the bishop of Durham (1071-80). Walcher, Symeon recorded, 'gave them the monastery of the blessed Paul, the apostle, formerly built by Abbot Benedict [Biscop] at Jarrow, the walls of which still stood, but roofless and hardly preserving any sign of its ancient dignity'. 31 The missionaries thatched the church and began to celebrate the divine office. Later, after they had failed to restore Whitby and Melrose,32 Walcher gave them Wearmouth. The monastery, Symeon wrote, 'was formerly famous and noble, as Bede, who lived there from infancy, testifies; but at this time it was hardly possible to appreciate its antiquity because of its ruined condition'. 33 The missionaries built themselves a shack and began to lead the religious life there. Walcher, who wanted to become a monk himself planned to found a monastic community in the see of St Cuthbert at Durham. 34 However, his scheme was ended by his murder. It was revived by his successor, William of St Carilef, who finally instituted a monastic community in Durham cathedral in 1083. William acted under the direct influence of Bede's Historia Ecclesiastica and prose Life of St Cuthbert. (His copy of the latter still survives in Durham cathedral library.)35 Symeon recorded that William, grieved at the desolate state of the see of St Cuthbert and the neglect of the saint's body, took advice how to amend the situation: He asked the wise old men of the bishopric how things were originally managed at St Cuthbert's: they replied that the see was on the island of 30
Symeoms Monachi Opera Omnia, i. 108. Ibid., i. 109. 32 Ibid., i. 111. 33 Ibid., i. 112. 34 Ibid.,i. 113. 35 MS B 11.35 in the chapter library at Durham; R. A. B. Mynors, Durham Cathedral Manuscripts to the End of the Twelfth Century, Oxford 1939, 41. 31
8
Legends, Traditions and History in Medieval England Lindisfarne and that monks served [St Cuthbert], both when he was alive and when he was dead, with due veneration; their assertion accords with the evidence in the Life [of St Cuthbert] and with the Historia Ecclesiastica Gentis Anglorum.36
The evidence cited shows that Bede's historical works made a positive contribution to the revival of monasticism in the North. It could even be claimed that they provided the initial incentive. The instigator of the movement, Aldwin, was an Englishman, and so were his two companions from Evesham, Elfwy and Reinfrid. They were continuing the tradition of the Anglo-Saxon Church, not only its missionary spirit but also the study of Bede. However, it is possible that Aldwin and his companions had another motive besides missionary zeal and the inspiration of Bede. Perhaps they intended to augment the reputation of the Anglo-Saxon Church, in order to help protect its traditions against the new Norman rulers. Walcher and William of St Carilef, although Normans, may have co-operated with the English missionaries partly to strengthen their own positions. Possibly they hoped to placate the hostile English in their diocese by showing an appreciation of the Anglo-Saxon achievement.37 Moreover, by increasing the prestige of Durham cathedral they may have hoped to be able to resist their enemies more effectively.38 Therefore, the appeal to Bede may have been one by-product of the threat to the tradition of the English Church posed by the new regime and of the political instability in the North after the Norman Conquest. 2. The use of Bede by chroniclers and controversialists in the Anglo-Norman period The Norman Conquest put in jeopardy the traditions of the whole Church in England as well as of each monastery. The threat impelled monks to write chronicles and saints' Lives in order to establish continuity with the past. Monks wrote chronicles covering the period from the earliest times to their own day in order to demonstrate that English church history was long and unbroken. In addition, they wrote local histories 36
Symeonis Monachi Opera Omnia, i. 120. An instance of the strength of the Bedan tradition in the Durham area is provided by Symeon who relates that a local man, Eadulf, died and came alive again, having had a vision of the afterworld; Symeon explicitly compares Eadulfs experience with Drythelm's; ibid., i. 114-15. A similar vision was experienced by a boy, Orm, who also lived in north-east England; an account was written in 1126 or soon after, and sent by Sigar, the parish priest of Newbald, in the East Riding of Yorkshire, to Symeon at Durham. See H. Farmer, 'The Vision of Orm', Analecta Bollandiana, Ixxv (1957), 72—82. For a pre-Conquest example of a man having a vision like Drythelm's see p. 402 and n. 24 above. 37 Walcher's attempt to placate the English in the political sphere was, ironically, one factor leading to his murder; see D. C. Douglas, William the Conqueror, London 1964, 240-1. 38 William of St Carilef could not rely on the support even of the king. For his quarrel with William Rufus, 1087-8, see A. L. Poole, From Domesday Book to Magna Carta, Oxford 1951, 100-4.
Bede's Reputation as an Historian in Medieval England
9
and hagiographies to prove the antiquity of their own monasteries. They hoped in this way to increase the prestige of the English Church and of the particular monasteries, so that they could the better withstand the innovations of their new ecclesiastical masters and the onslaughts of the new secular ones. In order to increase the prestige of the Church and of the monasteries, it was necessary to prove not only the continuity, but also the glory of their past history. Here writers faced a problem, since it was part of Norman propaganda to denigrate the eleventh-century Anglo-Saxon Church; one justification of the Conquest was that William i and Lanfranc had instigated a movement of ecclesiastical and monastic reform. A solution was that writers should turn to Bede, since he provided an account of a golden age of the Anglo-Saxon Church in the remote past. At no period did the chroniclers pay greater tribute to Bede than under William Rufus and Henry i. Chroniclers of general history, and some of the local historians, expressed admiration for both the sanctity of his life and the profundity of his learning. Some did so in passing, but others devoted a page or more to Bede's life (based on the Historia Ecdesiastica, v. 24, and Cuthbert's letter to Cuthwin) and achievements. Orderic Vitalis39 and Henry of Huntingdon, 40 who wrote about Bede at some length, commented especially on his saintliness and fame as a Father of the Church. Florence of Worcester concentrated mainly on his historiography. He recorded Bede's death, noting that he 'lucidly narrated the deeds of his people (his life and his History ended at one and the same time)'; 41 he noted Bede's knowledge of chronology, a subject in which he himself was interested.42 The longest, most eloquent and most comprehensive, encomium of Bede by a chronicler of general history is William of Malmesbury's in the Gesta Regum.^ Much of it is about Bede's saintly qualities and preeminence in biblical scholarship. Of Bede's fame in general, William wrote: He is a man easier to admire than adequately to extol. He was born in a remote corner of the earth. (Indeed, Britain is called another world by some because, being cut off by the sea, many cosmographers overlooked it.) And the place of his birth and education was in its remotest region, near 39
The Ecclesiastical History of Or dene Vitalis, ed., with an English translation, M. Chibnall Oxford Medieval Texts 1969-81), iii. 66, 68. 40 Henrici Archidiaconi Huntendunensis Historia Anglorum, ed. T. Arnold (R.S., 1879), 115-16. 41 Florentii Wigorniensis Monachi Chronicon ex Chronicis, ed. B. Thorpe (English Historical Society, 1848-9, 2 vols), i. 53. 42 ibid., i. 50. For other references to Bede see ibid., i. 44, 45, 46. For Florence's interest in chronology, see Gransden, Historical Writing, 144 and n. 56, 145 and n. 63, 146. 43 Willelmi Malmesbiriensis Monachi de Gestis Regum Anglorum . . ., ed. W. Stubbs (R.S., 1887-9, 2 vols), i. 58-67.
10
Legends, Traditions and History in Medieval England Scotland. Nevertheless, the bright light of his learning illuminated the whole world. 44
William commented twice on Bede's unique position in the historiographical tradition of the Anglo-Saxons, briefly at the beginning of the prologue to the Gesta Regum*5 and at some length in the encomium. In the latter he wrote: with [Bede] was buried almost all record of events until our own day; for there was no Englishman to rival his learning or imitate his virtues, none to emulate his glory and take up the thread of his teaching, now cut short. A few loved Jesus equally, but, although not ill-educated in letters, they spent their lives in graceless silence; others, after the merest taste of learning, preferred sloth and idleness. And thus, lazier men succeeded the lazy, and soon the love of scholarship grew cold throughout this island.46
Chroniclers could express their admiration of Bede in other ways than by eulogy. In some cases they explicitly stated that they wrote in order to continue Bede's work, and occasionally they imitated his method. Florence of Worcester wrote: We in truth, inspired by God, . . . have thought it fitting to leave a record of events from the death of Bede onwards, for our faithful successors.47
Similarly, William of Malmesbury wrote the Gesta Regum as a continuation of the Historia Ecclesiastica. Having deplored in the prologue the paucity of chroniclers after Bede's death, and the absence of any competent ones until his own day, he wrote: 'moved by love of my country and by the persuasion of others, I have decided to bridge the gap of interrupted time'.48 And Henry of Huntingdon stated in the prologue to his chronicle that, on the advice of Alexander, bishop of Lincoln, he had 'continued the Ecclesiastical History of the venerable Bede to the best of his ability'.49 Appreciation by imitation is particularly apparent in William of Malmesbury's chronicles. He began the Gesta Regum by listing his authorities in the prologue, in the same way as Bede had done in the Historia Ecclesiastical Moreover, the section on each bishopric in the Gesta Pontificum51 starts with a geographical description of the see, reminiscent of the description of Britain with which the Historia Ecclesiastica opens. Similarly, Henry of Huntingdon began his history with the 44
Ibid., 1.59. Ibid., i. 1-2. 46 Ibid.,1.66-7. 47 Chronicon ex Chronicis, i. 53. 48 Gesta Regum, i. 2. 49 Historia Anglorum, 3. 50 Gesta Re gum, i. 1-2. 51 Willelmi Malmesbiriensis Monachi de Gestis Pontificum Anglorum . . ., ed. N. E. S. A. Hamilton (R.S., 1870). 45
Bede's Reputation as an Historian in Medieval England
11
geography of Britain (partly based on Bede's).52 And Orderic (who made a copy of the Historia Ecclesiastica with his own hand) 53 may well have chosen the title of his work in imitation of Bede, and probably also borrowed from him (Historia Ecclesiastica, v. 24) the idea of appending an autobiography. 54 Thus Bede's life and learning inspired the Anglo-Norman historians to eulogy, and his achievements as an historian in particular inspired them to continue his work and even to imitate his method. However, it cannot be claimed that the influence of the Historia Ecclesiastica on historical method in the Anglo-Norman period was profound; the chroniclers were subject to numerous other influences. Probably it had more effect on the content of the chronicles than on their form. As a quarry of information for early Anglo-Saxon history the Historia Ecclesiastica had no rivals; it was extensively used by Florence of Worcester, William of Malmesbury, Orderic Vitalis and Henry of Huntingdon. In general their use of the Historia Ecclesiastica was uncritical. Chroniclers extracted information, or copied passages verbatim from the Historia Ecclesiastica without comment, sometimes putting them alongside material from other, usually less reliable, sources. (For example, Henry of Huntingdon interposed an account based on Nennius of the Britons' Trojan origins.)55 However, William of Malmesbury was on occasion scholarly in his approach to the Historia Ecclesiastica. In the Gesta Regum he pointed out a discrepancy between the estimated length of Ethelbert's reign as recorded in the Historia Ecclesiastica, of fifty-six years (ii. 5), and that given in the Anglo-Saxon Chronicle (s.a. 565), of fifty-three years; the matter, he wrote, must be left to the reader's judgement.56 In the Gesta Pontificum he commented that Bede's account of St Wilfrid's life is incomplete; therefore, he will use the Life by Eddius Stephanus.57 He also expressed doubt whether, as Bede recorded (Historia Ecclesiastica, iii. 6), St Oswald's arms were preserved at Bamburgh58 and wondered at Bede's silence on the foundation of Evesham by St Ecgwin.59 Those local historians who, in the Anglo-Norman period, turned to Bede had a more limited purpose than the chroniclers of general events; they used him merely to prove the antiquity of their houses. The historian of Ely Abbey described Bede as the Venerable' and 'most eloquent' doctor, and as 'a most truthful historian'.60 His main authority 52
Historia Anglorum, 5-7. Rouen MS 1343. See Bede's Ecclesiastical History of the English People, ed., with an English translation, B. Colgrave and R. A. B. Mynors (Oxford Medieval Texts, 1969), p. Ixi. 54 The Ecclesiastical History of Orderic Vitalis, ed. Chibnall, vi. 550-6. See ibid., vi. 556 n. i. 55 Historia Anglorum, 13. 56 Gesta Regum, i. 13. Cf. Historia Ecclesiastica, ed. Plummer, ii. 85 note. 57 Gesta Pontificum, 210; cf. ibid., 238-9. 58 Ibid., 293; cf. Historia Ecclesiastica, ed. Plummer, ii. 141, 158, notes. 59 Gesta Pontificum, 296. 60 Liber Eliensis, ed. E. O. Blake (Camden Society, 3rd ser., xcii, 1962), 2, 13. 53
12
Legends, Traditions and History in Medieval England
both for the abbey's foundation and for the life of its patron sahit, /Ethelthryth, was the Historia Ecclesiastical1 In addition, he used the universal chronicle in the De Temporibus, to provide St y£thelthryth's life with a comparative chronology.62 Symeon of Durham praised Bede at greater length than did the Ely writer. He included one eulogy in his history of the see of St Cuthbert (which covered the period from 635 to iog6) 63 and another in his letter about the early archbishops of York which he addressed to 'Hugh, dean of York'. 64 Bede, as the biographer of St Cuthbert, had especial importance to Symeon. And Bede himself was, in a sense, a patron saint of Durham; as has been seen, his relics had been removed to the cathedral in the mid-eleventh century, where they had become the centre of a cult.65 Symeon wrote of Bede in almost intimate terms, referring to him as 'our Bede'.66 He described Bede's 'study' at Jarrow in a way which suggests that he himself had visited it. It was 'a stone cottage; there Bede would sit, free from all disturbance, to meditate, read, dictate and write'. 67 He also recorded that the porticus 'on the north side of the church of St Paul at Jarrow was dedicated to Bede, to preserve his memory for the faithful'. 68 Symeon's two eulogies of Bede resemble William of Malmesbury's, but they are shorter and alone mention Bede's knowledge of Greek.69 Like William, Symeon seems to have been most impressed by the biblical commentaries,70 and, again like William, he commented on the fact that Bede's genius flowered in such a distant place: It seems unbelievable to many that in a remote corner of the world, a man, who never crossed the sea to acquire knowledge and never frequented the schools of philosophers, should shine with such erudition, and that his books should be famed throughout the world. 71 Moreover, Symeon wrote, Bede could describe far away lands which he had never visited: In his books he vividly describes the location, the nature, and the merits of many and diverse countries and regions, as if he himself had been there, 61
See ibid., pp. xxviii, liii, 2-50 passim. Ibid., 50-1. 63 Libellus de exordio atque procursu istius, hoc est Dunelmensis Ecclesiae, printed as Historia Dunelmensis Ecclesiae in Symeonis Monachi Opera Omnia, i. 3-135. 64 Printed in ibid., i. 222-8. One Hugh was dean from logo-c. 1109, and another from c. 1130-^. 1132; see J. Le Neve, Fasti Ecclesiae Anglicanae, Oxford 1854, 3 vols, iii. 120. (>5 See p. 402 above. 66 See his letter to Hugh; Symeonis Monachi Opera Omnia, i. 2 2 7 . 67 Ibid., i. 43. 68 Ibid., i. 42-3. 69 Ibid., i. 29, 30, 228. 70 Ibid.,i. 29. 71 The letter to Hugh; ibid., i. 227-8. See also the Libellus; ibid., i. 37. Cf. pp. 406-7 above. 62
Bede's Reputation as an Historian in Medieval England
13
although he had been reared from childhood in one monastery, where he spent his whole life until the day he died. 72
But Symeon, unlike William, was not surprised that Bede surmounted the geographical isolation of his home; it seemed perfectly explicable in view of the excellent library at Wearmouth/Jarrow and the high standard of learning during the Northumbrian renaissance.73 Moreover, he cited a passage from Bede's commentary on the Song of Songs, where Bede explained that he had had to rely on written authorities for his information about foreign lands, because he 'was born and nurtured far away in an island set in the midst of the sea'.74 Symeon's main sources for the early history of the see of St Cuthbert were the Historia Ecclesiastica and Bede's lives of St Cuthbert. 75 With their help he wrote a detailed account of the see's beginnings at Lindisfarne. His primary object was to augment in a general way the prestige of the see in its present site at Durham. However, in one instance Symeon defended the monastery on a particular issue, as if in response to a specific criticism, and appealed to Bede's authority to support his argument. He claimed that the appointment of the prior Turgot as archdeacon, with responsibility for spiritual matters throughout the bishopric, and the stipulation that subsequent priors should hold the same office were justified by precedent; the Life of St Cuthbert showed that Boisil, when he was 'praepositus' of the monastery (of Lindisfarne), had often gone on expeditions preaching to the people, following the example of St Cuthbert himself. Since 'praepositus' was merely the early name for 'prior', the precedent established by Boisil made it incumbent on those who succeeded him in the priorate to perform similar functions. 76 The use of Bede's authority to support contending parties was not uncommon during the reigns of William Rufus and Henry i, when the definition of jurisdictions caused frequent disputes between rivals, each claiming privileges at the expense of the other. Towards the mid-twelfth century a history of the archbishops of York was composed, covering the period from the early seventh century to the death of Thurstan (ii4o). 7 7 The author was a partisan of the Church of York, and his object was to prove the independence of the metropolitan see of York from subjection 72
Ibid., i. 41. See the letter to Hugh; ibid., i. 227-8. 74 Ibid., i. 228. Expositio in Cantica Canticorum, Lib. i (the Complete Works of the Venerable Bede, ed. J. A. Giles, London 1843-4, 12 vols, ix. 200). A copy (now lost) of Bede's commentary on the Song of Songs was among the books given by William of St Carilef to the cathedral; see C. H. Turner, 'The earliest list of Durham MSS.',/.7S., xixdgiS), 130, no. 40. 75 Passages are cited verbatim from the prose Life. For a direct reference to the verse Life see Symeonis Monachi Opera Omnia, i. 38 and n. a. 76 Ibid., i. 129. Cf. Historia Ecclesiastica, iv. 2 7 ; ed. Plummer, i. 269. 77 Printed in Historians of. . . York, ed. Raine, ii. 312-87. 73
14
Legends, Traditions and History in Medieval England
to the archbishop of Canterbury. 78 To achieve his end the York writer relied for the early period on Bede. He stated at the outset: Whoever reads the History of the English [i.e. the Historia Ecdesiastica] . . . will understand clearly that the exaction of profession [of obedience to the archbishop of Canterbury], is not only contrary to reason, but also undoubtedly contrary to the original institution of the church. 79 In order to defend the claims of York the chronicler sought to glorify the achievements of the province in general. He praised its famous men, including a section on Bede's life and works based mainly on extracts from Symeon.80 He called Bede the 'illustrious' and 'incomparable' doctor, whose 'immortal genius gave him and the whole of Britain everlasting fame', and whose 'writings were the admiration of Europe'.81 But the York chronicler's tendentious purpose necessitated more specific arguments. He cited Gregory's instructions to St Augustine from the Historia Ecdesiastica in order to prove that the pope originally instituted two equal metropolitan sees, London and York. Then, still relying on Bede, he demonstrated how the see of Canterbury gained importance instead of that of London and explained why there was no archbishop of York in St Augustine's time.82 Disputants might cite Bede in the course of proceedings, either verbally or in documents. Lanfranc cited the authority of the Historia Ecdesiastica in the council he held at Winchester in April 1072 to try to settle the question of the primacy. He used the Historia Ecdesiastica to prove that from the time of St Augustine until Bede's death, a period of nearly a hundred and forty years, his predecessors had ruled York and 'the whole of Britain and Ireland', ordaining bishops and holding councils, to which they had summoned the archbishops of York; St Augustine had not consecrated an archbishop of York because there was none in his day.83 Lanfranc's case was based on early history rather than on law. The evidence of Bede, although inconclusive, did provide, in the words of one historian, 'a satisfactory foundation for Lanfranc's claim to the historical precedence of Canterbury'.84 When the controversy was revived by Ralph d'6scures, archbishop of Canterbury (1114-22), the Canterbury party again appealed to early history, as recorded in the Historia Ecdesiastica. Archbisop Ralph, writing in 1119 to Calixtus n, in order to complain that Archbishop Thurstan 78
See ibid., ii, pp. xxi-xxii. Ibid., ii. 313. 80 Ibid., ii. 332-5. 81 Ibid., ii. 327-8, 331, 334. 82 Ibid., ii. 313-16. 83 See Lanfranc's letter to Alexander n; The Letters of Lanfranc Archbishop of Canterbury, ed., with an English translation, H. Clover and M. Gibson (Oxford Medieval Texts, 1979), 50-1. Cf. R. W. Southern, 'The Canterbury forgeries', E.H.R., Ixxiii (1958), 195. 84 A. J. Macdonald, 'Eadmer and the Canterbury privileges', J.T.S., xxxii (1931), 41. 79
Bede's Reputation as an Historian in Medieval England
15
had refused the oath of obedience, set forth Bede's evidence in detail.85 He admitted that Pope Gregory established two equal metropolitans, but pointed out that the sees in question were London and York, not Canterbury and York. He proceeded to cite the Historia Ecclesiastica to supply historical precedents for the superiority which Canterbury now claimed: he asserted that the bishops of Northumbria were suffragans of Canterbury (Historia Ecclesiastica, i. 29). He also demonstrated that St Benedict exercised authority over the British bishops (Historia Ecclesiastica, i. 27; ii. 2), that Laurence succeeded to St Augustine's power (Historia Ecclesiastica, ii. 4), and that Archbishop Theodore had authority over all England and was active in the province of York. Archbishop Ralph specifically stated that Canterbury, not York, had primacy over Scotland. He based this claim on the precedent of St Augustine's authority over the British bishops, arguing that the term 'British' included 'Scottish'.86 Eadmer became involved in this dispute when in 1120 Alexander i of Scotland appointed him to the bishopric of St Andrews. Accepting Canterbury's right without question, Eadmer demanded to do obedience to Canterbury not York. Alexander refused, and within six months opposition had forced him to resign.87 In the course of the dispute Eadmer wrote to Nicholas, prior of Worcester, for information to strengthen his case. In his reply Nicholas cited Bede to prove that York had no right to primacy over Scotland. He pointed out that it was Aidan who preached in Northumbria after Paulinus's expulsion, and that his four successors were all consecrated, and one even deposed, by the Scots.88 As is well known, the Anglo-Norman ecclesiastics and monks used forgery, besides legitimate weapons, in their disputes. Here again the importance of Bede as an historical source is manifest; he provided the forgers with essential information. Unlike the controversialists considered above, who cited Bede's authority openly, the forgers used the Historia Ecclesiastica covertly. They did not appeal to it to provide specific precedents relevant to the points at issue. (Indeed, had they been able to do so, they might not have considered forgery necessary.) Rather they borrowed phrases from the Historia Ecclesiastica, apparently with the intention of making the style of their spurious documents appropriate to the time when they were supposed to have been written. They also used Bede to provide a convincing historical background. The monks of Christ Church and of St Augustine's, Canterbury, 85 The letter is printed in Historians of. . . York, ed. Raine, ii. 228-51 (see esp. 228-39). See Southern, op. cit., 208-9. 86 Raine, op. cit., ii. 235-6. 87 Eadmeri Historia Novorum in Anglia, ed. M. Rule (R.S., 1884), 279-88. See R. W. Southern, Saint Anselm and his Biographer, Cambridge 1963, 236. 88 Nicholas's letter is printed in Councils and Ecclesiastical Documents relating to Great Britain and Ireland, ed. A. W. Haddan and W. Stubbs, Oxford 1869-78, reprinted 1964, 3 vols, ii. 202—4.
16
Legends, Traditions and History in Medieval England
resorted to forgery at much the same time, early in the 10708. The Christ Church monks wanted to prove the antiquity of their community in order to show that it existed already in the seventh century.89 For this purpose they forged, in about 1072, a bull of Boniface iv, which Lanfranc sent on their behalf as evidence to Alexander n.90 It purported to have been granted at the request of King Ethelbert and stipulated that Archbishop Laurence might establish monks at Christ Church, who were to live according to the Rule of St Bendict. The Historia Ecclesiastica supplied the form of the date of the bull and the historical background. The object of the monks of St Augustine's when they turned to forgery was to provide evidence to protect their privileges, notably of exemption from archiepiscopal control, against Lanfranc's encroachments. They produced a series of forged papal bulls and royal charters, granting St Augustine's various rights. Eleven of these documents still survive. Most of them purport to be of the seventh century; the Historia Ecclesiastica again supplied the historical background. It also provided occasional phrases.91 It can be seen, therefore, that the Anglo-Normans used Bede's reputation and his historical works in various ways. They used them as proof of the English Church's tradition: by eulogising Bede himself they added glory to the Northumbrian renaissance; by plundering his works they were able to describe in detail the achiqvements of the early Anglo-Saxon Church. Moreover, the historian of any monastery in existence in the seventh century used Bede's authority to increase its house's prestige by demonstrating its antiquity. In this way Bede made a vital contribution to salvaging the tradition of the English Church from the cataclysm of the Norman Conquest. In addition, the Anglo-Norman controversialists appealed to Bede for precedents in disputes over ecclesiastical jurisdiction and monastic privilege. And if they had to resort to forgery, Bede supplied them with background information and a stylistic model. 89 According to Eadmer (Historia Novorum, 18-19), Walkelin, bishop of Winchester, planned to replace the monks by secular canons not only in his own cathedral but also in Christ Church, Canterbury. Professor Knowles, however, doubted the reliability of Eadmer's information on this point; see M. D. Knowles, The Monastic Order in England, 2nd edn, Cambridge 1963, 130 and n. 2. 90 For the spurious bull of Boniface iv, see W. Levison, England and the Continent in the Eighth Century, Oxford 1946, 202-4, and C. N. L. Brooke, 'The Canterbury forgeries and their author', Downside Review, Ixviii (1950), 465-7. 91 See Levison, op. cit., 181 et seq. Professor Brooke argued that all the Canterbury forgeries were the work of one man, Guerno; Brooke, op. cit., 462-76, and the continuation of the same article, Downside Review, Ixix (1951) 210-31, passim. This is unlikely, since the forgeries produced at Christ Church in relation to the primacy dispute, which do not concern us here, were probably produced 1121-2. However, it is quite possible that Guerno was responsible for the earliest Christ Church forgery, the bull of Boniface iv, as well as for the St Augustine's forgeries; see Southern, op. cit., 193-4 and n. i.
Bede's Reputation as an Historian in Medieval England
17
III. c. iij6-c. 1200
Once the monasteries had been restored to comparative security and the readjustments entailed by the Norman Conquest made, historians no longer had the incentives provided by insecurity and controversy for the study of Bede. The influence of Bede declined. However, two factors assured that it retained some importance during the rest of the twelfth century. These factors will be considered separately. i. Bede as an inspiration for the Cistercian settlement in the North in the last half of the twelfth century
When the Cistercians settled in the North they found their ideals in harmony with the tradition of ancient Northumbria. Therefore, they adopted it in order to enrich their spiritual life, becoming, as it were, the spiritual heirs to the Northumbrian saints. Moreover, it is likely that it was not only in spiritual matters that the Northumbrian tradition was congenial to them. Northumbria had straddled what was in their day the border jSetween England and Scotland, including part of southern Scotland. And so it provided some precedent for the spread of the Cistercian order into Scotland. The centre of Bedan influence was Rievaulx. When Maurice, abbot from 1145 to 1147, moved there from St Cuthbert's, Durham, in 1132, he was already known as 'a second Bede'. His works, which have not survived, included one on the monastic life and one on the translation of St Cuthbert in 1104 (which Maurice himself may have witnessed).92 The most famous north-country Cistercian, Ailred of Rievaulx (abbot of Rievaulx 1147-67), came of a family already well-established in the ecclesiastical life of Anglo-Saxon England, belonging to an area which had been part of ancient Northumbria.93 Ailred's father, Eilaf, was hereditary priest of Hexham (he was ousted in 1113 to make room for a community of secular canons, when Thomas, archbishop of York, restored Wilfred's foundation); Eilaf s patron saint was St Cuthbert, and he became a monk of Durham in 1138 shortly before his death. And Ailred's great-grandfather was Alured, son of Westou, sacrist of Durham, the man responsible for the theft of Bede's relics from Jarrow. It is not, therefore, surprising that Ailred was imbued with the Northumbrian tradition. His loyalty to the Anglo-Saxon past was reinforced by his education in the court of King David of Scotland. Ailred retained an affection and admiration for the Scottish royal family, which 92
For Maurice and his works see F. M. Powicke, 'Maurice of Rievaulx', E.H.R., xxxvi (1921), 17-29. 93 For Ailred and his family see The Life of Ailred of Rievaulx by Walter Daniel, ed. F. M. Powicke Nelson's Medieval Texts 1950, pp. xxxiii-li passim. For his family tree see J. Raine, Hexham Priory (Surtees Society, xliv, xlvi, 1863-4, 2 vols), i. li-lii.
18
Legends, Traditions and History in Medieval England
descended from the Anglo-Saxon one.94 Like his father, he particularly venerated St Cuthbert, 95 whose austerities' his own resembled, and he inspired and helped Reginald of Durham to write a book about him. 96 He probably attended the translation of the Hexham saints in 1155; ne certainly wrote a tract for the occasion, the De sanctis ecclesiae Hagustaldensis. Ailred's principal source for this tract was Bede's Historia Ecdesiastica, and in his account of Acca he refers to Bede's commentaries on the gospels of Luke and Mark (because Bede had composed them at Acca's suggestion).97 Bede's Lives of St Cuthbert and the Historia Ecdesiastica were, therefore, of value to the north-country Cistercians as authorities on St Cuthbert and the other saints of the Northumbrianperiod. It could be argued, therefore, that Bede contributed as an historian to the success of the Cistercians in the North. However, undoubtedly, he was mainly of importance as a saint and Father of the Church; his influence was primarily spiritual, not historiographical. The Cistercians' use of the Historia Ecdesiastica was limited. As a result of the regulations of their order, which virtually prohibited all literary activity unless of a religious nature,98 the twelfth-century Cistercians in England produced no general chronicles. Thus they had no need of the Historia Ecdesiastica as a quarry of information for early Anglo-Saxon history as a whole. Nor, since their houses were of recent foundation, could they use it to increase the prestige of any of their monasteries in particular. 2. The use of Bede by critics of Geoffrey of Monmouth
The first writer who challenged, by implication, Bede's ascendancy as the primary authority for the early history of Britain was Geoffrey of Monmouth, whose Historia Regum Britanniae was published in 1136. However, it should be borne in mind that Geoffrey's work conflicted with the Historia Ecdesiastica only to a limited extent. Bede had written little, because he knew little, about the Anglo-Saxons' predecessors, the ancient Britons. Geoffrey began the Historia by observing that, except for Gildas and Bede, the chroniclers had totally ignored the British period, not even mentioning King Arthur and his many successors.99 And Henry of Huntingdon explained, in his letter answering a query from 'Warin the Briton', that he had started his history with Julius Caesar because he had no information for the previous centuries.100 It was this gap which 94
Powicke, op. cit., pp. xxxix-xlii. Ibid., pp. xxxvii-xxxviii. 96 See Reginaldi Monachi Dunelmensis Libellus de Admirandis Beati Cuthberti Virtutibus. ed. J. Raine (Surtees Society, i, 1835), viii, 4, 7, 32. 97 Raine, Hexham Priory, i. 33. 98 Consuetudines, c. Iviii; P. Guignard, Les Monuments primitifs de la Regie Cistercienne, Dijon 1878, 266. Cf. Knowles, The Monastic Order in England, 643-4 an d n. 6. 99 The Historia Regum Britanniae of Geoffrey of Monmouth, ed. A. Griscom, London 1929, 219. 100 Historia Anglorum, xx—xxi. 95
Bede's Reputation as an Historian in Medieval England
19
Geoffrey undertook to fill. But he wrote at fanciful length, not only about the Britons in the period before Caesar but also about them in the time of the Anglo-Saxon invasions and later, until King Arthur's final defeat. The narrative in the Historia Ecdesiastica conflicted with that in the Historia Regum Britanniae only to the extent that Bede was almost silent about the people whose exploits Geoffrey fully described. In addition, on some points Bede's statements were irreconcilable with Geoffrey's. The Historia Regum Britanniae soon became a 'best-seller', and its fictional narrative gained general acceptance as sober history. Significantly, it was in the North, where Bede's reputation was most flourishing, that the few voices of dissent were raised. The first may have been that of Ailred himself, whose Speculum Charitatis, written between 1142 and 1143, contains what appears to be an oblique reference to Geoffrey's work; Ailred lamented that novices 'weep more readily over fictitious tales about someone (I know not whom) called Arthur', than over pious books, and dismissed the stories as 'fables and lies'. 101 The historicity of the Historia Regum Britanniae was explicitly, if rather hesitantly, called in question by Alfred (or Alured) of Beverley, treasurer and sacrist of Beverley Minster, in his Annales sive Historia de Gestis Regum Britanniae.102 He probably composed the Annales in 1143: he explained that he wrote during a period of enforced idleness, when many people were under sentence of excommunication; this was probably a result of the legatine council held by Henry of Blois, bishop of Winchester, in H43.103 He turned to the study of history because everyone was talking of the new 'History of the Britons', and he was ashamed never to have read it. 104 Alfred copied extracts from standard histories and chronicles and arranged them in chronological order. He stated the source of each extract, and mentioned which other authorities agreed with it and in some cases cited an alternative account. One of his principal sources for the early part was the Historia Regum Britanniae; he put copious extracts side by side with those from Bede's Historia Ecdesiastica, and from other reputable works. He made no systematic attempt to evaluate the relative reliability of his sources. However, comparison of one with another raised questions in his mind. He wondered, for example, whether the Ambrosius Aurelianus in Bede (Historia Ecdesiastica, i. 16) was the same man as the Ambrosius Aurelianus mentioned by Geoffrey.105 More important, collation of sources combined with his common sense made him have doubts about some passages in the Historia Regum Britanniae. He noticed the anomaly that Trogus Pompeius, Suetonius and Orosius 101
Speculum Chantatis, lib. ii. c. 17: P.L. cxcv, col. 565. Aluredi Beverlacensis Annales sive Historia de Gestis Regum Britanniae, ed. T. Hearne, Oxford 1716. 103 Ibid., 1-2. Cf. J. S. P. Tatlock, The Legendary History of Britain, Berkeley, Gal. 1950, 210-11. 104 Aluredi Beverlacensis Annales, 2. 105 Ibid., 56. See Historia Regum Britanniae, 330, 360, 362, 365—6, etc. 102
20
Legends, Traditions and History in Medieval England
did not mention the British kings whose deeds Geoffrey extolled, 'neither do Gildas Sapiens nor Bede—all are equally silent'.106 And he admitted that he was uneasy about the historicity of King Arthur himself; 'neither the Roman nor the English historians record anything about the illustrious King Arthur, although he did such remarkable deeds with such skill and valour, not only in Britain against pagans, but also in Gaul against the Romans'. 107 Alfred was faced, therefore, with a dilemma: on the one hand was his respect for authority (he stated that he would not presume to question Geoffrey's veracity);108 and on the other was his doubt about the reliability of the Historia Regum Britanniae. His solution was to compromise. He decided, as he explained in the prologue, to borrow from the Historia Regum Britanniae 'only those passages which are not beyond belief'.109 The fact that he nevertheless made fairly free use of it was the result not only of his respect for authority, but also, as he mentioned in the prologue, of his desire to please his readers;110 like many other people, he recognised Geoffrey as a good read. A generation later the Historia Regum Britanniae found a more intelligent and uncompromising critic than Alfred of Beverley. At the end of the twelfth century William of Newburgh, a canon of the Augustinian priory of Newburgh in the North Riding of Yorkshire, wrote his Historia Rerum Anglicarum111 at the request of Ernald, abbot of Rievaulx.112 (Ernald was inhibited by the regulations of his order from asking one of his own monks to undertake the task.) Since William's chronicle began at the Norman Conquest, the narrative itself had no need of the Historia Regum Britanniae. However, the prologue started with a survey of the sources for early British and Anglo-Saxon history and a resume of the English settlements before it briefly described the purpose and scope of the actual chronicle (which continued to 1198). It is in the survey of sources that William demolished the credibility of the Historia Regum Britanniae. Indeed, this section and the short account of the English settlements seem to be distinct from the last part of the prologue, which constituted the true preface to the chronicle. The possibility must be considered that the prologue up to the 'true preface' was once a tract on its own.113 The question arises why anyone, presumably in this case William, should have written such a tract, which was subsequently incorporated 106
Aluredi Beverlacensis Annales, 24. Ibid., 76. 108 Ibid., 76. 109 Ibid., 2. 110 Ibid.,2-3. 111 Printed in Chronicles of the Reigns of Stephen, Henry II, and Richard /, ed. R. Hewlett (R.S., 1884-9, 4 vols), i. 3-408; ii. 411-500. 112 Ibid.,i. 3-4. 118 The suggested tract, as it survives in the prologue, might well have ended at 'ab omnibus respuatur' (ibid., i. 18, line 18), at which point William changes theme and starts leading up to his own chronicle. 107
Bede's Reputation as an Historian in Medieval England
21
in the prologue to the chronicle, to discredit Geoffrey of Monmouth. The reasons would probably have been political. The Celtic peoples could use the Historia Regum Britanniae as a weapon against the Angevin dynasty. It provided them with valiant predecessors, the greatest of whom was King Arthur. Arthur served as a focus of loyalty, becoming, as it were, their spiritual leader. Nor, indeed, was it certain that Arthur was dead; in his description of the wounded king's last refuge on the Isle of Avalon (Bk. xi. 2), Geoffrey left the possibility open that he would return once more and lead his people to victory. King Arthur became, therefore, the British 'hope'; and added zest to the rebellions of the Welsh and Bretons against the Angevin kings.114 William himself recorded the Welsh rising of 1157, commenting that the Welsh were descendants of the ancient Britons.115 But he dwelt at greater length on Breton disaffection. He recorded the posthumous birth of a son in 1186 to Henry n's son Geoffrey, by a daughter of the duke of Brittany. And he recorded that the Bretons had the child solemnly acclaimed and baptised Arthur—although King Henry had expressly ordered that he should be named Henry after him. 'Thus', wrote William, 'the Bretons, who are said long to have awaited the return of the fabulous Arthur, are now nurturing a real one, according to certain prophets learned in the great and famous Arthurian legend.'116 And almost at the end of the chronicle William noticed the Breton rebellion of 1196. The Bretons, 'who were rearing the boy Arthur under the mighty omen of his name', roused the French king to war against Richard i. The occasion, William explained, was Richard's request that the Bretons give him guardianship of his nephew, the young Arthur; by virtue of this concession Brittany would be bound to Richard against all enemies. But the Bretons, 'moved by suspicion rather than by prudence', refused and rallied to defend the child. 117 It is noteworthy that William's passages on dissidence among the Welsh and Bretons do not show an undue preoccupation with the subject. This fact seems to corroborate the view that the attack on the Historia Regum Britanniae in the prologue was once a tract on its own. It is not improbable that William would have been commissioned to write such a tract. A remark in the prologue that Gildas's De Excidio Britanniae had come into his hands 'several years previously' proves that he had been studying history for a while,118 and his reputation as an historian was sufficiently established for Ernald to ask him to write the chronicle. 114 See R. F. Treharne, The Glastonbury Legends, London 1967, 105-6, andj. S. P. Tatlock 'Geoffrey and King Arthur in 'Normannicus Draco', Modern Philology, xxxi (1933-4), 122 and n. 7, 123. 115 Chrons. Stephen, Henry II and Richard I, ed. Hewlett, i. 106-7. 116 Ibid., 1.235. 117 Ibid., 11.463-4. 118 Ibid.,1. 11.
22
Legends, Traditions and History in Medieval England
However, it seems unlikely that Ernald would have commissioned a political tract; it is more probable that someone connected with the royal court, even Henry n or Richard i himself, would have done so. Royal interest in anti-Arthurian propaganda was by no means unprecedented. Sometime between 1167 and 1169 the Draco Normannicus was composed in support of Henry n's suppression of the Breton rebellion.119 Its author, Etienne of Rouen, may well have written specifically to please the king.120 It relates the history of Normandy, drawing heavily on the Historia Regum Britanniae. But in it King Arthur is represented as a fay, and thus the Breton's hope of his return was reduced to an absurdity. 121 Again, when in 1191 the monks of Glastonbury exhumed the (supposed) bodies of King Arthur and Queen Guinevere, they may have acted at an earlier suggestion of Henry II. The purpose of this archaeological forgery was probably not only to increase the abbey's tourist trade; it may also have been intended to demonstrate, once and for all, for the instruction of the rebellious Welsh, that King Arthur was actually dead.122 It is likely, therefore that William's attack on Geoffrey in the prologue was a piece of political propaganda. He launched an eloquent tirade against the prophecies of Merlin and the Arthurian legends. Why, if the former was a greater prophet than Isaiah, and if the latter's victories were more glorious than Alexander's, had the ancient historians heard of neither? 123 Indeed, everything that 'that man' and others say concerning King Arthur is fictional; Geoffrey concocted his stories 'either because of his love of unbridled lying, or in order to please the Britons'. 124 In William's opinion, Gildas, despite his unpolished and illiterate style (the reason, William believed, why copies of Gildas's work were rare), was an author of integrity; 'for the sake of truth he did not spare his own people, but deplored their sins and shortcomings, showing them to be neither brave in war nor trustworthy in peace'.125 William twice repeated that the Britons of his own day were so stupid as to believe that Arthur would return; they could not bear to think of him as dead, preferring the tale that he had retreated wounded to Avalon. 126 Like the ecclesiastical and monastic controversialists, William took his stand particularly on the authority of Bede. He compared Bede's account of early Anglo-Saxon history with Geoffrey's and noticed a discrepancy; according to Geoffrey's chronology, King Arthur would have been a contemporary of King Ethelbert and would have been reigning when St 119
SeeTatlock, 'Geoffrey and King Arthur', 1-2, 124. Ibid., 124-5. 121 Ibid., 3, 113-23. 122 Cf. Treharne, op. cit. 105-6 and below p. 170. I2S Chrons. Stephen, Henry II and Richard I, ed. Hewlett, i. 17. 124 Ibid., i. 14. 125 Ibid., i. 11. 126 Ibid., i. 14, 18. 120
Bede's Reputation as an Historian in Medieval England
23
127
Augustine came to England. William stated that for this early period he would follow Bede's narrative, 'because the venerable Bede based it on historical truth and, therefore, it can be accepted as established beyond doubt'. 128 'Everyone', William declared, 'should reject without hesitation both the fable teller [i.e. Geoffrey] and his fables', and instead 'put complete trust in Bede, whose wisdom and sincerity can never be doubted.' 129 Bede should be accepted as the authority even for the ancient Britons: 'he narrated the famous deeds of our people accurately and concisely; he began right at the beginning, including the Britons, who are recognised as the first inhabitants of our island, in order to provide a fitting introduction to what especially interested him.' 130 Thus William drew the conclusion that all that can be known about the ancient Britons was what Bede and Gildas (whose work Bede used) recorded. Hence, since neither even mention the heroes, whose deeds Geoffrey so colourfully described, they were probably fictional—a supposition made the more likely by the fact that on Gildas's account the ancient Britons were cowardly and treacherous. The corollary of this conclusion was that the contemporary Britons had no glorious ancestors, and, since King Arthur had never lived, their belief that he would return was baseless. IV. c. I2OO—C. 1500 Bede's reputation as an historian from the thirteenth to the fifteenth century was less than it had been in the late eleventh and twelfth centuries, but it was not negligible. He and his historical works retained widespread respect among the writers of general histories. Both Matthew Paris and Ranulf Higden gave eulogies on Bede and detailed accounts of his life. But these lack the immediacy of their Anglo-Norman prototypes, partly because they were mainly derived from the latter. 131 The Historia Ecclesiastica remained an important source for early Anglo-Saxon history. It was most often used purely for information, but on occasion it served as an authority in disputes. Thus, Gerald of Wales, in his attempt to obtain the bishopric at St David's, appealed to it to prove that the Church of Wales was independent of Canterbury. (He was elected bishop in 1198, 127 128 129
Ibid., i. 15. Ibid., i. 14.
Ibid., i. 18. Ibid., i. 11. 131 Matthew Paris borrowed his account of Bede mainly from William of Malmesbury, but in addition he used Henry of Huntingdon; Matthaei Parisiemis, Monachi Sancti Albani, Chronica Majora, ed. H. R. Luard (R.S., 1872-83, 7 vols), i. 333-6. Higden also took his account mainly from William of Malmesbury; Polychronicon Ranulphi Higden Monachi Cestrensis, ed. C. BabingtonandJ. R. Lumby (R.S., 1865-8, 9 vols), vi. 218-26. 130
24
Legends, Traditions and History in Medieval England
but Archbishop Hubert Walter quashed the election; the consequent dispute lasted until 12O3.) 132 However, in this period the Historia Ecclesiastica was often cited alongside other, less respectable works, notably the Historia Regum Britanniae,133 Indeed, an author, having quoted Bede by name, might proceed to an unacknowledged borrowing from Geoffrey and so give the impression that Bede is its authority too. An early example of this method is supplied by Ralph Diceto, writing in the late twelfth century. He recorded (s.a. A.D. 178), citing the Historia Ecclesiastica, Bk. i, chap. 4, that Pope Eleutherius wrote to the British King Lucius to convert him to Christianity. He continued, here using Geoffrey but without acknowledgement, that Eleutherius then sent Fagan and Duvian who baptised Lucius. 134 The fifteenth-century authors Thomas Rudborne, monk of St Swithun's, Winchester, and John Rous, the antiquary, provided similar examples.135 Exceptions to this uncritical use of sources are rare. The Historia Regum Britanniae was accepted almost without question. Scholars neglected William of Newburgh's devastating criticism until the early sixteenth century, when the humanist Polydore Vergil adopted it as the basis of his own strictures on the Historia Regum Britanniae.136 Apparently, only three writers, Ranulf Higden, 137 Thomas Rudborne 138 and John Whethamsted (abbot of St Albans 1420-40, 1452-65),139 cast doubt on its historicity, and then only on some of its statements. Nor does it seem that the first two were independent critics; Higden, who was sceptical about the Arthurian legends mainly because of the silence of Gildas and Bede, may have been influenced by Alfred of Beverley,140 one of his sources, while 132
For appeals to the Historia Ecclesiastica in Gerald's letter (i 199) to Innocent HI, in his De Invectionibus (1205), and in his De Jure et Statu Menevensis Ecclesiae (c. 1218), see Giraldi Cambrensis Opera, ed. J. S. Brewer et alii (R.S., 1861-91, 8 vols), iii. 169-76 passim, 44-51, 111, respectively. 133 The Historia Ecclesiastica and the Historia Regum Britanniae are cited side by side by, for example, Matthew Paris, Higden and Gerald of Wales. 134 See Radulfi de Diceto Decani Lundoniensis Opera Historica, ed. W. Stubbs (R.S., 1876, 2 vols), i. 66. 135 Thomas Rudborne, Historia Major . . . Ecclesiae Wintoniensis, in H. Wharton, Anglia Sacra, London 1691, 2 vols, i. 180; John Rous, Historia Regum Angliae, ed. Thomas Hearne, Oxford 1716, 48. 136 See Historia Anglica, Basle 1546, 15-19 (for a reference to William of Newburgh's criticisms see ibid., 17, line 3). For the passage in the sixteenth-century English translation see Polydore Vergil's English History, ed. Henry Ellis (Camden Society, original ser., xxxvi, 1846, xxix, 1844, 2 vols), i. 26-33. 137 Polychronicon, v. 332-8. 138 Wharton, Anglia Sacra, i. 188. 139 Whethamsted in his Granarium, PL i, an encyclopaedia of historians and historical personages, disputes the Brutus legend; B.L. MS Nero C vi, fos. 33~33v. (His scepticism is noticed in L. Keeler, 'The Historia Regum Britanniae and four medieval chroniclers', Speculum, xxi (1946), 36.) However, he appears to accept the legend of Belinus and that of Brennus; ibid., fos. 28, 3OV-32. 140 See Polychronicon, i. 24.
Bede's Reputation as an Historian in Medieval England
25
Rudborne in his turn may have owed his doubts on the same subject to Higden.141 The comparative decline of Bede's influence as an historian is perhaps reflected by the printed editions of the Historia Ecclesiastica. The first edition was issued abroad, in a volume also including Eusebius's Historia Ecclesiastica, by Heinrich Eggestein of Strasburg, sometime between 1475 and 1482. 142 No other edition appeared before the end of the fifteenth century. (The first edition to be issued in England was that of 1643.)143 The absence of an early edition printed in England probably has little significance, since it was normal for Latin texts to be printed abroad. Possibly more relevant to our argument is the fact that no English translation was available which might have formed the basis of a printed edition. William Caxton issued, in 1480 and 1482 respectively, English versions of the Historia Regum Britanniae and of the Polychronicon, but the first translation of the Historia Ecclesiasticn to be printed was that by Thomas Stapleton, the Catholic polemicist; he published it in 1565 to provide proof that the Church of England was of papal origin.144 Bede's historiographical influence no doubt declined after the end of the twelfth century partly on account of contemporary circumstances. None of the factors which had encouraged writers of general history to study Bede's historical works in the early period existed in the later one: there was no monastic revival or new monastic order; there was no cataclysmic event comparable with the Norman Conquest to threaten the tradition of the Church of England; and there were no important controversies necessitating appeal to his authority. Nevertheless, in the late fourteenth and fifteenth centuries opinion in some quarters turned against monasticism; it was attacked by the Lollards, by other 'dispossessioners' and by the secular canons. The monks wrote in selfdefence. It was a matter of consolidating their position and guarding against general aspersions rather than of defending themselves in specific disputes. They composed tracts justifying monasticism on historical grounds, by demonstrating its long tradition. 145 Since the monastic movement originated before Bede's time and abroad, the Historia Ecclesiastica was not very useful to such authors. However, the case with a 141 Rudborne questioned the Arthurian legends because historic persons mentioned in them., notably the Emperors Lucius and Leo, did not rule at the time of King Arthur. He demonstrated these chronological flaws in the legends with some care, but Higden had already noticed them briefly. (For references see nn. 138, 137 above, respectively.) 142 See Catalogue of Books Printed in the Fifteenth Century now in the British Museum (Trustees of the British Museum, lithographic reprint, 1912-67, 9 pts), i. 71. 143 For a survey of the printed editions, see Bede's Ecclesiastical History of the English People, ed. Colgrave and Mynors, pp. Ixx-lxxii. 144 See M. McKisack, Medieval History in the Tudor Age, Oxford 1971, 39, and F. J. Levy, Tudor Historical Thought, Huntingdon Library*, California 1967, 110-12. 145 W. A. Pantin, 'Some medieval English treatises on the origins of monasticism', Medieval Studies presented to Rose Graham, ed. V. Ruffer and A. J. Taylor, Oxford 1950, 189-215.
26
Legends, Traditions and History in Medieval England
local historian might be different; if his monastery was founded in the Northumbrian period, Bede's evidence could? as always, be invaluable in augmenting its prestige. For this reason the Historia Ecclesiastica was studied as carefully as ever before at St Augustine's, Canterbury, and at St Cuthbert's, Durham. Thomas Elmham, monk of St Augustine's, Canterbury, wrote the Speculum Augustinianum shortly before 1414.146 He conceived this history of St Augustine's on such a massive scale that he completed it only as far as the year 803. He intended the work to be a definitive record of the abbey's past glory and of its privileges, and to silence any detractors. He was particularly at pains to demonstrate its independence of archiepiscopal authority. (In fact the pope had finally confirmed its exemption in 1397, thus ending the struggle with the archbishops which had driven the monks to forgery in the late eleventh century and to the employment sporadically thereafter of harassing tactics.)147 Bede's Historia Ecclesiastica was Elmham's principal literary authority for the abbey's foundation and early history. He praised Bede as an historian in fulsome terms (though he borrowed most of his words from William of Malmesbury), 148 and he specifically appealed to his authority to defend St Augustine's against any allegation contrary to its interests. He objected to Higden's assertion that Archbishop Theodore appointed Benedict Biscop abbot of St Augustine's. If Biscop had been abbot, and if, as certain 'detractors' alleged, he was Theodore's chaplain, St Augustine's would have been under the archbishop's control. Elmham asserted that there is no evidence in the Historia Ecclesiastica either that Biscop was abbot of St Augustine's, or that he was Theodore's chaplain.149 Similarly, Elmham 146 Printed as Historia Monasterii S. Augustini Cantuariensis, by Thomas of Elmham, ed. C. Hardwick (R.S., 1858). For the original title, Speculum Augustinianum, see F. Taylor, 'A note on Rolls Series 8', Bulletin of the John Rylands Library, xx (1936), 379-82. The Speculum must have been written before 1414 because in that year Elmham joined the Cluniac order, becoming prior of Lenton, near Nottingham; see J. S. Roskell and F. Taylor, 'The authorship and purpose of the Gesta Henrici Quinti: i', Bull, of the John Rylands Library, liii (1970-1), 436. 147 For St Augustine's claim to exemption from archiepiscopal control see M. D. Knowles, 'Essays in monastic history, iv. The growth of exemption', Downside Review, (1932), 401-15. For the dispute in the late eleventh century see pp. 411-12 and n. 91 above. The course of the struggle is described in the earlier chronicles of St Augustine's, the lost one by Thomas Sprott, which apparently ended in 1228, and William Thome's to 1397, both of which Elmham used; see Hardwick, op. cit., 77. Thome's chronicle is printed in Roger Twysden, Histonae Anglicanae Scriptores X, London 1652, 2 vols, i. cols 1757-2202; for a discussion of the chronicles of St Augustine's see William Thorne's Chronicle of Saint Augustine's Abbey, translated by A. H. Davis, with a preface by A. Hamilton Thompson, Oxford 1934, xx-xxvi. 148 Hardwick, op. cit., 309. 149 Ibid., 185-6, 202-5. Cf. Historia Ecclesiastica, iv. i (ed. Plummer, i. 204) and Polychromcon, vi. 78. Plummer (Historia Ecclesiastica, ed. Plummer, ii. 204) points out that Becle's Historia Abbatum §§3.4 (ibid. i. 366-7) has evidence that Biscop was once abbot of St Augustine's (for two years), despite Elmham's assertion to the contrary.
Bede's Reputation as an Historian in Medieval England
27
cites Bede to prove that Theodore in his capacity as papal legate, not as archbishop, appointed Hadrian to the abbacy; Hadrian, therefore, did not become his subordinate, but remained his friend and colleague.150 Elmham also cites Bede in order to defend Theodore against those who accused him of treating Wilfrid harshly. 151 His concern for Theodore's reputation may well have been partly the result of his burial at St Augustine's (Wilfrid lay in Canterbury cathedral). Elmham's interest in the identity of Cadwaller (he compares the evidence of Bede, William of Malmesbury and Geoffrey of Monmouth) may have owed something to the (supposed) burial of Cadwaller's brother, Mul, at St Augustine's. 152 Elmham's counterpart at Durham was John Wessington, prior from 1416 to 1446.153 The situation at St Cuthbert's was not unlike that at St Augustine's: serious conflict between the cathedral priory and its diocesan was over by the fifteenth century; indeed, it had ended in the first third of the fourteenth century.154 However, the monks were anxious to increase their prestige in general and to arm themselves against any new encroachment by the bishop. They too turned to past glory for present help. Bede's Historia Ecdesiastica and Lives of St Cuthbert were especially useful because they supplied a picture of St Cuthbert's see in its early days at Lindisfarne, when the bishop never interfered with the monastery.155 The popularity at Durham of Bede's prose Life of St Cuthbert is witnessed by the fact that the monks had a new copy made just before the Dissolution.156 Wessington wrote a history of St Cuthbert's which he intended to be definitive.157 His aims were similar to Elmham's; to increase his house's prestige by dwelling on its past achievements and to confirm its rights by putting them on record. He started with the see's origin in 635, but, like Elmham, he never finished his work; it ends in 1362. He included a long and eulogistic account of Bede's life and work: he borrowed most of it from Symeon of Durham, but added a passage from the 'Legenda Sanctorum';158 this addition recounts two 'miracles' purporting to 150 Hardwick, op. cit., 202-4. 151 Ibid.,279. 152 Ibid., 268-70. Cf. Historia Ecdesiastica, ed. Plummer, ii. 228, 265. 153 For Wessington's career see R. B. Dobson, Durham Priory 1400-1450, Cambridge 1973, 89-113. 154 Dobson, op. cit., 204-5, 222-4, and R- L. Storey, Thomas Langley and the Bishopric of Durham, 1406—1437, London 1961, 199—200. 155 See Dobson, op. cit., 203. 156 See (with further references) Dobson, op. cit., 32 and n. i, 350 n. 4, 360. 157 Wessington's history, the Libellus de exordio et statu ecclesie cathedralis quondam Lindesfarnensis, post Conchestrensis, demum Dunelmensis, ac de gestis pontificum eiusdem, is unprinted. Three manuscripts, all from Durham, are known to survive; one is cited in the next footnote. See H. H. E. Craster, 'The Red Book of Durham', E.H.R., xl (1925), 504-14. passim. For the place of the Libellus in the historiography of Durham see H. S. Offler, Medieval Historians of Durham, Durham 1958, 17. 158 B.L. MS Cotton Claudius D iv, fo. 14. The reference is presumably to John of
28
Legends, Traditions and History in Medieval England
explain why Bede was called 'venerable' instead of 'saint' everywhere except at Durham. Wessington also relied mainly on Symeon for the early history of St Cuthbert's; as has been seen, Symeon himself had based his narrative mainly on the Historia Ecclesiastica and on the Lives of St Cuthbert, but Wessington amplified it with further additions from Bede.159 The same motives which caused Wessington to compose the history of St Cuthbert's, made him write numerous tracts on the see's antiquities and rights, 160 and for them he used Bede when appropriate. One, for example, comprised copies of the descriptive captions which he himself wrote beneath the 'images' (i.e. the statues or pictures) of saints at the altar of SS Jerome and Benedict; Bede was his authority for the Northumbrian ones.161 And when Wessington collected evidences of the prior's rights to place before Thomas Langley, bishop of Durham, and his council in 1426, his 'declaration' began: 'the prior should have all the rights, offices and honours of an abbot, according to the dignity of the church of Lindisfarne, as set forth in Bede's De Gestis Anglorum, Bk. IV, chapter 25, and in the prose Life of St Cuthbert, chapter 16'; Wessington Tynemouth's Sanctilogium, compiled between 1327 and 1347, for which see V. H. Galbraith, 'The Historia Aurea of John, vicar of Tynemouth, and the sources of the St. Albans Chronicle (1327-1377)', Essays in History presented to Reginald Lane Poole, ed. H. W. C. Davis, Oxford 1927, 385. The passage cited by Wessington is almost identical with that in the Nova Legenda Anglie, ed. Carl Horstman, Oxford 1901, 2 vols, i. 111, and in the Historia Aurea. The latter is a mid-fourteenth century version of the Sanctilogium in which the saints' Lives are arranged chronologically (see Galbraith, op. cit., 382—5); a late fourteenth-century copy from Durham priory is now Lambeth Palace MSS 10, 11 and 12. It is possible that Wessington in fact copied the passage in question from the Historia Aurea from which he derived his copy of Pope Sergius's letter (Claudius D iv, fo. 14), and not directly from the Sanctilogium. This is suggested by the fact that both he and the Durham copy of the Historia Aurea (Lambeth Palace MS 12, fo. 50) have the marginal note in the hand of the scribe, 'Legenda Sanctorum in fine'. However, it is unlikely that he was using the copy now in Lambeth Palace; his version of the passage on why Bede was called 'venerable' has, besides a few slight variants from that in the Lambeth MS, a different beginning. Wessington begins: 'Licet enim Beda in sanctorum cathalogo computetur et a Dunelmensibus sanctus Beda nominetur, non tamen a pluribus aliis 'sanctus' sed 'venerabilis' appellatur'; Claudius D iv, fo. 14V. The Historia Aurea (and the Nova Legenda Anglie) begins: 'Licet enim Beda in sanctorum cathalogo computetur, tamen ab ecclesia non 'sanctus', sed 'venerabilis' appellatur'; Lambeth Palace MS 12, fo. 50. Wessington's account of Bede is in toto much fuller than that in the Historia Aurea. 159 Craster, 'The Red Book of Durham', 516-1 7. 160 A contemporary of Wessington, who listed his tracts, comments that Wessington 'compiled them not without labour and study for the perpetual preservation and defence of the rights, liberties and possessions of the church of Durham against the malice and machinations of would-be molesters'; Historiae Dunelmensis Scriptores Tres, Gaufridus de Coldingham, Robertus de Graystanes, et Willielmus de Chambre, ed. J. Raine (Surtees Society, ix, 1839), cdxviii-cclxix. Cf. Dobson, Durham Priory, 379 and n. 2, and Craster, 'The Red Book', 515 and n. 2. For Wessington's tracts in general see Dobson, op. cit., 382-4. 161 The text is printed in Rites of Durham, ed. J. T. Fowler (Surtees Society, cvii, 1903), 124-36.
Bede's Reputation as an Historian in Medieval England
29
proceeded with the extracts from Bede as quoted, and then gave copies of ten charters relevant to the issue.162 The cult of Bede flourished at Durham alongside that of St Cuthbert until the end'of the Middle Ages. In 1370 his relics were moved from the grave next to St Cuthbert's and re-interred in the Galilee chapel.163 Bishop Robert Neville (1438-57) had a new shrine built for him, decorated with gold and silver,164 and wanted to be buried in front of his altar (a wish which was not fulfilled).165 And St Bede, almost equally with St Cuthbert, became an object of pilgrimage.166 This survey provides some explanation for the high percentage of twelfth-century copies of the Historia Ecclesiastica in Laistner's table of the surviving manuscripts. At the outset the twelfth century inherited a vigorous tradition of Bedan studies from the Anglo-Saxon period, a tradition which was adopted with enthusiasm by the missionaries to the North in the late eleventh century. In the twelfth century itself the exceptional religious and political circumstances account for the blossoming of Bede's reputation. This survey does not, however, explain the relatively high percentage of fourteenth- and fifteenth-century manuscripts. Perhaps one of the explanations which Laistner offers for the rather similar survival pattern of manuscripts of Bede's biblical commentaries is applicable here too; simply that old copies needed replacing.167 162
Historia Dunelmensis Scriptores Tres, cclxx. The 'declaration', dated 17 June 1426, is now 1 . 7 . Pontificalia, no. 2, in the muniments of the dean and chapter of Durham. It is listed in R. B. Dobson, The Priory of Durham in the Time of John Wessington Prior 1416-1446 (unpublished Oxford D.Phil, thesis, 1962), 583, no. 22. 163 Rites of Durham, 46. 164 Ibid., 45, 233. 165 Ibid., 225. 166 In his account of his visit to Durham on his return from the court of James i of Scotland, Aeneas Sylvius Piccolomini (later Pius n) apparently mentioned only Bede's tomb, not that of St Cuthbert: 'Exinde Dunelmiam venit, ubi sepulchrum venerabilis Bedae presbyteri sancti viri hodie visitur, quod accolae regionis devota religione colunt'; Pii Secundi Max. Commentani Rerum Memorabilium, quae temporibus suis contigerunt, Frankfurt 1614, 5. This passage is mistranslated as 'Next he came to Durham, where today men go to see the tomb of the holy abbot, the Venerable Bede, which is piously revered by the inhabitants of the region', in The Commentaries of Pius II, translated by F. A. Gragg, ed. L. C. Gabel (Smith College Studies in History, xxii, nos 1-2, 1936-7), 20, The erroneous rendering of 'venerabilis Bedae presbyteri sancti viri' as 'the holy abbot, the venerable Bede', perhaps explains the assertion by a recent scholar that Aeneas Sylvius confused Bede's shrine with that of St Cuthbert's; Dobson, Durham Priory, 105. 167 Laistner, Handlist, 7.
ADDITIONAL NOTE Page 7. For a more detailed treatment of Bede's influence on Anglo-Norman monasticism see now R. H. C. Davis, 'Bede after Bede', in Studies in Medieval History presented to R. Allen Brown, ed. C. Harper-Bill, C. J. Holdsworth and J. L. Nelson (Woodbridge, 1989), 103-16.
This page intentionally left blank
2 Traditionalism and Continuity during the Last Century of Anglo-Saxon Monasticism
hose writing at the time, and subsequent historians, have tended to exaggerate the importance of the tenth-century monastic revival and of the reform movement which followed the Norman Conquest. During each period contemporary writers glorified the achievements of the reformation, of which they themselves were products, and belittled or even denigrated the religious life of the preceding era. This was partly because the hallmark of both reformations was the strict enforcement of the Rule of Benedict; the ideal of strict Benedictinism appealed to those writing during the reformations, since they themselves were strict Benedictines, and it has appealed to some historians in our own day. One result has been a tendency to emphasise the influence of continental models so much that it overshadows the importance of the Anglo-Saxon tradition. David Knowles makes continental influence on the tenth-century revival the theme of chapter i of his The Monastic Order in England. The fact that he starts his survey with a section which treats the preceding period, from the Northumbrian renaissance onwards, as
T
BJRL = Bulletin of the John Rylands Library; CCM = Corpus Consuetudinum Monasticarum ; Proceedings = Proceedings of the Battle Conference on Anglo-Norman Studies; A-SE = AngloSaxon England; Rev.Ben. = Revue Benedictine; EETS = Early English Text Society; PCAS = Proceedings of the Cambridge Antiquarian Society I am deeply indebted to Dr Simon Keynes and Dr Michael Lapidge for reading this article in typescript; they pointed out some errors and made a number of valuable comments and suggestions and provided me with useful bibliographical information. After writing my article I heard the paper by Patrick Wormald, '^Ethelwold and his continental counterparts: contact, comparison, contrast', at the Anglo-American Conference of Historians, now publ. in /Ethelwold of Winchester: studies in his career and influence, ed. Barbara Yorke, Woodbridge 1988, 13-42. Mr Wormald kindly allowed me to read the typescript of his article and read mine. He made some useful comments, in the light of which I have revised my article slightly. In general, Mr Wormald's article puts the case for continental influence on St Ethelwold, but concludes with a tribute to the contribution to the monastic revival of the Anglo-Saxon tradition (pp. 39-41), making a few of the same points which I had made.
32
Legends, Traditions and History in Medieval England
introductory matter, would itself have hampered appreciation of the revival's full significance. Knowles, indeed, makes some sweeping statements about the revival: the daughter houses of St Oswald's foundation at Westbury-on-Trym were 'pure reproduction[s] of Fleury'; the Life of St Oswald, written at Ramsey c. 1000, 'is in tone and style wholly a product of the school of Fleury'. He tends to treat the postConquest period in a similar manner: thus Eadmer was 'wholly of the school of Lanfranc and Anselm'. 1 Eric John lays even greater emphasis than Knowles on continental influence. For example, speculating on the origin of the monastic ideals of ^Elfheah, bishop of Winchester (who was himself a secular clerk), John suggests that they reached him from abroad, by way of Athelstan's court, since the king was in touch with Henry the Fowler of Germany. 2 However, historians have not uniformly under-rated the native tradition.3 Thomas Symons, in his discussion of the Regularis concordia issued by the reformers at Winchester c. 972, adopted a cautious and judicious approach to the problem of the origins, whether continental or indigenous, of the monastic and liturgical observances it contains.4 Richard Southern has done ample justice to the English contribution to the post-Conquest reformation, particularly with regard to the cult of saints.5 Moreover, historians increasingly recognise the achievements of the native tradition in the Viking Age and in the eleventh century prior to the Norman Conquest. Francis Wormald pointed out that 'from the reign of King Alfred onwards there was a steady stylistic development in the production of illuminated books in southern England'; this style was in the Anglo-Saxon tradition, although it was influenced by Carolingian models, and ' [prepared] the way.. for the much more 1
D. Knowles, The Monastic Order in England, 2nd edn, Cambridge 1963, 42, 65, 79. Eric John, 'The king and the monks in the tenth-century reformation', BJRL xlii ( I 959)j ^S n. 3. Another exponent of continental influence is D. A. Bullough, 'The continental background of the reform', in Tenth-century Studies, ed. David Parsons, London-Chichester 1975, 20-36. 3 Knowles, op. cit. 37—9 (cf. p. 83), 45 (monks' prayers for the king and queen) (but cf. Wormald, op. cit. 33), 45-6 (monks' importance in national life and the institution of monastic cathedrals). 4 Thomas Symons, ' Sources of the Regularis concordia', Downside Review xl (1941), 14-36, 143-70, 264-89 passim; see esp. pp. 32-3, 143-4, 168-70. However, much of what he says there has been superseded by the newly discovered customary of Fleury. See Lin Donnat, 'Recherches sur Tinfluence de Fleury au Xe siecle', in Etudes lige'riennes d'histoire et d* arche'ologie medievales, ed. Rene Louis, Auxerre 1975, 165-74; and Consuetudinum Saeculi X/ XI/XII Monumenta: Introductions, ed. Kassius Hallinger, CCM vii/i. 331-93. 5 R. W. Southern, 'Aspects of the European tradition of historical writing: 4. The sense of the past', Transactions of the Royal Historical Society xxiii (1973), 251-2; and idem, St Anselm and his Biographer, Cambridge 1963, 246 flf. Cf. e.g. A. Gransden, Historical Writing in England, [z], c. 550 to c. 7507, London 1974, ch. vii passim; idem, 'Cultural transition at Worcester in the Anglo-Norman period', British Archaeological Association Conference Transactions, I: Medieval Art and Architecture at Worcester Cathedral (1978), 1-14; idem, 'Baldwin, abbot of Bury St Edmunds, 1065-1097', Proceedings iv (1981), 65-76, 187-95z
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
33
spectacular changes' of the monastic revival in the second half of the century.6 J. Armitage Robinson demonstrated that the cult of saints and relics, an important feature of the Anglo-Saxon tradition (but one neglected by Knowles), flourished before the tenth-century revivalunder King Athelstan and, of course, earlier.7 Most recently Simon Keynes has discussed Athelstan's enthusiasm for relics which, he suggests, was the result of King Alfred's influence. 8 Knowles argued that, once the tenth-century revival was over, Anglo-Saxon monasticism recovered its distinctive form, and he pinpointed a few of its peculiarities.9 And Keynes and other modern scholars have brought to light much new evidence of remarkable cultural activity in many fields both before and after the tenth-century revival, while John Blair has demonstrated the survival of numerous clerical communities, ranging from the more or less monastic to 'loosely ordered groups of priests', and serving 'superior' churches - 'minsters'.10 It is not the intention of this article to minimise the importance of continental influence on the monastic reformations (although it must be remembered that at no time has England been immune to influences from abroad). However, it seems time to try to assemble the evidence for the continuity of the Anglo-Saxon monastic tradition, which had been founded in seventh-century Northumbria, and survived through the tenth-century monastic revival and into the Anglo-Norman period without interruption. Furthermore, because of this continuity and the vigour of the tradition in the tenth and eleventh centuries at times other than those of monastic reformation, we ought to re-evaluate the reformers' achievements; they should now be seen as rather less momentous than they have been in the past. Since the survival of the Anglo-Saxon tradition was a result of an essential feature of the Anglo-Saxon mentality, something should first be said about that subject in general. The Anglo-Saxons shared the backward-looking tendency so prevalent in the Middle Ages. This characteristic combined with intense pride in the achievements of the Anglo-Saxon people to produce a strong interest in their past. 'Nationalistic' traditionalism is very apparent at least as early as King Alfred's reign. It seems that Alfred and his circle wanted to recreate the 6
Francis Wormald, 'The 'Winchester School' before St ^Ethelwold', in England before the Conquest: studies in primary sources presented to Dorothy Whitelock, ed. Peter Clemoes and Kathleen Hughes, Cambridge 1971, 305-13. 7 J. A. Robinson, The Times of St Dunstan, Oxford 1923, 51-5, 71-80, 101-2. 8 Simon Keynes, 'King Athelstan's books', in Learning and Literature in Anglo-Saxon England, ed. Michael Lapidge and Helmut Gneuss, Cambridge 1985, 143-4. 9 Knowles, Monastic Order, 80-1 and n. 4 (property owning by monks), 83 and n. i. 10 See e.g. Simon Keynes's introduction to The Golden Age of Anglo-Saxon Art, ed. Janet Backhouse, D. H. Turner and Leslie Webster, London 1984, 12-13; David N. Dumville's introduction to The Historia Brittonum, j, The ' Vatican' Recension, ed. idem, Cambridge 1985, 18-23; John Blair, 'Secular minster churches in Domesday Book', in Domesday Book, a Reassessment, ed. Peter Sawyer, London 1985, 104-42 passim.
34
Legends, Traditions and History in Medieval England
Golden Age of Northumbria. 11 The late ninth-century Old English translation of Bede's Ecclesiastical History can be seen as one outcome of this desire. Possibly Alfred's rule itself was to some extent influenced by the Ecclesiastical History. Alfred, like the Northumbrian kings, saw the value of the Church as a support for kingship. A more particular instance of possible influence is provided by Asser in his Life of King Alfred (C. cii). He asserts that Alfred, having divided his annual revenues in half, further divided the second half, which was to be devoted to the service of God, into four parts. The first part was to go to the poor and 4 anyone who came to him'; the second to his two monastic foundations; the third to his school; and the fourth to monasteries throughout England and in Ireland and Brittany. 12 This quadripartite division is reminiscent of the first of Gregory the Great's Responsiones to St Augustine, which Asser could have read in the Ecclesiastical History (although he might have used an independent text). 13 Gregory states that, by custom, the pope instructs all newly consecrated bishops to divide their revenues into four parts. One was for the bishop and his household, to enable them to offer hospitality, the second for the clergy, the third for the poor and the fourth for the repair of churches. Alfred held Gregory's Pastoral Care in high regard and included it among the works which he translated for the benefit of priests. In his estimate of the Pastoral Care, and also in his recognition of the value of translations to the Church, Alfred was foreshadowed by Bede. Bede's letter to Bishop Ecgbert advises him to be guided by the Pastoral Care and advocates that the Creed and Lord's Prayer should be taught in the vernacular to those who knew no Latin.14 The monks of the later Anglo-Saxon period were backward-looking in their attitude to the Church in England; tradition and its preservation were a mental preoccupation. Like King Alfred they saw seventh-century Northumbria as laying the foundations of that tradition; this past gave healthy roots to the monasticism of their own day. It also provided the monks with examples of monastic observance and organisation which they could follow and advice which they should take; in addition, it 11 See Alfred the Great: Asserts Life of King Alfred and other contemporary sources, ed. and trans. Simon Keynes and'Michael Lapidge, Harmondsworth 1983, 25, 33. 12 Assess Life of King Alfred, ed. W. H. Stevenson, Oxford 1904, repr. 1959, 88-9. 13 Historia Ecclesiastica, i. 27(1), Venerabilis Baedae Opera Historica, 2 vols, ed. Charles Plummer, Oxford 1896, i. 48. For the existence of independent texts of Gregory's answers see Paul Meyveart, 'Bede's text of the Libellus Responsionum of Gregory the Great to Augustine of Canterbury', in Clemoes and Hughes, England before the Conquest, 23-8 (and nn. for further references). 14 Bede enjoins Ecgbert to meditate on the epistles of St Paul to Timothy and Titus 'et vcrbis sanctissimi papae Gregorii, quibus dc uita simul et uitiis rectorum siue in libro Regulae Pastoralis, seu in omeliis euangelii', and to impress the Creed and Lord's Prayer on the memories of his flock. 'Et quidem omnes, qui Latinam linguam lectionis usu didicerunt, etiam haec optime didicisse certissimum est; sed idiotas, hoc est, eos qui propriae tantum linguae notitiam habent, haec ipsa sua lingua discere, ac sedulo decantare facito': Letter to Egbert, §§3, 5, i. 406, 408-9.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
35
supplied precedents and thus justification for their own methods and proceedings. (The reformers were undeterred by, or perhaps unaware of, the fact that Northumbrian monasticism, except at Wearmouth and Jarrow, did not in fact conform in many respects to their own Benedictine ideals.) In order to use past history in the ways described, it was necessary to know about it. However, here inquirers were faced with almost insuperable difficulties. Material relevant to the state of monasticism in the Viking Age and, though to a lesser degree, in the eleventh century before the Conquest was very scarce. Neither, of course, did the monks have any aids to research to help them collect and evaluate what material there was. But, since reliance on tradition was such an essential part of their mentality, they had somehow to fill the vast gaps in their knowledge in order to reconstruct the past. One method of supplementing their defective information was to use literary topoi. The monks might use, for example, Bede's Ecclesiastical History and his two Lives of St Cuthbert (one in prose and one in verse). Later, in the post-Conquest period, works written during and shortly after the tenth-century revival could be used; ^Ifric's and Wulfstan's Lives of St Ethelwold proved particularly serviceable. Such works provided literary models: a writer might discover what happened in one instance and apply it to another; thus he might use verifiable evidence concerning the foundation of one monastery as a topos for the foundation of another about which he had no specific material. A conclusion reached by means of a topos is speculative. It is not necessarily false; on the contrary, because the Anglo-Saxons were so traditionalist, they must often have been tempted to imitate their forebears' acts. Nevertheless, since circumstances vary in different times and places, no conclusion obviously or possibly reached by means of a topos must be trusted unless supported by verifiable evidence. The historian may, of course, find it hard to decide whether or not a writer was using a literary model. Perhaps the substance of a later work has points in common with an earlier one, but other evidence makes its use as the exemplar problematical. For instance, the gist of some passages in the tenth-century reformers' writings is so very like parts of Bede's letter to Bishop Ecgbert that it is hard to believe that the reformers did not use it. However, there are no distinct verbal echoes, nor is it certain that Bede's letter to Ecgbert was readily available at the time of the revival. (It survives, as far as is known, only in three manuscripts, of the early tenth, early twelfth and fifteenth centuries.)15 Even if the parallels do not 15
M. L. W. Laistner, A Hand-List of Bede Manuscripts, Ithaca 1943, 120. Symons suggests that the authors of the Regularis concordia, when they prescribed daily communion, had in mind the passage in Bede's Letter to Ecgbert deploring the neglect of the eucharist in Northumbria and enjoining that people be taught the spiritual benefits of daily communion, Thomas Symons, 'Sources', 157-8; and idem (ed. and trans.), Regularis concordia, London 1953, 2 n.a. Patrick Wormald accepts that the reformers borrowed from the Letter to Ecgbert.
36
Legends, Traditions and History in Medieval England
actually prove direct borrowing, they certainly do prove that, on the topics which concerned them, the tenth-century reformers felt and thought very like Bede. The traditionalism of the Anglo-Saxon monks and the continuity of monastic history will be discussed in two parts: the tenth-century revival; and the period c. looo-c. 1100. I. The Tenth-century Revival From the reign of Edgar until that of William the Conqueror monasticism in England depended on royal patronage and protection. In fact King Edgar played as important a part in the tenth-century revival as did SS Dunstan, Ethelwold and Oswald. He must have believed in the revival's spiritual value. As a child he had spent some time in St Ethelwold's monastery at Abingdon,16 and St Ethelwold translated the Rule of Benedict for him and his queen, ^Elfthryth. 17 His alliance with the monks (typified by the fact that the royal palace at Winchester was situated immediately opposite St Ethelwold's cathedral)18 won him the benefit of their prayers; the Regularis concordia stipulates that the monks should pray daily for the king and queen.19 It also procured fitting burial places for royalty, since the monastic churches were the finest in the land. And the alliance gave him political advantages. The monks accorded an almost theocratic position to the king (the Regularis concordia even compares him with the Good Shepherd) ;20 and the daily prayers for the king and queen were constant reminders of their allegiance. The king also gained more mundane benefits: since the monasteries had a virtual monopoly of literate education, they were the recruiting ground for the episcopate; and the king chose his ministers from the abbots and bishops. Edgar, therefore, had both religious and political reasons for espousing the cause of the reformers. However, he was also probably influenced by the examples of earlier rulers. When he summoned the synod at Winchester £.972 in order to impose unity of observance on the monasteries, he may well have been following the example of Louis the Pious who had summoned a synod at Aachen for a similar purpose in 816. 16
See St Ethelwold's account of the establishment of the monasteries, in Councils and Synods, 1: AD 871-1204, ed. Dorothy Whitelock, Martin Brett and C. N. L. Brooke, 2 pts, Oxford 1981, i. 142-54; Eric John, Orbis Britanniae and Other Studies, Leicester 1966, 158-60. 17 Printed in Die Angelsdchsischen Prosabearbeitungen der Benediktinerregel, ed. Arnold Schroer, 2nd edn revised with suppl. by H. Gneuss, Darmstadt 1964. Cf. Mechthild Gretsch, Die Regula Sancti Benedict! in England und ihre Altenglische Ubersetzung, Munich 1973; and idem, ' Aethelwold's translation of the Regula Sancti Benedicti and its Latin exemplar', A-SE iii (1974), 125-51. 18 Martin Biddle, 'Felix urbs Winthonia: Winchester in the age of monastic reform', in Parsons, Tenth-century Studies, 138. 19 Regularis concordia, in Consuetudinum Saeculi X/XI/XII Monumenta non-Cluniacensia, ed. 20 Kassius Hallinger, CCM vii 3. 74, 83. Ibid. 70.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 37 It is certain that the reformers were influenced by the Capitula and Regula canonicorum issued by the synod of Aachen, since they borrowed from them for the Regularis concordia.21 Nevertheless, Anglo-Saxon history could provide an example of a king who had patronised the monks and summoned a reforming synod. Oswy, king of Northumbria (654-70), had done so. The purpose of the synod of Whitby (664) was to heal the rift in the Northumbrian Church between the Roman and Celtic parties. Edgar and SS Dunstan, Ethelwold and Oswald would have learned about Oswy from Bede's Ecclesiastical History (iii. 24 25), with which they were undoubtedly well acquainted (the prologue to the Regularis concordia draws on it in a number of places, and a long passage in St Ethelwold's treatise on King Edgar's establishment of the Monasteries also derives from it). 22 It is likely that the reformers turned to the Ecclesiastical History for precedents and advice. The prologue to the Regularis concordia states that they summoned monks from Fleury and Ghent to the synod because they recalled 'the letters in which our holy patron Gregory instructed the blessed Augustine that, for the advancement of the rude English church, he should establish therein the seemly customs of the Gallic churches as well as of Rome'. The author probably read the text of Gregory's Responsiones in the Ecclesiastical History (although he, like Asser, might have used an independent text). 23 Bede's letter to Bishop Ecgbert contains the model of a king, Ceolwulf, king of Northumbria (729—60/5), who, like Oswy, was a patron of monks. Bede tells Ecgbert to seek his help in the reform of his province and to summon a reforming synod.24 Bede's description of Ceolwulf would have fitted Edgar, or at least the reformer's image of him (which probably represented the type of king he aspired to be). It reads: qui et pro insita sibi dilectione religionis, quicquid ad regulam pietatis pertinet, firma protinus intentione adiuuare curabit, et maxime ilia, quae tu, quum sis propinquus illius amantissimus, bona caeperis, ipse, ut perficiantur, opitulari curabit. 25
Similarly, as Sally Vaughn has recently pointed out, a letter of Gregory the Great to King Ethelberht cited by Bede (HE i. 32) describes the attributes of a good Christian king. It tells Ethelberht to promote the Christian faith among his people and to follow the guidance of Bishop Augustine. Vaughn argues that this exemplum conditioned Eadmer's view of William the Conqueror and of his relationship with Lanfranc and Anselm. 26 Equally it could have influenced the tenth-century reformers' view of King Edgar. Bede's advice to Bishop Ecgbert on the creation of monastic sees and the 21
Ibid. 71 lines 7-11 and note. Cf. Symons, 'Sources', 165-70. 2:J Councils and Synods, ii. 143 and n. i . Reg. con., 71 2. 24 '-5 Ibid. Letter to Ecgbert, §9, i. 412. 26 Sally Vaughn, 'Eadmer's Historia novorum: a rcinterprctation Proceedings (1987), pp. 263-4. 22
38
Legends, Traditions and History in Medieval England
appointment of bishops resembles the regulations on these matters in the Regularis concordia. Bede reminds Ecgbert that Pope Gregory had stipulated that there should be twelve bishops in the province of York. Therefore, the reforming synod, in order to increase the existing number of bishops, should find monasteries suitable to serve as sees. Lest the abbot or community of a monastery chosen to be a see should oppose, the monks were to be allowed to choose one of their number to be bishop, and he was to rule both the monastery and the diocese; if there were no one suitable in the monastery, some suitable person in the diocese should be found. The passage reads: Quapropter commodum duxerim, habito maiori concilio et consensu, pontifical! simul et regali edicto prospiciatur locus aliquis monasteriorum, ubi sedes fiat episcopalis. Et ne forte abbas vel monachi huic decreto contraire ac resistere temptauerint, detur illis licentia, ut de suis ipsi eligant eum, qui episcopus ordinetur, et adiacentium locorum, quotquot ad eandem diocesim pertineant, una cum ipso monasterio curam gerat episcopalem; aut, si forte in ipso monasterio, qui episcopi ordinari debeat, inueniri nequeat, in ipsorum tamen iuxta statuta canonum pendeat examine, qui de sua diocesi ordinetur antistes.27 A passage in the Regularis concordia recalls this advice. It decrees that the monks of a cathedral monastery should elect the bishop, in the same way as monks of an ordinary monastery elect their abbot, from among their number, if there is a monk worthy of the office, and no one should oppose. But if, 'owing to their ignorance or sinfulness', there is no one suitable, a monk should be chosen from another monastery. Whoever is chosen should live with the monks and set an example of strict adherence to the Rule. The passage reads: Praefato equidem synodali conciliabulo hoc attendendum magnopere cuncti decreuerunt.. Episcoporum quoque electio uti abbatum, ubicumque in sede episcopali monachi regulares conuersantur, si Domini largiente gratia tanti profectus inibi monachus reperiri potuerit, eodem modo agatur; nee alio quolibet modo dum eiusdem sunt conuersationis a quoquam praesumatur. Si autem, imperitia impediente uel peccatis promerentibus, talis qui tanti gradus honore dignus sit in eadem congregatione reperiri non potuerit, ex alio noto monachorum monasterio [Regula Benedicti^ cap. Ixi. 31 ff.], concordi regis et fratrum quibus dedicari debet consilio eligatur. 28 Knowles saw this passage in the Regularis concordia as a supreme example of St Dunstan's statesmanship. It put the characteristically English institution of the monastic cathedral on a firm basis, at a time when, on the continent, reformers were trying to emancipate monasteries from episcopal control. Knowles pointed out that the idea that bishops should live in ordered communities originated in Gregory's Responsiones,29 but he does not mention the parallel in Bede's letter to Ecgbert. 27 29
28 Letter to Ecgbert, §10, i. 413. Reg. con., 74-5. Knowles, Monastic Order, 45-6. HE, i. 27; i. 48-9.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
39
Characteristic of the writings of the tenth-century revival is the invective against the clerical communities which preceded the new Benedictine monasteries. The Regularis concordia states that Edgar rejected 4 the abominations of the negligent clerks' ('eiectisque negligentium clericorum spurcitiis'). 30 Edgar's foundation charter to the New Minster, Winchester, describes in disparaging terms the clerks whom he expelled from that church.31 The earliest biography of St Oswald (c. 1000), which is attributed to Byrhtferth, thus depicts the character of the religious before the revival: In diebus illis non monastici viri, nee ipsius sanctae institutionis regulae erant in regione Anglorum, sed erant religiosi et dignissimi clerici qui tamen thesauros suos, quos avidis acquirebant cordibus, non ad ecclesiae honorem, sed suis dare solebant uxoribus.32 Jilfric and Wulfstan, the early biographers of St Ethelwold, cast similar aspersions on the clerks who served the Old Minster, Winchester, before its reform. They, too, refer to the clerks' wives. JLlfric writes: Erant autem tune in ueteri monasterio, ubi cathedra episcopalis habetur, male morigerati clerici, elatione et insolentia ac luxuria preuenti, adeo ut nonnulli eorum dedignarentur missas suo ordine celebrare, repudiantes uxores quas inclite duxerant et alias accipientes, gulae et ebrietati iugiter dediti.33 Possibly Bede influenced the terms of this invective. The passage in the Regularis concordia and that in Edgar's charter to the New Minster which 30 Reg. con., 70. Knowles accepts the truth of such aspersions on the clerks, op. cit. 41 n. 3, and so, with some reservation, does Hallinger, Reg. Con., 381-2. 31 Walter de Gray Birch, Cartularium Saxonicum, 3 vols, London [1883]—1893, repr. 1963, iii. 456 no. 1190; P. H. Sawyer, Anglo-Saxon Charters, London 1968, 240 no. 745. Cf. other passages in the same charter: 'vitiosorum cuneos canonicorum, e diversis nostri regiminis coenobiis Christi vicarius eliminavi', 'rebelliones omnipotentis voluntati obviantes possessionem domini usurpare non sustinens clericos lascivientes repuli': Birch, op. cit. iii. 459. John cites this charter and quotes passages from two other Winchester charters (Birch, op. cit. nos 1147, 1159; Sawyer, op. cit. 258-9 nos 817, 818) which contain virulent abuse of the clerks, Eric John, 'The church of Winchester and the tenth-century reformation', BJRL xlvii (1964-5), 420-1. I have not included these among my examples since the charters in question are of dubious authenticity and may be post-Conquest, at least in their present form; see F. E. Harmer, Anglo-Saxon Writs, Manchester 1952, 374 and nn. i, 2. If so, their invective could reflect Norman propaganda rather than tenth-century opinion. 32 Vita Oswaldi archiepiscopi Eboracensis, in The Historians of the Church of York and its Archbishops, ed. James Raine, Jr, 3 vols (Rolls Series, 1879-94), i- 4 1 1 33 ^Elfric's Vita S. &thelwoldi, in Three Lives of English Saints, ed. Michael Winterbottom, Toronto 1972, 22-3. Knowles, Monastic Order, 41 n. 3, takes this passage seriously. A slightly longer version occurs in Wulfstan's Vita glonosi et beati patris Athelwoldi, in Three Lives, 44. It should be noted that /Elfric's and Wulfstan's Lives are closely related; either /Elfric abbreviated Wulfstan's Vita, which is the fuller of the two, or Wulfstan expanded ^Elfric's. Recent scholars conclude that Wulfstan's is the original. See Barbara York, 'Introduction', in idem, Bishop .Ethelwold, 1-2; Michael Lapidge, '/Ethelwold as scholar and teacher', in ibid. 89 n. i. A new edition, with trans., of Wulfstan's Vita, by Michael Winterbottom and Michael Lapidge is forthcoming (Oxford).
40
Legends, Traditions and History in Medieval England
refer to the clerks' 'abominations' seem to echo the Ecclesiastical History, ii. 4: Bede relates that Pope Boniface obtained the Pantheon from the Emperor Phocas and, ' when all the abominations had been eliminated' ('eliminata omnia spurcitia'), converted it into a church.34 Moreover, there are parallels between the reformers' abuse and Bede's letter to Ecgbert. Bede had dwelt on the luxurious lives and licentiousness of the monks of his day and also inveighed against the immorality of young noblemen.35 In two particulars the resemblance between his invective and that of the reformers is striking. The accusation of the latter that the clerks were married, including the allegation in the Life of St Oswald that the clerks squandered 'treasure' on their wives, recalls part of Bede's tirade where he claims that the secular owners of monasteries divided their time between monastic concerns and their wives and children and gave land to their wives for the foundation of convents: Idem ipsi uiri modo coniugis ac liberorum procreandorum curam gerunt, modo exsurgentes de cubilibus, quid intra septa monasteriorum geri debeat, sedula intentione pertractant. Quin etiam suis coniugibus simili impudentia construendis, ut ipsi aiunt, monasteriis loca conquirunt, quae pari stultitia, cum sint laicae, famularum se Christi permittunt esse rectrices.36 The other parallel is Bede's attribution of the decline of monasticism to secularisation resulting from the power of the lay nobility. Noblemen obtained lands from the king in order to found monasteries; they then lived in them like lords, ruling a motley crowd of men unworthy of the monastic vocation: At alii grauiore adhuc flagitio, cum sint ipsi laici, et nullo uitae regularis uel usu exerciti, uel amore praediti, data regibus pecunia, emunt sibi sub praetextu construendorum monasteriorum territoria in quibus suae liberius uacent libidini, et haec insuper in ius sibi haereditarium regalibus edictis faciunt asscribi, ipsas quoque litteras priuilegiorum suorum quasi veraciter Deo dignas, pontificum, abbatum, et potestatum seculi obtinent subscriptione confirmari. Sicque usurpatis sibi agellulis siue uicis, liberi exinde a diuino simul et humane seruitio, suis tantum inibi desideriis, laici monachis imperantes, deseruiunt; immo non monachos ibi congregant, sed quoscunque ob culpam inobaedientiae ueris expulsos monasteriis alicubi forte oberrantes inuenerint, aut evocare monasteriis ipsi ualuerint; vel certe quos ipsi de suis a satellitibus ad suscipiendam tonsuram promissa sibi obaedientia monachica inuitare quiuerint. 37 The prohibition in the Regularis concordia on monasteries to acknowledge secular overlordship, with a particular reference to the ruin this had caused in the past, is reminiscent of this passage: Saecularium uero prioratum, ne ad magni ruinam detriment uti olim acciderat miserabiliter deueniret, magna animaduersione atque anathemate suscipi coenobiis sacris sapienter prohibentes.. 38 34 36 38
Op. Hist., i. 88. Ibid. §12, i. 416.
Reg. con., 7.
35 37
Bede's letter to Ecgbert, §§10-12, i. 413-16. Ibid. §12, i. 415-16.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
41
St Ethelwold, in his treatise on King Edgar's Establishment of the Monasteries, admonishes abbesses not to give 'God's estate to their kinsmen nor to great secular persons, neither for money nor flattery'. He also enjoins the king and secular lords not to seize the opportunity ' to rob God, who owns those possessions', if any abbess 'be convicted of a crime against church or state'. 39 On the evidence of these passages Eric John concludes that lay power alone accounted for lax observance in religious communities from the early eighth to the mid-tenth century. 40 He dismisses the Viking incursions as a factor. This conclusion, however, disregards the possible use in the Regularis concordia and Ethel wold's treatise of a topos derived from Bede. It also ignores the evidence of Asser and of Fulco in his letter to King Alfred, who both suggest the Vikings as another explanation. (Alfred, in the prologue to the Pastoral Care, treats the Vikings as instruments of divine punishment for the decline of monastic standards.) 41 Nor was the power of the nobility always inimical to monasticism: to cite just one example, during the revival itself Ramsey Abbey owed its site to a nobleman, ^Ethelwine, ealdorman of East Anglia. As its patron, he kept close contact with the monks, and he and one of his brothers were buried in the conventual church. During the reaction against the new monasteries which immediately followed the death of King Edgar, ^Ethelwine and his brothers and Byrhtnoth, ealdorman of Essex, fought in defence of the East Anglian houses. Nevertheless, the attacks were led by another thegn, JLlfhere, ealdorman of Mercia, and, while ^Ethelwine was regarded by the monks of his own monastery of Ramsey as 'amicus Dei', to those of Ely he was a bitter enemy (a usurper of their estates - in Ramsey's interest). 42 In order to justify their take-over of the religious system and to strengthen their position, the reformers resorted to propaganda. Perhaps sometimes they objected to a clerical community because it did not conform to the new continental standards of observance. But in many cases they had worldly reasons for wanting to discredit the clerks; if a clerical community owned a church and property which a reformer needed for his monks, it had to be dispossessed. The property of the clerical communities was partly built up from the inheritance of individual 39
40 Councils and Synods, 153-4. John, 'The king and the monks', 61. For the evidence of Asser and Fulco see Keynes and Lapidge, Alfred the Great, 103, 182-3. For Alfred's prologue to the Pastoral Care see ibid. 125. 42 For ^Ethelwine's relations with Ramsey see Vita Oswaldi, 427-9, 447, 468, 475. For his defence of the East Anglian monasteries see D. J. V. Fisher, 'The anti-monastic reaction in the reign of Edward the Martyr', Cambridge Historical Journal x (1952), 254, 258, 265-6, 267; cf. Wormald, 'Ethelwold', 36. For his bad relations with Ely sec Fisher, op. cit. 266-7. For lay lords as protectors of monasteries on the Continent in the late ninth and in the tenth century see H. E. J. Cowdrey, The Cluniacs and the Gregorian Reform, Oxford 1970, 11-12. For a careful assessment of yEHhcrc's probable motives (which were apparently mainly political) for attacking the new monasteries see Ann Williams, ' Princeps Merciorum gentis: the family, career and connections of /Elfhere, ealdorman of Mercia, 956-83', A-SE x (1982), 159-61, 166-70. 41
42
Legends, Traditions and History in Medieval England
clerks and from endowments given by their families. The clerks' relatives tended to be people of consequence in the neighbourhood and, apparently, to feel responsible for the communities with which they had a family connection. The clerks and their supporters were, therefore, powers to be reckoned with. 43 King Edgar had to resort to force to eject the clerks from the Old and New Minsters in Winchester and from Chertsey and Milton. 44 The reformers feared that such dispossessed clerks would try to recover their property, and the king in particular faced the danger of a hostile combination of them and their supporters. At least some people objected, in any case, to influence from abroad. The Anglo-Saxon Chronicle, in a eulogy of Edgar under 959, admits that he had one grave fault: 'he loved evil foreign customs and was excessively determined to introduce heathen manners into this land, and he encouraged foreigners and harmful people to come to this country'. 45 That the reformers' fears were not groundless is proved by the violence of the reaction against the new monasteries led by JElfhere, ealdorman of Mercia, and his followers. Apparently they despoiled a number of Mercian monasteries and, in a few cases, dispersed the monks.46 The conclusion that the reformers' abuse of the clerks was part of their propaganda is supported by the fact that it only occurs in writings of monastic origin (of Worcester, Ramsey and Winchester provenance). There is no trace of it in the Life of St Dunstan by the anonymous author, whose name began with 'B', who was probably a secular clerk, not a monk.47 Nor, in view of the achievements of the Anglo-Saxon Church in the tenth century before the monastic revival, is it likely that the clerks were as bad as the reformers painted them. Scholars from the tenth century until recently, while tending to exaggerate the impact of the tenth-century monastic revival, have in general paid too little attention to the fate of the old clerical communities which were replaced by Benedictine ones. In fact the reformers made no clean sweep. The abruptness of the transition from clerical to monastic community at Winchester, Milton and Chertsey was exceptional. In most cases the change was gradual. The reformers, indeed, had grown up in the old tradition. St Dunstan was consecrated to the priesthood by a clerk, his 43 For the clerks and their family connections see Fisher, op. cit. 255-64 passim; and D. J. Sheerin, 'The dedication of the Old Minster, Winchester, in 980', Rev. Ben. Ixxxviii (1978), 265, 269-70. 44 The Anglo-Saxon Chronicle, ed. Dorothy Whitelock with D. C. Douglas and S. I. Tucker, London 1961, 76.
45 Ibid - ?5-
46
See Fisher, 'Anti-monastic reaction', 254-70; Williams, op. cit. 167-70; Blair, 'Secular minster churches', 119. According to later tradition the monks of Evesham were expelled and the clerks reinstated - but not for long. See Chronicon Abbatiae de Evesham, ed. W. D. Macray (Rolls Series, 1863), 78. 47 I owe this comment about B's Life of St Dunstan to Dr Simon Keynes. The Life is printed in Memorials of St Dunstan, ed. William Stubbs (Rolls Series, 1874), pp. 3-52.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
43
48
kinsman, ^Elfheah, bishop of Winchester. St Oswald was sent to Fleury to study monasticism by another clerk, his uncle, Oda, archbishop of Canterbury (942—58), himself a notable reformer. Before going to Fleury St Oswald had ruled some clerical community at Winchester; we only have the word of his first biographer, who, being a monk of Ramsey, would have been a biased witness on the matter, that he found the clerks' mode of living repulsively luxurious. His biographer writes that St Oswald, having habitually worn silk and feasted sumptuously every day, became discontented: Cum his [clericis] mansitabat pius adolescens, velut Loth in Sodomis, quorum superfluas nenias cernens vir providus, coepit frequenter sollicitis cogitationibus, pulsari, et beatae mentis indagine praemeditari, quia carnis delicta magis augent quam debent, qui de alienis rebus vivunt et injuste agere non desistunt.. !49 The reformers must have recognised the merits of the old order and realised that its destruction, or any indiscriminate damage to it, would sap the vitality of their own movement. Their desire to minimise any break with the past is well illustrated by the fact that two of the foundations, Ely and Peterborough, were in fact refoundations of monasteries mentioned by Bede. Similarly, St Oswald attempted to re-establish a monastery at Ripon. His biographer vividly demonstrates his revivalist intention: Oswaldus vero impiger miles Christi sua ovilia perlustrans pervenit gaudens ad moenia monasterii quod dictum est Ripun, quod tune dirutum erat, quondam vero a reverentissimo viro Wilfridi celebriter constructum. 50 When the reformers wanted to dispossess a community of clerks, sometimes, at least, they acted with moderation and tact. Even at the Old Minster, Winchester, whence King Edgar 'drove the clerks' to make room for St Ethelwold's monks, there was some continuity of personnel; the king, according to St Ethelwold's biographers, ^Elfric and Wulfstan, gave the clerks the choice of conforming to the Benedictine Rule or of leaving. Three, Eadsige, Wulfsigeand Wulfstan, did in fact eventually conform: Misit quoque rex quendam ministrorum suorum famosissimum, Uulstanum vocabulo, cum episcopo, qui regia auctoritate mandauit clericis ocissime dare locum monachis aut monachicum suscipere habitum. At illi execrantes monachicam uitam ilico exierunt de aecclesia; sed tamen postmodum tres ex illis 48 Memorials of St Dunstan, 13-14. William of Malmesbury in his Life of St Dunstan alleges that ^Elfheah had been a monk of Glastonbury, ibid. 260. However, the earliest Life, that by 'B', makes no mention of this, and in any case even if William's statement were true (and it could have been a product of his desire to stress the antiquity of the monastery at Glastonbury), it would only signify that ,/Elmeah was a member of the 49 'clerical' community before Dunstan's reform. Vita Oswaldi, 411. 50 Ibid. 462. Cf. below p. 179.
44
Legends, Traditions and History in Medieval England
conversi sunt ad regularem conuersationem, scilicet Eadsinus, Uuilstanus. 51
Uulfsinus,
Two post-Conquest authorities, 'Florence of Worcester' (c. iioo) and the Ramsey chronicler (temp. Henry n), give similar accounts of St Oswald's procedure at Worcester, asserting that he gave the clerks in the cathedral the choice of conforming to the Rule of Benedict or of leaving.52 However, the Ramsey chronicler was probably here borrowing from 'Florence', and 'Florence' in turn was probably using as a topos ^Elfric's or Wulfstan's account of St Ethelwold's treatment of the clerks of the Old Minster. Rather more convincing, because it has no obvious model in ^Elfric or Wulfstan, is the Ramsey chronicler's statement that St Oswald sent one of the clerks from Worcester to Ramsey for training and then brought him back to rule the new monastic community in the cathedral: unus [clericorum] Winsinus nomine, qui caeteris et dignitate generis et opum copia insignior habebatur, saeculo renuncians et habitu monastico amictus, Rameseiam missus est, regularis ordinis disciplina instruendus et instituendus, ac deinceps inde revocatus Wigorniae prioratum est promotus.53 Moreover, St Oswald treated even those clerks who remained obdurate leniently. He left them in possession of the old cathedral, which in any case was too small, and built a new_one for his monks.54 51 jElfric's Vita S. ^Ethelwoldi, 23. For the similar passages in Wulfstan see ibid. 45. For Eadsig see jElfric's Lives of Saints, 2 vols, ed. W. W. Skeat (EETS, orig. ser. Ixxvi, Ixxxii, xciv, cxiv, 1881-1900), i. 442-7. Knowles, Monastic Order, 41, notes that these three clerks became monks of St Ethelwold's new monastery; however, he then continues with this sweeping statement: 'The expulsion of the clerks from the New Minster was effected in the next year, and followed by similar action in other places throughout the country.' Knowles comments that 'the measure, if its causes and consequences are studied in contemporary documents, needs no elaborate defence' and cites contemporary 'evidence' of the clerks' misdemeanours: ibid. 41 n. 3. 52 Florentii Wigorniensis monachi Chronicon ex Chronicis, 2 vols, ed. Benjamin Thorpe (English Historical Society, 1848-9), New York 1964, i. 141 (s.a. 969); Chronicon Abbatiae Rameseiensis, ed. W. D. Macray (Rolls Series, 1886), 41. 53 Ibid. 41-2. 'Florence of Worcester' states that Wynsige was a monk of Ramsey (which is not, however, incompatible with the Ramsey chronicler's narrative), Chron. ex Chronicis, i. 141. 54 SeeJ. Armitage Robinson, St Oswald and the Church of Worcester, London 1919, passim, esp. at pp. 3-6, 2O-J, 36-7. On the other hand John argues that the conversion at Worcester was more like that at Winchester rather than being peaceful, ' St Oswald and the Church of Worcester', in idem, Orbis Britanniae and Other Studies, Leicester 1966, 234-48. His case is based on the so-called Altitonantis charter purportedly of King Edgar to Worcester (Birch, Cart. Sax., iii. 377-81 no. 1135), the authenticity of which he defends in his Land Tenure in Early England, Leicester 1960, 80-139. However, John's views have not persuaded Professor R. R. Darlington, The Cartulary of Worcester Cathedral Priory (Pipe Roll Society Ixxvi (NS, xxxviii), 1962—3), pp. xiii—xxiv, esp. p. xviii n. 6, or Professor P. H. Sawyer, 'Charters of the reform movement: the Worcester archive', in Parsons, Tenthcentury Studies, 84-93. They both decide in favour of Robinson's conclusions. It may also be noted that the chronicle evidence cited by John is unreliable in this context.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 45 Nicholas Brooks has pointed out that, despite the tenth-century monastic revival, ' monastic cathedrals remained very much the exception in England'. St Dunstan, indeed, never replaced the clerks of Christ Church, Canterbury, by monks at all. The community's transformation took place probably in the early eleventh century, when it was brought about gradually. (The story that the clerks were dramatically expelled from the cathedral, which occurs in the forged 'foundation charter' of the second quarter of the eleventh century, is fictitious and was probably formulated under the influence of Edgar's procedure at Winchester). St Dunstan would have seen no reason to displace the clerks; by all accounts they fulfilled their duties in the cathedral well.55 In the north of England the community attached to the see of St Cuthbert remained clerical even longer - considerably longer - than that at Canterbury. It served St Cuthbert's shrine and the cathedral first at Chester-le-Street and then at Durham. Its prosperity is apparent in the Historia de Sancto Cuthberto, a history of the see written in the first instance in 946 but continued subsequently. The Historia records a visit to the shrine by King Athelstan in 934 on his way to campaign in Scotland. His gifts to the saint included extensive estates, rich treasures and precious books.56 One of the books was probably the splendidly illuminated volume containing Bede's two Lives of St Cuthbert and a liturgical office for the saint which still survives.07 The Historia asserts that Athelstan enjoined his brother, Edmund, that, if he himself were killed on the expedition, he should bear his body to St Cuthbert's and there commend his soul to God. In fact Athelstan survived, but Edmund, when king, visited St Cuthbert's for the sake of prayer, also on the way to Scotland. He placed two gold armlets (armillas aureas) and two Greek pallia on the body and confirmed 'peace and law' (pacem vero el legem] to all St Cuthbert's lands.58 Clearly at the start of the monastic revival the see of St Cuthbert was flourishing, untouched by the reformers. In 995 the clerks moved St Cuthbert's body to Durham for greater security. There their community survived, serving the shrine and cathedral until after the Norman Conquest, when, in 1083, it was finally replaced by a Benedictine monastery. At the same time the fortunes of the clerical community serving St Edmund's shrine at Beadericesworth (the later Bury St Edmunds) rose. Although the cult of St Edmund, king and martyr, had flourished in East 55 See Nicholas Brooks, The Early History of the Church of Canterbury; Christ Church from 597-1066, Leicester 1984, 255-7. 56 Historia de Sancto Cuthberto, in Symeonis Monachi Opera Omnia, 2 vols, cd. Thomas Arnold (Rolls Series, 1882-5), *• 211-12. For the Historia sec Edmund Crastcr, 'The patrimony of St Cuthbert', EHR Ixix (1954), 177-8. John Blair gives many examples of secular minsters which flourished in the tenth and eleventh centuries, 'Secular minster churches', 120-3. 57 The book is now Corpus Christi College, Cambridge, MS 183. See Keynes, 'King 5H Athelstan's books', 180-5. Historia de S. Cuthberto, 212.
46
Legends, Traditions and History in Medieval England
Anglia for a generation after the saint's 'martyrdom' by the Vikings in 869,59 it had subsequently lapsed. After the Norman Conquest the Bury monks contended that King Edmund (939-46) had granted to the clerks, who preceded them as guardians of St Edmund's shrine, jurisdictional privileges over Beadericesworth and the adjacent land. Although the charter, on which this claim rested, is of doubtful authenticity, 60 it is not unlikely that King Edmund would have held St Edmund, his namesake, in especial veneration and was a benefactor of the community at Beadericesworth. Certainly some of his relatives were its benefactors. His father-in-law, ^Elfgar, bequeathed to his daughter ^Ethelflaed, Edmund's second wife, an estate at Cockfield (Suffolk) with reversion to St Edmund's some time between 946 and 951. ^Ethelflaed did not directly bequeath the estate to St Edmund's but, together with an estate at Chelsworth (Suffolk), bequeathed it to her sister, ./Elflaed, and brother-inlaw, Brihtnoth, again with reversion to St Edmund's. ^Elflaed duly bequeathed Cockfield and Chelsworth and also Nedging (Suffolk) to St Edmund's in £.iooo. 61 The fourteenth-century Benefactors' Lists state that King Edwy, King Edmund's son (955-9),gave Beccles (Suffolk) 'but without a charter'. 62 Besides royal and royally connected benefactors, St Edmund's had one, or more probably two, episcopal patrons in the tenth century. Theodred, bishop of London (c. 909 x 926-951 x 953), who was on good terms with King Edmund, bequeathed estates at Nowton, Horningsheath, Ickworth and Whepstead (all in Suffolk) 'for the good of my soul'.63 The later Benefactors' Lists state that ^Elfric, bishop of East Anglia, gave estates at Soham, Bradfield, Brockford and Southwold (all in Suffolk), at Runcton (Norfolk) and at 'Thorpe'. One Benefactors' List identifies this Bishop ^Elfric with ^Elfric i (? x 970-970 x ?). 64 In addition, in about 961, °9 See C. E. Blunt, 'The St. Edmund memorial coinage', Proceedings of the Suffolk Institute of Archaeology xxxi (1970), 234-55. 60 For King Edmund's purported charter see Sawyer, Anglo-Saxon Charters, 191-2 no. 507; and C. R. Hart, The Early Charters of Eastern England, Leicester 1966, 54-8 no. 74. Hart believes that at least the part of the charter delineating the boundaries of St Edmund's jurisdictional area is pre-Conquest. 61 For ^Elfgar's, ^Ethelflaed's and j£flaed's wills see Sawyer, op. cit. 414-16 passim, 418 nos 1483, 1486, 1494; Dorothy Whitelock, Anglo-Saxon Wills, Cambridge 1930, 7-9, 35-43 nos ii, xiv, xv, 103-8, 137-46. ^Ethelflaed had acquired Chelsworth from King Edgar, Sawyer, op cit. 231 no. 703. 62 See the Benefactors' List, written in the last half of the thirteenth century, in Cambridge University Library MS Ff. 2. 33. fo. 5ov; The Pinchbeck Register, 2 vols, ed. Francis Hervey, Brighton 1925, ii. 284. Domesday Book, ii. 369^ shows that St Edmund's had held there before 1066 and still did so in 1086. 63 For Bishop Thcodred's will see Whitelock, Anglo-Saxon Wills, 2-5 no. i, 99-103. Abbo mentions Theodred's generosity to St Edmund, Passio S. Eadmundi, in Three Lives, 83. 64 St Edmund's certainly had landholdings at Bradfield, Brockford, Soham, Southwold and Runcton before the Norman Conquest, Domesday Book, ii. 362, 36ib, 368b, 37ib, 309 respectively. All these places except Southwold contributed a month's, or part of a month's, food-rent to the abbey, A. J. Robertson, Anglo-Saxon Charters, Cambridge 1956,
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
47
St Edmund's received a grant of land at Palgrave (Suffolk) from Wulfstan, a thegn, and between 961 and 995 the bequest of'Northo' from another thegn, Athene.6o By this time the reputation of St Edmund was well established in eastern England, sufficiently so to prompt the monks of Ramsey to commission Abbo of Fleury during his stay with them, c. 986—8, to write an account of his 'martyrdom' and early cult, the Passio Sancti Eadmundi.66 There is no doubt that no attempt was made during the monastic revival to reform St Edmund's clerical community, which survived well into the eleventh century. The community at Beadericesworth differed, it would appear, in an important respect from the one at Christ Church, Canterbury, and from that of St Cuthbert's: it was not attached to an episcopal see. However, there is a little evidence suggesting the possibility that Beadericesworth was a see for a time in the tenth century. In the eighth century East Anglia was divided into two dioceses. Bede records that the northern see was at Elmham and the southern one at Dunwich. 67 At some periods in the tenth and eleventh centuries the latter was at Hoxne; 68 its cathedral was dedicated to St Ethelbert, who, like St Edmund, was a martyred king of East Anglia. The evidence that Beadericesworth was ever a see is postConquest. It is to be found in literature produced during the quarrel between the monks of Bury St Edmunds and Herbert Losinga, bishop of East Anglia (1090 x 1091-1119), over the latter's attempt to move his see to Bury. In the course of the dispute Losinga's party produced a tract, a copy of which still survives, which supports the bishop's case partly by supplying an historical precedent; the tract 'proves' that Beadericesworth 94-200 passim. The abbot's claim to Southwold was disputed after the Conquest, see Hermann, De miraculis Sancti Eadmundi, in Memorials of St Edmund's Abbey, 3 vols, ed. Thomas Arnold (Rolls Series, 1890-6), i. 79. St Edmund's had landholdings before the Conquest at a number of places called 'Thorpe', 'Torp' or 'Torpa' in Domesday Book, namely Ixworth Thorpe and Thorpe Morieux (both within St Edmund's Liberty), and Abbot's Thorpe, Morningthorpe and Thorpland in Norfolk, Domesday Book, ii. 367 and 3^7b, 333, 369, 2iob, 212, 209 respectively (see also ibid. ii. 36ob). A 'Thorpa', together with Palgrave (Suffolk) provided a month's food-rent, Robertson, op. cit. 194, 200. Robertson identifies this as Westhorpe because it was the nearest 'Thorpe' to Palgrave, ibid. 443, and cf. Domesday Book, ii. 37ob, 371 (Westtorp). However, not all the landholdings which combined to provide a month's food-rent were close together, and both Ixworth Thorpe and Thorpe Morieux, unlike Westhorp, were within St Edmund's Liberty. Hart identifies Bishop yElfric as /Elfric m, IO39X 1043, commenting that the Benefactors' Lists call him 'the good', Hart, Early Charters, 248 no. 248. However, the Pinchbeck Register, having noticed the gift by ^Elfric cognomento bonus of the abovementioned landholdings, writes: 'Fuerunt enim tres venerabiles episcopi elmanenses uno nomine alfrici died unus bonus alter niger et tercius paruus ob differencia uocati'; clearly yElfric 'the good' was ^Ifric I, Pinchbeck Register, ii. 287. 65 For Wulfstan's grant see Sawyer, op. cit. 355 no. 1213. For Aetheric's bequest see ibid. 420 no. 1501; Whitelock, Wills, 42 no. 16/1, 146-8. 66 Printed in Three Lives, 67-88. 67 HE, ii. 15; iv. 5: i. 116-17, 217. Cf. ii. 108, 214-15. 68 See Bishop Theodred's will, Whitelock, op. cit. 2, 102; and Domesday Book, ii. 379-
48
Legends, Traditions and History in Medieval England
had once been an East Anglian see.69 It asserts that: Felix, a colleague (mentioned by Bede) of St Augustine, founded a cathedral dedicated to the Virgin Mary at Beadericesworth; King Cnut made ^Elfric. monk and prior of Ely, bishop of Beadericesworth, stipulating that he should replace the clerical community by a monastic one; after Cnut's death Harold transformed the cathedral priory into an abbey. This narrative is at variance with the account of the origins of the abbey propagated by the monks of Bury St Edmunds at the end of the eleventh century: they claimed that the abbey was founded by Cnut in 1020 on the advice of Bishop ^Elfwine, monk of Ely, to replace the clerical community serving St Edmund's shrine. V. H. Galbraith, who edited the Losinga tract, while concluding that it was 'an unconvincing perversion of the facts, so far as these are known 5 , 70 admits that it may contain a grain of truth. He concedes that, if the writer 'had limited his claim to the period from c. 850 to c. 960, it would have been hard to disprove him, so great is our ignorance concerning the East Anglian see at that time', Losinga had been abbot of Ramsey before becoming bishop of Thetford. The monks of Ramsey had had some part in the rise of Beadericesworth to importance as a cult centre and religious community - by asking Abbo to write the Passio. Therefore, some tradition about the early history of Beadericesworth could have survived at Ramsey. That the Losinga tract used Ramsey material is suggested by its reference to St Felix, whose body was translated to Ramsey in the eleventh century, 71 and by the dedication to St Mary; Ramsey Abbey was dedicated to St Mary (jointly with St Benedict) and so was St Oswald's Cathedral at Worcester.72 The tract differs from the Bury account not only in substance but also in one possibly significant detail: it calls the first bishop of Beadericesworth, jElfric, 'monk and prior of Ely'; as has been seen, the Bury tradition calls the bishop who advised Cnut to found the monastery ^Elfwine, describing him simply as 'monk of Ely'. ^Elfwine was bishop ? x 1019-23 and ^Elfric n 1023 x 1038-1038. Galbraith tacitly assumes that the author confused ^Elfric n with ^Elfwine. But perhaps the Losinga tract preserves here a piece of true information, that a bishop called ^Elfric had had his see at Beadericesworth. This could have been the bishop in St Oswald's time, ^Elfric i, bishop of East Anglia ? x 970-970 x ? who, as mentioned above, appears to have been a substantial benefactor of St Edmund's. The choice of Beadericesworth as a see in the mid-tenth century would not be surprising. It was a more important place, a villa regia according to 69 The tract is printed by V. H. Galbraith, 'The East Anglian see and the Abbey of Bury St Edmunds', EHR xl (1925), 226-8. 70 Ibid. 222. 71 Chron. Ram., 127-8. 72 Cf. Armitage Robinson, St Oswald, 3-6.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
49
Abbo,73 than Hoxne, and presumably St Edmund's cult under royal and episcopal patronage was more thriving than St Ethelberht's. It is known that Archbishop Oda, a zealous reformer, was concerned about the lack of ecclesiastical organisation in East Anglia, which had resulted from the Viking incursions of the ninth century. There is evidence suggesting that he revived the see of North Elmham, 74 and he may well also have turned his attention to East AnghVs southern see. Just possibly St Edmund attracted his attention for family reasons. His father was said to have been a soldier in the host of Ivarr, the Viking leader who, having conquered the north of England, attacked East Anglia and was, according to legend, responsible for St Edmund's 'martyrdom'. 75 Oda might have fostered St Edmund's cult in expiation for his father's complicity in rather the same way as Cnut was to promote the cult of St^Elfheah in order to expiate his murder by the Danes in 1012. It is not inconceivable that Oda planned to raise St Edmund's to a status like that of St Cuthbert's. His north-country connections lend weight to this possibility. In this context it must be noted that comparison of the St Edmund legend in Abbo with the St Cuthbert legend in the earlier Historia de Sancto Cuthberto shows a relationship between the two. It is more likely that the St Cuthbert legend influenced the St Edmund one than vice versa, and Abbo himself might well have been responsible for the borrowing. If so, perhaps he first learned of the St Cuthbert legend from Oda, whom he mentions as one of his patrons. 76 However, since Oda died in 958 Abbo's contact with him must have been about thirty years before his visit to Ramsey. This perhaps makes it more likely that he would have owed any information about St Cuthbert to St Oswald. The St Edmund legend resembles the St Cuthbert one in its emphasis on the uncorruption of St Edmund's body. Indeed, it explicitly compares St Edmund's uncorruption with St Cuthbert's. More specifically, Abbo tells two stories which seem to derive from a story in the Historia. One is about Leofstan, the proud thegn, who insisted on gazing at St Edmund's body and consequently went mad and died. The other relates that thieves were 73 Passio S. Eadmundi, 82. The description villa regia could, however, be a late eleventhcentury interpolation in the Passio] see Gransden, 'Baldwin', 72. 74 See Brooks, Early History, 223-4 and nn - 5 2 > 53- For the scale of the possibly tenth-century rebuilding of the cathedral at North Elmham see S. E. Rigold, 'The Anglian cathedral of North Elmham, Norfolk. Analysis and excavation by the Ancient Monuments Branch of the Ministry of Public Buildings and Works', Medieval Archaeology vi-vii (1962-3), 68, 104-8; and idem, in Peter Wade-Martins, Excavations in North Elmham Park 1967-1972, 2 vols (consecutively paginated), Norfolk 1980, i. 6-8, 137-48; ii. I am indebted to Professor Brooks for these references. 75 Passio S. Eadmundi, 73. For Oda's parentage see Vita S, Oswaldi, 404. 76 Abbo's acrostic poem addressed to St Dunstan includes these lines: 'Solus Odo pius cenSor qui jure sacerdoS/ Te pater ante fuiT, sat nos amplexus amaviT': Memorials of St Dunstan, 410. Cf. Armitage Robinson, St Oswald, 45. For St Oswald's probable interest in St Cuthbert see below p. 50.
50
Legends, Traditions and History in Medieval England
paralysed in the act of robbing St Edmund's shrine. 77 These passages are reminiscent of the story in the Historia of the proud Viking, Onalafball, who entered St Cuthbert's church, defied the saint, became paralysed and died.78 Oda's reforming activities had the full support of King Edmund. As has been seen, both he and his predecessor, King Athelstan, were benefactors of St Cuthbert's, and it is likely that he was also a patron of St Edmund's. Royal interest in these shrine churches was probably not only the result of piety. Two thriving Christian communities, one in the north, the other in East Anglia, headed by loyal bishops and in control of the surrounding areas, would have spread royal power. Robin Fleming has mentioned that, after the reconquest of the Danelaw, the monarchy by endowing monasteries created 'a network of..loyal ecclesiastical officials, and efficient administrative districts'. 79 St Cuthbert's and St Edmund's clerical communities served the same purpose. Moreover, they were Christian strongholds, where prayers could be said for the king's success in battle, and centres for the quick mobilisation of troops. Thus they served as outposts against invasion by heathens from Scandinavia, a constant threat at this time, exactly in those regions where the king most needed them. St Edmund was a particularly appropriate patron saint, since he had died defending the East Anglians against these very enemies. Fleming has demonstrated that, during the Viking Age, the monarchy had accumulated lands lost by the monasteries to create its own defensive blocs against the Danes.80 It seems that now the monarchy was endowing at least St Cuthbert's and St Edmund's for the same defensive purpose. It is clear, therefore, that the clerical communities at Chester-le-Street and Beadericesworth were flourishing at the beginning of the monastic revival and continued to do so throughout its course. There is no evidence that the reformers objected to them. St Oswald's veneration of the Northumbrian saints presumably included St Cuthbert. (His biographer at one point, apparently using Bede's prose Life of St Cuthbert as a source, compares a miracle worked by St Oswald with one of St Cuthbert's.) Perhaps he regarded St Edmund as the re-embodiment, so to speak, of his own namesake, St Oswald, king of Northumbria (634-42), another Anglo-Saxon king 'martyred' by heathens about whom he would have read in Bede. It was the monks of his foundation at Ramsey who commissioned Abbo to write the Passio Sancti Eadmundi, and St Dunstan himself, according to Abbo, provided information about St Edmund. 81 And although St Cuthbert's prospered in an area which was little affected 77 Passio S. Eadmundi, 68. I have suggested elsewhere that possibly the Passio's explicit comparison of St Edmund with St Cuthbert is a late eleventh-century interpolation, Gransden, 'Baldwin', 73-4. 78 Passio S. Eadmundi, 85, 83. Hist, de S. Cuthberto, 209. 79 Robin Fleming, 'Monastic lands and England's defence in the Viking Age', EHR c 80 (1985), 265. Ibid. 247-65 passim. 81 Passio S. Eadmundi, 67.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 51 by the monastic revival, St Edmund's flourished in eastern England where there was no lack of reforming activity. The shrine churches of St Cuthbert and St Edmund exemplify a marked characteristic of Anglo-Saxon religious life, the cult of saints. But that tradition was rooted in the Anglo-Saxon past, dating back to the period of the conversion to Christianity. Bede proves its existence in the seventh and eighth centuries. Although subsequently, until the mid-tenth century, evidence is scarce, it is known that King Alfred venerated saints: he apparently visited a saint's shrine in Cornwall to be cured of an infirmity. 82 The tenth-century reformers showed a lively interest in the cult of Anglo-Saxon saints. Perhaps their enthusiasm was fanned by Pope Gregory's letter to Mellitus, bishop of London, preserved in Bede's Ecclesiastical History'** Gregory recommended the use of relics as a means of encouraging popular worship. Bede himself informed the reformers about the seventh-century Anglo-Saxon saints. (It was under his influence that St Oswald excavated the bones of Northumbrian saints.)84 But the reformers were also influenced by the continental example. Fleury claimed to have the body of St Benedict, and it was a monk of Fleury, Abbo, who wrote St Edmund's Passio. The reformers recognised that relics gave prestige to individual churches. St Oswald used the relics of the Northumbrian saints for his attempted refoundation to Ripon. Having built a suitable church, he placed the bone of St Wilfrid, Ripon's original founder, together with those of the other saints, in a shrine. '^Edificauit nouae Hierosolymae portas, construxit ipsius coenobii noua fundamenta, quae ad perfectionem perfecit et cum simplicitate cordis Deo obtulit. '85 Of all the reformers, St Ethelwold did the most to exploit the cult of saints in his monasteries. Ely had, as it were, inherited the body of St ^theldreda; St Ethelwold included her feast, with full-page portrait, in his benedictional.86 He was also active in acquiring relics for those of his foundations which had none. A contemporary writes that St Ethelwold gained King Edgar's permission ' to move those bodies of saints which lay in desolate, neglected places, where formerly noble churches had stood, to the new monasteries so that they could be venerated by the faithful as was fitting'.87 He lavished relics on Thorney; 'every corner of 82
Asser's Life, cap. Ixxiv. 55. Cf. Keynes and Lapidge, Alfred the Great, 254-5. For the cult of relics in Anglo-Saxon England see Max Forster, ^ur Geschichte des Reliquienkultus in Altengland, Munich 1943. Cf. D. VV. Rollason, 'Lists of saints' resting-places in AngloSaxon England', A-SE vii (1978), 61-93 passim esp. at p. 81. 83 84 8S HE, . 30: i. 64-6. Vita S. Oswaldi, 462. Ibid. Cf. above p. 43. 86 The Benedictional of St. ^Ethelwold, facsimile ed, by G. F. Warner and H. A. Wilson, Oxford 1910, 37. 87 Liber vitae: Register and Martyrology of New Minster and Hyde Abbey, Winchester, ed. Walter de Gray Birch, London 1892, 286. Cf. Sheerin, 'Dedication of the Old Minster', 266 n. 4; and Alan Thacker, 'yEthelwold and Abingdon', in Yorke, Bishop ^Ethelwold, 59–63.
52
Legends, Traditions and History in Medieval England
the church was crammed with them'. 88 Of particular interest is St Ethelwold's promotion of the cult of St Swithun, bishop of Winchester (852 x 3-862 x 5), in the Old Minster.89 St Swithun, it was said, began to make known the presence of his body in the Old Minster in 968. Meanwhile, St Ethelwold was rebuilding the cathedral on a grand scale. On 15 July 971 he had St Swithun's body translated to a suitable shrine, and on 20 October 980 he solemnly dedicated the new cathedral, in the presence of King Ethelred and a vast concourse of people, including St Dunstan. The occasion and St Swithun's miracles were described by an eyewitness, the monk Lantfred, in the Translatio et miracula Sancti Swithuni.9Q It is not clear whether St Ethelwold in the first instance rebuilt the cathedral in order to provide a shrine church for St Swithun or whether he promoted the cult in order to increase his cathedral's prestige and help finance the rebuilding. However, the latter motives seem more likely; if they were his reasons, his choice of St Swithun could indicate an additional motive. Possibly he, in connivance with King Ethelred, intended to placate the clerks of the old dispensation, to whose class St Swithun had belonged.91 The magnificent ceremony of the dedication can be seen as symbolising the healing of the breach between the clerical party and the monastic one, a breach which had been expressed by open hostility during the antimonastic reaction following King Edgar's death. It can be said that, although the tenth-century monastic revival was inspired by influences from abroad, it marked no break with the AngloSaxon tradition. That tradition had survived into the tenth century in an attenuated form, in. the clerical communities. Benedictine monasteries replaced a few of the old communities, but the reformers, who had themselves grown up in the native tradition, did not treat the clerks as harshly as might at first appear. They preferred co-operation to confrontation; some clerks joined the new monasteries, and at least three clerical communities survived intact and prospered. The reformers turned for guidance mainly to the past, especially to the tradition's fountainhead, the Northumbrian renaissance, as revealed by Bede. Thus guided they were able to keep their movement in line with the indigenous tradition of Anglo-Saxon monasticism, at the same time supplementing it with what they learned from abroad. 88 Willelmi Malmesbiriensis monachi de gestis pontificum Anglorum libri quinque, ed. N. E. S. A. Hamilton (Rolls Series, 1870), 327. 89 Sheerin, op. cit. 264-70. Forthcoming is Michael Lapidge's important book, The Cult of St Swithun (Winchester Studies iv. 2), Oxford 1990. 90 Ed. E. P. Sauvage in AnaUcta Bollandiana iv (1885), 367-410. 91 Sheerin, op. cit. 269.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
53
II. C. IOOO-C. I IOO
St Ethelwold died in 984, St Dunstan in 988 and St Oswald in 992. The monks had already lost the patronage of a strong king with the death of Edgar in 975. Political instability had ensued, and law and order were soon further threatened by the resumption of Viking onslaughts. The extent of the injury which the Vikings inflicted on the monasteries is hard to assess, but it was greatest in East Anglia. Reputedly, they destroyed St Neots; the saint's body was moved to Crowland.92 (They also seem to have disrupted the clerical community at Beadericesworth; according to tradition the body of St Edmund was taken temporarily for safety to London.) 93 Among those killed at Assandun (1016) were Wulfsige, abbot of Ramsey,94 and a company of monks from Ely, who had gone there with their relics to pray for an English victory.95 Nevertheless, there was no breach in the Anglo-Saxon religious tradition. The monks endeavoured, with a large measure of success, to preserve and continue the work of the reformers. They were as conscious as the reformers themselves had been of the inheritance from the Anglo-Saxon past - the recent, the more remote and the very remote past - and as anxious to be guided by it. Bede remained the predominant authority. The cult of saints flourished in the late tenth and early eleventh centuries, as monastic liturgical calendars testify.96 Eadnoth, abbot of Ramsey (993-1006), inherited St Oswald's love of relics. Inspired by a vision, it was said, he discovered the (supposed) body of St Ivo at Slepe and brought it to Ramsey.97 As previously, cults were used to promote the prosperity of existing monasteries and to help establish new foundations. Eadnoth, continuing fn a modest way St Oswald's policy of expansion, founded a small monastery at Slepe, the site of St Ivo's invention.98 Perhaps he was also responsible for the translation of St Neot, whose shrine became the heart of another little monastery. This is suggested by the facts that the translation to Eynesbury apparently took place in Eadnoth's day and that St Neots, as the house was henceforth called, was endowed by Ramsey's lay patron, ^Ethelwine the ealdorman.' Mean92
Gesta pontificum, 321; Liber Eliensis, ed. E. O. Blake (Camden Society, 3rd series xcii, 1962), 102. Cf. Cyril Hart, 'The East Anglian chronicle', Journal of Medieval History vii (1981), 279; and most recently, The Anglo-Saxon Chronicles, a Collaborative Edition, ed. David Dumville and Simon Keynes, xvii = The Annals of St Neots with Vita prima Sancti Neoti, ed. David Dumville and Michael Lapidge, Cambridge-Woodbridge, 1984, p. Ixxxix. 93 Hermann, De miraculis S. Eadmundi, i. 40-6. 94 95 Chron. Ram. 118; Lib. Eli., 148. Ibid. 148. 96 See C. E. Fell, 'Edward king and martyr and the Anglo-Saxon hagiographic tradition', in Ethelred the Unready: papers from the millenary conference, ed. David Hill (British Archaeological Reports, British series lix, 1978), i ff. 97 98 Chron. Ram., 114-15; Lib. Eli., 141. Chron. Ram. 115; Lib. Eli., 141. 99 Ibid. 103-4; cf. Chron. Ram., 96. It was apparently in Eadnoth's day that the cult of St Neot first developed in the monastery at Eynesbury; see Marjorie Chibnall, 'History
54
Legends, Traditions and History in Medieval England
while, Eadnoth's family endowed both Slepe and the nunnery which Eadnoth had founded at Chatteris, making his sister abbess.100 Moreover, at this time the claims to sanctity of SS Dunstan, Ethelwold and Oswald received literary recognition. St Dunstan's Life was written by the anonymous 'B'. 101 Two of St Ethelwold's pupils, ^Elfric, monk of Cerne, previously a monk of the Old Minster, Winchester, perhaps the best known scholar of his day, and Wulfstan, reputedly precentor of the Old Minster, wrote biographies of him.102 At Ramsey an anonymous author, who has been identified as the scholar, Byrhtferth, wrote a Life of St Oswald.1®* These Lives were intended to do more than depict their subjects as saints; they are not just hagiographies but have many features of'straight' biography. They record the saints' achievements as reformers against the background of the tenth-century revival. They show particular interest in the saints' foundations and clearly intend the saints to shed reflected glory on them. The biographers were especially concerned in this respect with their own monasteries; for example, St Oswald's biographer, a monk of Ramsey, was at pains to include much information about that house.104 In varying degrees the Lives were intended to foster cults. ^Elfric and Wulfstan wrote partly to promote the cult of St Ethelwold in the Old Minster; therefore, they described the posthumous miracles which took place at St Ethelwold's shrine in the cathedral.105 Already, during the tenth-century revival itself, hagiographies had been produced to support specific cults: the monk Lantfred had contributed to St Ethelwold's promotion of the cult of St Swithun in the Old Minster, with his Translatio et miracula Sancti Swithuni; and Abbo of Fleury, while staying at Ramsey, c. 986—8, had composed the Passio Sancti Eadmundi to promote St Edmund's cult at Beadericesworth. Even earlier, in the reign of Ed red (946-55), a member of the clerical community at Ely, which preceded St Ethelwold's monastic foundation there, had written on the miracles of St Etheldreda.106 These eaVly eleventh-century biographers were reviving a wellof the priory at St Neots', PC AS lix (1966), 69; and Annals of St Neots with Vita prima, pp. Ixxxvii-xcii. It should be noted that Christopher Hohler rejects the view that St Neot was translated from Cornwall; he argues that the Cornish St Neot and the Huntingdonshire one were two separate people; see Gransden, Historical Writing, 49 n. 52. 100 See Hart, 'East Anglian chronicle', 277-9; anc* idem, 'Eadnoth, first abbot of Ramsey and the foundation of Chatteris and St Ives', PCAS Ivi-lvii (1964), 61-7. 101 Printed, and its authorship discussed, in Memorials of Saint Dunstan, pp. x-xxvi, 3-52. Cf. Brooks, Early History, 245-6. 102 ^Elfric's Vita and Wulfstan's Vita gloriosi are printed in Three Lives, 17-29 and 33-63. They are not independent authorities; see ibid. 2. 103 Printed in Raine, Historians of the Church of York, i. 399-475. The, case for Byrhtferth's authorship, based mainly on philological evidence, has been made most recently by Michael Lapidge, 'The hermeneutic style in tenth-century Anglo-Latin literature', A-SE iv (1975), 90-4; see 91 nn. 2-3 for references to the works of previous scholars who have discussed the problem of the authorship of the Vita S. Oswaldi. 104 See e.g. Vita S. Oswaldi, 429-34, 447, 468, 475. 105 106 Three Lives, 28, 62-3. Lib. Eli., p. xxxii.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
55
established, Anglo-Saxon, biographical tradition (It should, however, be remembered that, abroad, the tenth-century continental reformers were also the subjects of rather similar biographies.)107 The practice of writing the biography of a saint or other churchman shortly after his death began during the Northumbrian renaissance. The biographers of SS Ethelwold and Oswald knew their respective saints well ('B' may have been an exception in this respect), like the biographers of the Northumbrian period. Although the Lives of saints produced during the Northumbrian period can be classed as 'hagiographies', they contain much nonhagiographic material. The Lives of St Cuthbert by an anonymous monk of Lindisfarne and by Bede108 have much reliable information about St Cuthbert. Similarly, the anonymous Life of Abbot Ceolfrid™* Bede's Lives of the Abbots of Wearmouth/ jfarrowlw and the Life of Wilfrid, by Eddius Stephanus (or 'Stephanus' as it now seems we should call him),111 are all free of the miraculous element. Moreover, the Lives of St Cuthbert, the Life of Abbot Ceolfrid and the Life of Wilfrid have information respectively about the monasteries of Lindisfarne, Wearmouth/Jarrow and Wilfrid's foundations at Hexham and Ripon, while Bede's Lives of the Abbots of Wearmouth/Jarrow is virtually a history of those twin houses. Nevertheless, one purpose of some of the Northumbrian saints' Lives was to foster cults. Thus the Lives of St Cuthbert were written partly to promote his cult at Lindisfarne. An even better example is the Life of Pope Gregory the Great composed at Whitby early in the eighth century; 112 its principal intention was to promote St Gregory's cult at Whitby, which claimed to have a relic of him. It is clear, therefore, that the Northumbrian renaissance provided would-be biographers of the tenth-century reformers with a usefully varied biographical tradition. It is likely that the reformers consciously revived that tradition. The influence of Bede on ^Elfric is well known; for instance, ^Elfric drew heavily on the Ecclesiastical History for his Lives of Saints.11* Similarly, Byrhtferth was a Bedan scholar. He compiled a miscellany which strongly shows the influence of Bede. From this he derived most of the material which he used for his Enchiridion, which is 107 The continental Lives are fully described by Patrick Wormald in 'yEthelwold', passim. 108 Two Lives of St Cuthbert: A Life by an Anonymous Monk of Lindisfarne and Bede's Prose Life, ed. and trans. Bertram Colgrave, Cambridge 1940, repr. 1985; 'Bedas metrische Vita Sancti Cuthbert', ed. Werner Jaager, Palaestra cxcviii (1935). 109 no In Op. Hist., i. 388-404. ' Ibid. i. 364-87. 111 The Life of Bishop Wilfrid by Eddius Stephanus, ed. and trans. Bertram Colgrave, Cambridge 1927, repr. 1985. Cf. David Kirby, 'Bede, Eddius Stephanus and the Life of St Wilfred', EHR xcviii (1983), 101-14. 112 The Earliest Life of Gregory the Great by an Anonymous Monk of Whitby, ed. and trans. Bertram Colgrave, Lawrence, Kan. 1968. 113 See J. H. Ott, Ober die Quellen der Heiligenleben in /Elfrics Lives of Saints, i. inaugural dissertation, Halle 1892, 44-7; and Ruth Waterhouse, '/Elfric's use of discourse in some saints' lives', A-SE v (1976), 83-91. For ^Elfric's use of Bede's De temporum ratione, De temporibus and De natura rerum for his De temporibus anni see &lfric s De temporibus anm, ed. Hcinrich Hcnel (EETS, orig. ser. ccxiii, 1942), pp. liii-lvi.
56
Legends, Traditions and History in Medieval England
virtually a commentary on Bede's books on time.114 The Life of Oswald has what appears to be a borrowing from Bede's prose Life of St Cuthbert.115 The Anglo-Saxon religious tradition continued in the late tenth and early eleventh centuries despite the facts that, after the death of Edgar, the monks lacked a strong royal patron and then suffered to an indeterminate degree from Viking incursions. After the accession in 1016 of Cnut the monks once again had a staunch royal supporter and one who imposed peace on the country. Even before he became king, Cnut had connections among the monks: ^Ethelwine, abbot of Abingdon (1016—30), was his friend, and ^Elfweard, first a monk of Ramsey and then abbot of Evesham, his relative.116 The chronicles of Ramsey, Abingdon and Ely have passages eloquent in his praise.117 (Although these three chronicles were compiled after the Conquest, in Henry n's reign, they contain much pre-Conquest material.)118 The Ramsey chronicler appreciated that Cnut's merit lay, not only in his respect for the Church and his patronage of the monks, but also in his success in the secular sphere - in establishing the rule of just laws throughout his empire; in the same way his personal virtue combined piety and love of peace with martial skill. The chronicler writes: Interea Cnuto rex Christianissimus, nulli praecedecessorum suorum regum comparatione virtutum vel bellica exercitatione inferior, coepit sanctam ecclesiam enixissime venerari et religiosorum causis virorum patrocinari, eleemosynis profluere, justas leges vel novas condere vel antiquitus conditas observare. Quumque non solum Angliae sed et Daciae simul et Norguegiae principaretur, erat tamen humilitate cernuus, usus venerei parcus, alloquio dulcis, ad bona suadibilis, ad misericordiam proclivis, amatorum pacis amator fidissimus, in eos autem qui vel latrocinio vel depraedatione jura regni violassent ultor severissimus.119 Cnut, indeed, had political motives as well as religious ones for patronising the monks. No doubt he was pious and approved of monks, but there were practical reasons for winning their favour. Since he was by 114 Byrhtferth's miscellany survives only in a late eleventh-century copy, St John's College, Oxford, MS 17. It contains many of Bede's works. See P. S. Baker, 'Byrhtferth's Enchiridion and the computus in Oxford, St John's College 17', A-SE x (1982), 123-42. I am grateful to Dr Lapidge for calling my attention to this manuscript. The Enchiridion is printed in Byrhtferth's Manual, ed. J. S. Crawford (EETS clxxvii, 1929). For the use of Bede in the Manual see Crawford's footnotes and Heinrich Henel, 'Byrhtferth's Preface: the Epilogue of his Manual9, Speculum xviii (1943), 290. For the influence of Bede on Byrhtferth in general see Michael Lapidge, 'Byrhtferth of Ramsey and the early sections of the Historia regum attributed to Symeon of Durham', A-SE x (1982), esp. at pp. 120-1. 115 Vita S. Oswaldi, 448. Raine suggests, probably wrongly, that Bede's metrical Life of St Cuthbert was the source, ibid. 448 n. i. 116 Chronicon Monasterii de Abingdon, 2 vols, ed. Joseph Stevenson (Rolls Series, 1858), i. 433. Chron. Eve., 83. 111 Chron. Ram., 125-6 (extract cited below); Chron. Ab., i. 433; Lib. Eli., 152-4. Henry of Huntingdon also eulogises Cnut, Henrici Archidiaconi Huntendunensis Historia Anglorum, ed. Thomas Arnold (Rolls Series, 1879), 188-9. 118 See Gransden, Historical Writing, 273-5. ' ^nron- Ram-, 125-6.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 57 origins a Viking warlord and had seized the throne by force, he needed support; the monks were invaluable allies. Possibly Cnut saw himself, and wished to be seen, as a second King Edgar. One version of Cnut's laws states in the preliminary matter that his councillors were determined to 'zealously observe Edgar's laws'.120 In fact, although his codes draw partly on the laws of Edgar, they are more indebted to those of Ethelred. This disparity between claim and actuality suggests that Cnut wanted to be identified with the successful Edgar, not with the unfortunate Ethelred. Cnut was a generous benefactor of the monasteries. He gave lands, for example, to the Old Minster, Winchester,121 Bury St Edmunds 122 and Sherborne.123 Sometimes he helped monasteries, for instance Abingdon and Ramsey, to acquire lands. And he made gifts of relics and other treasures, for example to Abingdon,124 Evesham,125 the New Minster, Winchester,126 and to Christ Church, Canterbury. 127 According to postConquest tradition Cnut founded two monasteries, St Benet of Holme and Bury St Edmunds. However, the evidence for his foundation of these houses is by no means beyond dispute.128 The evidence that he founded a nunnery at Ramsey is more convincing. It occurs in the Ramsey chronicle; the relevant passage reads: .. Ramesensem ecclesiam, circumquaque suavem spirantem jam bonae opinionis odorem, plurimum diligens, in ipsa insula quandam aliam ecclesiam juxta primam, eiusdem formae sed quantitatis paulo dissimilis, datis de fisco regio sumptibus, in honorem Sanctae Trinitatis construi fecit. In qua quum coetum monialium aggregare decrevisset, ut, sicut in altero monasteries collegium virorum, sic in hoc quoque chorus feminarum sedulum Deo obsequium exhiberet..129 120
The so-called D version of 1018. See A. G. Kennedy, 'Cnut's law code of 1018', A-SE xi (1983), 62, 64, 72 (I owe this reference to Dr Simon Keynes). Whitelock argues that this version of Cnut's laws and also Cnut I and II were the work of Wulfstan i, bishop of Worcester 1002-16 and archbishop of York 1002-23; Dorothy Whitelock, 'Wulfstan and the laws of Cnut', EHR Ixiii (1948), 433-52. Her conclusion is strongly supported with additional evidence by Kennedy, op. cit. 57-66. Whitelock, op. cit. 442-3, also contends that Wulfstan regarded Edgar's reign as a Golden Age. Since Wulfstan wrote the codes for Cnut and was the king's friend, it can be assumed that Cnut agreed with his views. 121 Sawyer, Anglo-Saxon Charters, 292 no. 972. 122 Ibid. 293-4 no- 9&°- Only P art of this charter is apparently authentic; see Harmer, 123 Anglo-Saxon Writs, 433-4. Sawyer, op. cit. 292 no. 975. 124 125 Chron. Ab., i. 433-4. Chron. Eve., 83. 128 A picture, executed c. 1020, of Cnut and Emma placing a large gold cross on the altar of New Minster, Winchester, is BL MS Stowe 944, fo. 6; reproduced, e.g. Frank Barlow, Edward the Confessor, London 1970, repr. 1979, plate facing p. 40. For a vivid (probably idealised) description of Cnut's demonstration of piety and his generosity when he visited St Berlin's and St Omer's, at St Omer in Flanders, see Encomium Emmae reginae, ed. Alistair Campbell (Camden Society, 3rd series Ixxii, 1949), 36. His gift to the altar of each monastery was so large that it had to be brought not 'shut up in a bag 1 but 'wrapped in the folds of a cloak'. 127 Harmer, op. cit. 168 and n. i. 128 See A. Gransden, 'The legends and traditions concerning the origins of the Abbey 129 of Bury St. Edmunds', below pp. 90-104 passim. Chron. Ram., 126.
58
Legends, Traditions and History in Medieval England
Despite the Ramsey chronicle's late date (temp. Henry n), there seems no good reason for doubting its accuracy on this point. In this instance the chronicler had no obvious motive for fabrication. On the contrary, since his subsequent narrative shows that he was strongly opposed to any i lea of a nunnery next to a monastery, he might surely have been tempted to suppress such a piece of information. He follows the notice of the foundation with the statement that Cnut never fulfilled his plan; this was 4 by divine providence', since 'it was not unknown how useful, or rather damaging, the close proximity of the sexes can be'. He observes that the crypt, which had lain underneath the high altar of the 'great church', could still be seen in his own day in the monks' cemetery.130 We only have the chronicler's word for it that Cnut never brought his scheme to completion. Perhaps the chronicler was indulging in wishful thinking, unhampered by certain knowledge. By his time the climate of opinion tended to be hostile to * double monasteries' - those comprising two houses, one for each sex - of the Anglo-Saxon type. It is not impossible that a nunnery did exist for a while beside the monastery at Ramsey and that it subsequently languished to extinction from lack of support, leaving no trace except this notice in the Ramsey chronicle.131 Although Edward the Confessor is credited with no new monastic foundation, he restored and lavishly endowed Westminster Abbey,132 besides being a generous benefactor of other houses.133 In their patronage of the monks both Cnut and Edward the Confessor were supported by their queens, Emma and Edith, who were worthy successors of Edgar's queen, ^Elfthryth, as patronesses of the religious. The Regularis concordia had made the queen responsible for the nuns; 134 Edith built a new church for those at Wrilton.135 Emma did not patronise the nuns in particular (unless her influence was responsible for Cnut's foundation of the nunnery at Ramsey), but she was generous to the monks. Goscelin, in his Life of St Wulsin, which he wrote towards the end of the eleventh century, relates 130
'Ex providentia tamen Dei, quam quid utilitatis aut damni ex vicinitate sexuum amborum provenire posset non latebat, propositum non implevit. Porro crypta, quae subtus majus ipsius ecclesiae altare fabricata fuerat, ejusdem aedificii testis et index, in coemiterio nostro hodieque indemnis perdurat': Chron. Ram., 126. 131 Dr Sally Thompson, whose book on post-Conquest nuns and nunneries is to be published by Oxford University Press, informs me that a number of post-Conquest nunneries disappeared leaving hardly any evidence of their previous existence, and that there must have been others which left no trace at all. Moreover, in the twelfth century it was not unusual for a monastery to act as protector to a group of nuns within its precincts (for example Christine of Markyate and her followers were sheltered by Geoffrey, abbot of St Albans). Later the tendency was for a nunnery under monastic protection to be founded at a distance from the monastery (an example is the nunnery of Sopcwell, which was dependent on St Albans). 132 The Life of King Edward who Rests at Westminster, ed. and trans. Frank Barlow, London 1962, 44-6. 133 E.g. Bury St Edmunds, see Harmer, Anglo-Saxon Writs, 148-9, and Abingdon. See below p. 187 and n. 137. 134 Reg. con. vii. 70. 135 Life of King Edward, 46-9.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 59 that Emma, during a visit with Cnut to Sherborne Abbey, was so shocked at the state of the roof of the church that she paid for its repair.136 Queen Edith was also a benefactress of the monks. The Abingdon chronicler records her distress, when she visited Abingdon with Edward, to find that the monks, because of their poverty, could give the children of the cloisters nothing better to eat than bread. Therefore she persuaded Edward to grant jointly with her the vill of Lewknor as an endowment to supply the boys with proper meals.137 Nevertheless, the position of the monks in religious life and in politics changed in the course of the eleventh century. No spiritual leaders of calibre comparable with that of Dunstan, Ethelwold and Oswald appeared. There were very few new foundations of importance. The most notable was Bury St Edmunds; some time fairly early in the century the clerical community serving St Edmund's shrine at Beadericesworth was replaced by a Benedictine one.138 The tradition of the foundation of monasteries by the lay nobility just survived until the middle of the century. In about 1002 Wulfric Spott founded Burton, and in 1045 Leofric, earl of Mercia, and his wife Godgifu founded Coventry abbey.139 Meanwhile, the monks' influence on the central government was on the wane. They lost their virtual monopoly of the episcopal bench.140 One reason for this was that Edward the Confessor, who had spent the first thirty years or so of his life in Normandy and France, tended to give bishoprics to foreign clerics. Some of these men were trained in the Lotharingian school, whose reforms included the establishing of disciplined clerical communities in episcopal sees. Such were Edward's priests, Hermann, Leofric and Giso, to whom Edward gave Ramsbury (1045), Crediton (1046) and Wells (1060) respectively. However, the main reason why the monks lost their monopoly was the development of royal government. The king needed a permanent staff, and it was more convenient to use the secular priests of his own chapel rather than monks who belonged to their individual monasteries. Since the king had to reward the chaplains for their services, he gave some of them bishoprics. This development was already apparent in Cnut's reign. The erosion of the monks' importance at the centre of national power encouraged the growth of localism, a trend not countered by any contrary force. The Rule of Benedict did not legislate for a unified order but for virtually autonomous monasteries; it provided no links between houses. It 138 C. H.Taylor, k The Life of St. Wulfsin of Sherborne by GoscelirT, Rev. Ben. Ixix (1959)> 81 (caP- xiii). Emma also bequeathed a manor, Kirby Cane, to the Abbey of Bury St Edmunds, Harmer, Anglo-Saxon Writs, 148-9, 159 nos 16, 17. 137 Chron. Ab. i. 460—1 ; ii. 283. See Victoria County History, Oxfordshire, 11 vols, 1939—83 (in progress), viii. 100. 138 This is the argument I put forward, in preference to the traditional ascription of the foundation of the Benedictine monastery to Cnut, in the article cited above, n. 128. 139 fYank Barlow, The English Church 1000-1066, 2nd cdn, London 1979, 316. 140 See R. R. Darlington, 'Ecclesiastical reform in the late Old English period', EHR li (!93 6 )> 395-6-
60
Legends, Traditions and History in Medieval England
is true that, in England during the tenth-century revival, monasteries founded by one or other of the spiritual leaders had been in close touch with each other. But when the founders were dead, their 'families' of houses tended to drift apart. Although Cnut and Edward the Confessor were patrons of the monasteries, neither attempted to impose uniformity on them by holding a synod comparable to Edgar's synod of Winchester. Centrifugal tendencies had appeared early in the revival itself; otherwise, the Regularis concordia, which was a product of a desire to impose greater uniformity, would not have been necessary. Localism is a marked feature of the Life of St Oswald (c. 1000); it is evident in the Life's concern for, and extensive information about, the early history of Ramsey Abbey.141 The cult of saints encouraged local attachments; a monastery's patron saint became, as it were, its persona, and its relics in general added to its individual reputation. The popularity of cults suffered no abatement in the course of the eleventh century. Cnut himself was an enthusiast.142 (As has been seen, Ramsey under Abbot Eadnoth played an important part in the dissemination of particular cults.) Nor did Eadnoth's endeavours slacken after he became bishop of Dorchester (1007 x 9-1 o 16); his burial of the body of St ^Elphege in London marked the start of the cult of that saint.143 Another Ramsey monk, ^Elfweard, on becoming abbot of Evesham (c. 1014-44), purchased relics of St Odulf for his new home.144 In this way the prestige of individual monasteries grew. A monastery's prestige attracted gifts partly in the form of landed property; and an increase in monastic estates led in its turn to an increase in prestige. Indeed, during the period between the tenth-century revival and the Norman Conquest endowment of the Anglo-Saxon foundations was more lavish than at any other time in the history of medieval England. In regard to monasteries in the east Midlands in general, and to Thorney and Crowland in particular, a recent scholar has w r ritten: The ultimate wealth and standing of any house was largely determined by its date of foundation. Crowland and Thorney, like neighbouring Ramsey, Ely and Peterborough, were among the richest because they were established early and enjoyed the fruits of Anglo-Saxon patronage. It was in the pre-Conquest period that they achieved their enduring control over land in the east midlands. Never again was so much property to pass into the hands of the Church.145 The growth of monastic estates was not only owing to the patronage of Anglo-Saxon kings, ecclesiastics, nobles and thegns. It was also indebted to the ability of a number of outstanding abbots. For example, ^Ethelwig, abbot of Evesham (1058-77), was spectacularly successful in increasing 141
Vita S. Oswaldi, 429-34, 447, 463-8, 475. For Cnut's gifts of relics to monasteries see above p. 185 and nn. 125, 127. Cf. Chron. 143 Ab. i. 433, and Chron. Ram., 127. Chron. Ram., 115. 144 Chron. Eve., 83. 145 Sandra Raban, The Estates of Thorney and Crowland, Cambridge 1977, 88. Cf. below n. 147. 142
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 51 146 Evesham's landed property. The result of the combination of the generosity of benefactors and the business acumen of abbots was that the holdings of at least a number of pre-Conquest foundations reached their greatest extent, despite fluctuations caused by Viking invasions, the rapacity of neighbours and political instability.147 Moreover, to this period belongs the earliest evidence of the elaborate system of estate management which was to become common after the Conquest. The first surviving list of food-farms was drawn up for the Ely estates by Leofsige, abbot of Ely from 1029 to IO 35- The system of food-farms was in fact well established and quite widespread before the Conquest; it certainly existed at Westminster, Abingdon, Bury St Edmunds, Peterborough and Ramsey, as well as at Ely.148 At the same time the separation of the bishop's property from that of the monks was taking place in some cathedral monasteries. The similar division of property between abbot and convent was also probably evolving in the greater abbeys.149 The power of individual houses was increased by the grant of papal and royal privileges. It is impossible to be certain about the exact nature of these early privileges because the evidence tends to be scanty and its meaning obscure, and some of it is of dubious authenticity.150 Already 146 Chron. Eve., 88-90, 94-6. For JLthel wig's career after the Norman Conquest see below p. 193. 147 For the growth of the holdings of Thorney and Crowland in the late Anglo-Saxon period see Raban, op. cit. 6-29. Of Ramsey Professor Raftis writes, 'The long list of properties pertaining to Ramsey Abbey enumerated in [Domesday Book] is a fitting epitaph to the first century of growth.. This Domesday map marked the end of an era of geographical expansion. It was the substantially complete ground plan upon which may be traced movements in agrarian history for the next four and one half centuries so that the list of properties compiled by Cromwell's inquisitors tallies markedly with that of his eleventh-century predecessor': J. A. Raftis, The Estates of Ramsey Abbey, Toronto 1957, 21. Dr Smith wrote of Christ Church, Canterbury, 'It is quite clear that the cathedral priory was primarily indebted to Saxon kings, nobles and thegns, for its vast endowment': R. A. L. Smith, Canterbury Cathedral Priory, Cambridge 1943, 9. The lands of Ely suffered considerable fluctuation during the period from the tenth-century monastic revival until the Conquest. They reached their greatest extent in the first decades of the eleventh century; see Edward Miller, The Abbey and Bishopric of Ely, Cambridge 1951, 16-25. For the formation of the Peterborough estates 966-1066 see Edmund King, Peterborough Abbey 1086-1310, Cambridge 1973, 6-11. 148 See Miller, op. cit. 38; Raftis, op. cit. 34-5 and n. 42; Reginald Lennard, Rural England 1086-1135, Oxford 1959, 130-1 and n. i, 132-3; and (with further references) Margaret Howell, 'Abbatial vacancies and the divided mensa in medieval England', this 149 JOURNAL xxxiii (1982), 174 and n. 5. Ibid. 175-6 and nn. 150 For the question of monastic exemption and protection in England and on the Continent, from the sixth to tenth centuries, see with further references, Wormald, '^EthelwokT, 21-2 and nn. 33-4, 23-4 and nn. 42-3, 34 and nn. 88, 89; and, for the early period, H. H. Anton, Studienzuden Klosterprivilegien derPdpsteim Frilhen Mittelalter, Berlin-New York 1975. I am indebted to Dr Rosamund McKitterick and Mr Wormald for help on this subject. Knowies's conclusions have, in fact, been more or less corroborated by later research. He believed that, in practice, a number of English monasteries before 1066 were free from subjection to the diocesan, though their de facto 'exemption' was rarely put to the test, M. D. Knowles, 'Essays in monastic history IV - the growth of exemption',
62
Legends, Traditions and History in Medieval England
during the Northumbrian renaissance popes had granted privileges which gave virtual exemption from episcopal and royal authority. Benedict Biscop had obtained a privilege from Pope Agatho for Wearmouth/Jarrow ('pro tuitione sui monasterii'), stipulating that it should be free from outside interference ( c ab omni prorsus extrinseca irruptione tutum perpetuo reddit ac liberum'). Pope Sergius confirmed this privilege at the request of Biscop's successor, Abbot Ceolfrid.151 During the .tenth-century revival the purpose of the privileges seems to have been to protect monasteries from oppression by the diocesan bishop (but probably not to prevent just episcopal supervision) and to give various degrees of freedom from temporal burdens. John xv apparently granted a privilege to Glastonbury, giving the monks the right to elect their own abbot and hold the Isle free from any outside interference.152 Ramsey also claimed to have obtained a privilege from Pope John.153 Moreover, during the revival, and in the eleventh century, kings alienated regalian rights to some of the greater abbeys. King Edgar granted extensive rights of jurisdiction to Worcester Cathedral and Ely.154 Later such grants became more specific. Chut gave royal rights of jurisdiction to Christ Church, Canterbury, 155 and probably to the Old Minster, Winchester.156 Edward the Confessor was particularly generous to Bury St Edmunds, granting it the jurisdictional area of the eight-and-a-half hundreds, exemption from taxation and also the right to mint coins.157 Although there was great variation between monasteries, many flourished in the half century or so'before the Norman Conquest, growing in prestige, wealth and power. The historian of Peterborough, then known as the 'Golden City', could write of Abbot Leofric (iO52-autumn 1066): 'He did much for the benefit of the monastery of Peterborough, Downside Review xxxi (1932), 211, 213, 225—6, 396, 401, 420—1. He concludes that 'it has become quite clear that the origins of exemption must be sought long before the Conquest, and that the changes within our period are precisions rather than developments': ibid. 423-
151 Bede's Lives of the Abbots, vi, xv, in Op, Hist. i. 369, 380. Noticed by: Wilhelm Schwarz, 'Jurisdictio und Condicio. Eine Untersuchung zu Privilegia libertatis der Kloster', ^eitschrift der Savigny-Stiftung fiir Rechtsgeschichte. Kanonistische Abteilung xlv Ixxvi (1959), 69-70; Heinrich Appelt, 'Die Anfange des papstlichen Schutzes', Mitteilungen des Instituts fur Osterreichische Geschichtsforschung Ixii (1954), 106—7; Anton, op. cit. 62, 65. 152 The Early History of Glastonbury.. William of Malmesburfs De antiquitate Glastonie ecclesie, ed. and trans. John Scott, Woodbridge 1981, 129, 204. Noticed by Willy Szaivert, 'Die Entstchung und Entwicklung der Klostercxcmtion', Mitteilungen des Instituts fur 153 Osterreichische Geschichtsforschung lix (1951), 295. Chron. Ram., 48. 154 Unfortunately, certainty about the nature of the liberty of Oswaldslaw is impossible because our knowledge is dependent' on charters of dubious authenticity, e.g. the Altitonantis charter, John, Land Tenure, 8off. and above n. 54. For Ely sec Miller, Ely, 25-35. 155 Sawyer, Anglo-Saxon Charters, 295 no. 985. Gf. Harmer, Anglo-Saxon Writs, 79, 168-71, 181-2 no. 26; and most recently and especially Brooks, Early History, 288-90. 156 Sawyer, op. cit. 292 no. 976. Cf. Harmer, op. cit. 382, 397-8 no. 109. 157 Ibid. 145-8 passim, 151, 156-65 passim. Cf. Gransden, 'Legends and traditions', below p. 92.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
63
with gold and silver vestments and land, more indeed than any before or after him.'158 Corporate pride served as a spur to the development of localism; monks were intent upon the interests, whether spiritual or temporal, of their own houses. Undoubtedly, this inward-looking mentality had some bad results. It could lead to corporate selfishness; one consequence of this could be unseemly quarrels between monasteries over land, relics and the like.159 Nevertheless, the evils of localism can easily be exaggerated. Despite the existence in some monasteries of distinctive customs (for example, one which came perilously near to property owning by individual monks), 160 there is no evidence that, in general, the Rule of Benedict was flouted. ^Ethelwig, abbot of Evesham (1058-77), held maundies and was a generous alms-giver in accordance with the Rule.161 There is even evidence that, at least at Christ Church, Canterbury, the Regularis concordia was not forgotten.162 (As will be seen, the post-Conquest strictures on the Christ Church monks were ill-founded.) Moreover, on occasion monks resisted royal control. In 1052 those of Christ Church attempted, but failed, to elect one of their number to be head of their house - which meant, of course that he would also be archbishop.163 In the same year the monks of Abingdon successfully insisted on their right to elect their abbot.164 Within the monasteries monks continued and developed the AngloSaxon cultural tradition, which had been revived during the tenthcentury revival. Scholars now recognise the monks' achievements in many fields in the century prior to the Conquest. Evesham was especially famous for goldsmiths' work and sculpture under Abbot Mannig (1044-58).165 The art of book illustration continued to thrive in a number of centres; the tenth-century technique of line-drawing survived, though its representational style was modified by Viking influence and by the infiltration of Romanesque art from France.166 Vernacular literature, a tradition rooted in eighth-century Northumbria and promoted by King Alfred, also flourished.167 The Anglo-Saxon Chronicle, which is disappointingly brief for the tenth century, has many detailed annals for the eleventh century in excellent prose, some of impressive power.168 iss Whitelock, Anglo-Saxon Chronicle, 142 (s.a. 1066). Cf. Darlington, 'Ecclesiastical 159 reform', 402. See e.g. Chron. Ram., 118-19, 127-8, 166; Lib. Eli., 135-6. 160 See Chron. Ab., i. 477. Cf. Knowles, Monastic Order, 80-1 and n. 4. 161 Chron. Eve., 90—3. Cf. Sancti Benedicti Regula monachorum, cap. liii. 162 The only two surviving complete texts of the Regularis concordia, one of them probably post-Conquest, are from Christ Church; see Symons's edn, pp. liii-lix. For continued interest in it at Christ Church after the Conquest, see Southern, St Anselm, 245-8 163 passim. Life of King Edward, 18-19. 164 l65 Chron. Ab., i. 463-4. Chron. Eve., 86-7. Cf. Barlow, English Church, 336. 166 Francis Wormald, English Drawings of the Tenth and Eleventh Centuries, London 1952,
49-53167
See Peter Clemoes, 'Late Old English literature', in Parsons, Tenth-century Studies, 103-14, 230-3 nn. 168 See Cecily Clark, 'The narrative mode of the Anglo-Saxon Chronicle before the Conquest', in Clemoes and Hughes, England before the Conquest, 230-3.
64
Legends, Traditions and History in Medieval England
Particularly remarkable is a chronicle from 983 to 1044 composed retrospectively in the early 10408 at Abingdon; a copy soon went to St Augustine's, Canterbury, where it was incorporated into the Anglo-Saxon Chronicle. Its unusual feature is its strong political bias (in favour of the house of Godwin), which is in marked contrast to the non-committal tone of most annalistic writing. 169 Pride in the possession of relics, themselves a source of prestige, was expressed in, and fostered by, the writing of hagiographies. The revival in the late tenth and early eleventh centuries of the Northumbrian tradition of sacred biography was sustained later in the eleventh century. A monk of Ramsey, probably Byrhtferth, supplied Evesham with a Life of its patron saint, Ecgwin.170 (Contact between Ramsey and Evesham must have been encouraged by the succession c. 1014 of ^Elfweard, a Ramsey monk, to the abbacy of Evesham.) It was probably later, in the mideleventh century, that a Life of St Neot was written for St Neot's.171 Hagiography, however, formed only part of the Anglo-Saxon tradition of sacred biography; a man did not need to be a saint to merit biographical attention. Although there is no evidence that complete biographies of churchmen other than saints were composed in the late Anglo-Saxon period, it is likely that biographical material highlighting particular activities of some such men was produced. It apparently concerned the abbots and benefactors of the specific monasteries where the authors wrote. An author's intention would have been partly to commemorate the individuals concerned but also, by recording in chronological sequence their associations with, and contributions to, the welfare of the monastery for which he wrote, to compose a virtual ' house history'. The prototype of 'house histories' structured round abbatial biographies was of course Bede's Lives of the Abbots of Wearmouth/ Jarrow. Hitherto, scholars have neglected the possible existence of this type of literature in the late Anglo-Saxon period. The evidence for it is in the preConquest sections of some post-Conquest chronicles. Just occasionally a chronicle refers to what seems to have been an early narrative source of this kind which is otherwise unknown. But the identification of a passage borrowed from a pre-Conquest abbatial history is nearly always suggested by another type of evidence: a passage may stand out from the adjoining narrative because of its precise and vivid detail and the sense of 169
See David Dumville, 'Some aspects of annalistic writing at Canterbury in the eleventh and early twelfth centuries', Peritia ii (1983), 26-31, 38; Simon Keynes, 'The declining reputation of King ^Ethelred the Unready', in Hill, Ethelred the Unready, 227-53170 See Lapidge, 'Hermeneutic style', 90—4; and idem, 'Byrhtferth and the Vita S. Ecgwini Medieval Studies xli (1979), 331-53. For the possibility that Byrhtferth also wrote for Ramsey Passions of SS Ethelred and Ethelbert, whose relics vEthelwine the ealdorman gave to the abbey, Chron. Ram., 55, see idem, 'Byrhtferth of Ramsey', 119-20. 171 See Annals of St Neots with Vita prima, pp. xciv-xcvi. Cf. Hart, 'East Anglian chronicle', 277-9; and idem, 'Eadnoth, first abbot of Ramsey and the foundation of Chatteris and St Ives', PCAS Ivi-lvii (1964), 61-7.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 65 immediacy it conveys. Such a passage could, of course, derive from some lost work written shortly after, and not before, the Conquest, based on oral information and the author's own observation. However, the impression that some post-Conquest chronicles did use pre-Conquest abbatial histories is deepened by comparison with other post-Conquest writings. For instance, Hermann's De miraculis Sancti Eadmundi, written by a monk of Bury St Edmunds c. 1100, shows the haziest knowledge of the abbots who preceded Baldwin (1065-98); this suggests that Bury had no pre-Conquest biographies of its abbots.172 Even if we accept that a few monasteries did produce abbatial histories in the late Anglo-Saxon period, any such text can only be reconstructed very tentatively, since it could have been revised and interpolated at any time before the post-Conquest chronicler borrowed from it. The chronicle of Abingdon, compiled in Henry n's reign and covering the period from the abbey's foundation to 1154, has considerable detail about the abbots from Siward (1030-44) to Ordric (io52-January io66).173 For instance, it has a graphic account of how the disreputable Abbot Spearhafoc (c. 1047-51), a goldsmith, packed the gold and jewels, which Edward the Confessor had given him to make a crown, into 'various receptacles' and absconded, never to be seen again.174 The Abingdon chronicler certainly used the evidence of charters, many of which he copied in full.175 He may also have drawn on the oral traditions of his house. But the fact that he was writing a century after the period in question, when memories were no longer fresh, combined with the precise detail and vivid touches in some of his narrative, strongly suggest that, in addition, he used some now lost abbatial history, possibly a pre-Conquest one. At Evesham a Life of Abbot ^Ethelwig was written shortly after his death (1077). This Life no longer survives as an independent work, but it was incorporated in the early thirteenth-century chronicle of Evesham which is still extant.176 The latter also contains well-informed accounts of ^Ethelwig's predecessors, ^Elfweard (c. 1014-44) and Mannig (1044-58).177 The section on Mannig is particularly colourful and detailed. The biographer of ^Ethelwig testified to the truth of his narrative about ^Ethelwig and 'the others' ('tarn de illo [i.e. ^Ethelwig] quam de aliis'). His authorities were, he states, ancient charters, oral information and his own observation ('Partim namque in antiquis cards huius loci reperimus, partim a fidelissimis viris audivimus, partim nos ipsi oculis nostris perspeximus'). 178 However, the words 'de aliis' are imprecise and need not refer to ' the other [abbots]'; they could mean ' other things' or 'other people' mentioned in the ^Ethelwig narrative. Therefore, the 172
173 Cf. Gransden, 'Baldwin', 65. Chron. Ab. i. 443-5, 451-2, 461-4. 175 See e.g. ibid. i. 434-42, 446-50, 452-7. Ibid. i. 462-3. 176 Chron. Eve., 87-96. Cf. for this Life R. R. Darlington, 'Aethelwig, abbot of 177 Evesham', EHR xlvii (1938), 1-22, 177-98. Chron. Eve., 81-7. 178 Chron. Eve., 94. 174
66
Legends, Traditions and History in Medieval England
possibility remains that the chronicler used some pre-Conquest abbatial history for ^Elfweard and Mannig. The Ramsey chronicle, compiled in Henry n's reign, has a much fuller pre-Conquest section than either the Abingdon chronicle or the Evesham one. This section includes much biographical material, which is interspersed with other types of entry (notably copies of charters). The description of the abbey's foundation and its early years centres on the joint roles of St Oswald and Earl ^Ethelwine.179 The chronicle then has entries, some substantial, about the abbots from Eadnoth (993-1006) to Athelstan (1020-43).18° Most remarkable are a number of passages, some quite long and amounting in all to about twenty-five printed pages, about Athene, bishop of Dorchester (1016-34), who had been educated at Ramsey and was buried there. The great detail of these passages and their lively tone suggest that the chronicler copied them from some now lost Life of ^Etheric. Whether this is so or not, they are now an integral part of the Ramsey chronicle (they overlap with the passages concerning individual abbots of Ramsey, which will be discussed below). The first passage about .Etheric relates how he, when a scholar of Ramsey, climbed with three fellow pupils among the great bells of the church and set them ringing discordantly, so that one cracked.181 The abbot forgave the boys, despite the monks' anger with them, partly out of kindness, but also because he believed that the boys, all of noble birth, would amply recompense the abbey later in life. The next passage182 begins with a notice of ^Etheric's succession to the bishopric of Dorchester. It then relates how he visited Ramsey to hear the complaints of Abbot Wythman (1016-20), a German by birth, against his insubordinate monks who were exasperated by the severity of his rule. ^Etheric, perhaps moved by loyalty to previous companions, decided in the monks' favour and reprimanded Wythman for his unreasonable behaviour. Also relating to ^Etheric is the passage, quoted above, recording Cnut's foundation of a nunnery at Ramsey; Cnut took this action at ^Etheric's urging ('cuius hortatu'). 183 The same is true of the notice that the body of St Felix was translated from Soham to Ramsey; this was done with Cnut's support, again at ^Etheric's instance ('Cnutone rege precibus ^Etherici episcopi favente'). 184 The chronicle also notes the death of jEtheric and his burial at Ramsey, referring to him in eulogistic terms.185 Before this final entry about Athene are five long passages which form a highly distinctive part of the chronicle. They describe in vivid detail how ^Etheric acquired estates for Ramsey.186 He purchased or otherwise obtained them, not necessarily by scrupulous means, but always with Cnut's help, from local Scandinavian landowners. An example is the story 179 181 183 185 186
18 Chron. Ram., 29-45, 85-108. ° Ibid. 109-10, 112-19, 121-5, 127-8, 155. 182 Ibid. 112-14. Ibid. 120-6. 184 Ibid. 126. Ibid. 127-8. Ibid. 147. Ibid. 128-44. This section of the Ramsey chronicle is not cited in Raftis, Estates.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
67
of his acquisition of Ellington (Hunts). He was accompanying Cnut as he travelled through his kingdom, but on this occasion there was not room for him to stay in the same vill with the king. He, therefore, together with four royal clerks, lodged with a Dane, who gave them generous hospitality. The Dane himself became very drunk and entered into a wager with Athene; he undertook to sell him his vill if ^Etheric could raise the (nominal) purchase money of fifty marks by morning. As soon as his host fell asleep in drunken torpor, ^Etheric sent one of the clerks riding post haste to Cnut. The clerk, who interrupted Cnut at a game of chess, obtained a loan from the king which enabled ^Etheric to buy the estate.187 Another story, which tells how ^Etheric acquired Therfield (Herts) for Ramsey, reveals that ^Etheric employed a local agent to purchase land as opportunity arose.188 Besides illustrating how monastic landholdings were built up, these stories throw light on relations between the Anglo-Saxons and Scandinavian settlers. ^Etheric acquired two vills, Therfield and Shillington (Beds), because the Danish landowners fled through fear of the English.189 The Ramsey chronicle attributes Athene's generosity to Ramsey to his wish to recompense the abbey for cracking the bell.190 In addition, the chronicle has information about the careers and achievements of the three other bell-crackers: Athelstan, who became abbot of Ramsey;191 Eadnoth, who, after being a monk of Ramsey, succeeded ^Etheric as bishop of Dorchester;192 and Oswald, St Oswald's nephew, who became a monk of Ramsey and a noted writer and poet.193 Bishop ^Ednoth was, like ^Etheric, a generous benefactor of his old home,194 and Oswald, together with Abbot Wythman, persuaded Edward the Confessor to give lands and rights, which are specified in the chronicle, to Ramsey.195 The question arises as to the possible nature of the early biographical sources of the pre-Conquest section of the Ramsey chronicle. (The chronicler's use of charter evidence and of oral information is well known, although the extent to which he used the latter is impossible to determine.)196 Information from some of these literary sources may have reached him indirectly through an intermediate source or sources. One intermediate source can be tentatively identified. An appreciable portion of the pre-Conquest section is in flowery Latin, characterised by the use of unusual words and verbose sentences, biblical quotations and direct speech. If passages in this prose style are isolated, they can be roughly 187
188 Chron. Ram., 135-40. Ibid. 142. Ibid. 140, 143. 190 Ibid. 128-9, !46. Similarly, the chronicle attributes the gifts of another of the bellcrackers, Bishop Eadnoth, to penitence for his part in the incident. 191 192 Ibid. 124, 127, 155. Ibid. 148, 159. 193 Ibid. 159-60. For Oswald see Lapidge, 'Hermeneutic style', 94-5. 194 195 Chron. Ram., 159. Ibid. 159-60. 196 In my emphasis below on the likelihood that the Ramsey chronicle used preConquest written narrative sources, I revise the view expressed in my Historical Writing, 275, that it leaned heavily on oral evidence. 189
68
Legends, Traditions and History in Medieval England
divided into two groups. The first mainly concerns the close co-operation of St Oswald and Earl ^Ethelwine in the foundation and early years of Ramsey. The second group includes the description of the bell-cracking incident and the culprits' later careers. Both groups have some thematic unity. The theme of the first group is the close friendship and co-operation of St Oswald and ^Ethelwine, who are jointly credited with Ramsey's foundation and successful early years; their partnership is given visual expression by the image of the twin towers of the abbey church.197 The theme of the second group is provided by the bell-cracking incident itself, which forms the starting point for the accounts of the benefits done to the abbey by the four boys later in life. The ornate prose of the passages is reminiscent of the literary style of the forged charter, purportedly of King Edgar, granting privileges to the abbey. A copy of the charter occurs in the chronicle, introduced and followed by passages in the same style.198 It seems likely, therefore, that some of the chronicle's pre-Conquest sources suffered embellishment at about the time when the Edgar charter was concocted (perhaps even by the forger himself). Pierre Chaplais attributes the forgery to the Westminster monk, Osbert of Clare, who died c. 1170.199 If Osbert did forge the Edgar charter, the dates of his period of literary activity make him a possible candidate also for the authorship of some of the flowery passages in the chronicle; the latter has an account of the rule of Abbot Walter (i 135-60) which is partly in flowery prose.200 However, it would be extremely rash to argue that all instances of this style in the chronicle must be post-Conquest, rasher still to attribute them, even tentatively, to one author. Ornate Latin was characteristic of prosewriting at Ramsey (and in many other monasteries) in the late AngloSaxon period and was fairly widespread after the Conquest and throughout the Middle Ages. Most of the chronicle's flowery passages concerning St Oswald and ^Ethelwine were elaborations of information found in the first Life of St Oswald.2®1 Other early written sources probably lie behind much of the rest of the narrative in the pre-Conquest section, both flowery and that in simple prose. The wealth of apparently contemporary detail leads to this conclusion. The latter is, moreover, supported by explicit references. Thus, the chronicler admits that he does not know why Abbot ^Elfweard 197 The towers of the late tenth-century church at Ramsey were obviously impressive; they are described in Chron. Ram., 41. The main tower of the first stone church cracked and had to be rebuilt, ibid. 85-8. The image of St Oswald and Earl ^Ethelwine as the two towers originated in the early Life of St Oswald, i. 469; it states that a monk had a vision of the fall of the two towers, presaging the deaths of St Oswald and Earl ^Ethelwine. This idea is elaborated in Chron. Ram., 102-3 passim. 198 Ibid. 68-70 and n. 2 For the charter itself see ibid. 181-9; and Sawyer, Anglo-Saxon Charters, 254-5 no. 798. 199 pierre Chaplais, 'The original charters of Herbert and Gervase abbots of Westminster (1121-1157)', in A Medieval Miscellany for Doris Mary Stenton (Pipe Roll Society Ixxvi, NS xxxvi, 1960), 92. 200 201 Chron. Ram., 325-36. See ibid. p. xxiv.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
69
was afflicted with illness (he suffered from leprosy), because ' the chronicles are silent' ('Cuius valetudinis causam quamlibet chronica taceant'). 202 The nature of this reference suggests that 'the chronicles' were domestic histories of the abbey. Again, to substantiate his statement that St Oswald acted as abbot of Ramsey during his lifetime, the chronicler appeals to a document containing monastic professions made in St Oswald's presence ('schedae quaedam in ecclesia nostra hodieque reperiuntur professiones virorum temporis illius in presentia ejus celebratas continentes'). 203 Whether this document included any biographical details about the monks is impossible to say. However, another of the chronicler's references is perhaps more relevant to the investigation of his sources. For his account of Athelstan the 'Half-King', father of Ramsey's lay patron, ^Ethelwine, and his family, he cites the authority of a certain ancient document containing the names of kings and other benefactors of the .abbey, noting their gifts: Fidejubent sermonis nostri veritatem quaedam in archivis ecclesiae nostrae repertae vetustissimae scedulae, eorundem regum nomina et quibusdam personis factas ab eisdem terrarum donationes continentes; quae donationes etiam ab ipsis personis postmodum ecclesiae nostrae in perpetuam eleemosynam cum earundem scedularum munimento sunt collatae, in quarum singulis vir ille inter alios nobiles earundem donationum testis invenitur ascriptus.204 Since Athelstan was not (so. far as is known) himself a benefactor of Ramsey, he must have been included in the document because he was ^Ethelwine's father. A very rough parallel would be the inclusion of a Life of Oda, St Oswald's uncle, in the Life of St Oswald.™ His inclusion suggests that the document was quite a substantial work. Possibly the passages in the chronicle about ^Etheric, including the narratives describing his acquisition of landholdings for Ramsey, were copied from it. If, indeed, it did contain this amount of detail, it would be comparable to the Libellus quorundam insignium operum beati &thelwoldi episcopi composed to record St Ethelwold's acquisition of property for his monastery at Ely.206 (A similar, but much shorter and more concise work was written to record St Ethelwold's gifts to another of his foundations, Peterborough.)207 A distinctive feature of the Libellus is its use of charters containing lively narratives of how (purportedly, at least) St Ethelwold acquired the land.208 The narratives in the Ramsey chronicle describing 202
203 Chron. Ram., 157. Ibid. 41-2. Ibid. 12-13. Cf. with references to the Ramsey chronicle, including this passage, Cyril Hart, 'Athelstan 'Half King' and his family', A-SE ii (1973), 115-44. 205 Vita S. Oswaldi, 401-10. Cf. Armitage Robinson, 'St Oswald', 38-42. 206 The Libellus quorundam insignium operum beati /Ethelwoldi episcopi, which was the source for much of book ii of the Liber Eliensis, was a translation of an OE work composed in the late tenth century, Lib. Eli., pp. ix-x, li. (A new edn, with trans, and discussion, is being prepared by Simon Keynes and Alan Kennedy.) For two documents from Rochester, which contain graphic narrative similar to that found in the documents copied in the Ely Libellus, see Robertson, Anglo-Saxon Charters, 84-7, 122-5. 207 208 Printed ibid. 72-5. Cf. ibid. 75-83. Lib. Eli., pp. ix-x. 204
70
Legends, Traditions and History in Medieval England
^Etheric's land acquisition could derive ultimately from similar charters. Some such charter could also underlie the account in the Abingdon chronicle of how Abingdon obtained Lewknor (Oxon) from Queen Edith and Edward the Confessor.209 It has been argued above that there is a distinct possibility that at least some of the biographical material in the pre-Conquest sections of the chronicles of Abingdon, Evesham and Ramsey was written before the Conquest. If so, it indicates considerable historiographical activity on a local level. But even if some, or all, of this material is post-Conquest, the information it contains is of value. The written record, as such, is no more reliable than near-contemporary, oral testimony, while personal observation carries more weight than either. The details preserved in these chronicles show that there were many admirable abbots and generous benefactors of the monks in the eleventh century prior to 1066. They give a generally favourable picture, which is what the other evidence leads one to expect. The Norman Conquest could have resulted in irreparable damage to the monasteries. Not only did their estates suffer depredation and usurpation, but their cultural and intellectual inheritance was threatened with extinction. Although, of the thirty-five or so pre-Conquest abbots, only about half a dozen fled or were deposed, when, one by one in the course of nature, the others died, the Anglo-Saxon succession lapsed; the last Anglo-Saxon abbot to die was Wulfstan of Worcester in iog5.210 At the same time the episcopal bench had undergone Normanisation. Reform was in the hands of foreigners. Significantly enough, Lanfranc's Constitutiones are based on continental sources;211 they betray no debt to the Regularis concordia, though they were intended for Christ Church, Canterbury, which possessed at least two copies of it. Lanfranc and other newcomers, both ecclesiastical and lay, had no knowledge of, nor sympathy for, the saints whom the Anglo-Saxons venerated.212 Moreover, monastic prestige suffered from Norman apologists. The latter contended that William won at Hastings because he had God's support and apparently considered that one of the most praiseworthy results of the Conquest (and by implication one of its justifications) was the reform of 209
Chron. Ab., 459-61. See above p. 59., See Knowles, Monastic Order, 103-6, 111 ff. 211 Lanfranc's principal sources were the customs of Cluny; see Rose Graham, 'The relation of Cluny to some other movements of monastic reform', JTS xv (1914), 179-95 at pp. 184-5. The article is reprinted in idem, English Ecclesiastical Studies, London 1929, 1-29. Cf. The Monastic Constitutions of Lanfranc, ed. David Knowles, London 1951, pp. xii-xiii. 212 For Lanfranc's attitude see The Life of St. Anselm Archbishop of Canterbury by Eadmer, ed. and trans. R. W. Southern, London 1962, 50-4. For aspersions cast by Norman lords on St Edmund and his cult see Hermann's De miraculis S. Eadmundi, i. 97, 86. Cf. most recently James Campbell, 'Some twelfth-century views of the Anglo-Saxon past', in idem, Essays in Anglo-Saxon History, London 1986, 209 ff. 210
71
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 213
the Church carried out by Lanfranc with William's backing. To highlight the reform movement Anglo-Norman writers denigrated the condition of the Anglo-Saxon Church on the eve of the Norman invasion.214 Wholesale denigration injured the reputation of all the monasteries, but some houses suffered even worse because they were subjected to particular criticism. However, the complete Normanisation of any monastery took time, since the pre-Conquest monks remained for the term of their lives. The amount of dislocation depended on how soon it acquired a foreign abbot and on what kind of man he was. But, without exception, the monasteries had to defend their privileges, landed property and prestige. The monks were successful partly because they had built up the strength of their houses earlier in the eleventh century. They defended privileges and landholdings by appealing to pre-Conquest charters, often revised and rewritten to meet new needs, by forging more charters if necessary and by obtaining royal confirmations. They defended their houses' prestige by proliferating hagiographies of patron saints.215 Thus, they used the past to protect them in the present. Indeed, they met the challenge of the Norman settlement by infusing fresh vitality into their own tradition. The Anglo-Saxon practice of writing a Life of a recently dead head of a house is known to have survived in two monasteries. At Worcester two monks, Hemming and Coleman, each wrote a Life of Wulfstan, bishop of Worcester (d. 1095) j216 an<^ at Evesham a monk wrote the Life of Abbot SEthelwig (d. 1077). 217 Hemming's Life ofWulfstan and the Life of JEthelwig record in detail, in typically Anglo-Saxon fashion, their subjects' acquisition of property for their monasteries. Coleman's Life of Wulfstan has another feature characteristic of a number of pre-Conquest biographies: it is divided into sections according to subject matter. There are three sections concerning: first, Wulfstan's early life and education; 213
It seems fair to assume that Eadmer's attitude reflects Norman propaganda; for his account of Lanfranc's reforms etc. see Eadmeri Historia novorum in Anglia, ed. Martin Rule (Rolls Series, 1884), 12-17. For the Norman view of Lanfranc see Guillaume de Poitiers: Histoire de Guillaume le Conquerant, ed., with French trans., Raymonde Foreville, Paris 1952, 126-9. 214 Eadmer asserted that the monasteries were almost totally destroyed in Edward the Confessor's reign, Hist, nov., 5. William of Malmesbury denigrated the pre-Conquest Church in general, including the monasteries. Eadmer particularly attacked the monks of Christ Church, Canterbury. 215 For Osbern's and Eadmer's hagiographies, with references, see Southern, Saint Anselm, 248-51, 277-85. For those of the professional hagiographer, Goscelin, see Life of King Edward, 91 -11 i. 216 Hemingi Chartularium ecclesiae Wigorniensis, 2 vols, ed. Thomas Hearne, Oxford 1723, 403-8. Coleman's work, which was in OE, only survives in William of Malmesbury's Latin translation, The Vita Wulfstani of William of Malmesbury, ed. R. R. Darlington (Camden Society, 3rd ser. xl, 1928). 217 Chron. Eve., 87-96. Cf. Darlington, '^Ethelwig', 1-22, 177-98. For this biographical form, of which the earliest example in England is Asser's Life of King Alfred, see Gransden, op. cit. 51-2, 56, 88.
72
Legends, Traditions and History in Medieval England
secondly, his episcopate; and, thirdly, his character, personal habits and miracles. These divisions closely resemble those of ^Elfric's Life of St Ethelwold. But most remarkable is the fact that Coleman's Life of Wulfstan was in Old English (it survives only in William of Malmesbury's Latin translation) - surely a piece of conscious revivalism. At the same time the Anglo-Saxon tradition of historical writing was continued. It has recently been shown that Christ Church, Canterbury, was a hive of historiographical activity, all within the bounds of the Anglo-Saxon tradition; new annals were written and the Anglo-Saxon Chronicle was continued.218 A copy of the Chronicle went from Christ Church to Peterborough, where it was continued until the accession of Henry n.219 Worcester probably had more than one version of the Chronicle, one of which may have continued until ii3O. 2 2 0 Some version (or versions) formed the basis of the Latin chronicle by the Anglo-Norman monk of Worcester, known as 'Florence', and of his continuator, John of Worcester.221 A measure of the monks' success was the quick recognition by the new regime of their immediate value and of the value of their past tradition. William the Conqueror soon realised that they were potentially useful allies because of their knowledge of, and power in, the localities. He used both Wulfstan of Worcester and ^Ethelwig of Evesham as advisers and administrators to help maintain his authority in the difficult regions bordering on Wales.222 The lavishness of his grants of privilege to Bury St Edmunds suggests that he intended to increase the power of this loyal outpost in the rebellious Fens and against the threat of invasion from Scandinavia.223 The abbot, Baldwin, a Frenchman appointed by Edward the Confessor, having been Edward's physician, became William's. At the same time the new ecclesiastical establishment came to accept some parts of the Anglo-Saxon tradition. It realised that, on a national level, the 218 See Dumville, 'Aspects of annalistic writing', 23-57 passim. For the Christ Church versions of the Anglo-Saxon Chronicle in particular see ibid. 40 ff.; Whitelock, Anglo-Saxon Chronicle, pp. xi-xii. 219 The Peterborough Chronicle, 7070-7/54, ed. Cecily Clark, 2nd edn, Oxford 1970. Moreover, the Anglo-Saxon tradition of line-drawing continued to flourish at Christ Church in the generation after the Conquest. See Francis Wormald, 'The survival of Anglo-Saxon illumination after the Norman Conquest', Proceedings of the British Academy xxx (1944), 127-45. Professor Wormald later slightly revised the views which he had expressed there, English Drawings, 53 and n. i, 54-8. 220 VVhitelock, Anglo-Saxon Chronicle, pp. xiv-xvi, xx. 221 Chron. ex Chronicis; and The Chronicle of John of Worcester, 1118-1140, ed. J. R. Weaver, Oxford 1908. For the use of the Anglo-Saxon Chronicle by the Worcester chronicler/ chroniclers see English Historical Documents, 120; Martin Brett, 'John of Worcester and his contemporaries', in The Writing of History in the Middle Ages. Essays presented to Richard William Southern, ed. R. H. C. Davis and J. M. W'allace-Hadrill, Oxford 1981, 111 and n. 3, 123-4 and n. i. For the attribution of the chronicle to 1118 to Florence see Brett, op. cit. 104 and n. 3; below, pp. 116-17.' 222 Vita Wulfstani, pp. xxvii-xxviii; Chron. Eve., 89. 223 Gransden, 'Baldwin', 67.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
73
Church in England and, locally, each individual church, could only profit from having an ancient and often glorious history. Anglo-Saxon saints regained respect, and the Anglo-Saxon biographical and historiographical traditions were turned to Anglo-Norman uses. Thus, Eadmer, monk of Christ Church, Canterbury, continued and developed the Anglo-Saxon biographical tradition. The subject of his biography, St Anselm, Lanfranc's successor as archbishop of Canterbury and previously prior of Bee, was one of the newcomers. Like earlier AngloSaxon biographers Eadmer knew his subject well; he was a member of Archbishop Anselm's household and keeper of his chapel and relics. He divided his biography, in typically Anglo-Saxon fashion, according to subject matter; it is in two parts: the Life ofSt Anselm concerns St Anselm's spiritual life and the History of Recent Events his 'public' career.224 Although the two parts jointly constitute the biography, Eadmer has developed the Anglo-Saxon structural mode to such an extent that they form two separate, though complementary, books. Similarly, the AngloSaxon historiographical tradition was useful to the new generation. But it had to be made accessible and attractive to the new, French-speaking monks. One purpose of the spate of annalistic writing in Latin at Christ Church was to enable them to learn about Anglo-Saxon history225 in general and about the history of Christ Church in particular. This must also have been one reason why a parallel Old English and Latin version of the Anglo-Saxon Chronicle was made there.226 At the same time 'Florence', writing at Worcester, while using the Anglo-Saxon Chronicle as the principal source of information about English history for his (Latin) chronicle, included appropriate material about Norman and French history derived from continental sources (mainly from Marianus Scotus) to suit new interests. There is no better illustration of Norman co-operation with the AngloSaxons in preserving tradition and continuity than the mission to the north of England in loyy. 228 An Anglo-Saxon, Aldwin, prior of Winchcombe, was inspired to missionary activity by reading Bede and decided to refound the monasteries of Wearmouth and Jarrow. He was joined by two Anglo-Saxon monks from Evesham. The Norman bishop of Durham, Walcher, gave them the sites of Wearmouth and Jarrow, where .
224
.
997
Printed respectively: Life of St Anselm; Hist. nov. See Dumville, 'Aspects of annalistic writing', 54-5. F. P. Magoun (ed.) in 'Annales Domitiani Latini: an edition', Medieval Studies ix (1947), 235-95. I* should, however, be remembered that the translation of the AngloSaxon Chronicle into Latin had begun well before the Conquest. Both Asser and /Ethelweard translated substantial portions for their works. See Asser's Life of King Alfred, pp. Ixxxii-lxxxix; Keynes and Lapidge, Alfred the Great, 55-6; The Chronicle of ^thelweard, ed. and trans. Alistair Campbell, London 1962, pp. xxiii, xxxvii. 227 Brett, 'John of Worcester', no-n and n. i. 228 Symeonis monachi Opera Omnia, i. 9-11, 108 ff. Cf. Knowles, Monastic Order, 166 ff. For Bede's influence on the revival see above pp. 7 and n. 30, 8 and Davis, op. cit. (above p. 29). 225 226
74
Legends, Traditions and History in Medieval England
they built huts for themselves and those who joined them under the ancient ruins. Walcher was himself a student of Bede and planned to refound a monastery in St Cuthbert's see at Durham, but it was his successor, another Norman and also a student of Bede, William of St Carilef, who finally, in 1083, did so. Thus, Anglo-Saxons and Normans together attempted to restore Northumbrian monasticism, and also to continue the work of the tenth-century reformers, by promoting the re florescence of monasticism in the north. Anglo-Norman historiography reflects both the initial hostility of the two peoples' mentalities and their partial reconciliation. The Normans' implied claim that they invaded in order to reform a corrupt Church was in actuality contradicted by those monks who were intent on recording, even glorifying, the achievements of their abbots. In the next generation Anglo-Norman historians adopted a view of the past which, to some extent, reconciled Norman propaganda with Anglo-Saxon pride. Its chief architects were Eadmer and William of Malmesbury. Both were of mixed parentage; they were at once Normans and fervent believers in the reforms of their own times and Englishmen acutely conscious that their heritage was in peril. Their divided loyalties must have made them the more anxious to produce some kind of historiographical reconciliation. Eadmer saw a pattern, which William of Malmesbury elaborated, in the Ristory of Anglo-Saxon monasticism. The foundations of the tradition were laid during the Northumbrian renaissance, when monasticism reached a height of excellence never to be surpassed in England. Then followed a period of decline, leading to the extinction of monastic life. The latter was revived in the last half of the tenth century, a period of outstanding achievements. But again decline followed, the result in the first instance of the Viking incursions. This decline reached its nadir on the eve of the Norman Conquest and was followed by another great period of reform, that initiated by Lanfranc and carried through with William the Conqueror's help.229 This historiographical pattern had the advantage that it allowed the Anglo-Saxon Church a glorious tradition but did not detract from the importance of the Anglo-Norman reformation. The glory of the Anglo-Saxon tradition depended on the two peak periods, of which even the most recent was well in the past, and the Anglo-Norman reformation put an end to years of decadence. Moreover, the reformation was in line with the two great periods of the Anglo-Saxon era; it was their worthy successor. Eadmer's scheme had much truth in it and was, as a whole, perfectly plausible. The towering figure of Bede stood witness to the Northumbrian renaissance, and there was no doubt about the subsequent decline of strict monastic observance. Available evidence about the tenth-century revival 229 See Hist, nov., 3-5. Cf. Southern, Saint Anselm, 309-12; Willelmi Malmesbiriensis monachi De gestis regum Anglorum libri quinque, 2 vols, ed. William Stubbs (Rolls Series, 1887-9), i- 3°4~6.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism
75
was quite sufficient to substantiate the claims in its favour, while that for the period which followed was too scanty to disprove the theory of decline. When William of Malmesbury did research for his Gesta pontificum, a virtual gazetteer of monastic England, he could discover hardly anything about the eleventh century. He saw the very lack of contemporary sources, representing the monks' silence, as damning, an indication of intellectual stagnation.230 There is ample evidence for Anglo-Norman interest in the Northumbrian renaissance and in the tenth-century revival. Bedan studies underwent a dramatic resurgence231 and were accompanied by the proliferation of copies of the Ecclesiastical History. Bede himself was eulogised, notably by Symeon of Durham and William of Malmesbury, and the Ecclesiastical History, the principal source for early Anglo-Saxon history, was the means of informing the Norman settlers of their new country's origins. Historians saw themselves as Bede's continuators, even as his humble imitators. Bede's influence on the actual progress of the Anglo-Norman reformation is well illustrated by the above-mentioned renewed monastic impetus in the north. Anglo-Norman interest in the tenth-century revival is shown, for example, by the fact that a number of important works written then were recopied in the late eleventh and early twelfth centuries. Indeed, to this period belongs the apparently sole surviving manuscripts of ^Elfric's Life of St Ethelwold232 and of St Ethelwold's treatise on King Edgar's Establishment of the Monasteries,233 one of the only two extant manuscripts of the Regularis concordia and an important manuscript of St Ethelwold's translation of the Rule of Benedict.234 The three earliest manuscripts of Abbo's Passio Sancti Eadmundi are late eleventh-century, 235 and the five earliest manuscripts of Wulfstan's Life of St Ethelwold are all twelfth-century. 236 The AngloNorman historian, Orderic Vitalis, produced a reworked version of the latter. 237 230 Ibid. i. 1-2; Gesta pontificum, 328, 331. 231 por more details about Bede's influence in the Anglo-Norman period, with 232 references, see above pp.7-16. Three Lives, 6. 233 p rmtec j most recently in Councils and Synods, 142-54. Armitage Robinson, Times ofSt Dunstan, 160-7, argues that the treatise is post-Conquest. Dorothy Whitelock, however, contends that it is by St Ethelwold, Whitelock, 'The authorship of the account of King Edgar's establishment of the monasteries', Philological Essays, Studies in Old and Middle English Language and Literature in Honour of Herbert Dean Meritt, ed. J. L. Rosier, The Hague-Paris 1970, 127-36. Whitelock makes a good case for dating the composition of the work to the revival. Nevertheless, she ignores the passage in the narrative which, as Robinson points out, is reminiscent of one in William of Jumieges; this parallel suggests that the narrative was at least revised after the Conquest. 234 Cotton MS Faustina A X; N. R. Ker, Catalogue of Manuscripts Containing Anglo-Saxon, Oxford 1957, 194-5, 196 no. 154 B, art. i. Printed in Schroer, Die Angelsachsischen 235 Prosabearbeitungen. See Gransden, 'Baldwin', 72, 75 and nn. 146-8. 236 Three Lives, 7. 237 Printed in Jean Mabillon, Ada sanctorum Ordinis S. Benedicti, 9 vols, Paris 1668-1701, vii. 608-24; PL cxxxvii. 81-104. Michael Winterbottom. 'Three Lives of St. Ethelwold', Medium £vum xli (1972), 196-9.
76
Legends, Traditions and History in Medieval England
The Anglo-Norman historians were undoubtedly influenced by the tenth-century revival. Eadmer, in fact, linked his own times with the tenth-century revival by means of a theory of history. This theory underlies his historiographical pattern described above in so far as the latter relates to late Anglo-Saxon monasticism. It is based on an interpretation of an alleged prophecy of St Dunstan to King Ethelred.238 St Dunstan said that, because Ethelred had seized the throne by shedding the blood of his brother, Edward the Martyr, his reign would be one of bloodshed; he would suffer constant invasion from abroad and his kingdom would be continually wracked by the cruellest oppression and devastation. The Viking invasions had followed (the initial cause of monastic decline) and then the Norman Conquest, and so St Dunstan's prophecy had been fulfilled. Eadmer also alleged that St Dunstan appeared in a vision to Lanfranc to encourage him during his struggle to defend the privileges of the church of Canterbury. 239 He saw St Dunstan's co-operation with King Edgar as foreshadowing that of Lanfranc and Anselm with William the Conqueror.240 Similarly, the Anglo-Norman historians were influenced in their writing of local history by what information they had about the tenthcentury revival. They continued the Anglo-Saxon practice of seeking topoi from the past in order to fill gaps in their knowledge. Eadmer, William of Malmesbury and others, who were determined to prove the antiquity and unbroken history of particular monasteries, could often discover nothing about a house's foundation, unless it were early enough to have been recorded by Bede. Ignorance about monastic history from the late eighth century until the tenth-century revival was general and profound. Therefore, local historians turned to the revival itself for plausible hypotheses about what had happened. The entry under 964 in the AngloSaxon Chronicle saying that King Edgar expelled the clerks from the Old and New Minsters at Winchester and from Milton and Chertsey was a favourite topos. Another seems to have been ^Elfric's and Wulfstan's statement that, on Edgar's authority, St Ethelwold gave the clerks of the Old Minster the choice of becoming monks or leaving and that the clerks left, although three of them later returned to join the new community.241 Thus 'Florence of Worcester' (c. iioo) asserts (s.a. 963) that Edgar ordered the reformers to expel clerks living in communities throughout Mercia and to replace them by monks or nuns. He also states that, as a result, St Oswald expelled those clerks from Worcester Cathedral who refused to become monks and gave the habit to those who agreed, 238
239 Hist, nov., 3-5. Ibid. 18. Ibid. 3, 5. See Sally Vaughn, 'Eadmer's Historia novorum Proceedings x (1987), 269, 276, 286. Professor Vaughn also argues that Eadmer found a number of archetypes for Lanfranc and Anselm in Bede's HE, ibid. 263-85. The evidence for Eadmer's use of such archetypes is implicit, not openly expressed. Nevertheless, Professor Vaughn's argument conforms with my idea of the Anglo-Saxons' backward-looking mentality and certainly, in the case of St Dunstan, is convincing. 241 Three Lives, 23, 45. 240
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 77 appointing a monk of Ramsey, Wynsige, to be their prior.242 Writing later, the Ramsey chronicler gives much the same account as Florence of Worcester, but in one respect he seems to follow ^Eflric more closely; he asserts that Wynsige was a clerk of the old community whom St Oswald sent for training to Ramsey, before restoring him to Worcester to rule the new monastery.243 William of Malmesbury claims that St Dunstan expelled the clerks from Malmesbury Abbey and introduced monks; similarly he alleges that Wulfsige, bishop of Sherborne (? 993-1001/2), expelled the clerks from, and established monks in, his cathedral. 244 Again, the historian of Ely (temp. Henry n) asserts that St Ethelwold expelled the clerks from Ely in order to introduce monks.245 The case of Bury St Edmunds presented a special problem since two weighty authorities, Abbo and ^Elfric, made it clear that in their time, the late tenth century, St Edmund's shrine was still served by a clerical community. 246 The monastic foundation was, therefore, post-Edgarian. Nevertheless, the abbey's first historian, Hermann (writing c. noo), and his successors seem to have used topoi drawn from the tenth-century revival for their accounts of the foundation. In fact they may well have had virtually no information about it. Hermann claims that the clerical community was replaced by a monastic one in Cnut's reign, with the king's consent and on the advice of Bishop ^Eflwine.247 The Bury monk, who interpolated a copy of the chronicle of'Florence of Worcester' in Henry I's reign, elaborated on this. He alleged that Cnut, on ^Elfwine's advice, moved the clerks 'to other places' and established monks.248 Here he was probably using the topos supplied by ' Florence' himself, who presumably had derived it from his principal source for pre-Conquest history, the Anglo-Saxon Chronicle.249 The Bury interpolator also claims that Cnut gave the clerks the choice of becoming monks and staying or of leaving with adequate provision.250 Once more his immediate model may have been ' Florence'; the ultimate source was probably ^Ifric or Wulfstan. According to Hermann's narrative, one of the clerks, the 242 Edgar ordered SS Dunstan, Oswald and Ethel w o l d ' u t , expulsis clericis, in majoribus monasteriis per Merciam constructis monachos collocarent. Unde S. Oswaldus, sui voti compos effectus, clericos Wigorniensis ecclesiae monachilem habitum suscipere renuentes de monasterio expulit; consentientes vero hoc anno, ipso teste, monachizavit, eisque Ramesiensem coenobitam Wynsinum, magnae religionis virum, loco decani 243 praefecit': Chron. ex Chronicis, i. 141. Chron. Ram., 40-1. 244 For the ' refoundation' of Malmesbury Abbey see William of Malmesbury's Vita S. Dunstani, 301. William says nothing about the expulsion of clerks from Malmesbury in the Gesta pontificum, an earlier work; see ibid. 407. For Sherborne see ibid. 178. 245 Lib. Eli., 74. The passage stating that he expelled the clerks was an interpolation, presumably temp. Henry n, into the late tenth-century Libellus. 246 Three Lives, 82 ff.; ^Elfric's Lives of Saints, ii. 327 ff. 247 Hermann, De miraculis S. Eadmundi, 47. See Gransden,' Legends and traditions', 10 ff. 248 Memorials of St Edmund's Abbey, i. 341-2. 249 Chron. ex Chronicis, i. 141. 250 'aut in eodem loco ad religionis culmen erexit, aut dads aliis rebus de quibus abundantius solito victum et vestitum haberent': Memorials of St Edmund's Abbey, i. 342. Cf. above p. 172.
78
Legends, Traditions and History in Medieval England
sacristan ^Ethelwine, did in fact stay.251 Here again the model was probably ^Elfric, who states that Eadsige, sacristan of the previous clerical community in the Old Minster, Winchester, became a monk there.252 The attribution by Eadmer and other Anglo-Norman historians of the decline of monasticism, which followed the tenth-century revival, in the first instance to the Viking incursions may be another example of literary borrowing. But the model was not provided by the tenth-century revival; at that time the preceding decline was ascribed, possibly partly under Bede's influence, to excessive aristocratic power, which resulted in the secularisation of monasteries and the loss of their property. Perhaps attribution to such a cause would have been distasteful to the AngloNorman historian, since it would have seemed suspiciously like an innuendo against the rapacity of the feudal lords of their own day. The model for the idea of the Vikings as the primary cause was probably St Dunstan's prophecy as described in Osbern's Life of St Dunstan.2™ Osbern was an Anglo-Saxon monk of Christ Church, who was already of mature age at the time of the Norman Conquest. After a difficult period of readjustment to the new era, he became the first monk to write Lives of the Canterbury saints.254 Eadmer's well-known description of the preConquest monks of Christ Church, Canterbury, should be seen in the light of this tendentious historiography, a historiography which often relied on literary models rather than on first hand observation or other sound evidence. Eadmer asserts that the monks lived in omni gloria mundi, auro videlicet, argento, variis vestibus ac decoris cum pretiosis lectisterniis, m diversa musici generis instrumenta, quibus saepe oblectabantur, et equos, canes et accipitres, cum quibus nonnunquam spatiatum ibant, taceam, more comitum potius quam monachorum vitam agebant.255 William of Malmesbury embroidered this description and claimed that it was applicable to most Anglo-Saxon monks before the Conquest.256 Knowles took Eadmer's account seriously,257 but Professor Darlington and Sir Richard Southern have pointed out that it does not accord with what other information we have about the Christ Church monks.258 In general, historians have regarded the passage as reflecting Norman propaganda on the corruption of the late Anglo-Saxon Church. In fact, it could be at least partly a result of Eadmer's tendency to treat the AngloNorman monastic reformation as a reproduction of the tenth-century 251
Ibid. i. 34-7, 53-4. Even ^Ethelwine's name is suspect, see Gransden, 'Baldwin', 73, 252 74. Cf. idem, below p. 94. ^Elfric's Lives of Saints, i. 422-7. 253 Memorials of St Dunstan, 127. Cf. ibid. 115; and Southern, Saint Anselm, 311. Asser, and Fulco, archbishop of Rheims, in his letter to King Alfred, had previously suggested that the Viking onslaughts were a cause of the monastic decline. Alfred, in his preface to Gregory's Pastoral Care, considers the invasions to have been God's punishment for the 254 decline. See above p. 41 and n. 41. Southern, op. cit. 248-52. 255
257 258
Vita S. Dunstani, 237-8.
256
Gesta pontificum, 70-1.
Knowles, Monastic Order, 79. Darlington, 'Ecclesiastical reform', 402 n. 2; Southern, Saint Anselm, 247 and n. i.
Traditionalism and Continuity in Late Anglo-Saxon Monasticism 79 revival. Possibly Eadmer based his criticisms on those levelled by late tenth-century writers against the clerks of the old communities whom the reformers ousted. Admittedly, ^Elfric and the author of the first Life of St Oswald include among their strictures one on the clerk's non-celibate state, a criticism which Eadmer does not level at the Canterbury monks. However, he (and his audience) knew that the Anglo-Saxon monks were not married and believed them to have been celibate, but perhaps the mention of sumptuous bed-hangings (a detail omitted by William of Malmesbury) is an oblique reference to concupiscence. Conclusion Historians have seen the tenth-century monastic revival and the AngloNorman reformation as two peaks of monastic excellence, each rising steeply from a plain of decadence. They have tended to attribute these peak periods to continental influence and to concentrate their research on the nature of that influence. At both periods the reformers themselves helped promote such a view. No doubt they saw their achievements in this light - and certainly wanted others to do so. This was because, in each case, they needed to justify, and gain support for, their takeover from the pre-existing religious 'establishment'. The Anglo-Norman reformers made this picture of the tenth-century revival even more spectacular; they regarded the reformers as their own prototypes and stressed the contrast between their high level of achievement and the low level of the previous era. At neither period could the reformers afford to jettison the AngloSaxon tradition. To do so would have done irreparable damage to the prestige of the Church of England. Fortunately, there was no obstacle to prevent the tenth-century reformers from regarding the seventh century as a Golden Age; it served to emphasise the subsequent decline, which King Alfred had only partly, but they themselves had successfully, reversed. The Anglo-Norman reformers treated the seventh century in similar fashion, besides praising the tenth-century revival itself. In this way the reformers during both periods were able to shed reflected glory on their own movements and, by keeping within the Anglo-Saxon tradition, to make them more acceptably. However, this historiographical scheme distorted historical reality. It resulted in a neglect of the achievements by religious communities in the two 'low level' periods and, worse still, led to their propagandist denigration. After the Norman Conquest a few local historians, urged on by the necessity at that hazardous time of defending the rights and prestige of their own monasteries, altered the picture in their particular regard. They showed much that was creditable to late Anglo-Saxon monasticism. This and other evidence suggest the conclusion that we should see English monastic history, at least before c. noo, not in terms of plain and mountains, but of hills separated by gently undulating country.
This page intentionally left blank
3
Legends and Traditions concerning the Origins of the Abbey of Bury St Edmunds D A V I D K N O W L E S AND R. N E V I L L E H A D C O C K were aware of the difficulties which beset any scholar who tries to establish the precise date of the foundation of a pre-Conquest monastery. Indeed the origins of such houses were various and usually obscure.1 The Scandinavian invasions caused a hiatus in English monastic history, and it is generally agreed that there was virtually no continuity between the period of Northumbrian monasticism and tenth-century revival. The post-Conquest houses were to all intents and purposes founded during and after that revival. The term, however, 'to found', is misleading. It denotes a multiple procedure: the formation of a monastic community, its endowment by charter, the construction of a church and monastic buildings, and the dedication of the church. The word is certainly applicable to Ramsey abbey, founded by St Oswald (bishop of Worcester 961-92, archbishop of York 972-92), in conjunction with ^Ethelwine the ealdorman. But is is only partially applicable to most of the other houses. Here, therefore, we use it to mean only the constitution of a regular community observing the Rule of St Benedict. The tenth-century revival marked the restoration of some ancient monasteries, notably Abingdon, Ely and Peterborough. It also saw the establishing of Benedictines in houses where there were already congregations of priests or 'canons' living semimonastic lives. This, it seems, was the situation at Christ Church, Canterbury, Evesham, Glastonbury, Malmesbury, Winchester and Worcester. Little is known about these early congregations; preConquest evidence is rare and, as will be argued below with reference to Bury St Edmunds, post-Conquest evidence should not be relied on. The priests served a cathedral or church, under the aegis of a patron saint whose relics they possessed. Three congregations, indeed, claimed to have the perfect, uncorrupt body of their patron saint; there was St ^Etheldreda at Ely, St Cuthbert successively at Lindisfarne, Chester-le-Street and Durham, and St Edmund at Beodricesworth. Only in a few cases is it known with any degree of i. David Knowles and R. Neville Hadcock, Medieval Religious Houses, England and Wales (znd edn, London, 1971), p. i.
82
Legends, Traditions and History in Medieval England
certainty whether members of the pre-existing congregation stayed on, adapting themselves to the new regime, or whether they were replaced wholesale by Benedictines from elsewhere.1 The historian undertaking research into the origins of a preConquest monastery is faced with the problem both of the paucity of contemporary evidence, and of the often biased nature of that and later evidence. The monks themselves distorted the historiography of their houses. In view of the importance attached to tradition in the Middle Ages, it is not surprising that they were eager to know their houses' history. They were interested in normal circumstance, but interest deepened in times of especial religious fervour, and of trouble. Then the monastic historian aimed to increase his monastery's fame, power and wealth by glorifying its patron saint, and when necessary by proving its rights to privileges and possessions. He would, therefore, embellish what facts he could discover with unwarranted detail, and if evidence was scanty or absent, he might use forged documents and resort to literary fabrication. Crowland, Glastonbury, Westminster and St Swithun's, Winchester, are all notorious for their tainted, tendentious histories. At first sight the origins of St Edmund's abbey present less problems than those of most houses. Scholars accept that it was founded nearly fifty years after the tenth-century monastic revival, by which time light had surely dawned. Indeed there are today a number of unquestioned beliefs about the abbey's foundation, and also about the preexisting secular congregation, and its raison d'etre, the cult of St Edmund — and even about the martyrdom itself. It is the purpose of this paper to examine the evidence for these beliefs, treating the subject chronologically, in order to discover how securely they are based. The Anglo-Saxon Chronicle records s.a.Syo: 'that winter King Edmund fought against [the Danes], and [they] had the victory, and killed the king and conquered all the land'. The phrase 'J)one cining of slogan' does not exclude the possibility that St Edmund was killed in battle and not martyred at all.2 Numismatics prove that within twenty years of St Edmund's death, King Alfred promoted his cult and that it flourished in the Danelaw.3 But after a while enthusiasm for the cult seems to have declined; no more St Edmund coins were minted 1. Above pp. 42-4, below p. 94. 2. The Anglo-Saxon Chronicle; a revised translation ed. Dorothy Whitelock with D. C. Douglas and S. I. Tucker (London, 1961), p. 46. Professor C. E. Fell has called my attention to the fact that St Olaf, king of Norway, was killed in battle (1030), but the tradition grew that he was martyred (Bollandists, Ada Sanctorum•, July, vol. vii, p. 107). For a fourteenth-century picture of St OlaPs 'martyrdom', see Islandske hdndskrifter og dansk kultur. Udstillung pd statens Museum for Kunst (Copenhagen, 1965), p. 83. For another similarity between the St Edmund legend and that of St Olaf, see infra 87 and n. 6. 3. C. E. Blunt, 'The St Edmund memorial coinage', Proceedings of the Suffolk Institute of Archaeology, xxxi (1970), pp. 234—55.
Legends and Traditions concerning Bury St Edmunds
83
after about 91 o or perhaps a little later, and St Edmund does not appear in the surviving liturgical calendars of the ninth, tenth and early eleventh centuries. However, at the time of the tenth-century monastic revival at least the potential of St Edmund's cult was sufficiently recognized for the monks of Ramsey to commission a Life of the saint, Abbo of Fleury's Passio Sancti Hadmundi.1 The information in the Passio that the cult centred on a shrine of St Edmund at Beodricesworth is corroborated by the chronicle of ^Ethelweard the ealdorman and by Resting Places.2 By this time the cult was popular enough for JElfric to include an English rendering of the Passio in his Lives of Saints.3 So far the evidence for the death of King Edmund and for the development of the cult to r.iooo is trustworthy. But Abbo puts flesh on this skeleton. He claims the highest authority. The prologue^ dedicates the work to St Dunstan and reveals that Abbo wrote it during his visit to Ramsey (985-7), at the request of the monks.4 It also claims that Abbo's authority for St Edmund's martyrdom was St Dunstan himself: the latter when a boy had heard St Edmund's armour bearer telling King Athelstan, and Abbo had heard St Dunstan telling the bishop of Rochester and the abbot of Malmesbury. It was normal in the middle ages for a religious community to commission a Life of its patron saint, and the question arises why the monks of Ramsey should have commissioned one of a saint buried elsewhere. The prologue suggests that St Dunstan himself may have originated the idea, or a request might have come from the congregation at Beodricesworth. However, although the Passio shows strong local attachment, it is not only to Beodricesworth - and to Ramsey — but also to the Fenland, on which it has a paean.5 It is not improbable that Ramsey's founder, St Oswald, suggested that Abbo should write the Passio. His interest in the cult of saints is well known. As archbishop of York he searched for and found the relics of a number of Northumbrian saints. One Northumbrian saint, his namesake, St Oswald king and martyr, was of course an obvious parallel with St Edmund.6 He might well have promoted St Edmund's cult 1. The most recent edition is in Three 'Lives of 'English Saints; ed. Michael Winterbottom (Toronto Medieval Latin Tests, 1972), pp. 67-87, the edition cited here. Another edition is in Memorials of St. Edmund's Abbey, ed. Thomas Arnold (Rolls Series, 1890—96, 3 vols.), i. 3-25. 2. Passio, p. 82; The Chronicle of JEthelweard, ed., with an English translation, A. Campbell (Nelson's Medieval Texts, 1962), p. 36; Die Heiligen England*, ed. F. Liebermann (Hanover, 1889), pp. 13, 14; D. W. Rollason, 'Lists or saints' resting-places in Anglo-Saxon England', Anglo-Saxon England, vii (1978), pp. 61-93. 3. Printed in jElfric's Lives of Saints, ed., with translation into modern English, W. W. Skeat (Early English Text Society, original series, Ixxvi, Ixxxii, xciv, cxiv, 1881-1900, 2 vols. in 4 pts), ii. 314-3 5, the editon used here. Another edition is in JElfric, Lives of Three English Saints, ed. G. I. Needham (London, 1966), pp. 43—59. 4. Passio, p. 67. 5. Passio, pp. 69-70. 6. The Historians of the Church of York, ed. James Raine, jr (Rolls Series, 1879-94, 3 vols.), i. 462.
84
Legends, Traditions and History in Medieval England
in order to add another illustrious name to the catalogue of East Anglian saints, an intention which would have appealed to the Ramsey monks. He may even have wanted to put East Anglia on a par with Northumbria. Abbo himself emphasizes the parallel between St Edmund and St Cuthbert, saying that if anyone doubts the uncorruption of St Edmund's body, he should remember the case of St Cuthbert.1 Abbo relates that the Danish leader Ivarr captured St Edmund at 'Haeglesdun' (or, as here, 'Haegilisdun'),2 had him tied to a tree, scourged, shot at with arrows, and finally beheaded. To prevent a decent burial, the Danes, as they returned to their boats, threw the saint's head into a thicket in 'Haeglesdun' wood.3 Christians in the neighbourhood buried the body and when peace was restored searched for the head. At length the head attracted their attention by calling 'here, here, here', and they found it between the paws of an immense wolf, which was guarding it from other wild animals. The country folk, followed by the wolf, took the head back to their village. On arrival the wolf returned to the wood, and the villagers fitted the head back on to the body; the two miraculously reunited. The villagers buried the perfect uncorrupt body, which began to work miracles, and built a simple church over it. When the times were favourable (ubi tempus oportunum invenif) the body was moved to the royal vill (villa regta, presumably a vill owned by the king of East Anglia),4 a suitable church built to house it, and lavishly endowed.5 Abbo shows that the shrine was served by some kind of semi-monastic congregation. He also asserts that shortly before the time of writing (Paulo ante nostra moderna tempord) the body was cared for by a holy woman called Oswen, who cut its hair and nails every year.6 Historians have of course dismissed the miraculous elements in this story. However, they have in general treated the rest of it uncritically. The account of the martyrdom itself has been used as evidence for the way Danes executed their prisoners,7 even for ritual sacrifice by 1. Passio, p. 68. None of the extant texts of the Passio was written before the Norman Conquest and the emphasis on the similarity between St Edmund and St Cuthbert could be the result of post-Conquest revision; see A. Gransden, 'Baldwin, abbot of Bury St Edmunds, 1065-1097', Proceedings of the Battle Conference on Anglo-Norman Studies, iv, 1981 (Woodbridge, 1982), pp. 72, 73, 192, 193 nn. 99-108, 120-21, and cf. infra n. 4, and p. 93. 2. Passio, p. 73. 3. Ibid. p. 79. 4. villa regia: possibly a post-Conquest addition; see infra, p. 93 and n. 2. Cf. supra, n. i. 5. Passio, p. 82. 6. Ibid. pp. 82-3. 7. Dorothy Whitelock, 'Fact and fiction in the legend of St. Edmund', Proceedings of the Suffolk. Institute of Archaeology, xxxi (1970), pp. 221-2. Professor Whitelock states that 'the main facts of the martyrdom are likely to be true'. She cites as a parallel the martyrdom of St yElfheah in i o 12 by the Danes, who used him for target practice. But the parallel is not close since ^Ifheah apparently was killed by being pelted with bones, stones and pieces of wood; Anglo-Saxon Chronicle, ed. Whitelock, pp. 91-2, and Thietmar von Merseburg Chronik, ed. Werner Trillmich (Ausgewahlte Quellen zur Deutschen Geschichte des Mittelalters, ix, Berlin, 1962), p. 400. Whitelock ('The legend of St Edmund', p. 222 and n. 18) also states that 'the removal of the head is a well-evidenced practice among primitive peoples'; she cites three examples, none of which, however, resembles the case of St Edmund at all closely.
Legends and Traditions concerning Bury St Edmunds
85
1
blood eagle. This disregards the fact that there is no evidence in the literary sources that the body in the shrine had been riddled with arrows and beheaded. Abbo, in his account of St Edmund's translation to Beodricesworth, emphasizes that the body was perfect (integer et uiuenti simillimus), but had, as a sign of martyrdom, a thin red line like a thread of silk around its neck (Tantum in eius collo ob signum martyrii rubet una tenuissima riga in modum fill coccinei}.2 Allegedly the coffin was next opened and the body examined by Abbot Leofstan (?1045-1065), but more probably when St Edmund was translated to Abbot Baldwin's new church in 1095.3 The occasion is described by Hermann who wrote c. 100 and was apparently a monk of Bury at that time. His De Miraculis Sancti Eadmundi* stresses the perfection of the body (Jacet integer ut dor miens ).5 He makes no mention of any mark around the neck, although he relates an improbable tale of how Abbot Leofstan conducted an experiment to see if the head had truly grown back on the body.6 However, he does describe the body's clothing as blood stained and perforated with arrow holes (Exuitur itaque sanctus sancti martyrii vestibus partim rubeis rubore sanguinis, partim perforates ictibus telorum crebris). The clothing was removed 'for restoration' for the salvation of the faithful (sed tamen reponendis, saluti credentium profuturis), and placed in the treasury.7 Jocelin, of Brakelond, in his account of the examination of the body by Abbot Samson, similarly stresses the body's perfection and makes no allusion to any wounds.8 The signs of the martyrdom, the thin red line, the blood-stained, perforated clothing, mentioned in the sources, could obviously all have been fabricated by St Edmund's guardians. The absence of early evidence that the body in the shrine had suffered in the way Abbo 1. A. P. Smyth, Scandinavian Kings in the British Isles 8jo-SSo (Oxford, 1977), pp. 211-12, and J. M. Wallace-Hadrill, The Vikings in Francia (Reading, 1975), p. 10. Since I completed the present article, an article has appeared which throws grave doubt on the trustworthiness of accounts of the rite of the 'blood-eagle' in skaldic verse, and which discards the martyrdom of St Edmund as an example of the alleged rite for substantially the same reasons as I do (infra ]87 and nn. i, 2; see Roberta Frank, 'Viking atrocity and skaldic verse: the rite of the BloodEagle', ante, xcix (1984), 332-43. 2. Passio, c. 14, 11. 3-10 (p. 82). 3. A parallel would be the opening of St Cuthbert's tomb during his translation in 1104; Symeonis Monachi Opera Omnia, ed. T. Arnold (Rolls Series, 1882-5, 2 vols.), i. 247-61. Cf. ibid. i. 229 n.a., and Symeonis Dunelmensis Opera Omnia, ed. H. Hinde (Surtees Society, li, 1868), i. 188-97. For my reasons for doubting the story of the opening of St Edmund's tomb by Abbot Leofstan see 'Baldwin', p. 73. The story is accepted by N. Scarfe, 'The body of St. Edmund', Proceedings of the Suffolk Institute of Archaeology, xxxi (1970), 308-9. 4. Printed Memorials of St Edmund's Abbey, ed. Arnold, i. 26-92 (henceforth referred to as 'Hermann'). For Hermann see infra 89 and n. 7. 5. Hermann, p. 53. 6. Ibid, p. 54. I inadvertently overlooked this passage when I stated (citing ibid. p. 56) in 'Baldwin', p. 193, n. 116, that Hermann does not connect Leofstan's gout of the hands with this irreverent act. 7. Ibid. pp. 5 3-4. 8. Chronicle of Jocelin of Brakelond, ed., with an English translation, H. E. Butler (Nelson's Medieval Classics, 1949), p. 114.
86
Legends, Traditions and History in Medieval England
describes has led to the suggestion that it was not St Edmund's.1 On the other hand if it was St Edmund's, this fact undermines the credibility of the martyrdom account in the Passio. The latter interpretation is the most likely, since the narrative is little more than a hotchpotch of hagiographical commonplaces, a fact neglected by recent historians. Abbo himself remarks on the resemblance of St Edmund's martyrdom to that of St Sebastian who was also made a target for arrows.2 St Sebastian was eventually cudgelled to death, while St Edmund was beheaded. Usuard provides many examples of martyrs who, having survived a series of torments, were at last decapitated.3 Abbo's specific models may have been the Passio SS. Dionysii, Rustici et Eleutherii by Fortunatus (fl. 700) and the Vita S. Dionisii by Hilduin (d.84o). It is not unlikely that Abbo of Fleury would have turned for a model to the legend of St Denis, the patron saint of the abbey of St Denis, Paris, and of the kings of France, the most venerated saint in the Ile-de-France, especially since St Edmund was a king and one whose cult was, as will be seen, associated with royalty. (Hilduin's Vita at least was known in England by the late tenth century; it is in the 'Cotton-Corpus' Legendary, of which Kline's L/w of Saints is in effect a translation.)4 Hilduin relates that St Denis was stretched on a rack, scourged, flung into a fire, and thrown to lions — but God saved him, as He had Daniel, by making the lions lie down peaceably before him. The reference to Daniel supports the suggestion that Hilduin was one of Abbo's models, since it could explain the presence in the Passio of an inapposite reference to the Daniel story: the passage relates that the country folk, on finding St Edmund's head guarded from other wild creatures by a wolf, decided that his merits must resemble Daniel's.5 Hilduin has the topos of the saint's body living after the severance of the head (St Denis walks for over two miles carrying his head until he reaches the place where he wishes to be buried). And Fortunatus has the topos of the head speaking.6 The claim that St Edmund was a victim of the 'blood eagle' rests mainly on one passage in the Passio; Abbo states that when St Edmund was taken almost dead from the stake for execution, his ribs had been laid bare by numerous arrow wounds, as if 'he had been stretched on a rack or pierced by iron hooks' (retectis costarum latebrispraepunctionibus credis ac si raptum equuleo 1. Cf. Scarfe, op. tit. p. 309. 2. Passio, pp. 78-9. 3. For examples see Migne, PL, cxxiii, cols. 623-4, 673-4, 681-2, 959-60; ibid, cxxiv, cols. 17—18, 169—70. 4. See P. H. Zettel, 'Saints' Lives in Old English: Latin manuscripts and vernacular accounts: ^Elfric', Perifia, i (1982), 17-37. I am indebted to Dr Michael Lapidge for this reference. For ^Elfric's rendering of Hilduin see jE/fric's Lives of Saints^ ed. Skeat, ii. 168-91. 5. Passio^ c. 12, 11. 47-8 (p. 81). Cf. Migne, PL, cvi, cols. 44-5. 6. MGH, Auctores Antiquissimi, iv, pt 2, p. 104, 1. 13. Cf. Edmund Colledge and J. C. Marler, 'Cephalologie, a recurring theme in classical and medieval lore', Traditio, xxxvii (1981), 41 If (pp- 418-19 for especial reference to the St Denis legend).
Legends and Traditions concerning Bury St Edmunds
87
1
aut seuis tortum ungulis). But racks and hooks are hagiographical commonplaces. For example St Denis, as observed above, suffered on a rack, and according to Usuard four of the martyrs who shared the same anniversary as St Sebastian (20 January) were tortured with hpoks as well as the rack.2 An exact parallel for the wolf story has proved elusive. However, the Life of St Mary of Egypt (in the Vitae Patrum) has a rather similar tale, although in this instance the animal in question is a lion. When Zosimus, a devout monk, went in search of St Mary in the desert, he found her dead body guarded by a huge lion, which, having helped him dig a grave, returned 'as quietly as a lamb' to its desert solitude. Since Abbo no doubt knew that there were no lions in England, he might well have considered a wolf, a savage beast well-known in western Europe, a more suitable guardian for St Edmund's head.3 Abbo's story of Oswen, the holy woman who tended St Edmund's body, is not apparently paralleled in all its details elsewhere. The topos itself, of a pious woman caring for a saint's body, was wellknown and seems to have originated in the legend of St Sebastian, whose body was looked after by St Lucina (30 June).4 It occurs in the St Denis legend; after his martyrdom St Denis's body was rescued and buried and his shrine built and served by a matron called Catulla.5 The detail that Oswynn cut St Edmund's hair and nails is paralleled in the twelfth century legend of St Cuthbert and in the legend of St Olaf of Norway (d.io3o).6 Rather earlier (in 1000), when Charlemagne's tomb was opened, it was recorded that the emperor's nails had grown and needed cutting.7 It can, therefore, be argued that Abbo's Passio is a patchwork of borrowings from wellknown hagiographies, which Abbo adapted as he thought fit. Almost certainly he knew virtually nothing about St Edmund's death and 1. Passio, c. 10, 11. 27-8 (p. 79). 'seuis tortum ungulis' should be 'tortured with cruel hooks', not 'torn by savage claws', as in Dr Smyth's translation of the passage: Smyth, op. cit. pp. 2 1 1 - 1 2 .
2. Migne, PL, cxxiii, cols. 673—4. Cf. also St Alexander, ibid, cxxiv, cols. 15—16. 3. I am indebted to Sister Benedicta Ward for drawing my attention to the parallel between the legend of St Mary of Egypt (Migne, PL, Ixxiii. cols. 688-9) and that of St Edmund. Isidore (Etymologia, XII, cap. 2) makes a comparison between the wolf and the lion. Dr Michael Lapidge has pointed out to me that the.legend of St Mary of Egypt in the Lives of Saints was not by JEAfric but occurs in early versions of the Lives, notably in BL Cotton Ms. Otho B X, an eleventh century text from Worcester, and in BL Cotton MS. Julius E VII, a twelfth century text from Bury St Edmunds: JElfric's Lives of Saints, ed. Skeat, ii. 446-7; cf. ibid. ii. ix, xiv, xv, and N. R. Ker, Medieval Libraries of Great Britain (second edition, London, 1964), pp. 20, 207. 4. Bollandists, Acta Sanctorum, January, vol. ii, p. 257. 5. MGH, Auctores Antiquissimi, iv, pt 2, p. 104, 1. 20; Migne, PL, cvi. cols. 48-9). 6. See, respectively, Reginald! Monachi Dunelmensis Libellus de admirandis Beati Cuthberti Virtutibus, ed. James Raine, sr (Surtees Society, i, 183 5), p. 57, and Bollandists, Acta Sanctorum, July, vol. vii, p. no. For another similarity between the St Edmund legend and that of St Olaf, see supra, p. 2, n. 2. Cf. The Relics of St Cuthbert, ed. C. F. Battiscombe (Oxford, 1956), pp. 45-6. For the relationship between the legends of St Edmund and St Cuthbert see infra, p. 21, n. 2. 7. Robert Folz, Le Souvenir et la Legende de Charlemagne dans I'TLmpire germanique medieval (Paris, 1950), p. 92. I owe this reference to Professor R. A. Markus.
88
Legends, Traditions and History in Medieval England
early cult; nor could succeeding hagiographers add any new credible information. The depths of their ignorance is betrayed by their failure to agree on an important point, the exact location of the supposed martyrdom. Abbo places it at 'Haeglesdun' while later tradition (to be discussed below) places it at Hoxne. All that is certain is that by c. 1000 a shrine containng a body, supposedly that of St Edmund, was the centre of a cult at Beodriceworth. So far we have discussed the development of the hagiology of St Edmund in the late tenth century. The next important period of growth was the late eleventh century, to which time also belongs the genesis of the legends and traditions concerning the abbey's origins. It is essential to consider their evolution against the background of post-Conquest politics and society. It should also be remembered that the abbot, Baldwin (1065—97), was an ex-monk of St Denis and had been an intimate of Edward the Confessor (he was his physician). After the Conquest the monks of Bury had to defend themselves against Normans who tried to seize their estates2 and even cast doubt on the actual presence of St Edmund's body in the abbey.3 But their worst enemy was the bishop of East Anglia, first Arfast (1070—^.1085) and then Herbert Losinga (1090/1—1119).4 Arfast tried to move his see from Thetford to Bury. The dispute came to a head in 1081, when, after more than a decade of contention involving appeals to Rome, and the intervention of William the Conqueror and Lanfranc,5 the case was heard in the royal court.6 Although judgement was given in the abbey's favour, Arfast and Losinga continued the struggle until the latter moved the see to Norwich in 1094/5. Nevertheless he and at least one of his successors towards the end of the twelfth century tried to assert authority over the abbey.7 In the course of the dispute both parties appealed for support to historical precedents. Since very little was known about the history in Anglo-Saxon times of the East Anglian see or of the cult of St Edmund, the way was open for the disputants to interpret what little information there was as served them best, and to invent corroborative 'evidence'. Our knowledge of the controversy - its course and the protagonists' arguments - is fragmentary. Nevertheless, it undoubtedly had a profound influence on the development of St Edmund's hagiology and on that of the 1. Passio, pp. 73, 79. 2. Feudal Documents from the Abbey of Bury St Edmunds, ed. D. C. Douglas (London, 1932), nos. 9, 10; Hermann, pp. 58-9, 75-6, 79. Cf. Gransden, 'Baldwin', pp. 67, 68, and nn. 3. Hermann, p. 86. Cf. Gransden, 'Baldwin' , p. 73. 4. For the dispute see David Knowles, 'Essays in monastic history, IV - the growth of exemption', Downside Review, 1 (1932), 208-12, and Gransden, 'Baldwin', pp. 69-72. 5. Hermann, p. 65. Cf. infra, p. 91. 6. Hermann, pp. 65-7. An account of the trial is in the forged charter of privilege, allegedly of William I, to St Edmund's; see Douglas, op. cit. p. xxxiv, and Gransden, 'Baldwin', p. 71. 7. See The Customary of the Benedictine Abbey of Bury St Edmunds in Suffolk, ed. Antonia Gransden (Henry Bradshaw Society, xcix, 1973, henceforth referred to as Bury Customary), pp. xl, xli.
Legends and Traditions concerning Bury St Edmunds
89
abbey's historiography, an influence which recent historians have ignored or underestimated.1 The popularly held belief that St Edmund was martyred and first buried at Hoxne, which by the late fourteenth century was current even at Bury2 (despite the contrary statement in Abbo's Passio), seems to have originated as part of the bishop of East Anglia's case; the charter which Herbert Losinga granted to his new cathedral priory at Norwich in r.noo includes among the endowments the church of Hoxne 'with the chapel of St Edmund, where the martyr was killed'.3 The tradition was accepted at St Albans in the thirteenth century; it appears in the chronicles of both Roger of Wendover and Matthew Paris.4 It could have reached St Albans from the latter's cell at Wymondham, within ten miles of Norwich, and then reached Bury from St Albans. Possibly it owed its origin to a simple substitution of the name of St Edmund for that of another 'martyred' king of East Anglia, St Ethelbert (who was beheaded in 794 by order of King Offa of Mercia); Ethelbert was, at latest by the mid-tenth century, the patron saint of the church at Hoxne, which had presumably acquired relics of him and was served by some kind of community.5 The original purpose of such a substitution might well have been to provide the bishop of East Anglia's party with a precedent; Hoxne had been the bishop of East Anglia's see in the ninth century and again under Edward the Confessor;6 therefore St Edmund's present burial place (Bury) should be the present bishop's see. Meanwhile at Bury itself Hermann, who claimed to have been in Bishop Arfast's service before becoming, as it would appear, a monk of St Edmund's,7 wrote the De Miraculis as a piece of propaganda for his new home. His object was to increase St Edmund's and the abbey's prestige in order to fortify it against its enemies, especially the bishop of East Anglia. He also wrote to provide a piece justificative for the monks in their dispute with Losinga. Hermann, indeed, is the 1. See e.g. p. 90,n. i infra. 2. Bodley 240 (for which see infra, pp. 15-16 passim), pp. 631, 632. 3. W. Dugdale, Monasiicon Anglicanum, ed. J. Caley, H. Ellis and B. Bandinel (London, 1817-30, 6 vols. in 8), iv. 16. Cf. Whitelock, 'The legend of St. Edmund', p. 223. 4. Roger of Wendover alleges that St Edmund was buried at 'Hoxe' while Matthew Paris gives 'Hore'; Matthew Paris, Chronica Majora, ed. H. R. Luard (Rolls Series, 1872-83, 7 vols.), i. 400 and n. 4. For the Hoxne legend see V. H. Galbraith, 'The East Anglian see and the abbey of Bury St Edmunds', ante, xl (1925), 223 and n. 3, and Gransden, 'Baldwin', pp. 70, 190, nn. 75-9. In modern times the legend was believed by, for example, Thomas Arnold (Memorials, i. xxi, 10, 16), and A. J. Robertson, -Anglo-Saxon Charters (Cambridge, 1956), p. 425. 5. See Anglo-Saxon Wills, ed. and translated Dorothy Whitelock (Cambridge, 1930), no. i. Cf. C. R. Hart, 'The East Anglian chronicle', journal of Medieval History, vii (1981), p. 264. St Ethelbert was buried at Hereford; see Rollason, 'Resting places', pp. 64, 89. 6. Whitelock, Wills, pp. 4, 99, 102, and Domesday Book .., ed. A. Farley (Record Commission, 1783, 2 vols.), ii. 379. 7. Hermann, pp. 62, 63, 67. Cf. The Letters of Lanfranc Archbishop of Canterbury, ed., with an English translation, Helen Clover and Margaret Gibson (Oxford Medieval Texts, 1979), p. 152 and n. 6. For Hermann as a propagandist for Bury, see Gransden, 'Baldwin', pp. 72-3.
90
Legends, Traditions and History in Medieval England
earliest authority for the universally held belief1 that St Edmund's abbey was founded in 1020 by King Canute. However, in view of the tendentious nature of his narrative this statement deserves more attention than it has hitherto received. The relevant passage reads: Quo [i.e. Canute] tetrarchizante, sanctique martyris [Eadmundi] veneratione pullulante, clericalis ordo famulatus sancto in ordinem monachicum mutatur in eodem loco, indagine veritatis talia commutando felici commercio, ut rex et martyr venerandus frequentiori famulatu necnon digniori veneraretur. Nee hoc absque regis concessu et optimatum ejus fuit assensu, sed et monitu sani consilii ^Elfwini praesulis diocesiani, Eliensis quidem monachi, amatoris autem ordinis sancti, tune tempore concessis et datis martyri sancto multis donariis, anno millesimo xxmo Domini generationis, comitatu vero Thurcilli comitis, ad honorem sancti suppeditantis.2
There is nothing improbable in the story that Canute 'founded' St Edmund's abbey. His retribution for the murder by the Danes of another famous Englishman, Archbishop ^Elfheah, is well known.3 Moreover, he was the friend and benefactor of both Ely and Ramsey4 — and in the vicinity of the latter he planned to found another religious house, a nunnery.5 And he endowed the community at Beodricesworth, whether the pre-monastic or monastic one is not clear. Hermann states, in the paragraph preceding the foundation account quoted above, that Canute protectorem suum post Deum invisens sanctum [Eadmundum], actu regali xeniauit locum donis ac redditibus propriis munificauit, liberumque omni consuetudine chyrographizauit.6
Since the surviving diploma in Canute's name to St Edmund's contains a grant of exemption from episcopal control, it is generally believed that the monks of Bury had it forged for the 1081 trial in 1. The story is accepted, with varying degrees of detail and precision, by: David Knowles, The Monastic Order in England (2nd edn, Cambridge, 1963), p. 70; Knowles and Neville Hadcock, Medieval Religious Houses, England and Wales, p. 75; F. Hervey, The History of King Eadmund the martyr and the early years of his abbey (London, 1929), p. 32; A. Goodwin, The Abbey of St Edmundsbury (Oxford, 1931), p. 5; F. E. Harmer, Anglo-Saxon Writs (Manchester, 1952), p. 138, n. i; Robertson, Anglo-Saxon Charters, p. 425; R. H. C. Davis, 'The monks of St Edmund, 102 i-i 148', History, xl (195 5), 232-3; Scarfe, 'The body of St Edmund', pp. 307-8; Hart, 'East Anglian chronicle', p. 274, and the same author's The Early Charters of Eastern England (Leicester, 1966), p. 64. 2. Hermann, p. 47. 3. Anglo-Saxon Chronicle, s.a. 1023. 4. Liber Eliensis, ed. E. O. Blake (Camden third series, xcii, 1962), pp. i^-^-j passim, and Chronicon Abbatiae Rameseiensis, ed. W. D. Macray (Rolls Series, 1886, henceforth referred to as Ramsey Chron.}, pp. 125-6, 133-4, 139, 267. (However, many of the charters in Canute's name to monasteries are spurious; see P. H. Sawyer, Anglo-Saxon Charters (London, 1968), pp. 28 5-96 passim.} Another East Anglian chronicler, Henry of Huntingdon, writing in the mid-twelfth century, eulogized Canute; Historia Anglorum, ed. Thomas Arnold (Rolls Series, 1879, repr. Kraus 1965), pp. 188-9. 5. Ramsey Chron. p. 126. 6. Hermann, p. 46.
Legends and Traditions concerning Bury St Edmunds
91
the royal court.1 However, Miss Harmer suggests that part of it was copied from a genuine charter.2 This section is the grant of fish due annually to Canute, a fishery, and all rights from pleas in the vills owned by St Edmund's, and the confirmation of Queen Emma's gift of 4,000 eels a year. Emma herself had bequeathed Kirby Cane in Norfolk to St Edmund's.3 A cause and/or the result of Canute's foundation could have been the boom in the cult of St Edmund which, the liturgical calendars show, occurred in the lozos.4 However, too much weight should not be attached to this evidence since at this period the cult of a number of other Anglo-Saxon saints underwent spectacular development.5 Canute's claim in fact rests on Hermann's unsupported evidence — tendentious evidence written over eighty years after the supposed act of foundation. There is no mention of it in the forged diploma in Canute's name, nor in that in William the Conqueror's name which was probably also fabricated for the 1081 trial and includes an apparently reliable account of the proceedings.6 When, as Hermann records, Lanfranc (sometime in the loyos) held an inquiry to try to settle the dispute between the Bury monks and the bishop, he appealed for evidence to the aged abbot of Ramsey, ^Elfwine (1043—79/80), whose memory stretched back to the time of Canute; he testified, without any allusion to the foundation story, that already in those days St Edmund's had enjoyed its privileges.7 And neither of the contemporary authorities, the Encomium Emmae Reginae and the Anglo-Saxon Chronicle, mention any such matter - nor does the later but often authoritative Florence of Worcester (^.1118). It seems possible that the Bury monks adopted Canute as their 1. Sawyer, Anglo-Saxon Charters, no. 980. Printed B. Thorpe, Diplomatarium Anglicum JEvi Saxonici (London, 1865), pp. 305—8. 2. Harmer, Writs, pp. 140-1, 433-5, nn. passim. 3. Ibid. pp. 148, 159-60 (no. 17), 440. 4. St Edmund is in the calendars of Christ Church, Canterbury (1012—23), anc^ *n both of those of New Minster, Winchester (1023—35, 1035 respectively); English Kalendars before A.D. 1100, i, Texts, ed. Francis Wormald (Henry Bradshaw Society, Ixxii, 1934), pp. 124, 138, 180. St Edmund also has a special mass in the sacramentary of c. 1020-2 5 which was almost certainly composed at Ely; The Missal of Robert of Jumieges, ed. H. A. Wilson (Henry Bradshaw Society, xi, 1896), pp. xxviii, 225-6. For the provenance of the sacramentary see J. B. L. Tolhurst, 'An examination of two Anglo-Saxon manuscripts of the Winchester School: the Missal of Robert of Jumieges, and the Benedictional of St. ^Ethelwold/ Archaeolgia, Ixxxiii (1933), 27-41. St Edmund's omission from the calendar is almost certainly accidental; Tolhurst, loc. cit. p. 30 and n. i. 5. See C. E. Fell, 'Edward king and martyr and the Anglo-Saxon hagiographic tradition', Ethelred the Unready: Papers from the Millenary Conference, ed. David Hill (British Archaeological Report, British Series, lix, 1978), p. 7. 6. Although Douglas's argument in favour of the authenticity of the diploma is unconvincing, his contention that its account of the trial is sound seems indisputable; Feudal Documents from the Abbey of Bury St Edmunds, p. xxxiv. 7. regis jussu Lanfrancus Cantuariae praesul, hac pro re terminato ibidem novem comitatuum ccetu, ^Ifwino Ramesiensi abbate tune pleno dierum, ac sene, cujus testimonium ex tempore regis Cnuti prolatum, voceque novem comitatuum obfirmatum; abbatia viguit praenominata tune temporis libertate testificata; Hermann, p. 65.
92
Legends, Traditions and History in Medieval England
putative founder in the late eleventh century. Since he was the benefactor of their own and other monasteries he could be seen as a latter-day King Edgar, the patron of English Benedictinism. The monks' objective would have been to increase the abbey's reputation and in particular to assert its independence of the bishop of East Anglia. To have a royal founder, as well as a royal patron saint, was particularly prestigious. In addition, to be under royal protection and at the king's disposal, to be in effect a royal eigenkloster, ensured immunity from any other authority, notably the diocesan's. Since Abbot Baldwin's former home, St Denis, was such a royal foundation and one throughout its history closely associated with the French monarchy, he must have appreciated the advantages of the royal connection. Nor would this idea have represented any break with the abbey's tradition since it already had an established link with royalty. Possibly the earliest royal benefactor of the community at Beodricesworth was King Edmund (opinions are divided about the authenticity of his charter).1 As has been seen, Canute himself was probably a benefactor of St Edmund's, and his wife Emma certainly was. But perhaps at no time throughout its history were St Edmund's ties with royalty closer than under Edward the Confessor. The latter regarded St Edmund as his ancestor (two of his surviving writs refer to St Edmund as 'my kinsman'),2 and was a substantial benefactor. He granted the abbey regalian rights, including the right to have a mint;3 most important was the grant to the second abbot, Leofstan, of the liberty of eight and a half hundreds where abbatial authority virtually replaced the king's;4 this liberty, the charter states, had previously been held by Edward's mother, Emma.5 And in 1065 he appointed the St Denis' monk, his own physician, Baldwin, to succeed Abbot Leofstan. David Knowles has already suggested that Edward brought with him from Normandy the idea that a ruler might exercise proprietary rights over an abbey under his patronage, and that he regarded Westminster as his eigenkloster.6 It seems that he saw St Edmund's in a similar way. The question arises why Abbot Baldwin would have preferred Canute to Edward the Confessor as the putative founder of St Edmund's. One answer would obviously be that Canute offered the abbey a more ancient tradition. The notion could have been suggested by Canute's well-known generosity to the East Anglian monasteries, which included his grant to St Edmund's, and because of Canute's wife, Emma; she was not only a benefactor 1. See Sawyer, Anglo-Saxon Charters, no. 507, for further references. 2. Harmer, Writs, pp. 160-1 (no. 18), 164—5 (no- 24)3. Ibid. pp. 150-1, 165-6 (no. 25). 4. For the grant see ibid. pp. 145—7, T 54~5 (no- 9)> 435~95. For Queen Emma's 'miniature shire' of the eight and half hundreds, see H. W. C. Davis, 'The liberties of Bury St Edmunds', ante, xxiv (1909), 418-19. 6. Knowles, Monastic Order, pp. 5 79—80. For the protection and freedom enjoyed by a royal eigenkloster, see ibid. pp. 569, 584.
Legends and Traditions concerning Bury St Edmunds
93
of St Edmund's but also it was her estate, the liberty of the eight and a half hundreds, which became the nucleus of the abbey's property. It remains to discuss the part played by ^Elfwine, bishop of Elmham 1022—1023 x 1038. Hermann asserts that he, 'a monk of Ely and lover of the monastic life', advised Canute to introduce monks at Beodricesworth. V. H. Galbraith and Florence Harmer have already indicated their surprise that the bishop should have advised Canute to establish a highly privileged monastery within his own diocese.1 Even supposing that the whole story is true and Hermann merely mistook the date, it is odd that he should admit ^Elfwine's part. As apologist for the Bury monks against the East Anglian bishop, it was risky to ascribe a crucial role in their abbey's foundation to the diocesan. Such an admission would be particularly surprising if, as I have suggested elsewhere, the monks actually revised the Passio during the dispute to reduce the bishop's part in St Edmund's origins.2 Possibly the solution of these problems is to be found in the propaganda purveyed in support of the bishop's case. In the course of the dispute Losinga's party produced a tendentious history of the East Anglian see and of the church of Beodricesworth.3 Having alleged that Felix, a colleague of St Augustine, founded a cathedral there dedicated to the Virgin Mary (it has no reference to St Edmund), it states that Bishop ,/Elfwine (called ^Elfric presumably in error) replaced the secular clerks by monks and that after Canute's death King Harold transformed the cathedral priory into an abbey. The passage concerning ./Elfric's foundation reads: Is [i.e. Canute ] Alfricum sic monachum et priorem Elyensis monasterii ascitum in ecclesia apud Beodrichesworth pontificem elegit, obsecratus ut clericos qui ibidem irreligiose vivebant expelleret et monachos constitueret.4
Losinga's tract, therefore, makes ^Elfwine the virtual founder. In this it is supported by the Liber Eliensis, which states that ^Elfwine ipsius [Canuti] regis precepto in Betricheswrde primum monachorum adduxit catervam .. eisque affluenter subsidia detulit,ausilium impendente Thurchillo comite, insuper rebus et ornamentis ipsi loco de sua parte collatis quamplurimis, eterne libertati donavit . .5
However, in this instance little weight can be put on the evidence of the Liber Eliensis: it is a late authority (it was compiled in Henry IPs reign); the passage cited appears to be based on Hermann, only 1. Galbraith, 'East Anglian see', p. 225 and n. i; Harmer, Writs, p. 140 (Galbraith refers to the bishop of East Anglia as '^Elfric' instead of 'vElfwine', presumably having been misled by his source, the 'Losinga' tract, for which see infra). 2. 'Baldwin', p. 72. 3. Printed Galbraith, 'East Anglian see', pp. 226-8. For the manuscript see infra, p. 22 and n. 5. 4. Ibid. p. 226. 5. Liher Eliensis, p. 1 5 5 .
94
Legends, Traditions and History in Medieval England
^Elfwine is substituted for Canute as the prime mover; and, as will be seen, the compiler's information was contaminated by Norwich propaganda. The idea of attributing such a role to ^Elfwine could well have originated among Arfast's apologists; by demonstrating St Edmund's abbey's initial debt to the diocesan, it strengthened the bishop's case against the Bury monks in their struggle for exemption. Hermann might have heard the story when in Arfast's service, brought it with him to Bury, and there incorporated it, suitably modified, in the De Miraculis. What happened to the clerks who formed the pre-monastic congregation at Beodricesworth is problematic. Hermann indicates that one member of the previous congregation, ^Ethelwine, the sacristan, remained and joined the new community. His evidence accords with that of the Bury interpolator of Florence of Worcester who asserts that Canute expelled only those clerks who did not take the habit: presbyteros vero qui inibi inordinate vivebant aut in eodem loco ad religionis culmen erexit, aut datis aliis rebus de quibus abundantius solito victum et vestitum haberent, in alia loca mutavit.1
However the evidence of neither Hermann nor the interpolator is trustworthy on this point. Hermann claims that it was ^Ethelwine who carried St Edmund's bier to safety in 1010, and who later, when Abbot Leofstan opened the coffin, identified the body as that of St Edmund. He was, therefore, a vital witness that in Hermann's day the body in the shrine was indeed the martyr's.2 Thus ^Ethelwine's role suggests the possibility that Hermann 'invented' him to serve his propagandist purpose.3 The statement of the Bury interpolator of Florence of Worcester is equally suspect. It could be based on Hermann's evidence, or more probably on that of Florence of Worcester himself. The latter describes how St Oswald constituted the cathedral priory at Worcester. (His account derives ultimately from one in either ^Elfric's or Wulfstan's Vita S. JEthelivoldi - their respective texts are at this point, as in so many other places, almost identical — which tells how St Ethelwold expelled the clerks from the Old Minster, Winchester.) Florence asserts that St Oswald. clericos Wigorniensis ecclesiae monachilem habitum suscipere renuentes de monasterio expulit; consentientes vero hoc anno, ipso teste, monachizavit.4
It became the accepted tradition at Bury, probably by Henry IPs 1. Memorials •, i. 341—2. 2. Hermann, pp. 52-3. Cf. pp. 40-6 passim. 3. y£thel wine's name itself is suspect; see Gransden, 'Baldwin', pp. 73, 74, and infra, p. 21, n. 2. 4. Chronicon ex Cbronicis, ed. Benjamin Thorpe (English Historical Society, 1848-9, 2 vols.), i. 141. Cf. ^Ifric's Vita S. jEthelwoldi in Chronicon Monasterii de Abingdon, ed. Joseph Stevenson (Rolls Series, 1858, 2 vols.), i. 260-1, and Wulfstan's Vita S. jEthelwoldi in Acta Sanctorum, i August, p. 92.
Legends and Traditions concerning Bury St Edmunds
95
reign, that Canute, as a preliminary to introducing Benedictine monks at Beodricesworth, expelled all the clerks.1 (Apparently the tradition, which does not concern us here, that Canute formed the clerks, twelve in number, into a gild, the dou^egild^ which later Abbot Baldwin provided with a parish church, originated at about the same time.)2 However, as in the case of the story of partial expulsion, that of wholesale expulsion cannot be substantiated. It first appears in the Losinga tract, which states that Canute appointed JElftic (sic for ^Elfwine) to the see of Beodrices worth ut clericos .. expelleret? Losinga's apologist may well have borrowed the idea from Florence of Worcester who, in his turn here using the Anglo-Saxon Chronicle, states (5^.969) that King Edgar S. Dunstano Dorubernensis, et B. Oswaldo Wigorniensis, et S. ^Ethelwoldo Wintoniensis ecclesiae episcopis praecepit, ut, expulsis clericis, in majoribus monasteriis per Merciam constructis monachos collocarent.4
Thus, there is no reliable evidence either that only some of the clerks were expelled, or that all were. A corollary of the accepted tradition of total expulsion was that the monastery was settled by a colony of monks from elsewhere. The full story of such colonization of Beodricesworth does not appear at Bury until the late fourteenth century, when the Life and Miracles of St Edmund (now Bodley Ms. 240) was compiled. This vast compilation includes a narrative describing the abbey's foundation. The relevant passage states that in 1020 Bishop ^Elfwine 'quosdam de Hulmo sancti Benedicti, et quosdam de ecclesia Eliensi monachos ibi collocauit', citing (in the margin) as its authority 'the chronicles of Ely'. It proceeds, here citing 'the chronicles of Holme': Placuit enim [Canute] regi, ut de ecclesia beati Benedicti de Hulmo antiquitus constituta pars dimidia fratrum ad ecclesiam Beodricensem dirigeretur. Ea vero tempestate xxvj. monachi, viri religiosi et bonae famae, in ecclesia de Hulmo Deo et sancto Benedicto famulabantur. De ilia namque parte dimidia fratrum ad Beodricesworthe directa vir prudens et honestus Uvius erat, qui in ecclesia beati Benedicti officium prioratus agebat, et in ecclesia S. Edmundi apud Beodricesworthe abbas extitit primus . .5
A passage almost identical to, and possibly the source of, the citation 'from the chronicles of Ely' is in the Liber Eliensis (temp. Henry II).6 The substance of the citation 'from the chronicles of 1. 2. 3. 4. 5. 6.
Bury Customary, p. 123 (cf. pp. xl, xli). Ibid. p. 59, n. 6 for further references. See also Douglas, Feudal Documents, pp. 161-2. Supra, p. 13. Chronicon ex Chronicis, i. 141. Memorials, i. 359. Liber Eliensis, p. 15 5.
96
Legends, Traditions and History in Medieval England
Holme' is in the two chronicles composed at Holme in the late thirteenth and early fourteenth centuries respectively, that is 'John de Oxenedes' and the so-called Cronica Buriensis* and also in the register of Holme, of c. 1300.2 The story at this stage has a detail not in the Life and Miracles of St Edmund; it states that Holme supplied St Edmund's with half of its own books, vestments, sacred vessels and the like. But there is no trace at Bury of any tradition that monks from Ely and Holme colonized St Edmund's abbey until the late thirteenth century. Neither Hermann nor the interpolator of Florence of Worcester have any such story. Nevertheless, the interpolator does mention Ufi: he asserts that Canute 'praefecitque eis patrem et abbatem nomine Uvium, virum scilicet humilem, modestum, mansuetum et pium.'3 This passage, however, is suspiciously like the one in Florence of Worcester about the appointment by St Oswald of the first prior of Worcester: 'eisque Ramesiensem coenobitam Wynsinum, magnae religionis virum, loco decani praefecit.'4 The earliest mention that Ely helped colonize St Edmund's is the passage in the Liber Eliensis* but thereafter there is silence until the compilation of the Life and Miracles of St Edmund. As in the case of the Ely tradition, the one connecting Holme with the colonization of St Edmund's first appears elsewhere than at Bury itself, in fact in a Norwich source - the Losinga tract; this states that Bishop ^Elfwine appointed a monk of Holme to be prior of his new cathedral priory at Beodricesworth ('Holmensis monasterii monachum priorem constituit').6 The story appears at last at Bury late in Henry Ill's reign, in a register, the Liber Albus.1 There a short historical narrative has the same story as was to occur shortly afterwards in the Holme sources. The statement by a recent scholar that the special position afforded by the Customary of St Edmund's (^.1234) to the abbot of Holme in the chapter at Bury testifies to Holme's part in the foundation of St Edmund's abbey, is not sound.8 The relevant passage states: 'Our prior shall yield place to no bishop or abbot in chapter, save to our
i.Chronica Johannis de Oxendes, ed. H. Ellis (Rolls Series, 1859), p. 19, and Memorials, iii. i. For the provenance of the Cronica Buriensis see infra, pp. 240-4. 2. BL MS. Cotton Galba E II, fo. 36*; printed in St Benet of Holme 1020-1210. The Eleventh and Twelfth Century Sections of Cotton MS. Galba E ii. The Register of the Abbey ofSt Benet ofHolme, ed. J. R. West (Norfolk Record Society, ii, iii, 1932, 2 vols), i. 35-6, and in Dugdale, Monasticon, iii. 135. This together with the confraternity between Bury and Holme (see infra, p. 9 7), is the source of the account of the foundation of Bury in Davis, 'The monks of St. Edmund', pp. 232-3. 3. Memorials; i. 341. 4. Chronicon ex Chronicis, i. 141. 5. Supra, p. 95 . 6. Galbraith, 'East Anglian see,' p. 226. 7. BL MS. Harley 1005, fos. 35, 3 5 v . Cf. Gransden, 'The 'Cronica Buriensis' and the abbey of St Benet of Hulme', p. 78 and n. 6. 8. Davis, 'The monks of St. Edmund', p. 233.
Legends and Traditions concerning Bury St Edmunds
97
own abbot or to the abbot of Holme and the like, who by right ought to hear the secrets of the chapter, and to have competence to correct any abuses, according to our customs.'1 It will be noticed that the reference to the abbot of Holme is followed by words 'and the like', presumably indicating other heads of houses to whom similar respect was due. In fact the passage probably concerns the heads of those houses with which Bury had entered into confraternity. The surviving text of the confraternity between St Edmund's and St Augustine's, Canterbury, stipulates that if the abbot of one house visited the other, he should enjoy the same rights in that house as he did in his own.2 By the mid-thirteenth century Bury had confraternities with Peterborough, St Albans and Westminster as well as with St Augustine's.3 There is no positive evidence that it had one with Holme before ^.1234, the date of the Customary, although it certainly did late in Henry Ill's reign.4 The passage cited from the Customary suggests that there was an earlier confraternity. Indeed the text of the extant Bury/Holme confraternity could well be based on an earlier one. It includes the provision that any monk who committed a crime (barring treachery to his house or abbot) and, therefore, fled or was expelled, might take refuge in the other house where he would live on equal terms with the monks there. This kind of provision was characteristic of early thirteenth century confraternities, or 'confederacies', as Knowles calls such agreements; Knowles describes them as 'mutual insurance policies' and tentatively attributes their proliferation in the last half of the twelfth and early thirteenth century to the 'democratic' movements which then took place in a number of houses — including Bury.5 These could result in the flight or exile of dissidents for whom a confederate house would provide asylum.6 There is, therefore, no evidence of any tradition at Bury that St Edmund's abbey was colonized from Holme until late in Henry Ill's reign. This, combined with the even later appearance at Bury of the tradition that Ely also supplied monks, indicates that the credibility of the colonization story needs investigation. The monks of Ely 1. Prior noster nulli episcopo, nulli abbati cedat in capitulo nisi abbati proprio vel abbati de Hulmo seu tali, qui secreta capituli de iure debeat audire et excessus regulariter sciat secundum consuetudinem nostram corrigere; Bury Customary, p. 18, lines 7-10. 2. Historiae Anglicanae Scriptores X, ed. Roger Twysden (London, 1652), p. 1843. 3. A mid-thirteenth century list of the confraternities enjoyed by St Edmund's is in Harley 1005, £250. R. M. Thomson, The Archives of the Abbey of Bury St Edmunds (Suffolk Records Society, xxi, 1980), p. 18, dates the index of charters, in which this list occurs, to after 1254. 4. Harley 1005, fos. 35, 35 v . A text is also in the 'Cronica Buriensis' (Memorials, iii. 2) and in the Holme register (West, St Benet of Holme 1020-1210, i. 3). 5. Knowles, Monastic Order, pp. 474-5 and nn. For Bury see A. Gransden, 'A democratic movement in the abbey of Bury St. Edmunds in the late twelfth and early thirteenth centuries', Journal of Ecclesiastical History, xxvi (1975), 25—39. 6. St Albans had cells to which the abbot could exile dissidents; Abbot John (1195-1214) did so (Gesta Abbatum Monasterii Sancti Albani, ed. H. T. Riley (Rolls Series, 1867-9, 3 vols.), i. 251). Bury had no cells. Cf. infra pp. 103, 243-4.
98
Legends, Traditions and History in Medieval England
might have claimed that their abbey helped colonize St Edmund's in order to gain reflected glory for their house; in the course of the twelfth century St Edmund's became one of the most renowned monasteries in England. However, it is certainly possible that Ely, an ancient and famous house little over twenty five miles from Beodricesworth, did provide monks for the new community, especially if Bishop ^Ifwine, a monk of Ely, was instrumental in the foundation. Nevertheless, the lack of early evidence testifying to Ely's part is disturbing. In the case of Holme's participation the weight of probability is against the tradition. St Benet's was neither famous nor particularly near to St Edmund's, being nearly fifty miles away. Situated in the marshes on the Norfolk coast about eight miles north of Norwich, it emerged in the course of the eleventh century and enjoyed in Anglo-Saxon and post-Conquest times a merely local reputation. The only reliable evidence for its pre-Conquest history are a few wills containing bequests to the abbey.1 (The charter of King Canute may well be a forgery, and that of Edward the Confessor certainly is.)2 Scholars have accepted that Canute founded the abbey in 1019, although the earliest authority for this belief is a brief notice in 'Oxenedes'.3 The latter, indeed, reads like a contrivance designed to make plausible the claim, which immediately follows, that Holme monks were sent in 1020 to form part of the new community established by Canute at Beodricesworth. The tradition that monks from Holme formed part of the first monastic community at Beodricesworth could have been the result of misidentification. Ramsey abbey, as well as Holme, was dedicated to St Benedict. (Although its full dedication was to St Benedict and St Mary, charters of the Anglo-Saxon period usually call it St Benedict of Ramsey.)4 It is certainly possible that Ramsey, one of the most influential centres of the tenth century monastic revival, helped colonize St Edmund's monastery at Beodricesworth. It had played a 1. Whitelock, Wills, nos. 25, 28, 31, 33, 34 (p. 90). 2. Sawyer, Anglo-Saxon Charters, nos. 984, 1055. See West, St Benet of Holme 1020-1210, i. 1-2 and n. 5, 3-5. For the post-Conquest charters of Holme, which are of interest for tenurial history, see F. M. Stenton, 'St Benet of Holme and the Norman Conquest', EHR 37 ^1922), 225-35. 3. VCH, Norfolk, ii. 330; Knowles and Hadcock, Medieval Religious Houses, England and Wales, p. 75; West, op. cit. ii. 191. The principal sources cited by the KCHfor the early history of Holme and the succession of its abbots are: the history of Holme to 1275 in BL MS. Cotton Nero D II (printed in Oxenedes, pp. 267-76); 'Oxenedes'; the Bury/Holme confraternity (see supra). All these sources are, at least in their present form, late thirteenth century and should not be relied on. They are also cited for the abbatial succession in David Knowles, C. N. L. Brooke and Vera London, The Heads of Religious Houses, England and Wales, 940—1216 (Cambridge, 1972), p. 67. The latter cites in addition the more detailed information given by William Worcester, writing in 1479. Worcester had a particular interest in Holme; in the fifteenth century it was popular with the East Anglian nobility, and one of its patrons was his own patron, Sir John Fastolf. See William Worcestre, Itineraries, ed., with an English translation, J. H. Harvey (Oxford Medieval Texts, 1969), pp. 2, 92, 220-2, 232-4. 4. See Harmer, Writs, nos. 58-60, and Whitelock, Wills, nos. 31, 39.
Legends and Traditions concerning Bury St Edmunds
99
crucial role in St Oswald's constitution of Benedictine communities at Worcester and Winchcombe. St Oswald's precise method of procedure with regard to Worcester is debatable. According to the Ramsey chronicler,Wynsige, the first prior of Worcester, had previously been a member of the clerical congregation there whom Oswald had sent to Ramsey for instruction before returning him to Worcester.1 But Florence of Worcester asserts that he was a monk of Ramsey.2 Florence is probably right, since Winchcombe would provide a parallel; St Oswald sent Germanus, prior of Ramsey, to reform the clerical congregation at Winchcombe, appointing him abbot.3 Ramsey's reforming activity did not end with St Oswald's death. It was continued by his successor at Ramsey, Abbot Eadnoth (993- 1006) who established Benedictine communities at Slepe (afterwards St Ives, Hunts.)4 and at Chatteris.5 And later still, a monk of Ramsey, ^Elfweard, was abbot of Evesham (^.1014—1035). 6 Besides its contribution to the actual establishing of Benedictine monasteries, Ramsey took an active part in fostering the cult of saints, a preoccupation no doubt encouraged by the example of Fleury, allegedly the resting place of St Benedict himself. Ramsey's lay patron, ^Ethelwine the ealdorman, acquired for it the relics of two Kentish martyrs, SS. Ethelbert and Ethelred,7 and Abbot Eadnoth shared St Oswald's interest in relics; this is shown by his 'invention' of the body of St Ivo, preliminary to the foundation of the monastery of Slepe.8 And Abbot .^Elfweard of Evesham almost certainly translated the relics of St Wigstan to Evesham and bought the relics of St Odulf for the abbey.9 To promote the growth of cult centres, Ramsey's greatest scholar, Abbo's pupil Byrhtferth, apparently wrote saints' Lives. He was probably the author of the Life of Ramsey's patron saint, St Oswald,10 and of St Ecgwin, patron saint of Evesham.11 Just possibly he also wrote a Passio of the Kentish martyrs, SS. Ethelbert and Ethelred.12 (There were also Lives of St 1. Ramsey Chron. pp. 41-2. 2. Chronicon ex Chronicis, i. 141. 3. Ramsey Chron. p. 42. 4. Goscelin, Vita S. Ivonis, Migne, PL, civ. cols. 85 et seqq.; Liber Eliensis, p. 141 and n. i; Ramsey Chron. pp. 114-15. For the endowment of St Ives and Chatteris see C. R. Hart, 'Eadnoth, first abbot of Ramsey, and the foundation of Chatteris and St Ives', Proceedings of the Cambridge Antiquarian Society, Ivi—Ivii (1964), 61—7. 5. Liber Eliensis, 141 and n. 5. 6. Chronicon Abbatiae de Evesham, ed. W. D. Macray (Rolls Series, 1863), p. 81. 7. Ramsey Chron. p. 5 5 . 8. Supra, n. 4. 9. Chron. Evesham, pp. 36-8, 81-5, 314. 10. See Michael Lapidge, 'The hermeneutic style in tenth-century Anglo-Latin literature', Anglo-Saxon England, iv (1975), pp. 9 I ~311. Michael Lapidge, 'Byrhtferth and the Vita S. Ecgwini Mediaeval Studies, xli (1979), 331-5312. Michael Lapidge, 'Byrhtferth of Ramsey and the early sections of the Historia Regum attributed to Symeon of Durham', Anglo-Saxon England, x (1981), 119-20.
100
Legends, Traditions and History in Medieval England
Neot, patron saint of St Neot's,1 and St Ethelbert, patron saint of Hereford and Hoxne.)2 An important part in the expansion of Ramsey's influence was played by its lay patron, the ealdorman y£thelwine. He and his family not only endowed Ramsey but also Crowland and St Neots.3 He died in 992 but his work was continued by Abbot Eadnoth who, with the help of his own family, endowed Slepe and Chatteris.4 Thus Ramsey's influence spread in the Fenlands and farther afield, to Worcester and the Severn basin. In East Anglia, however, its expansion was not without setbacks. King Edgar's death (in 975) was followed by a reaction against the monasteries, and when ^Ethelwine the ealdorman died (991), the ties between Ramsey and its satellites were weakened. But most serious were the renewed Viking attacks of the early eleventh century culminating in the battle of Assandurr in 1016, in which the abbot of Ramsey was killed.5 It is against this background that the origins of Bury must be seen. Clearly the links between Ramsey and Beodricesworth were close. Not only was Ramsey less than forty miles away, but also some of its estates adjoined those of 'St Edmund's.6 In the late 9808 the Ramsey monks were already sufficiently interested in Beodricesworth as a cult centre to ask Abbo to write the Passio of its patron saint, a work which can be seen as a prototype of Byrhtferth's hagiographies. St Edmund figures in the late tenth century metrical calendar of Ramsey, which indicates that he must also have been in Ramsey's (now lost) liturgical calendar.7 His martyrdom is also recorded in the late tenth century annals probably compiled at Ramsey.8 And after the Norman conquest the monks of Bury must have believed that some knowledge of their abbey's early history was preserved at Ramsey, for they appealed to Abbot ^Elfwine to give evidence at Lanfranc's inquest. There is indeed no direct evidence from Ramsey that it helped colonize St Edmund's. There is no reference to such an event in the Ramsey chronicle. However, the chronicle is a comparatively late authority, belonging to the reign of Henry II, and does 1. It has been suggested that this Life of St Neot was composed shortly after the saint's translation to Huntingdonshire towards the end of the tenth century; Hart, 'East Anglian chronicle', pp. 265-7, 278~92. Ibid. pp. 264—5. 3. Ibid. p. 279. 4. Supra, p. 99 n. 4. 5. Ramsey Chron. p. 118. 6. Hart, 'East Anglian chronicle', p. 177. 7. See the forthcoming article by Michael Lapidge, 'A metrical calendar from Ramsey', Revue Benedictine, xcv (1985). 8. C. R. Hart, 'The Ramsey Computus'>EHR Ixxxv (1970), 42. The 'Ramsey annals' were probably copied from the 'York annals', a copy of which possibly reached Ramsey through the agency of St Oswald, founder of Ramsey and archbishop of York; see Lapidge, 'Byrhtferth of Ramsey and the early sections of the Historia Regum', pp. 115—16. The version of the 'York annals' preserved by the Durham chronicler does not include the notice of St Edmund's martyrdom; Hart, loc. cit. p. 37.
Legends and Traditions concerning Bury St Edmunds
101
not mention Ramsey's foundation of the albeit very minor houses at Slepe and Chatteris. And there is one piece of evidence which does suggest the possibility that some such tradition did linger at Ramsey into the Anglo-Norman period. As observed above,1 the first hint of the Holme story, the statement that Bishop ^Elfwine appointed a monk of Holme to be prior of his new monastery at Beodricesworth, is in the Losinga tract. Herbert Losinga before he became bishop of Thetford had been abbot of Ramsey for four years. Perhaps he knew a tradition that St Benedict of Ramsey helped colonize St Edmund's, but substituted St Benedict of Holme as the mother house for propagandist reasons. (My suggestion that the Norwich party substituted St Edmund for St Ethelbert as the patron saint of the church of Hoxne, would provide a parallel to the substitution of St Benedict of Ramsey for St Benedict of Holme.) It would have been inexpedient for Losinga in his struggle with the Bury monks to associate the origins of their abbey with so powerful a house as Ramsey and one, moreover, outside his diocese. St Benedict of Holme, having little prestige and being within his own diocese, was open to no such objections. Meanwhile it seems that the monks of Bury were either ignorant or indifferent, or both, concerning the provenance of the original occupants of their abbey. Ignorance is a perfectly plausible hypothesis. The story of ^thelwine, the sacristan of the clerical congregation at Beodricesworth, bearing St Edmund's bier to London for safety in 1010 is, as I have argued elsewhere,2 unlikely to be true at least in its present form. Nevertheless, it surely preserves some tradition that sometime during the Viking incursions the martyr's body had been moved to a safer place. In fact the incursions must have disrupted the life of the community at Beodricesworth, and by doing so disrupted its corporate memory.3 Therefore, if the monastery was already in existence, knowledge of its origins could have been destroyed. The haziness of the monks' recollections in AngloNorman times, and the absence of pre-Conquest written evidence 1. Supra, pp. 95-6. 2. 'Baldwin, p. 74. However, I do not sufficiently emphasise there the complexity of the relationship between the legend of St Edmund and that of St Cuthbert. The earliest source for the latter's cult, the Historia de Sancto Cuthberto (in Symeon of Durham, Opera Omnia, ed. Arnold, i. 196-214) is mid-tenth century, but the earliest surviving text is mid-eleventh century. The legend was elaborated by Symeon in his Libellus, c. 1100 (ibid. i. 3—135) and again by Reginald of Coldingham in the mid-twelfth century (supra, p.87n. 6). Thus the St Cuthbert legend evolved at the same time as the St Edmund one, and mutual cross-fertilization must be expected, especially as the bishop of Durham presided over the translation of St Edmund in 1095 (cf. Battiscombe, op. tit. p. 46). Nevertheless, it seems fair to assume that the story of ^Ethelwine the sacristan carrying St Edmund's bier was influenced by the St Cuthbert legend, and not vice versa, since there is no convincing evidence that yEthelwine the sacristan was a real person, while ^Ethelwine, bishop of Durham, certainly was. 3. For the decline of monastic fervour and the monks' sufferings, which were real though passing, in the late tenth and early eleventh centuries, owing to the Viking raids, see Knowles, Monastic Order, p. 69.
102
Legends, Traditions and History in Medieval England
about the abbey's earlier history, are demonstrated by the appeal in the trial in the royal court in 1081 to the memory of the aged Abbot ^Ifwine of Ramsey. Moreover, since their main preoccupation was to establish the royal origin and connections of the abbey, the provenance of its first inmates was not their concern. The questions arise how the Bury monks learnt the Holme story and why they adopted it late in Henry Ill's reign. The information could well have reached them from Norwich at that period, since there was a close historiographical connection between Norwich and Bury; from 1258 to 1263 the Norwich chronicler extracted from the chronicle composed at St Edmund's in the last half of the thirteenth century.1 Probably Bury lent a copy to Norwich for transcription and might well have received historical material in exchange.2 There is also evidence that the sacrist of Bury, William of Hoo (i 280—94), with whose office the composition of the last section of the Bury chronicle is to be associated, had previously been precentor of Norwich.3 His migration to Bury, perhaps a result of the Norwich riots of 1272, would have created another historiographical link.4 The vehicle for the transmission of the Holme story could have been the Losinga tract. A possible indication that this was the source is the fact that the only known copy of the tract is in the Bury register, the Liber Albus, in a hand dateable to the 12608 and izyos. 5 Although the tract only states that Bishop ^Elfwine made a monk of Holme prior of the new monastery, this could be the seed of the story. The Bury monks, who of course believed that the new monastery was an abbey and not a cathedral priory, would have substituted 'abbot' for 'prior', Supplied the name Ufi, since they believed he was their first abbot, and presumably invented the other details. The Bury monks probably adopted and elaborated the story of colonization from Holme in order to provide historical justification for the exceptionally binding terms of the confederacy which they made with Holme in the last decade of Henry Ill's reign.6 The 1. See Bartholomew Cotton, Historia Anglicana, ed. H. R. Luard (Rolls Series, 1859), pp. lii-lviii and The Chronicle of Bury St Edmunds 1212-1301, ed., with an English translation, A. Gransden (Nelson's Medieval Texts, 1964, henceforth referred to as Bury Chron.), p. xxvii. 2. For an example of an early manuscript of the Bury chronicle being 'edited' for use by the chronicler of another house, in this case Peterborough, see Bury Chron. p. xliv. 3. Cotton, Hist. A.ngl. p. 149; cf. A. Gransden, Historical Writing in England, i, c.jj 0—1307 (London, 1974), pp. 399-400. 4. Both Bartholomew Cotton and the Bury Chronicle have graphic (but not identical) accounts of the attack on Norwich cathedral. Cotton, Hist. Angl. pp. 146-9, Bury Chron. pp. 50-1. 5. BL MS Harley 1005, fos. 197, i97 v . Printed Galbraith, 'East Anglian see', pp. 226—8. This text is by a scribe active at Bury in the 12605 and 12705, who wrote a number of items in Harley 1005, for example, the best surviving text of the Bury Customary, besides additional customs compiled after 1267/8 (Bury Customary, p. xxxviii, Appendix II). The text of the Losinga tract has a late fourteenth century gloss by Henry Kirkstead (Galbraith, op. cit. pp. 226, 228), for whom see R. H. Rouse, 'Bostonus Buriensis and the author of the Catalogus Scriptorum Ecclesiae', Speculum, xli (1966), pp. 471-99 (esp. p. 484). 6. Supra, p. 97 and n. 4.
Legends and Traditions concerning Bury St Edmunds
103
confederacy has an apparently unique provision: in case of disaster from fire, war or other grave misfortune, half the monks of the stricken house might reside in the other house, as full members of the community until their own house was restored to prosperity. The earliest text of the confederacy is in the hand of the abovementioned scribe of the Losinga tract, which dates it to the 12608 and i zyos. Indeed, it is very likely that the extant confederacy was drawn up in or after 1264, as a result of the troubles with the town which beset the abbey then and thereafter. At the time of the Barons' War and again during the widespread disorders which accompanied the deposition of Edward II in 1326/7, the townsmen of Bury constituted a serious threat to the abbey's security, opposing the monks with actual violence.1 The Bury monks must also have suffered vicarious alarm in the summer of 1272, when the citizens of Norwich attacked the priory and set fire to the cathedral. In such circumstances the remote site of Holme could provide a welcome asylum to frightened monks from Bury.2 It is in fact known that the sacrist and other monks fled there in 1327 and stayed until it was safe to return. This inquiry into the legends and traditions concerning the origins of St Edmund's abbey does not prove that they are false, but it does suggest that they are open to grave doubt. The hagiographical character of the Passio, the subsequent absence of Anglo-Saxon sources apart from the charters and wills of Edward the Confessor's reign, and the propagandist bias of Anglo-Norman and later sources combine to warn us against rash acceptance of these legends and traditions. They also developed elsewhere than at Bury, and wherever they grew they too were susceptible to extraneous pressures. Such legends and traditions, far from remaining isolated, contributed to the formation of those evolving at Bury itself. The legend that St Edmund was martyred and first buried at Hoxne seems to have originated as part of the propaganda used by the bishop of East Anglia in the struggle with Bury. Once that conflict was resolved, the legend could gain acceptance at Bury with impunity; by the late fourteenth century it had become an undisputed element in St Edmund's hagiology. The tradition of Bishop ^Elfwine's role in the abbey's 'foundation' may also derive from the bishop's propaganda; perhaps Hermann adopted it, suitably modified, because he thought, in Nfiew of its plausibility, that it would be rash to jettison it outright. Similarly, the seed of the tradition that St Edmund's was colonized from St Benet of Holme is to be found in the bishop's propaganda. It was only adopted at Bury in the late thirteenth century when the 1. See M. D. Lobel, Borough of Bury St Edmunds (Oxford, 1935), pp. 126-8, 142-4. 2. Gransden, 'The 'Cronica Buriensis' and the abbey of St Benet of Hulme', pp. 81-2. In similar circumstances the monks of St Albans used their cell at Tynemouth - some of them took refuge there during the Peasants' Revolt; Thomas Walsingham, Triistoria Anglicana^ ed. H. T. Riley (Rolls Series, 1863-4, 2 vols.), i. 469. See infra, pp. 240 and n. 6, 243-4.
104
Legends, Traditions and History in Medieval England
monks' conflict with the town gave them a motive for fostering it. In the absence of evidence corroborating the accepted account of the origins of St Edmund's abbey, we are justified in suggesting an alternative'which fits the facts at least as well. I suggest that Ramsey was responsible for the reform of the clerical congregation at Beodricesworth. (The dedication of Holme to St Benedict suggests that it too owed its origin as a Benedictine monastery to Ramsey.) This reform could have happened during St Oswald's rule at Ramsey, in connection with Abbo's composition of the Passio. But, since the spectacular growth in the cult of St Edmund did not take place until the I020S, it seems more likely that the reform was the work of St Oswald's successor, Abbot Eadnoth. Perhaps Eadnoth, imitating St Oswald's practice as exemplified at Worcester and Winchcombe, sent a monk of Ramsey to reform the clerical congregation at Beodricesworth, appointing him abbot. Life in St Edmund's new monastery would have been disrupted by the Viking incursions, but have recovered under Canute. At this point we are, so to speak, on dry land. Florence records that the church was dedicated in 103 2,! and charters show that both Canute and Emma made grants to St Edmund's. More lavish endowment followed in the reign of Edward the Confessor, who seems, indeed, to have given the monastery his especial patronage. But although Abbot Baldwin exploited this royal connection after the Norman Conquest in the struggle against the bishop of East Anglia, its importance soon began to decline; ultimately it was the pope who was to become the abbey's principal protector.2 These hypotheses are, like the generally accepted legends and traditions concerning the origins of St Edmund's abbey, unprovable on the present evidence. However, their plausibility at least demonstrates that the abbey's origins are obscure and problematical - the alleged certainty about them is chimera; as is the case with a number of other pre-Conquest foundations, they deserve further scholarly attention.
1. Chronicon ex Chronicis, i. 189. 2. Knowles, 'Essays in monastic history, IV - the growth of exemption', pp. 210-13 passim. I am indebted to Mr Christopher Hohler and Dr Michael Lapidge for reading the typescript of this article and for making valuable comments and suggestions, especially about my discussion of the St Edmund legend. I alone, however, am responsible for any errors and for the opinions expressed.
1 David harping, from St Wulfstan's portiforium. (Cambridge, Corpus Christi College, MS 391 Jo. 24v)
This page intentionally left blank
4
Cultural Transition at Worcester in the Anglo-Norman Period
The Norman Conquest posed the monasteries of England with a problem, how to preserve the continuity of their religious traditions. Abbacies falling vacant were filled with Normans, and neither they nor the ecclesiastical authorities readily recognised the claim of the Anglo-Saxon saints to sanctity.1 As Archbishop Lanfranc said: 'These Englishmen among whom we are living have set up for themselves certain saints whom SHORTENED TITLES USED GestaPontifaum: GestaRegum: GRANSDEN: Hemingi Chartularium: MACRAY, Chron. Evesham: Memorials ofSt Dunstan: SOUTHERN (1963): SOUTHERN (1970): Vita Wulfstani: WEAVER, Chron. Worcester:
Willelmi Malmesbiriensis Monachi de Gestis Pontificum Anglorum ed. N.E.S.A. Hamilton (Rolls Series, LII, 1870) Willelmi Malmesbiriensis Monachi de Gestis Regum, ed. W. Stubbs (Rolls Series, XC, 1887, 1889) A. Gransden, Historical Writing in England c. 550-c. 1307 (London 1974) Hemingi Chartularium Ecclesiae Wigorniensis, ed. T. Hearne (Oxford 1723) Chronicon Abbatiae de Evesham, ed. W.D. Macray (Rolls Series, XXIX, 1863) Memorials ofSt Dunstan, ed. W. Stubbs (Rolls Series, LXIII, 1874). R.W. Southern, St Anselm and his Biographer (Cambridge 1963) R.W. Southern, 'The Place of England in the Twelfth Century Renaissance', Medieval Humanism and Other Studies (Oxford 1970) The Vita Wulfstani of William of Malmesbury, ed. R.R. Darlington (Camden Society, XL, 1928) The Chronicle of John of Worcester 1118-1140, ed. J.R.H. Weaver (Anecdota Oxoniensa, Oxford 1908)
1 Cf. SJ. Ridyard, 'Condigna veneratio: Post-Conquest Attitudes to the Saints of the Anglo-Saxons', Anglo-Norman Studies, ix, Proceedings of the Battle Conference, 1986 (Woodbridge, 1987), 179-206.
108
Legends, Traditions and History in Medieval England
they revere. But sometimes when I turn over in my mind their own accounts of who they were, I cannot help having doubts about the quality of their sanctity'.2 The Anglo-Saxon saints were on trial. Walter, the Norman abbot of Evesham (1077-1104) put the saints' relics preserved in his abbey to the test of fire: only those spared by the flames were to be accounted genuine — luckily they proved non-inflammable. 3 By the second generation after the Conquest, Anglo-Norman writers had come to the rescue of the Anglo-Saxon saints. The loss of their reputations would have been damaging to the monasteries. The cult of saints was lucractive business. Relics attracted pilgrims and they gave offerings to the monks, an important source of income.4 At Christ Church, Canterbury two monks, first Osbern and then Eadmer, wrotes Lives of a number of the saints connected with the cathedral. They included material relating to the history of Christ Church itself.5 Here they were following the practice of earlier Anglo-Saxon hagiographers who, because they treated the patron saint of a monastery as the persona., as it were, of the place, had used hagiographies to some extent a vehicles for local history. The scholar who did most to salvage the reputations of the AngloSaxon saints and to write the histories of the houses associated with them was William of Malmesbury. In 1125 he published his Gesta Pontificum Anglorum, which was virtually a survey of ecclesiastical England. It is divided into sections, one for every bishopric, and each section ends with an account of the religious houses, including Lives of their saints - thus nearly half of his account of Worcester is devoted to the life and miracles of St Wulfstan (bishop 1062-1095).6 Indeed William's declared purpose was to remedy the deficiencies of the Anglo-Saxon historians and hagiographers; since Bede, he asserts, they had neglected to record their country's past and the achievements of its great men.7 Malmesbury Abbey, together with the great abbeys of the Severn valley, including Worcester, was in the early 12th century the centre of a remarkable cultural revival.8 Other writers besides William were engaged
2 The Life ofSt Anselm Archbishop of Canterbury by Eadmer, ed., with an English translation, R.W. Southern (Nelson's Medieval Texts, 1962), 50-2. Cf. Gransden, 105. 3 Macray, Chron. Evesham, 323-4. 4 See G.R. Cheney, 'Church-Building in the Middle Ages', Bulletin of the John Rylands Library, XXXIV (1951-52), 29, 32. 5 Gransden, 127-31. 6 Gesta Pontificum, 279-89, 301-3. 7 Ibid., 4, and Gesta Regum, 1,2. 8 For scriptural studies in the Severn Valley see Hugh Farmer, 'William of Malmesbury's Commentary on Lamentations', Stadia Monastica, IV (1962), 284-5; for the contributions to the study of science by Abelard of Bath see C.H. Haskins, Studies in the History of Mediaeval Science (Cambridge 1924), 20-42, and by Walcher of Malvern see below p. 111 and nn. 25, 27. For the historians see below.
Cultural Transition at Worcester
109
in establishing links with the past. At Evesham the prior Dominic wrote the Life of the patron saint, Ecgwin, and of two other pre-Conquest saints, St Odulf and St Wistan, whose relics were preserved in the abbey.9 Moreover, he or another monk of Evesham wrote a history of the abbey from its foundation in the 8th century until the rule of Abbot Walter.10 This history resembles in form Bede's Lives of the Abbots ofWearmouth and Jarrow. The author had no doubt read the work; certainly Bede's Ecclesiastical History was studied at Evesham. Already in 1073 or 1074, it had inspired two of the monks to go with the prior of Winch combe (which was at that time ruled jointly with Evesham) to visit the holy places of ancient Northumbria and emulate the asceticism of the Bedan saints. At the persuasion of Walcher bishop of Durham they refounded Jarrow, and two years later a group from the south refounded Wearmouth. 11 Bede was also studied at Worcester. Wulfstan dedicated a church to Bede. His Life., which will be discussed later, notes that after the dedication, Wulfstan preached a sermon 'which sparkled with the same eloquence as had formerly moved the tongue of Bede, . . . that prince of letters among the English people'.12 And later when Nicholas, who was prior of Worcester from c. 1113 to c. 1124, wrote to Eadmer (of Christ Church, Canterbury) on the rights of the see of York in Scotland, he cited the Ecclesiastical History as evidence.13 A recent scholar has written: 'Of all the Old English monasteries, Worcester was the most successful in preserving its links with the past'.14 The cathedral priory flourished during the Anglo-Norman period. This may have been partly because its fame and prosperity were well established in early Anglo-Saxon times. It was already a cultural centre in the time of King Alfred who had summoned Waerferth bishop of Worcester and three of his fellow Mercians, Plegmund (later archbishop of
9 Macray, Chron. Evesham, 1-17, 313-37. See J.C. Jennings, 'The Writings of Prior Dominic of Evesham', E.H.R. LXXVII (1962), 298-304. 10 Macray, Chron. Evesham, 1-98. For the suggestion that Dominic wrote the History, including the Life of Abbot ^Ethelwig which it incorporates (for which see R.R. Darlington, '^Ethelwig, abbot of Evesham', E.H.R. xlviii (1933), 1-22, 177-98) see Jennings, op. cit., 302-3. His case, however, rests on cross-references between Dominic's Life of St Ecgwin and the History, which could have been inserted by Thomas of Marlborough who revised and continued the History in the early 13th century. Therefore, his arguments do not vitiate those I put forward against Dominic's authorship in Gransden, 113 (cf. ibid., 89, 111-13, and M.D. Knowles, Monastic Order in England (Cambridge 1963), Appendix VIII, 704-5. 11 For the post-Conquest monastic settlement of the north see above/below pp. 7-8, with further references. 12 Vita Wulfstani, 20. 13 Councils and Ecclesiastical Documents, ed. A.W. Haddan and W. Stubbs (Oxford 186971), II, 203, 204. 14 Southern (1970), 168.
110
Legends, Traditions and History in Medieval England
Canterbury), ^Ethelstan and Werwulf, to help him in his educational reforms: at his order Werfrith translated the Dialogues of Pope Gregory from Latin into Anglo-Saxon.15 The standard of literacy at Worcester was probably maintained at least for some years after Alfred's death.16 And during the 10th-century monastic revival it again achieved pre-eminence during the episcopate of St Oswald who became bishop of Worcester in 961 and held the see together with that of York from 972 until his death in 992. The priory had the benefit both of St Oswald's religious reputation and of his administrative ability which created the immunity of the Oswaldslow. However, during the Danish domination in the early llth century Worcester suffered badly. The area was ravaged by Swein, and the cathedral was despoiled and finally sacked.17 It owed its recovery to Wulfstan, one of the greatest Anglo-Saxon bishops and one whose rule at Worcester bridged the Norman Conquest. Having been loyal to the old order, he transferred his support to William I shortly after the Conquest. He was high in the favour of both William I and William II and was often at the royal court. Locally he helped enforce order: he assisted in suppressing the baronial revolts of 1075 and 1088 and played an important part in the government of the shire, acting as a judge at least once.18 His power in the area was only rivalled by that of ^Ethelwig abbot of Evesham from 1059 to 1077. Wulfstan contributed much to the welfare of the cathedral priory. He regained estates lost during the disruptions caused by the Danes and by the Norman Conquest,19 and, in order to record the priory's rights and property, he commissioned the subprior Hemming to write a cartulary. 20 This included a short biography of Wulfstan himself which is devoted mainly to his work for the priory.21 According to legend, soon after the Norman Conquest Wulfstan had nearly been deposed by Archbishop Lanfranc in the presence of William I in an ecclesiastical council at Westminster. He was charged with 'simplicity and ignorance' but a miracle saved him; he refused to surrender his staff to Lanfranc, but placed it on the tomb of Edward the Confessor, from whom he had received it. The stones opened and held the end fast, so that
15 Asser's Life of King Alfred, ed. W.H. Stevenson (Oxford 1904, reprinted 1959 with a contribution by Dorothy Whitelock), 62-3 (ch. 77), 303-5. 16 The evidence for this is the excellent series of legal records in English bearing the names of the ealdorman ^Ethelred and of his wife ^Ethelflaed, King Alfred's daughter; P.M. Stenton, The Latin Charters of the Anglo-Saxon Period (Oxford 1955), 45. 17 For a summary of the history of Worcester in the 10th and 11th centuries see V.C.H., Worcestershire, II, 95-7. 18 Vita Wulfstani, xxvii-xxviii. 19 See Hemingi Chartularium, II, 406-8, 418-19. Cf. Vita Wulfstani, xxvi. 20 See Hemingi Chartularium, I, 1. Cf. Vita Wulfstani, xlii. 21 Hemingi Chartularium, II, 403-8. Cf. Gransden, 87 and n. 158.
Cultural Transition at Worcester
HI
no one but Wulfstan himself could take it up: this was interpreted as indicating that God wanted him to retain the bishopric.22 In view of known facts about Wulfstan it seems unlikely that the charge of ignorance was justified. He started his monastic career as schoolmaster,23 and indeed later contributed both directly and indirectly to Worcester's cultural achievements. It was he who was responsible for the introduction at Worcester of the new movement in science which resulted from the acquisition by north-western Europe of the learning of the Arabic world and the Iberian peninsular.24 This knowledge reached Worcester from Malvern where Walcher, a Lotharingian, was prior from 1091 to his death in 1125. He was noted as a mathematician and astronomer and used the works of the converted Spanish Jew, Peter Alphonso.25 Already by 1092 Walcher was using the astrolabe and writing on scientific matters. His treatises show his transition from the cumbersome Roman fractions to the more precise measurements of degrees, minutes and seconds. He describes his observation of a lunar eclipse he saw in Italy in 1091 and, in more detail, one he saw in England in 1092.26 A manuscript (now Bodleian, Auct. F.I.9) containing his lunar observations, the most up-todate researches of the Spanish astronomers and some excellent diagrammatic illustrations, was made at Worcester c. 1130.27 Possibly Wulfstan was also interested in chronology, an interest which could well have originated with the study of Bede, the first English chronologist,28 and been fostered by his friendship with the learned Robert bishop of Hereford (who like Walcher of Malvern was a Lotharingian). Their friendship is attested by a story given in rather different versions by a number of authorities. William of Malmesbury relates that when Wulfstan was on his death-bed, he appeared in a vision to Robert telling him to hasten to Worcester if he wanted to see him alive
22
The story first appears in Osbert of Clare's Vita beati Edwardi Regis, ch. XXIX, ed. Marc Bloch in Analecta Bollandiana, XLI (1923), 64-6. Cf. Vita Wulfstani, xxxi-xxxii. On Wulfstan's alleged simplicity see Emma Mason, 'Change and Continuity in EleventhCentury Mercia: the Experience of St Wulfstan at Worcester', Anglo-Norman Studies, VIII, Proceedings of the Battle Conference, 1985 (Woodbridge 1986), 169-70, 173-4, and idem, 'St Wulfstan's staff: a legend and i;s uses', Medium SEvum, LIII (1984), 166-7. 23 Vita Wulfstani, 9. 24 Southern (1970), 168. 25 For Walcher see C.H. Raskins, 'The Reception of Arabic Science in England', E.H.R. xxx (1915), 56-9. 26 See the next note. 27 Extracts from Bodleian MS Auct. F.I.9, including Walcher's descriptions of the lunar eclipses, are printed in Haskins, loc. cit., 57-9. One of the illustrations (f. 88) is reproduced by Southern (1970), plate VIII. (For another scientific illustration from Worcester see page 8 and n. 75.) For the Worcester provenance of the manuscript see N.R. Ker, Medieval Libraries of Great Britain (London 1964), 208 and nn. 5, 6. 28 For the influence of Bede on Wulfstan see above p. 109.
112
Legends, Traditions and History in Medieval England
again. On the way Robert had another vision of Wulfstan, who announced his own death and said that Robert too would die shortly. Robert arrived at Worcester, buried Wulfstan and was presented by Prior Nicholas with the lambswool cope 'in which [Wulfstan] was wont to ride', as a token of their long friendship. But before Robert could leave he was taken ill with cold shivers and died 'with tears and sighs'.29 Bishop Robert had imported from Germany a copy of the universal chronicle of Marianus Scotus, an Irish monk, who lived as a recluse first at Fulda and then at Mainz.30 His chronicle was written in annalistic form and covered the period from the Creation until A.D. 1082 (in which year Marianus probably died). It has one pecularity: Marianus believed that the accepted date of the Incarnation was twenty two years too late — and demonstrated the chronological justification for his view at some length. Therefore he dated events doubly; according to Dionysius and according to his own computation.31 The introduction of Marianus's chronicle into England was important because it was the most comprehensive and up-to-date universal history yet known here. Its information on European affairs was particularly valuable because they had been neglected by the Anglo-Saxon chroniclers. It received the immediate attention of scholars. Bishop Robert himself wrote an abbreviated version,32 and William of Malmesbury described it both in the Gesta Pontificum and in the Gesta Regum, commenting that there was much to be said for its chronological system although the latter had not been widely accepted.33 No doubt on account of Bishop Robert's friendship with Wulfstan, a copy soon reached Worcester, and there at the command of Wulfstan, one of the monks began interpolating it with material relating to England, bringing it up to date.34 Wulfstan himself was interested in history. He was regarded as an authority on the pre-Conquest period. Eadmer, enumerating his virtues, concludes: 'He was above all imbued with knowledge of the ancient
29
Gesta Pontificum, 302-3. Cf. Vita Wulfstani, xv, n. 2, 60-3, and Florentii Wigorniensis Monachi Chronicon ex Chronicis, ed. Benjamin Thorpe (English Historical Society, 1848, 1849), II, 36-7. 30 William of Malmesbury records that Robert first brought Marianus's chronicle to England: Gesta Pontificum, 301. 31 See W.H. Stevenson, 'A Contemporary Description of the Domesday Survey', E.H.R. XXL (1907), 73 and n. 3. 32 See ibid, 72 and n. 2, 73. 33 Gesta Pontificum, 301; Gesta Regum, II, 345. 34 This is the chronicle from the Creation of the World to A.D. 1140, which to 1118 is attributed to Florence of Worcester (see n. 29 above and pp. 6 etsqq.) and thereafter to John of Worcester (Weaver, Chron. Worcester). For the statement that Wulfstan commissioned the chronicle see below p. 117 and n. 62.
Cultural Transition at Worcester
113
customs of the English'.35 And Lanfranc's successor Anselm wrote to him to discover what the customs were relating to the archbishop of Canterbury's rights to dedicate churches on his estates outside his diocese.36 Moreover, Wulfstan was interested in his cathedral priory's pre-Conquest history, particularly in the saints associated with it. During the trial of the dispute between him and Thomas archbishop of York, concerning the latter's claim to supremacy over the bishopric of Worcester and to some of its property, Wulfstan had with him the Lives of St Dunstan (bishop of Worcester from 960 until 962 when he succeeded to Canterbury) and of St Oswald.37 He translated the relics of St Oswald when he rebuilt the cathedral between 1084 and 1088 or 1089,38 and refounded the priory originally established by Oswald at Westbury on Trym.39 After his death the cult of St Oswald continued to flourish at Worcester, and in the early 12th century the monks commissioned Eadmer to write a new Life of St Oswald. However, the researches of hagiographers were not the most characteristic feature of cultural continuity at Worcester (Oswald's career was too recent and well known to need much study). On the other hand two literary forms typical of the Anglo-Saxon period were revived in the cathedral. Wulfstan himself contributed indirectly to the revival of one, and directly to the revival of the other. The first which will be considered is the biography arranged according to subject matter; Wulfstan's own posthumous biography was arranged in this way. The second literary form in question is the vernacular annals (that is the Anglo-Saxon Chronicle); Wulfstan himself commissioned one of his monks to write Latin annals which were partly a translation of these. First, the bipartite biography. The earliest secular biography written in Anglo-Saxon England, Asser's Life of King Alfred, was divided into two parts. The first relates mainly to Alfred's public life, his career as king, and the second part to his private life — his character, family and the like. This division of the biography into two parts according to subject matter originated with Suetonius's Lives of the Caesars and was transmitted to Asser by Einhard's Life of Charlemagne.^1 It was also used in two
35
6.
36
Eadmeri Historia Novorum in Anglia, ed. Martin Rule (Rolls Series, LXXXI, 1884), 45-
Eadmer preserves a copy of the letter; ibid., 46-7. Vita Wulfstani, 25. 38 Ibid., xxxix, 52. 39 Ibid., xxxix, 52. 40 Eadmer's Vita S. Oswaldi is printed in Historians of the Church of York, II, ed. James Raine (Rolls Series, LXXI, 1886), 1-59. Cf. p. below. 41 For Asser's arrangment of his work and use of Einhard see Asser's Life of King Alfred, ed. Stevenson, Ixxix-lxxx, 54 (ch. 72), 294, and Marie Schutt, 'The Literary Form of Asser's Vita Alfred?, E.H.R. LXXII (1957), 209-20passim. 37
114
Legends, Traditions and History in Medieval England
subsequent royal biographies written in Anglo-Saxon times: the biography of King Canute (the Encomium Emmae Regime)42 and the Life of King Edward (the Confessor).43 On the other hand most of the sacred biographies, that is Lives of saints and ecclesiastics (often the same people), written in this period were unitary - they were arranged chronologically and not divided according to subject matter. However, one saint's Life, ^Elfric's Life ofSt Ethelwold is divided according to subject matter:44 first it deals with St Ethelwold's parentage, infancy and career; then with his character, death and miracles. This approximates to the Suetonian division. It was the practice among the Anglo-Saxons for the biography of a famous bishop to be written by someone who knew him well, within a few years of his death. This practice survived the Conquest —such biographies were written of Leofric of Exeter (1050-1072), 45 Gundulf of Rochester (1077-1108),46 Anselm of Canterbury (1093-1109)47 - and of Wulfstan of Worcester. The latter concerns us here. The author was Coleman, precentor of Worcester Cathedral, who had been Wulfstan's chaplain for fifteen years and the bishop's chancellor.48 For the Life which he wrote sometime between 1095 and 1113, he adopted an arrangement very similar to ^Elfric's Life ofSt Ethelwold. Book I deals with Wulfstan's birth, parentage, education and rule as bishop up to the Norman Conquest. Book II concerns his episcopate after the Conquest, and Book III relates to his private life, his death and miracles. Coleman painted a picture of a saintly bishop of the Anglo-Saxon type, not dissimilar to St Cuthbert. Wulfstan was personally holy and assiduous in his religious life and episcopal duties. Like St Cuthbert he was constantly riding from place to place, fulfilling his pastoral duties. The Life says that c he always carried in his bosom a psalter and book of prayers', and he recited the hours, psalms and prayers with his attendants. 49 Just as St Cuthbert's portable altar has survived, so also has the very portiforium, or breviary, of St Wulfstan. It is a small book, less 42 The Encomium is in three books; the first two are a biography of King Canute (the last relates to Emma and her son Harthacanute). The biography, having recounted Canute's conquests and secular activities, ends with his spiritual life and relations with the church; Encomium Emmae Reginae, ed. Alistair Campbell (Camden Society, LXXII, 1949), 35-9. 43 See The Life of King Edward, ed. Frank Barlow (Nelson's Medieval Texts, 1962), xx. 44 See Gransden, 79. 45 Printed in The Exeter Book of Old English Poetry, ed. R.W. Chambers, M. Forster and R. Flower (London 1933), 8-9. 46 Printed in Henry Wharton, Anglia Sacra (London 1691), II, 273 et sqq. 47 See p. 108 and n. 2 above. 48 For Coleman's career and the composition of the Life of Wulfstan see Vita Wulfstani, viii. 49 Ibid., 95 and n. 2. Cf. ibid., 49, and Dame Laurentia McLachlan, 'St Wulstan's Prayer Book', Journal of Theological Studies, xxx (1929), 176.
Cultural Transition at Worcester
115
than nine inches by six, and is now in Corpus Christi College, Cambridge (MS391) 50 (P1. 1). Coleman also stressed that Wulfstan (like earlier Anglo-Saxon bishops) was of consequence in national affairs.51 And in the Anglo-Saxon manner, Coleman worked the local history of the priory into his biography. He mentions Wulfstan's reforms as prior before he became bishop, and his building activities.52 More especially he gives a detailed account of Wulfstan's dispute with Thomas archbishop of York; thus the Life served as a record of Wulfstan's success and of the independence of the bishops of Worcester from the authority of the archbishops of York.53 But Coleman's debt to the Anglo-Saxon tradition went further than the arrangement of his work and the treatment of the subject matter. He wrote the Life in the Anglo-Saxon vernacular. This may well have been a piece of conscious revivalism because within a generation the monks could no longer understand it. Therefore they asked William of Malmesbury to translate it into Latin — which he did, dedicating his translation to Warin, who was prior from c. 1124 to c. 1143.54 The fact that Coleman wrote in AngloSaxon may also have contributed to the loss of his work — for his text we have to rely on William of Malmesbury's translation.55 It has been suggested that the only copy of Coleman's Life was sent to the pope in the early 13th century as evidence in the proceedings to secure Wulfstan's canonisation and was never returned. 56 Next to be considered is the survival of annalistic historiography at Worcester. The Anglo-Saxon Chronicle itself was firmly rooted there. It is very likely that at least two copies were continued by the monks until after the Conquest. There is good evidence that the so-called D version which went down to 1079 was written at Worcester.57 Moreover, another version 50 See PL 1. For the importance of this picture as evidence for the survival at Worcester of the Anglo-Saxon technique of line drawing see C.M. Kauffmann, Romanesque Manuscripts 1066-1190 (London-Boston 1975), p. 45. The volume is described by McLachlan, loc. cit., 174-7. 51 Vita Wulfstani, 22-4. 52 Ibid., 11, 15,21,52. 53 Ibid., 16-20, 24. 54 Ibid., 1. 55 William of Malmesbury claims (ibid., 2) to have adhered closely to Coleman's text, and Professor Darlington argues that he made few alterations or additions (ibid. ix). I have accepted this view although Mr Farmer contends that William allowed himself considerable latitude; see Hugh Farmer, 'Two Biographies by William of Malmesbury' in Latin Biography, ed. T.A. Dorey (London 1967), 165-74. 56 Vita Wulfstani, viii n. 4, xlvii. 57 The Worcester provenance of the D version of the Anglo-Saxon Chronicle is supported by G.N. Garmonsway; The Anglo-Saxon Chronicle (London 1955), ed. G.N. Garmonsway, xxxvii-xxxix. However Dorothy Whitelock and F.M. Stenton suggest York (The Anglo-Saxon Chronicle, ed. D. Whitelock, D.C. Douglas and S.I. Tucker (London 1961), xv; F.M. Stenton, Anglo-Saxon England (Oxford 1947), 681). Charles Plummer
116
Legends, Traditions and History in Medieval England
now lost, reaching to 1130, appears to have been used by the AngloNorman chroniclers at Worcester who are discussed below.58 Nevertheless at Worcester as elsewhere the Norman Conquest ultimately killed the Anglo-Saxon Chronicle. English was no longer a language used by the literate; it was replaced by Norman French for the less formal kinds of literature, Latin of course remaining the learned language. The survival of annalistic historiography at Worcester was partly the result of the introduction of Marianus Scotus's chronicle, and Wulfstan's command that it should be supplemented and continued, as already mentioned.59 Thus was begun the Worcester chronicle, the so-called Chronicon ex Chronicis (the Chronicle of Chronicles), which was subsequently continued there to 1140. Nevertheless, although Marianus was responsible for the inception of the Worcester chronicle, the Anglo-Saxon Chronicle played a crucial role in its actual composition. It provided the entries on English history which the chronicler interpolated into Marianus's work, and also much of the material after Marianus ended, probably down to 1130.60 Perhaps even more significant, it encouraged the writing of contemporary annals. Marianus, though he brought his chronicle up to date, relied mainly on earlier works; his gift was as a compiler, not a contemporary annalist. The Anglo-Saxon chroniclers, on the other hand, were distinguished as contemporary annalists; from the reign of King Alfred they had added entries to the Chronicle year by year, when news reached them. Such a chronicle was not a formal literary work, but an ever growing record of recent events. This aspect of the AngloSaxon Chronicle was adopted at Worcester; although until the early 12th century the Worcester chronicle is mainly a compilation, from 1121 it is a contemporary record, and from 1130 (when the lost version of the AngloSaxon Chronicle apparently ended), it is wholly independent of all known chronicles. The name of the monk whom Wulfstan commissioned to write it is open to dispute. The chronicle itself has an entry under 1118 reading: On the nones of July died Dom. Florence monk of Worcester. This chronicle of chronicles excels all others because of his profound erudition and studious application.61
suggests Evesham (Two of the Saxon Chronicles Parallel, ed. C. Plummer (Oxford 1892, 1899), II, Ixxvi-lxxvii). 58
Pp. 116-20. P. 112 and n. 34. For the Worcester chronicle cf. below pp. 203-4 and nn. 21-4. 60 See Whitelock, Douglas and Tucker, op. cit. xx. Cf. R.R. Darlington and P. McGurk, 'The 'Chronicon ex Chronicis' of 'Florence' of Worcester and its Use of Sources for English History before 1066; Anglo-Norman Studies, V, Proceedings of the Battle Conference, 1982 (Woodbridge, 1983), 185-96 passim. 61 Weaver, Chron. Worcester, 13. 59
Cultural Transition at Worcester
117
The Worcester chronicle to 1118 has consequently gone under the name of Florence of Worcester. However when sometime early in the 12th century (the exact year is unknown), the chronicler Orderic Vitalis visited Worcester from Normandy, he found a monk called John writing the chronicle. Orderic writes: John was an Englishman by birth, entered the monastery of Worcester as a boy and won great repute for his learning and piety. He continued the chronicle of Marianus Scotus and carefully recorded the events of William's reign and of his sons William Rufus and Henry up to the present.. at the command of the venerable Wulfstan bishop and monk.62
If Orderic had merely stated that John was writing the chronicle at the time of his visit, there would be no problem, because the chronicle itself has evidence that the latter part of it at least was by a monk called John: in the annal for 1139 it has some verses apostrophising the reader and asking for corrections to be made 'if John has offended in any way'.63 However, it is hard to reconcile Orderic's explicit assertion that John started the chronicle, writing at Wulfstan's command, with the entry in the annal for 1118 that Florence was responsible for it up to that date. It could be assumed that Florence wrote to 1118 and John added the continuation to 1140, and that the passage in Orderic was simply the result of confusing the name of John, the writer at the time of Orderic's visit, with that of Florence, the earlier one. However, there is no stylistic indication of a break in the chronicle at 1118 — the whole reads like the work of one man.64 A possible solution is that the chronicle was written by John using material collected by the assiduous Florence. In its present form the chronicle was composed between 1124 and 1140. It contains some information about John himself. It records in the annal for 1132 that he was standing next to Uhtred, precentor of Worcester, at Mass when Uhtred was seized by his final illness; the author was deeply grieved because Uhtred had 'loved me like a foster-father'.65 Probably soon after this the author must have temporarily fallen into disgrace, for he was exiled to Winchcombe Priory; while there he met Henry I's physician Grimbald who told him of the three visions experienced by the king while in Normandy in 1130.66 He must have been back at Worcester
62
The Ecclesiastical History of Orderic Vitalis, ed., with an English translation, Marjorie Chibnall, II (Oxford Medieval Texts, 1969), xxi, 187-8. 63 Weaver, Chron. Worcester, 49. 64 See Gransden, 144 and n. 55, and Martin Brett, 'John or Worcester and his Contemporaries', The Writing of History in the Middle Ages. Essays presented to Richard William Southern, ed. R.H.C. Davis and J.M. Wallace-Hadrill (Oxford, 1981), 104 and n. 3, 110. 65 Weaver, Chron. Worcester, 36. 66 Ibid., 10, 33. For the visions see p. 119 and Pis. 2, 3.
118
Legends, Traditions and History in Medieval England
by 1139 at latest, for on 7 November of that year he was at Lauds in the cathedral when the Empress Matilda's forces entered and sacked the city.67 The Worcester chronicle is significant in the history of historical writing for a number of reasons. It transmitted the Anglo-Saxon Chronicle to the post-Conquest period. Its extensive borrowings from the Chronicle were of course in Latin translation. The translation of the Chronicle into Latin had begun almost at the time of its inception in Alfred's reign — Asser translated the portions relating to Alfred for his Life of King Alfred. Later, in about 1000, ^Ethelweard the ealdorman had translated much of it into Latin for his chronicle.68 and soon after the Norman Conquest a bilingual version, in Anglo-Saxon and Latin, was written at St Augustine's, Canterbury. 69 Nevertheless, despite these earlier translations, it was the Worcester chronicle which contributed most to making the text available in Latin to the Anglo-Normans. Besides transmitting much of the contents of the Anglo-Saxon Chronicle, the Worcester chronicle also encouraged the keeping of contemporary annals in other monasteries. It was comparatively popular among the monks of the day. A copy reached Durham probably soon after 1118, where it formed the basis of the Historia Regum, which is commonly attributed to Symeon of Durham, for the years from 848 to 1118.70 Soon after 1130 copies up to that date were made for the abbeys of Abingdon, Bury St Edmunds, Gloucester and Peterborough.71 The Worcester chronicle had features making it particularly attractive to the monks of other communities. Its literary form was different from the two other important histories written in the Anglo-Norman period, Eadmer's History of Recent Events, and William of Malmesbury's Gesta Regum. Both these works have literary shape, and deal with topics discursively even if this involved disregarding the chronological sequence. They could not, therefore, be interpolated without disturbing the structure. The Worcester chronicle on the other hand has no literary pretensions; it is a purely chronological record of events. Thus it could be interpolated at any point or continued without altering its nature, and so
67
Ibid., 57. See The Chronicle of JEthelweard, ed., with an English translation, Alistair Campbell (Nelson's Medieval Texts, 1962), xxii-xxxvii. 69 The old English text of F (BL MS Cotton Domitian A VIII) is printed in The Anglo Saxon Chronicle, ed. Benjamin Thorpe (Rolls Series, XXIII, 1861), I, 3-329 passim; the Latin text is printed and described F.P. Magoun, jn., 'Annales Domitiani Latini: an Edition', Mediaeval Studies, IX (1947), 235-95. For Thorpe's text see ibid., 235 n. 3. Cf. Garmonsway, op. cit., xli and n. 2. 70 See P. Hunter Blair, 'Some Observations on the Historia Regum attributed to Symeon ofDurham', Celt and Saxon, ed. N. K. Chadwick (Cambridge 1963), 107-11. 71 See Weaver, Chron. Worcester, 5-9 passim. 68
Cultural Transition at Worcester
119
monasteries which acquired copies could interpolate material relating to affairs of local interest and omit entries which exclusively concerned Worcester. For example the Gloucester copy inserts a notice of the death of Roger de Berkeley, a benefactor of St Peter's abbey, in 1137, and an account of King Stephen's visit to Gloucester in 1138 but omits the details of the death of the precentor Uhtred at Worcester.72 Similarly some of the copies were continued - the Peterborough copy goes down to 1295.73 In addition to its usefulness in providing a framework and starting point for the chronicles of other monasteries, the Worcester chronicle has merits as a history. It includes European history and is more than a series of dry annals. To some extent it reflects the cultural achievements of Worcester at that time. It shows an interest in astronomy, having an enthusiastic outburst in favour of the Arab astronomer, Al-Kharismi.74 It has a number of astronomical details; it records, for example, the sun spots seen on 8 December 1128, and has a good diagrammatic picture of them.75 It is also concerned with chronology; it discusses at some length under A.D. 703 and 725 Bede's views on chronology expressed in De Temporibus, and gives chronological details relating to a number of dates.76 Moreover, the Worcester chronicle has in places the merit of readability. It includes some amusing stories. For example, there are three tales of German origin which must have been intended to entertain.77 (It is noteworthy that William of Malmesbury included similar stories, in order, his editor suggests, to float the heavier portions of his narrative.) 78 And there is the graphic account of the visions experienced by Henry I in Normandy which were interpreted as divine admonitions to him to rule more justly. First, as he lay asleep he was threatened by a band of rustics who vanished when he leapt up to punish them. Second, a group of armed knights threatened the sleeping king; when the king attacked them with his sword, he did not wound any. And third, his bed was surrounded by prelates expostulating and waving their staffs, on account of Henry's ill-treatment of the Church. On his return to England he was nearly drowned in a storm in the channel, but was spared on vowing to stop exacting Danegeld for seven years and to go on pilgrimage to Bury St Edmunds. 79 The visions and voyage are illustrated 72
Ibid., 6-7, 25 n. 3, 36 n. 4, 48 n. 4. Printed in Florentii Wigorniensis Monachi Chronicon ex Chronicis, ed. Thorpe (English Hist. Soc., London 1848-9), II, 136-279. 74 Weaver, Chron. Worcester, 53. 75 Ibid., 28. See Southern (1970), 168 and pi. VII. 76 See Thorpe, op. cit., I, 45; II, 36, and Weaver, Chron. Worcester, 34, 53. 77 Weaver, Chron. Worcester, 34-6, 45-8. 78 Gesta Regum, I, xc-xci. Stubbs suggests (ibid., I, Ixxxix) that some of William of Malmesbury's stories of German derivation reached him from Walcher of Malvern; John of Worcester could have obtained his from the same source. 79 Weaver, Chron. Worcester, 32-4. 73
2 The visions of Henry I, from the chronicle of John of Worcester: Henry is beset by peasants and by knights. (Oxford, Corpus Christi College, MS 157, p. 382)
3 The visions of Henry I, from the chronicle of John of Worcester: Henry is beset by bishops; his stormy crossing from Normandy. (Oxford, Corpus Christi College, MS 157, p. 383)
120
Legends, Traditions and History in Medieval England
by four pictures, together covering nearly two full pages, of high quality, testifying to the abbey's artistic achievements (Pis 2, 3). The Worcester chronicle is apparently the earliest surviving illustrated chronicle produced in western Europe.80 The dissemination of the Worcester chronicle must also have been partly the result of Worcester's reputation as a cultural centre. It was in the Anglo-Norman period the venue of scholars. The visit of Orderic Vitalis, who was no doubt attracted by its fame, and the contact of both Eadmer and William of Malmesbury with Worcester, have already been mentioned.81 A key person in Worcester's cultural connections was Prior Nicholas, who was interested in history. Nicholas supplied Coleman with details concerning Wulfstan's piety which Coleman included in his Life.82 He himself figures in one anecdote: it relates how Nicholas disregarded Wulfstan's prohibition of drinking in the priory (because of the brawls to which it gave rise), and was divinely punished by a nightmare which prevented him sleeping until he had confessed to Wulfstan. 83 William of Malmesbury himself visited Worcester and became a personal friend of Nicholas. Evidence of his visit is provided, for example, by his eyewitness description of Wulfstan's tomb.84 Besides at the monks request translating Coleman's Life of Wulfstan into Latin, he also used it as his main source for the account of the see of Worcester in the Gesta Pontificum85 He gives an anecdote about Nicholas as a young man, which he must surely have heard from the prior himself: Nicholas, afterwards prior of Worcester, was a favoured pupil of Wulfstan. [One day when] he was sitting at his feet, the bishop, overflowing with joy, stroked the youth's head with a gentle hand, where the flowing hair deceptively covered a bald patch. 'I think', he said, 'you will go bald'. But the youth, who already in adolescence grieved that he was aging in that respect, lamented the ill fortune of falling hair. 'Why', he said, 'don't yOu hold it on for me?' The bishop replied laughing, 'Believe me, never as long as I live will the remainder fall out'. It was as he said, but in the very week when he departed this life, it all vanished - I don't know where - leaving the scalp bare.86
Eadmer too visited Worcester, perhaps when Nicholas was elected 80
The four pictures are reproduced in Weaver, Chron, Worcester, frontispiece, as well as below pis. B, C. For the importance of the Worcester chronicle as perhaps the earliest example of an illustrated chronicle see C.M. Kauffmann op. a/., pp. 43-50. 81 Pp. 109, 115. 82 See Vita Wulfstani, 51, 52, 54. 83 Ibid., 56. 84 Gesta Pontificum, 288. 85 Ibid., 278-88. Cf. Vita Wulfstani, ix. 86 Gesta Pontificuam, 287.
Cultural Transition at Worcester
121
prior. It may have been on this occasion that the monks commissioned him to write the Life ofSt Oswald?1 In it he mentions that he had seen and handled St Oswald's chasuble, which had been found intact when the saint was translated twelve years after his death, and was still in use at the time of Eadmer's visit.88 Relations between Worcester and Christ Church, Canterbury, were obviously close. Ties had no doubt been strengthened by Wulfstan's victory, with Lanfranc's help, over Thomas archbishop of York, which had put the diocese beyond dispute under the control of the archbishop of Canterbury. 89 Eadmer corresponded with the Worcester monks - in 1123 or 1124 he wrote exhorting them to elect a monk to the bishopric; he pointed out that in those evil times wicked men were seeking to exclude monks from the episcopal bench90 (nevertheless, the king appointed a secular, Simon). The Worcester monks also benefited from Eadmer's work as an historian. Not only did he write the Life ofSt Oswald for them, but also soon after 1121 a copy of his History of Recent Events was at Worcester where it was extracted for the Worcester chronicle.91 But perhaps Eadmer's debt to the Worcester monks was even greater than their debt to him. There is certain evidence that they supplied him with information for his Life ofSt Dunstan, which he wrote partly to correct the errors in the Life by Osbern, Eadmer's predecessor as hagiographer at Christ Church. 92 He acknowledges at some length in the prologue his obligation to Ethelred, monk of Worcester, who had formerly been subprior and precentor of Christ Church.93 There is also independent evidence that Prior Nicholas helped him. A long letter from Nicholas to Eadmer survives.94 It was written in reply to a query concerning the mother of Edward the Martyr, and explains that Edward's mother, ^Ethelflaed, though never crowned queen, was legitimately married to King Edgar: on the other hand his second wife ^Elfthryth was both crowned and legitimately married. Nicholas based his conclusions 'on the ancient authority both of chronicles and of songs which are known to have been composed in the vernacular at that time by learned men, and on the
87
See above p. 113 and n. 40, and Southern (1963), 283 n. 2. Vita S. Oswaldi, 50. 89 Seep. 115 and n. 53. 90 Printed in Wharton, Anglia Sacra, II, 238. Cf. Southern (1963), 286, and D.L. Bethell, 'English Black Monks and Episcopal Elections in the 1120s', E.H.R. LXXXIV (1969), 681 (Bethell reprints Wharton's copy of the letter, ibid., 697-8). 91 For John of Worcester's use of Eadmer's Historia Novorum see Weaver, Chron. Worcester, 3-6passim. For the date of the Historia see Southern (1963), 298-309 passim. 92 See Eadmer's Vita S. Dunstani in Memorials ofSt Dunstan, 162-3. Cf. Southern (1963), 281 n. 2. 93 Memorials of St Dunstan, 163-4. 94 Printed in ibid., 422-4. 88
122
Legends, Traditions and History in Medieval England
testimony of other writings'. This information enabled Eadmer to correct one of Osbern's errors95 — he had stated that Edward's mother was a nun and not Edgar's wife.96 Eadmer also corrected another of Osbern's mistakes, the assertion that in Dunstan's time Worcester cathedral bore the same dedication as it did subsequently, that is to St Mary.97 Eadmer denied this, claiming that it was dedicated then to St Peter and explaining how (in his opinion) the later dedication to St Mary arose. He writes: The holy and religious Oswald, who succeeded to the rule of this church when Dunstan became archbishop of Canterbury, because he could not convert the clerks living there from their depravity, nor drive them hence (since they were of noble birth and powerful), built a church nearby dedicated to Mary mother of God, where he himself would serve with the monks whom he planned to unite with Christ.98 Besides help with his work, Eadmer received advice on his career from Prior Nicholas. In 1120 he was invited by Alexander I, king of Scotland, to become bishop of St Andrews. However, because he demanded to pay obedience to Canterbury, not York, disputes ensued, and he resigned six months later without having been consecrated.99 In the course of the quarrel he wrote to Nicholas for evidence that the archbishop of York had no right to consecrate the bishops of the Scottish sees. Nicholas's reply, which still survives,100 illustrates both his historical interests and ability, and his friendship for Eadmer. He demonstrates, with appeals to precedent, that the archbishops of York had never consecrated the bishops of Scotland. He advises Eadmer to obtain consecration in Rome — and asks him to bring back as many 'white pearls' as possible, and at least four large ones. It can, therefore, be seen that cultural life at Worcester was active during the Anglo-Norman period. The monks played an important part in
95 96
Ibid, 214.
See Osbern's Vita S. Dunstani in ibid., 112. Ibid., 103, 106. 98 Ibid., 197. A similar passage is in Eadmer's Vita S. Oswaldi in Historians of the Church of York, ed. Raine, II, 25. J. Armitage Robinson, 'St Oswald and the Church of Worcester', British Academy Supplemental Papers, V (1919), 3-4 cites the similar explanation for the change of the dedication of Worcester cathedral given by William of Malmesbury in his Vita S. Dunstani (Memorials of St Dunstan, 303-4); he overlooks the testimony of Eadmer which was almost certainly used by William of Malmesbury (see Gransden, 130 n. 170). Dr Eric John cites charter evidence which is apparently incompatible with Eadmer's explanation; see Eric John, 'St Oswald and the Church of Worcester', Orbs Britanniae and Other Studies (Leicester 1966), 234-48. 99 Southern (1963), 236. 100 Haddan and Stubbs, op. cit., II, 202-4. Cf. p. 109 and n. 13. 97
Cultural Transition at Worcester
123
saving the heritage of the Anglo-Saxons from obliteration after the Conquest. Without the work of such men much more would have been swept away by the Normans, for, in the words of St Wulfstan. 'Wretches that we are, we destroy the work of saints because we think in our pride that we can do better'.101
Note On pages 116 and 117 above I state that John of Worcester's chronicle ends in 1140. Dr Martin Brett has pointed out to me that there is no evidence that it did end in 1140 since the earliest manuscript, Corpus Christi College, Oxford, MS 157, ends incomplete. However, it must have been completed before 24 September 1143 since it mentions Henry of Blois, bishop of Winchester, as papal legate 'non tune sed nunc'. Weaver, Chron. Worcester, 38 and n. 7. (Henry was a papal legate from 1 March 1139 to 24 September 1143.)
101
Vita Wulfstani, 52. Cf. Gesta Pontificum, 283.
This page intentionally left blank
5
Prologues in the Historiography of Twelfth-Century England
The medieval tradition of the literary prologue had its roots in Greek and Roman times.1 The classical prologue was adapted in the early Christian era to conform to the new religious ideals.2 In a prologue the author introduces himself to the reader, and tries to put him in a receptive frame of mind; he informs him about the purpose and scope of the work, and demonstrates his own rhetorical skill. To achieve these ends he uses a variety of literary commonplaces, topoi. One species of prologue was the prologue to an historical work. However, although an historical prologue might well have distinctive features, it had much in common with prologues to other classes of work: therefore, generalisations about it may also be appropriate to them. It is the purpose of this paper to examine historical prologues as a genre. The subject will be discussed in two parts. First, the characteristics of historical prologues will be treated generally. Secondly, those w r ritten in twelfth-century England will be examined in more detail, to see how they conform to the norm. Usually a prologue dedicates a work to some important person, and often it claims that that person, or some other named individual, ordered or persuaded the author to write.3 Next the author may well try to disarm the reader by a declaration of modesty.4 He is unequal to the task, unable to write good enough prose to do justice to his subject. (He may even use a 1 For the classical prologue tradition see Tore Janson, Latin Prose Prefaces, Studies in Literary Conventions (Acta Universitatis Stockholmiensis. Studia Latina Stockholmiensia XIII, Stockholm et al., 1964). 2 For the medieval tradition to c.1200 see Gertrud Simon, 'Untersuchungen zur Topik der Widmungsbriefe mittel-alterlicher Geschichsschreiber bis zum Ende des 12. Jahrhunderts', Archiv fur Diplomatik, iv (1958), 52-119, v-vi (1959-60), 73-153. 3 Simon, op. cit., iv, 54-63; E. R. Curtius, European Literature and the Latin Middle Ages, translated by W. R. Trask (London, 1953), p. 85, D. W. T. C. Vessey, 'William of Tyre and the art of historiography', Mediaeval Studies, xxxv (1973), 436-8 (cf. below, pp. 56-7). For the classical origins of this topos see Janson, pp. 116-20. Cf. below, pp. 129, 131-3. 4 Simon, iv, 98-119 passim, Curtius, op. cit., pp. 83-4, Vessey, op. cit., p. 439.
126
Legends, Traditions and History in Medieval England
diminutive of himself or his book.) He promises to write briefly, 5 and in a simple, even rustic, style, but will compensate for his lack of literary skill by his industry-perhaps claiming to have studied by night as well as by day.6 And sometimes he will mention the work of previous historians, thus parading his learning, and discuss his sources.7 He may admit the especial difficulty of writing contemporary history: if he criticises someone he will be accused of malice, but if he praises, he will be accused of flattery, which must be abhorrent to any historian.8 To tell the truth, prologues commonly insist, is the historian's primary duty. 9 His narrative should be accurate, its concrete facts right and, moreover, it should be free from bias. Often an historian declares that his purpose was to preserve the memory of past deeds for posterity.10 This done, history provides examples of good and bad conduct for the reader to emulate or eschew.11 Above all, it shows God's dominion over man, especially how He punishes wrongdoers.12 Such are the principal topoi which tend to occur in historical prologues, treated with varying degrees of elaboration. Some authors, indeed, add little of their own. Therefore, it is debatable whether historical prologues in general are of much value as evidence to the present day historian. If the writer was merely repeating a set form, his prologue, it can be argued, tells us little about his mentality, except that he was well acquainted with the rules of rhetoric, and that he thought it appropriate to write a prologue in rhetorical style. It would tell us nothing reliable about his real thoughts about history and the historian's task, nor about the immediate occasion for his writing. This is the view taken by D.W.T. Vessey in his discussion of
5 6
7 8 9 10 11 12
Bernard Guenee, 'Les premiers pas de 1'histoire de 1'historiographie en Occident au Xlle siecle', Comptes Rendus de I'Academic des Inscriptions et Belles-Lettres (1983), pp. 146. For classical examples see Janson, pp. 124-49 passim, 159. Cf. below, pp. 137-8. For promises of brevity see Simon, op, cit., v-vi, 82-8, Curtius, op. cit., pp. 85, 487-94. For classical examples see Janson, pp. 96-7, 154-5. Cf. below, 138. See especially Bernard Guenee, 'L'histoire entre 1'eloquence et la science. Quelques remarques sur le prologue de Guillaume de Malmesbury a ses Gesta Regum Anglorum', Comptes Rendus de I'Academic des Inscriptions et BellesLettres (1982), pp.359, 367-9, See also Vessey, op. cit., p.441. The topos of historians' industriousness in pursuit of learning had well-established classical origins; see Janson, pp. 97-8, 147-8. Cf. below, pp. 139-40. See Guenee, 'L'histoire entre 1'eloquence et la science', p. 368, idem, 'Les premiers pas', pp. 136-44. For the classical origins see Janson, pp. 97, 152-3, 155-6. Cf. below, pp. 138-9, 141. Simon, op. cit., iv, 88-9, Vessey, op. cit., p. 442-3. Cf. below, p. 143 & nn. 127, 128. Simon, op. cit., v-vi, 89-94, Vessey, op. cit., p. 441. For the classical origins of the 'truth' topos see Janson, p. 67. Cf. below, pp. 128-9, 141-6. Simon, op. cit., iv, 78-81, v-vi, 94-111 passim. Cf. below, p. 134. Simon, op. cit., v-vi, 94-111 passim. Cf. below, pp. 135-6. See below, pp. 135-6, 147-8.
Prologues in the Historiography of Twelfth-Century England
127
William of Tyre's Gesta A ma lrici( written in the 1170s). Thus, he writes: 'A prologue was, above all, a place where rhetoric and commonplace were inevitable and expected'; and again, 'It would be wrong to interpret [William's] prologue as a highly significant statement of his motives and intentions or as a declaration of deeply-felt principle.'13 Dr Vessey questions William's claim that he wrote at King Amalric's command, and also shows that, contrary to William's assertion that he speaks the 'truth', the Gesta is a very sententious work.14 (Similarly, Jeanette Beer has shown that, despite the recurrent claims in their works to veracity, William of Poitiers, the anonymous author of the Gesta Francorumet aliorumHierosolimitanorum, and Villehardouin were much more concerned to persuade readers and listeners of their overall argument than to record events with factual accuracy.15) However, Gertrud Simon, in her detailed study of historical prologues before 1200, is more cautious. It is a problem, she admits, whether a prologue is merely a collection of topoi, or whether the author's own feelings predominate.16 Other recent scholars have decided more positively in favour of the value of historical prologues. J.O. Ward writes: The obvious starting point for an analysis of what twelfth century historians thought they were doing is recognised to be the exordium prefaced to most histories. Recent research suggests that these exordia were meant as quite precise guides to contemporary historical ideas.17 The scholar who has worked most extensively on medieval historical prologues, Bernard Gienee, has no doubt of their value: II est bien vrai que ces morceaux de bravoure que sont souvent les prefaces ont parfois des developpements conventionnels qui justifient le peu d'interet qu'y prennent les lecteurs. Mais souvent aussi un auteur a mis beaucoup de lui-meme dans sa preface, et il y attache une grande importance. .. En fait, dans sa preface, 1'auteur prend souvent soin de dire 1'historien qu'il croit etre et 1'histoire qu'il veut faire. Et comme je voulais approfondir et preciser mes recherches sur ce qu'avaient ete 1'histoire et 1'historiens au Moyen Age, 1'idee s'est imposee a moi qu'il fallait systematiquement etudier ces miroirs plus 13 Vessey, op. cit., pp.440, 444-5. 14 Ibid., pp. 436-55 passim. 15 J. M. A. Beer, Narrative Conventions of Truth in the Middle Ages (Etudes de philologie et d'histoire, xxxviii, Geneva, 1981), chapters 1-3. 16 See Simon, op. cit., iv, 52-4. 17 J.O. Ward, 'Some principles of rhetorical historiography in the twelfth century' in Classical Rhetoric and Medieval Historiography, ed. Ernst Breisach (Studies in Medieval Culture, xix, Medieval Instituate Publications, Western Michigan University, Kalamazoo, 1985), p. 106 and n.23 for further references.
128
Legends., Traditions and History in Medieval England
au moins fideles que sont les prefaces des oeuvres historiques d'occident medieval.18 There are, therefore, divergent views about the value of prologues as historical evidence. In fact, it seems best to treat them with caution, but notwithstanding to take them seriously. Indisputably, they contain many topoi. Guenee points out 19 that Sallust's historical prologues, the earliest ones in Latin, and especially that to Bellum Catalinae, were models of primary importance throughout the Middle Ages. How7 meaningful particular topoi are is often hard to decide. Especially difficult are those concerning an author's state of mind. For instance, we cannot be sure how modest a specific author was, despite any protestations to that effect. Probably he thought he ought to be modest, but if, as well he may, he also shows pride in his calling,20 that would be hard to reconcile with actual humility. And if humility made an historian declare that he would write briefly, for fear of boring, it did not necessarily make him do so. Author's claims that they have written the truth are problematical in cases where their works are biased, and perhaps contain falsehoods. However, the medieval historians' concept of 'truth' differed from ours.21 They considered that the overall truth of a work was more important than the factual accuracy of every detail. The author had to prove that what he believed to be true was correct, for example that God was on the side of the crusaders, or that a ruler was justified in conquering another country if he intended to reform the church. In the interests of such truths an author was justified in omitting discordant facts and filling gaps in knowledge with convenient probabilities. Moreover, the historian in the Middle Ages had different standards of evidence from those of an historian today. He trusted most what he himself had seen, but this of course left much unknown. He, therefore, turned to what people of repute had written or said,22 and accepted the truth of what was probable. Probability, that is Verisimilitude', was not established by 18 Guenee, 'L'histoire entre 1'eloquence et la science', p. 357. 19 Ibid., p. 357, idem, 'Histoire, memoire, ecriture. Contribution a une etude des lieux communs', Comptes Rendus de VAcademic des Inscriptions et BellesLettres (1983), pp. 442-3. 20 See below, pp. 138-9. 21 It must, however, be admitted that exactly what authors meant by 'truth' is often far from clear. The question is discussed by, for example: Beer, Narrative Conventions of Truth in the Middle Ages, esp. pp. 10-11; Roger Ray, 'Rhetorical scepticism and verisimilar narrative in John of Salisbury's Historia Pontificates' in Breisach, op. cit., pp. 77-83 passim', D. J. Wilcox, The Measure of Times Past (Chicago UP, 1987), chapter 8 passim. 22 For the medieval historians' attitude to 'authority' in assessing data see especially Bernard Guenee, Histoire et Culture historique dans {'Occident medieval (Paris, 1980), pp. 129-33.
Prologues in the Historiography of Twelfth-Century England
129
invention, but by careful research and deductions drawn from concrete evidence. It was allowable to insert such probabilities as true facts without warning the reader of type of evidence used. These ideas about truth, and the evidence necessary to establish it, gave an historian considerable latitude. In the service of an overall argument, he could distort the evidence by suppression and unfair emphasis. His authorities themselves might give him false information, and the process of discovering Verisimilitude' was inevitably partly subjective: it was a matter of opinion whether the evidence was sufficient to support any conclusion at all; the line between a reasonable deduction and one derived from inadequate evidence was a fine one.23 An author might decide that, for instance, a certain turn of events was probable, and could, therefore justifiably be presented as concrete fact, simply because he would have liked it, or thought that it ought, to have happened. It is not, therefore, surprising that some narratives are so slanted that it is hard to reconcile them with the prohibition of bias, which was part of the 'truth' topos, even if we consider them in the light of medieval ideas about truth and evidence.24 Nevertheless, despite this reservation, bearing those ideas in mind, we should read with sympathy the claims in prologues to write truth. This is especially so because they are often supported by plenty of evidence, both in prologues and in the works themselves, of great assiduity in the pursuit of information and in the assessment of its accuracy. Indeed, there is much that can be said in defence of the value of historical prologues. They certainly should not be dismissed as mere verbiage. There is no reason to assume that an author did not believe at least most of what he wrote. It is safest to accept the truth of a statement, unless there is strong evidence to the contrary. Thus, an author's claim that he wrote at the request of a named individual should normally be believed (remembering, however, that he may have had more than one motive for writing). And, even if such a statement is false, the identity of the alleged patron would be of interest. Prologues can tell us much about what historians in the Middle Ages thought about themselves and their calling. The very currency of the topoi which the historian had at hand would have helped shape his ideas. His choice of them is itself significant. Changes in the popularity of one or other topos suggest changes in historians' attitudes.25 Thus, Guenee argues that as the twelfth century progressed authors laid less stress on the importance of an eloquent, persuasive style, and more on their industry and pursuit of learning: their tastes were moving from rhetoric to 23 For William of Malmesbury's methods of arguing see below, pp. 144-6. 24 For examples see above, pp. 126-7. 25 On the value of topoi as historiographical evidence see especially Guenee, 'Histoire, memoire, ecriture', pp. 441 et seqq.
130
Legends, Traditions and History in Medieval England
scholarship.26 An author's acknowledgement of debt to previous writers was conventional,27 but the selection of sources cited was his own. And an author could, and often did, include his own observations in the prologue. He might, for instance, explain any distinctive structure he had adopted for his work.28 A prologue, indeed, provided a place for miscellanea and creative writing. Guenee has shown that the historical prologues of the twelfth century are of especial interest. He emphasises the primary importance of Hugh of St Victor's universal history, published in 1130.29 Its influence, Guenee believes, was largely responsible for a shift from eloquent to erudite historiography. Guenee surveys prologues written throughout the medieval West, including England. Our intention here is to look at the principal English examples in greater detail. The twelfth century was, indeed, a golden age of historiography in England, as elsewhere in Europe. Particularly famous historians are Eadmer, William of Malmesbury, Geoffrey of Monmouth, Henry of Huntingdon, Ralph Diceto, Roger of Howden, William of Newburgh, and, at the turn of the century, Gervase of Canterbury. Among the AngloNorman historians Orderic Vitalis, writing in Normandy, should also be numbered. All except Roger Howden30 start with prologues of comparable interest, although of varying lengths and sophistication. Moreover, one of the lesser twelfth-century historians, Alfred of Beverley, begins his Annales (which Gross considered a worthless compilation)31 with an illuminating prologue.32 Clearly authors recognised prologues' usefulness. William of Malmesbury wrote (c.l 125) substantial prologues to each of the five books which together comprise his Gesta Regum,35 and did the same for his Gesta Pontificum.34 Later, towards the end of the century, Diceto included in the prologue to his Abbreviations Chronicorum long passages from the 26 Guenee, 'L'histoire entre 1'eloquence et la science', pp. 357-70 passim. Cf. above, p. 126 & n. 6 and below, pp. 150-1. 27 See above, p. 126 & n. 7 and below, p. 141. 28 See below, pp. 146-50. 29 Guenee, 'Les premiers pas', pp. 137 et seqq. 30 See below, p. 151. 31 Charles Gross, A Bibliography of English History to 1485, ed. E.B. Graves (Oxford, 1975), p. 405 (no. 2795). 32 Aluredi Beverlacensis Annales sive Historia de Gestis Regum Britanniae, ed. Thomas Hearne (Oxford, 1716), pp. 1-10. 33 Willelmi Malmesbiriensis Monachi de Gestis Regum Anglorum Libri Quinque.., ed. William Stubbs (Rolls Series, 1887-9, 2 vols), i, 1-2, 103-4; ii, 283-4, 357-8, 465-6. 34 Willelmi Malmesbiriensis Monachi de Gestis Pontificum Anglorum Libri Quinque, ed. N.E.S.A. Hamilton (Rolls Series, 1870), pp. 3-5,139,208-10,277, 330-1.
Prologues in the Historiography of Twelfth-Century England
131
prologues of a number of previous writers, from Justin to Ivo of Chartres.35 And William of Newburgh made use of the prologue of his HistoriaRegum Anglicarum*6 to accommodate extraneous matter-to discuss a subject about which he apparantly felt very strongly: he inserted a full-scale attack on Geoffrey of Monmouth's Historia Regum Britanniae; this, since Geoffrey's work concerned the ancient Britons, had no relevance to William's history, which did not deal with them at all, but began at the Norman Conquest.37 In our examination of the twelfth-century English historical prologues, notice will be taken of topoi used, and where possible their immediate source suggested. At the same time what the prologues say will be treated as evidence. Consideration will be given to the light they throw on various historiographical matters: the motives, both immediate and long term, for the writing of history; the status of historiography, and its accepted spiritual value and worldly uses: historians' view of themselves and their task; their attitude to, and treatment of, their sources; and the methods they adopted to structure their works. Finally, we will consider whether two of Guenee's general conclusions about the twelfth-century historical prologues of the medieval West apply equally to the English examples: first was the influence of Hugh of St Victor as important in England as on the continent? Secondly, was there a move from rhetoric to scholarship? There are plenty of English examples of prologues which state that an individual work was written in response to a commission or request. Sometimes an author wrote for his monastic superior (or superiors) or fellow monks. Symeon of Durham was ordered (c.1100) to write by his monastic superiors,38 and Orderic Vitalis started work at the command of 35 The Historical Works of Master Ralph deDiceto, Dean of London, ed. William Stubbs (Rolls Series, 1876, 2 vols), i, 20-4. 36 Chronicles of the Reigns of Stephen, Henry II, and Richard I, ed. Richard Hewlett (Rolls Series, 1884-9, 4 vols), i, 11-19. 37 I suggest elsewhere the possibility that William originally wrote the section of the prologue which attacks Geoffrey's Historia (ibid, from p. 11, line 2, 'Qui nimirum', to p. 18, line 18, 'ab omnibus respuatur') as a propaganda tract in the royal interest to discredit the Arthurian legend; the Bretons (like the Welsh) believed that King Arthur would return and lead the British people to victory against their Angevin rulers. They seem to have identified the legendary Arthur with Henry II's grandson, the son of Geoffrey and a daughter of the duke of Brittany. He was born in 1186, baptised Arthur and reared, according to William of Newburgh (ibid., i, 235) 'under the mighty omen of his name'. Near the end of his chronicle William records the Breton rebellion of 1196. See A. Gransden, 'Bede's reputation as an historian in Medieval England', Journal of Ecclesiastical History, xxxii (1981), above, pp. 20-2. 38 The so-called Historia Dunelmensis Ecclesiae in Symeonis Monachi Opera Omnia, ed. Thomas Arnold (Rolls Series, 1882-5, 2 vols), i, 3. The evidence for Symeon's authorship of the Historia is not beyond dispute; A. Gransden,
132
Legends, Traditions and History in Medieval England
Roger, abbot of St Evroult (1091-1123), and completed it for his successor, Guerin des Essarts.39 Eadmer was urged by friends to write the Historia Novorum and its companion volume, the Vita Sancti Anselmi,40 and similarly friends persuaded William of Malmesbury to write Book IV of the Gesta Regum (despite his reluctance because of the difficulties presented by near contemporary historiography).41 And Gervase of Canterbury wrote for 'Brother Thomas and our humble little family'. 42 Some authors wrote for distant friends. John of Salisbury wrote the Historia Pontificalis (? 1164) at the request of his 'dearest friend and master', Peter de Celle, abbot of St Remigius, Rheims.43 And Richard of Devizes, a monk of St Swithun's, Winchester, wrote his Cronicon de TemporeRegisRichardiPrimi'( 192-8) at the request of his friend ('his venerable father, always his master') Robert, formerly prior of St Swithun's, but then a Carthusian monk of Witham.44 It seems likely that occasionally a request amounted to a commission in return for some benefit to the author, or, presumably, if the latter was a monk or regular canon, to his community. The benefit may have been money or patronage. William of Newburgh, himself an Augustinian canon of Newburgh, explains that Ernald, abbot of the Cistercian abbey of Rievaulx, asked him to write.45 William of Malmesbury dedicated the second edition of his Gesta Regum46 and his Historia Novella to Robert earl of Gloucester (whose patronage Malmesbury abbey badly needed during the anarchy of Stephen's reign),47 while Geoffrey of Monmouth
39 40 41 42 43 44
45 46 47
Historical Writing in England, [i], C.550-C.1307 (London, 1974), pp. 115-16 and nn. 66-70 for further references. The Ecclesiastical History of Orderic Vitalis, ed. Marjorie Chibnall (Oxford, 1969-80, 6 vols), i, 130-2. Eadmeri Historia Novorum in Anglia, ed. Martin Rule (Rolls Series, 1884), p. 1; The Life of St Anselm Archbishop of Canterbury by Eadmer, ed. with an English translation R. W. Southern (Nelson's Medieval Texts, 1962), p. 1. Gesta Regum, ii, 357. Chronica in The Historical Works of Gervase of Canterbury, ed. William Stubbs (Rolls Series, 1879-80, 2 vols), i, 89: 'tibi, mi frater Thoma, et nostrae familiolae pauperculae scribo*. loannis Saresberiensis Historia Pontificalis, ed. with an English translation Marjorie Chibnall (Nelson's Medieval Texts, 1956), p. 3: 'Vnde voluntati tue, dominorum amicorumque karissime, libentius acquiescens..'. The Chronicle of Richard of Devizes of the Time of King Richard the First, ed. with an English translation J.T. Appleby (Nelson's Medieval Texts, 1963), p. 1: 'Venerabili patri et semper domino Roberto..'. Richard says he writes partly to remind Robert of his friend; ibid., p. 2. Chrons. Stephen, Henry II, and Richard I, i, 3. The dedication to Abbot Ernald in fact is in a dedicatory letter, preceding the prologue. See Gesta Regum, i, Ivi-lviii; ii, 518-21. The Historia Novella of William of Malmesbury, ed. with an English translation K. R. Potter (Nelson's Medieval Texts, 1955), p. 1. Cf. below, p. 146 and n. 141.
Prologues in the Historiography of Twelfth-Century England
133
dedicated various copies of the Historia Re gum Britanniae to a number of great men, including King Stephen, Robert earl of Gloucester, Waleran count of Meulan, and Alexander bishop of Lincoln.48 One reason why some people liked writing history was because it gave them a worthwile and enjoyable occupation. Prologues express this sentiment in various commonplace ways. William of Newburgh was grateful to Abbot Ernald for saving him from the dangers of idleness.49 Sallust had claimed that the study of history was a more fruitful occupation for one's leisure than to spend it 'in idleness and sloth, or, by turning to farming or the chase, to lead a life devoted to servile [i.e. physical, not intellectual,] employment'.50 William of Malmesbury seems to echo Sallust on this matter (although possibly he also had in mind the Biblical prohibition of 'servile works' on Sunday): in the prologue to Book IV of the Gesta Regum, he writes that, after a period of leisure and silence, his love of learning has returned; he cannot be idle and spend his time on outdoor pursuits and on those unworthy of an educated man.51 Idleness was in any case unacceptable to any good Benedictine monk, since the Rule taught that it 'is the enemy of the soul'.52 Gervase of Canterbury actually quotes the Rule: 'because I know that idleness is the enemy of the soul, I have occupied myself with this work' (i.e. the writing of his chronicle).53 Alfred of Beverley, a secular clerk, wrote his Annales to prevent himself from wasting time. He explains in the prologue that, owing to an interdict (i.e. in 1143), the celebration of the divine office in his church had temporarily stopped. This and other sufferings of the clergy nearly drove him to despair. 'But merciful God saved me from the chasm of desperation, and led me back 48 The Historia Regum Britanniae of Geoffrey ofMonmouth, ed. Acton Griscom (London, 1929, repr. Slatkine, 1977), pp. 219-20. Cf. J.S.P.Tatlock, The Legendary History of Britain (University of California Press, 1950), pp. 436-7, 444. 49 Chrons. Stephen, Henry II, and Richard /, i, 4. 50 Sallust, Bellum Catilinae, iv, ed. T. E. Page, E. Capps and W. H. D. Rouse, with an English translation by J. C. Rolfe (Loeb Classical Library, 1931), p. 8: ' . . . non fuit consilium socordia atque desidia bonum toium conterere, neque vero agrum colundo aut venando servilibus officiis intentum aetatem agere..' Cf. Leviticus 23.7: 'Dies primus erit vobis celeberrimus, sanctusque; omne opus servile non facietis in eo'. I am indebted to Professor Lewis Kelly for calling my attention to this Biblical archetype. 51 Gesta Regum, ii, 357: 'sed dum aliquamdiu solutus inertia vacassem, rursus solitus amor studiorum aurem vellit et manum injecit, propterea quod nee nil agere possem, et istis forensibus et homine litterato indignis curis me tradere non nossem'. 52 Rule of St Benedict, Cap. XLVIII. 1, from the Rule of the Master. La Regie de Saint Benoit, ed. A. de Vogue (Sources Chretiennes, Paris, 1971-7, 7 vols), ii, 598 and note. 53 Gerv. Cant., ed. Stubbs, i, 89-90: 'Et quia novi quod otiositas inimica est animae, otium meum hoc negotio curavi occupare.'
134
Legends, Traditions and History in Medieval England
to reason.' He decided to spend his time on the not ignoble occupation of reading;54 and he goes on to say that what he read in particular was history - Geoffrey of Monmouth's Historia Regum Britanniae. Historiography, however, did more than remove its practitioners from the temptations of idleness. It had other virtues and values. Most importantly, it was a rational study, as Gervase of Canterbury explains.55 Sallust wrote that 'it behoves men not to go through life in silence like beasts', but to strive to excel intellectually.56 Henry of Huntingdon applied this maxim directly to the study of history: knowledge of the past distinguishes men from brutes, whether men or beasts; the former do not know, and do not want to know, their people's origins, nor the history of their country; the latter, of course, are incapable of such knowledge, or of wishing for it.57 History, Henry asserts, makes the past present (the lives and deaths of brutish men are assigned to everlasting oblivion).58 It was commonplace that historiography preserves the past for posterity.59 But why should we remember the past? Above all history demonstrates the workings of Providence on earth. This idea, of Biblical origin and prescribed for the Middle Ages by St Augustine's De Civitate Dei and Orosius' Historia contra Paganos, underlay all medieval Christian 54 Annales, p. 2. 55 Geru. Cant., ed. Stubbs, i, 87: 'nichilque aliud comprehendere nisi quod historiae de ratione videtur competere'. 56 Bell. Cat., i (Loeb edition), p. 2: 'Omnis homines qui sese student praestare ceteris animalibus summa ope niti decet ne vitam silentio transeant veluti pecora, quae natura prona atque ventri oboedientia finxit. Sed nostra omnis vis in animo et in corpore sita est; animi imperio, corporis servitio magis utimur; alterum nobis cum dis, alterum cum beluis commune est. Quo mihi rectius videtur ingeni quam virium opibus gloriam quaerere, et, quoniam vita ipsa qua fruimur brevis est, memoriam nostri quam maxume longam efficere; nam divitiarum et formae gloria fluxa atque fragilis est, virtus clara aeternaque habetur.' This paragraph is quoted here in full because it almost certainly influenced Henry of Huntingdon; see below and next note and p. 65 and n. 63. 57 Dedicatory epistle to Bishop Alexander; Henrici Archidiaconi Huntendunensis Historia Anglorum, ed. Thomas Arnold (Rolls Series, 1879), pp. 2-3: 'Habet quidem et praeter haec illustres transactorum notitia dotes, quod ipsa maxime distinguat a brutis rationabiles; bruti namque homines et animalia unde sint nesciunt, genus suum nesciunt, patriae suae casus et gesta nesciunt, immo nee scire volunt. Quorum, homines quidem illos infeliciores judico; quia quod bestiis ex creatione, hoc illis ex propria contingit inanitione; et quod bestiae si vellent rion possent, hoc illi nolunt cum possint.' 58 Ibid., p. 2: 'Historia igitur praeterita quasi praesentia visui repraesentat;..' Ibid., p. 3: 'Sed de his [hominibus brutis] jam transeundum est, quorum mors et vita sempiterno dotanda est silentio.' 59 Eadmer, Hist. Novorum, p. 1; William of Malmesbury, Gesta Regum, i, 3 (see below p. 69 and n. 101); idem, Hist. Novella, p. 1; Orderic Vitalis, i, 132; John of Salisbury, Hist. Pont., 3; William of Newburgh, dedicatory epistle to Abbot Ernald, Chrons. Stephen, Henry II, and Richard I, i, 3.
Prologues in the Historiography of Twelfth-Century England
135
historiography.60 John of Salisbury gives it explicit expression: chroniclers have always had but one intention, to relate noteworthy matters, 'so that what was done may show the invisible God'.61 One way in which historiography could show God's power was by recording instances of the transitory nature of all earthly glory. This was one of Henry of Huntingdon's favourite themes.62 Sallust had remarked that 'the renown which riches or beauty give is frail and fleeting'. He adds that 'only mental excellence is a splendid and lasting possession'.63 Henry of Huntingdon, on the other hand, saw only spiritual glory, which leads to salvation, as eternal.64 Richard of Devizes also wrote partly to show the fickleness of fortune.65 Moreover, the study of history should do more than increase faith; it should also improve behaviour. One facet of the belief in providential history was the idea that God rewards the good and punishes wrongdoers. This idea was familiar from the Bible (for example from Psalms 7, 36 and 90, and the Book of Amos), and was central in Orosuis' Historia. Moreover, it figures large in Bede's prologue to the Historia Ecclesiastical he hoped that readers would be morally, as well as spiritually, improved by the examples of this aspect of divine power to be found in his history.66 The twelfth-century historians probably had these models in mind in their treatment of the same idea. 'What', asks William of Malmesbury, 'is more to the advantage of virtue, and more conducive to justice, than to learn about God's beneficence to the good, and vengeance on traitors?'67 Similiarly, John of Salisbury asserts that he, like other chroniclers, wrote 'so that examples of [divine] rewards and punishments may make men more zealous in fear of God and the pursuit of justice'.68 He adds that 'whoever is ignorant of the past rushes blindly to the future. Nothing (except the grace and law of God) teaches the living more correctly and forcefully than knowledge of the deeds of the dead.'69 And John supports his argument 60 SeeBenoitLecroix, Orose etsesldees(Montreal/Paris, 1965), esp. pt 2, chapters I and II. 61 John of Salisbury, Hist. Pont., p. 3. 62 See the verses at the end of the epistle to Bishop Alexander and the prologue to Book II, pp. 3-4, 37. Cf. Gransden, Historical Writing, i, 197 andnn. 107-12. 63 See the last two lines of the passage in Bell. Cat. cited above, p. 134 n. 56. 64 Hen. Hunt., p. 37. 65 Richard of Devizes, p. 2. 66 Venerabilis Baedae Historia Ecclesiastica Gentis Anglorum.., ed. Charles Plummer (Oxford, 1896, 2 vols, repr. in one vol. 1961), i, 5. 67 Historia Novella, p. 1. 68 Hist. Pont., p. 3. For John of Salisbury's extensive and intelligent use of exempla see Peter Von Moos, 'The use of exempla in the Policraticus of John of Salisbury' in The World of John of Salisbury, ed. Michael Wilks (Studies in Church History, Subsidia 3, Oxford, 1984), pp. 207-61. 69 Hist. Pont., pp. 3-4.
136
Legends, Traditions and History in Medieval England
with a quotation from Cato's suppositions work, the Disticha de Moribus: 'The lives of others are our teachers/70 Gervase of Canterbury pursues the same theme, expounding at length the edificatory value of historiography. In his introduction (ingressus) to his prologue, addressed to his dear brother (presumably his fellow monk Thomas), he says that three things deliver men from evil - prohibition, precept and example, and example is generally the most effective. Knowledge of the punishments which God inflicts on wrongdoers scares men into trying to avoid similar fates, while the examples of the lives of saints and the good deeds of many others inspire emulation and love of God. Gervase's prologue takes up the theme. Historical writings are repositories of such examples; Gervase writes, therefore, for his audience's edification.71 Historical writings had worldly uses as well as spiritual and moral value. In very practical terms, they could, Eadmer points out, in certain circumstances provide helpful precedents. As John of Salisbury says: 'chronicles are useful for establishing or abolishing custom, for strengthening or weakening privileges'.72 But historiography also had more general secular uses. William of Malmesbury claims that the examples in his narrative of the energy, enterprise, bravery and other merits of great men will encourage others to similar behaviour.73 Moreover, historical writings could boost a nation's self-awareness and pride in itself. 'It is idle and ignoble', writes William of Malmesbury, 'to know about distant lands and not about one's own great men.'74 As already mentioned (p. 64), to Henry of Huntingdon a man's ignorance of his own country's history was a piece of brutish stupidity, which doomed him to eternal oblivion. Finally, a few of the prologues testify to the perennial value of history as a source of pleasure. William of Newburgh was grateful to Ernald for 70 Ibid., p. 3. Disticha de Moribus, iii, 13: 'uita est nobis aliena magistra'. 71 Gerv. Cant., i, 84-7. 72 Eadmer, Hist. Nov., p. 1: he wrote partly 'ut posterorum industriae, si forte quid inter eos emerserit quod horum exemplo aliquo modo juvari queat, parum quid muneris impendam'. John of Salisbury, Hist. Pont., p. 3: 'Valet etiam noticia cronicorum ad statuendas uel euacuandas prescriptiones et priuilegia roboranda uel infirmanda.' 73 Gesta Regum, ii, 283: 'de [Willelmo] talia narrabo libenter et morose quae sint inertibus incitamento, promptis exemplo, usui praesentibus, jocunditati sequentibus'. Ibid., ii, 358: William says he will write about the Crusade 'quia tarn famosam his diebus expeditionem audire sit operae pretium et virtutis incitamentum'. Gesta Pontificum, p. 4: 'Quid enim dulcius quam majorum recensere gratiam, ut eorum acta cognoscas a quibus acceperis et rudimenta fidei et incitamenta bene vivendi.' Hist. Novella, p. 1: 'Quid porro iocundius quam fortium facta uirorum monimentis tradere litterarum, quorum exemplo ceteri exuant ignauiam, et ad defendendam armentur patriam?' 74 Gesta Pontificum, p. 4.
Prologues in the Historiography of Twelfth-Century England
137
imposing on him an easy, pleasant task (one unfitting for the monks of Ernald's own house, bound by the strict Cistercian rule). He had asked him 'to wander in the avenues of history, as if for the recreation of the spirit'.75 Eadmer starts his prologue to the Historia Novorum by describing history as a solace; in his day men investigate the past in order to find 'comfort and strength'.76 Henry of Huntingdon, at the beginning of his prologue, expresses (with echoes of Boethius) a similar sentiment: while all kinds of learning are 'a mitigation of trouble and consolation in grief, history deserves pride of place as the noblest and most delightful of studies.77 Despite all that historians wrote in favour of historiography, ostensibly they regarded it as a comparatively lowly pursuit. They were not, in the words of William of Newburgh, concerned with 'lofty matters and divine mysteries'.78 Moreover, many historians made a show of their own humility in the conventional way. Rhetoric set a standard of eloquence hard to attain. Sallust had said that the writing of history was one of the most difficult tasks, partly because 'the style and diction must be worthy of the deeds recorded'.79 William of Malmesbury was very conscious of the need for eloquence. He praises previous historians who wrote simple, persuasive Latin (especially Bede, but also Eadmer),80 and condemns those who were verbose and obscure (notably Eddius Stephanus and y£thelweard).81 And he felt that he himself was insufficiently eloquent to do justice to the momentous affairs of Henry I's reign. ('Hardly Cicero, whose eloquence was venerated throughout the Latin world, would attempt it, or even the Mantuan bard.')82 Orderic Vitalis similarly protests his lack of 'literary skill and eloquence',83 and William of Newburgh asserts that many monks in Ernald's own monastery could have written a history 'with greater ease and elegance than he'.84 Gervase of Canterbury's humility was truly abject. He divided historical writing into two categories, histories and chronicles, which were different in narrative mode and structure. Histories were written in diffuse and elegant style. Chronicles, on the other hand, were simple and concerned with dates, and the record of the acts of kings 75 Chrons. Stephen, Henry II, and Richard /, i, 4. 76 Hist. Nov., p. 1. 77 Hen. Hunt., p. 1. 'Cum in omni fere literarum studio dulce laboris lenimen et summum doloris solamen, dum vivitur, insitum considerem, turn delectabilius et majoris praerogativa claritatis historiarum splendorem amplectendum crediderim'. 78 Chrons. Stephen, Henry II, and Richard I, i, 4. 79 Bell. Cat., iii (Loeb edition) p. 6: '..in primis arduum videtur res gestas scribere; primum quod facta dictis exaequanda sunt,..' 80 Gesta Regum, i, 1. 81 Gesta Pontificum, p. 210; Gesta Regum, i, 1, 3. 82 Ibid., ii, 465. For the sequel to this passage see below, p. 141 and n. 109. 83 Orderic, i, 130. 84 Chrons. Stephen, Henry II, and Richard I, i, 3.
138
Legends, Traditions and History in Medieval England
and princes, and other events.85 Gervase berated chroniclers who exceeded the limits of the genre by writing eloquent prose- who 'make broad their phylacteries and enlarge the borders of their garments' (Matthew, xxiii.5.88). He did not claim for his 'little self even the title of chronicler. He was a mere compiler, and his work of the utmost insignificance.86 In keeping with such protestations of humility were authors' assertions that they were afraid of boring.This alleged fear was in turn one reason for their expressed intention of writing briefly. Since in fact they did not write briefly, such statements must be regarded as rhetorical flourishes. Sallust had promised to be brief about the Catilinarian conspiracy.87 William of Malmesbury made a similar promise and explained his motives: 'I shall write as briefly as I can, bearing in mind the tastes of the studious, and the scorn of the disdainful.'88 Henry of Huntingdon claims that he wrote briefly about the Roman Emperors, because 'greater prolixity would have been boring'.89 But brevity should not be carried to excess. Ralph Diceto writes that he will indulge neither in 'profusion and tedious chatter, nor in obscure brevity'.90 Historians may have been, or at least thought they ought to be modest, but a number by implication betray self-esteem and pride in their calling.91 They tend to see themselves as heirs to a noble tradition. (Indeed, on the continent Robert of Torigni justified writing his chronicle by appealing to great predecessors, from Cyprian to Sigebert of Gembloux, whose names he recites.)92 Both William of Malmesbury and Henry of Huntingdon declare themselves to be successors of Bede; they would continue where Bede left off.93 Orderic Vitalis begins his prologue with a list of those 'who had previously studied and recorded all ages, the good and evil fortunes of men, for the benefit of posterity'.94 He then lists seven great historians (including Bede), from Moses to Paul of Monte Cassino, and says that 'following the 85 Gerv. Cant., i, 87. 86 Ibid., i, 89:'.. aliqua gestorum praeteriti temporis et futuri compilare potius quam scribere cupio, '; 'Me autem inter cronicae scriptores computandum non esse censeo, quia non bibliotecae publicae sed tibi, mi frater Thoma [continues as above p. 62 n. 42].' Ibid., 91: 'Cronicam parvulam pusillus ego conscripturus, . . . ' 87 Bell. Cat., iv (Loeb edition), p. 8: '..de Catilinae coniuratione quam verissume potero paucis absolvam'. 88 Gesta Pontificum, p. 139: 'breviterque omnia digerens, quantum potero consulam studiosorum notitiae et fastidiosorum nausiae'. 89 Hen. Hunt., p. 37. 90 Diceto, i, 34. 91 See Guenee, 'Les premiers pas', 142-8 passim. Professor Guenee argues that historians' pride grew in the course of the twelfth century. 92 Chrons. Stephen, Henry II, and Richard I, iv, 61-2. 93 Gesta Regum, i, 2; Hen. Hunt., p. 3. 94 Orderic, i, 130.
Prologues in the Historiography of Twelfth-Century England
139
95
early Fathers, I shall write about ecclesiastical affairs'. Like Bede he called his work Historia Ecclesiastica. John of Salisbury starts with an even more elaborate list of historians and includes a few biographical details and brief comments on their works. Of Cassiodorus he remarks that he was a Gentile turned Christian, a senator turned monk, and he had 'many successors distinguished in this learning (studium)'. In his own age, John asserts, 'many wise men have laboured in this way for the benefit of contemporaries'.96 He ends with Sigebert of Gembloux, sketching the scope of his works. The longest and most elaborate such list of historians compiled in England was by Ralph Diceto. His prologue includes a list of forty-two famous historians with notes of when and what they wrote.97 He starts with Trogus Pompeius and ends with himself ('the most modern of the modern') 'Ralph, dean of the church of London, began the book called Ymagines Historiarum at A.D.I 147 and continued to 1190.'98 As Guenee points out, John of Salisbury and Diceto compiled their historiographical surveys under the influence of Hugh of St Victor99 (whom both include in their lists). The prologue to Hugh's universal chronicle has a list of thirtyfour historians.100 Although disordered and inaccurate, it was the first list of its kind, and exerted a seminal influence on many later historians. In fact, it was an important step forward in the history of historiography. Historians in the twelfth century, like their classical prototypes, might protest their lack of literary skill, but were happy to mention their industtriousness. 'I hope', writes William of Malmesbury,' to have credit with posterity for industry if not talent.'101 Historiography was, indeed, regaining its 95 Ibid., i, 130. Orderic himself made copy of Bede's Historia Ecclesiastica; see Bede's Ecclesiastical History of the English People, ed. with an English translation B. Colgrave and R. A. B. Mynors (Oxford Medieval Texts, 1969), p. Ixi. Possibly Orderic borrowed the idea of appending an autobiography to his own work from Bede; see Orderic, vi, 550-6 and n.l. 96 Hist. Pont., p. 2. 97 Diceto, i, 20-4. 98 Ibid., i, 23-4. 99 Guenee, 'Les premiers pas', 139-40. See also G.A. .Zinn, 'The influence of Hugh of St Victor's Chronicon on the Abbreviationes Chronicorum by Ralph Diceto', Speculum, Hi (1977), 38-61. 100 Hugh of St Victor's universal history is unpublished: see W. M. Green, 'Hugh of St Victor De tribus maximis circumstantiis gestorum', Speculum, xviii (1943), 484-93. The list of historians (noticed ibid., 493) is printed in G. Waitz, Archiv der Gesellschaft fur d'ltere deutsche Geschichtskunde, xi (1858), 307-8. Cf. Guenee, 'Les premiers pas', 137. 101 Gesta Regum, i, 3: habiturus, ut spero, apud posteros, postdecessum amoriset livoris, si non eloquentiae titulum, saltern industriae testimonium'. Gesta Pontificum, 4: 'Quod cum fecero, videbor mihi rem nulli attemptatam consumasse, non meo fretus ingenio sed dignae cognitionis allectus studio.' Cf. Guenee, 'L'Histoire entre 1'eloquence et la science', 359, 367-8, and above p. 56 and n. 6.
140
Legends, Traditions and History in Medieval England
status in England of a learned subject, a status lost after the death of Bede. Importance was now attached to the accumulation and evaluation of data. Bede had set an example of assiduity in the search for material, as he reveals in his prologue. Of the twelfth-century historians perhaps William of Malmesbury took the greatest pains in this respect. He attached much importance to chronicle evidence: chronicles, he writes, 'shine like a beacon to light my path'.102 His prologue to Book II of the Gesta Regum states that he 'sought chronicles far and wide' and even 'procured some histories of foreign peoples'. He was acutely aware of gaps in his knowledge. 'I confess', he sadly admits, 'I profited little from all this industry.'103 And if he was short of material for the Gesta Regum, he was even shorter of it for the Gesta Pontificum. For that he found virtually no chronicles to help him; 'almost totally destitute of such solace, I feel my way through the shadows of gross ignorance, with no light of history to guide me'.104 He especially laments his lack of information about saints, the result of the absence of saints' Lives, or of their domicile in distant lands.105 He even asks readers to help him make the Gesta Regum more comprehensive by sending additional information; he will insert it in the margin if unable to fit it into the appropriate place in the text.106 William's claims that he took such trouble collecting data have been amply justified by recent research. He wras extraordinarily widely read, and did not claim to have used wrorks wrhich he did not use, at least indirectly.107 The very fact that most of his account of the Anglo-Saxon church after Bede's time is so brief indicates that in general he accepted the limitation imposed by lack of
102 Gesta Pontificum, p. 4: with reference to the Gesta Regum. William writes 'Siquidem ibi aliquid de cronicis quae prae me habebam mutuatus, velut e sullimi specula fulgente facula, qua gressum sine errore tenderem, ammonebar.' (For the sequel to this passage see below, n. 104). For William as a learned man see below, n. 107. 103 Gesta Regum, i, 104: 'Chronica longe lateque corrogavi, sed nihil propemodum hac, fateor, profeci industrial Ibid., i, 103: 'Itaque, cum domesticis sumptibus nonnullos exterarum gentium historicos conflassem, familiari otio quaerere perrexi si quid de nostra gente memorabile posteris posset reperiri.' 104 Gesta Pontificum, p. 4 (continuing the passage cited above, n. 102): 'Hie autem, pene omni destitutus solatio, crassas ignorantiae tenebras palpo, nee ulla lucerna historiae praevia semitam dirigo.' (For the sequel to this passage see below, p. 73 n. 124.) 105 Ibid., p. 277 (prologue to Book IV). For the lack of written evidence about his abbey's patron, St Aldhelm, see ibid., p. 330 (prologue to Book V). 106 Gesta Regum. i, 104: 'ut meo stylo apponantur saltern in margine quae non occurrerunt in ordine'. 107 See R.M.Thomson, William of Malmesbury (Woodbridge, 1987), esp. chapters 2, 3, 6-10 and appendix I.
Prologues in the Historiography of Twelfth-Century England
141
evidence. Moreover, the autograph manuscript of the Gesta Pontificum shows his use of the margins for late additions.108 Literary sources were, of course, of no help for contemporary history; reliance had to be placed on oral and documentary evidence. William of Malmesbury expresses regret in the prologue to Book V of the Gesta Regum that he will not be able to deal with Henry I's reign fully, because he is 'far from the secrets of courts', and so does not know all the king's doings and great deeds.109 (Orderic Vitalis appeals explicitly to his monastic status as a reason for areas of ignorance: he cannot explore 'Macedonian, Greek or Roman history, or other noteworthy matters, because, being, of my own free choice, a cloister monk, I am bound irrevocably to the monastic observance'.110) It is well known that many twelfth-century historians made extensive use of documents. John of Salisbury's prologue mentions the searches he undertook in order to write the history of the popes after 1148 (when Sigebert of Gembloux's continuator ended): 'I could discover no chronicle, although I have found in church archives some notes of memorable events, which could be of help to any future writer.'111 Symptomatic of historians' concern with their sources of information was the development of the practice of discussing them in the prologue. Bede provided a model. However, since Bede had had hardly any literary sources to serve his purpose, his prologue mainly details oral ones (although it also describes his acquisition of documents). William of Malmesbury was the first historian in the twelfth century to adopt this practice. In the prologue to Book I of the Gesta Regum he briefly surveys the literary sources for English history to which he was indebted; he starts with Bede, 'the most learned and least proud' ('vir maximus doctus et minimus superbus'), and ends with Eadmer.112 Such an account of sources can, of course, be indistinguishable from a list of eminent historians, of the kind mentioned above.113 Hugh of St Victor's list combines both roles, and so to a limited extent does Ralph Diceto's. However, such coincidences are exceptional. For example, John of Salisbury 's Historia Pontificalis begins in 1148, the year when, John states, Sigebert of Gembloux (in fact his continuator), the last author in his eminent historians' list, ended. The English historians' attitude to truth was that typical of the Middle Ages. Sallust had declared that he hoped to write truthfully. 114 Perhaps he was William of Malmesbury's model for his claim that he means to write 108 Gesta Pontificum, pp. xii, 16 n. 7. 109 Gesta Regum, ii, 465 (prologue to Book V, here continuing the passage cited above, p. 137 and n. 82. 110 Orderic, i, 130. 111 Hist. Pont., p. 2. 112 Gesta Regum, i, 1-2. 113 Above, p. 138. 114 See above, p. 138 n. 87.
142
Legends, Traditions and History in Medieval England
the truth in Gesta Regum and Historia Novella.115 Chronology is of course inextricably linked to historical truth: to have historical validity an event must be correctly dated. Ralph Diceto and Gervase of Canterbury make their consciousness of this clear. Diceto cites a dictum of Hugh of Fleury in his prologue: 'an event not attributable to a specific time cannot be accepted as history, but must be dismissed as an old wives' tale'.116 Gervase, declaring that he wishes above all to keep the path of truth, confesses that the study of chronology is full of pitfalls. He laments that many errors arise because of the variety of ways used to calculate the years. To underpin their statements historians appealed to authority. Their immediate model may have been Bede, who in his prologue disclaims responsibility for the truth of statements in the Historia Ecclesiastica, because he had trusted reliable witnesses.117 William of Malmesbury refuses to vouch for his information about the distant past; veracity, he says, rests with his written authorities. For recent times he relied on what he himself had seen, or heard from trustworthy persons.J18 And Geoffrey of Monmouth tried to give credibility to his specious narrative by appealing to a (supposed) authority, 'an old book in the British language'.119 Occasionally an authority was questioned. William of Malmesbury observes that Bede's account of St Wilfred has many omissions, and warns that Abbot Faritius' Life of St Aldhelm is untrustworthy, because Faritius was a foreigner (a Tuscan) and ignorant of the Anglo-Saxon language. 12° A few historians compared authorities in order to establish the truth. Alfred of Beverley clearly had doubts about the veracity of Geoffrey of Monmouth's Historia Regum Britanniae. He claims to have treated it with judicious caution. He asserts that he compared its text with other histories, noting agreements and disagreements; he had, he alleges, only copied into 115 See Gesta Regum, i, 284: 'Mihi haec placet provincia, ut mala, quantum queo, sine veritatis dispendio extenuem.' Historia Novella, 1: 'Itaque primo uocata, ut decet, in auxilium Diuinitate, rerum ueritatem scripturus, nichilque offense daturus aut grade, ita incipiam.' 116 Diceto, i, 15: 'Res gestae quae nulla regum ac temporum certitudine commendantur non pro hystoria recipiuntur; sed inter aniles fabulas deputantur.' Diceto attributes this passage to Aimoin of Fleury, but Stubbs notes (ibid., 15 n. 3) that it is by Hugh of Fleury. 117 See Eccl. Hist., prologue (ed. Plummer), i, 6: 'Ut autem in his, quae scripsi, uel tibi, uel ceteris auditoribus siue lectoribus huius historiae occasionem dubitandi subtraham, quibus haec maxime auctoribus didicerim, breuiter intimare curabo.' (Bede's account of his authorities follows.) And ibid., i, 8: 'Lectoremque suppliciter obsecro, ut, siqua in his, quae scripsimus, aliter quam veritas habet, posita reppererit, non hoc nobis imputet, qui, quod uera lex historiae est, simpliciter ea, quae fama uulgante collegimus, . . ' 118 Gesta Regum, i, 3. 119 Hist. Brit., i, 1 (ed. Acton Griscom), p. 219. 120 Gesta Pontificum, pp. 210, 331 respectively.
Prologues in the Historiography of Twelfth-Century England
143
his Annales passages corroborated in this way.121 William of Newburgh knew no such moderation. He pilloried Geoffrey as an unbridled liar and totally demolished his credibility. He did so mainly be using the same comparative method as Alfred. Bede was his principle yardstick. ('Trust must be put in Bede, whose wisdom and sincerity are beyond doubt.'122) Gervase of Canterbury turned to authority to solve the problem of the multiplicity of chronological systems. He decides to follow earlier chronicles (notably the Worcester ones), and adopt the Dionysian era; anyone who prefers to date by the Evangelists need only subtract twenty-two years from any date. Moreover, to avoid the confusion caused by the fact that different historians might begin the year on different days (for example, the Annunciation , Nativity or Passion), he himself will start at Christmas, as most of his predecessors had done.123 In addition, historians used probability in their search for truth. William of Malmesbury says that he will do his best to write the truth in the Gesta Pontificum by 'following the light of reason, since there are no chronicles to light the way'.124 And William of Newburgh attacked Geoffrey of Monmouth's Historia partly on grounds of improbability.125 Historians were anxious to show that they recognised the need for accuracy and had taken great trouble to avoid error. Exceptionally, one might even ask the reader for help. Bede explained that he sent King Ceolwulf the Historia Ecclesiastica for criticism.126 Ralph Diceto declares that he does not mind having any chronological or other errors, or stylistic infelicities, corrected.127 William of Malmesbury's prologues show his awareness of the danger of bias entering an historical narrative. He was afraid that his account of the Norman Conquest might suffer from the bias in his authorities: Norman writers extol William the Conqueror to the skies, without discrimination, while the Anglo-Saxon ones denigrate him, because he subjugated their land.128 But the danger of bias was worst when writing contemporary history. The problem was that an author might make enemies if he wrote the truth, and was, therefore, tempted to slant his narrative, in order to avoid the risk. Sallust asserted that, since returning 121 122 123 124
125 126 127 128
Annales, pp. 2-3. See above, pp. 19-20. Chron. Stephen, Henry II, and Richard I, i. 18. Cf. above, pp. 20-2, 61. Geru. Cant., i. 88-90 passim. Gesta Pontificum, p. 4 (continuing the passage cited above, p. 70 n. 104): 'Aderit tamen, ut spero, lux mentium, ut et integra non vacillet veritas et instituta conservetur brevitas.' The substance here is very like that of the passage in Sallust, Bell. Cat., iv, cited above, p. 68 n. 87. See especially Chrons. Stephen, Henry II, and Richard I, i, 15-18. Hist. Eccl., prologue (ed. Plummer), i, 5. Diceto, i, 19. Gesta Regum, i, 283.
144
Legends, Traditions and History in Medieval England
from public life, his determination to write history was strengthened, because at last his mind was at liberty, free from hope, fear or partisanship.129 Perhaps William had this statement in mind when he wrote that friends had warned him of the peculiar danger of writing about the present: 'truth is often ship-wrecked and falsehood praised, because to speak ill of contemporaries is perilous, and good laudable'.130 The fact that William revised the Gesta Regum and Gesta Pontificum in order to modify his criticisms of the Anglo-Norman kings and other great men of his day, indicates that his fears were real, and he acted on them.131 There is plenty of evidence in the prologues that historians subscribed to the ideal of historical veracity. But was this a matter of mere words? In fact, as long as they had no motives for distortion, most of them clearly took trouble to be accurate. However, historians used the latitude allowed them by the medieval concept of 'truth' and ideas of evidence outlined above (pp. 128-9), and a few wrote very biased narratives. William of Malmesbury is a good example of an historian who made a little information go a long way. His objective was to produce a coherent, meaningful narrative, rather than a string of disconnected facts, and to make it more convincing by filling gaps in his knowledge with plausible (but unverifiable) statements. A further objective was to write an account which supported, as will be seen, a certain point of view. To achieve his ends William sometimes resorted to fallacious arguments. 129 Bell. Cat. iv (Loeb edition), p. 8: ' eodem regressus statui res gestas populi Romani carptim, ut quaeque memoria digna videbantur, perscribere; eo magis, quod mihi a spe, metu, partibus rei publicae animus liber erat'. For the topos see above, p. 56 and n. 8. 130 Gesta Regum, ii, 357 (prologue to Book IV): 'Scio plerisque ineptum videri quod gestis nostri temporis regum scribendis stylum applicuerim; dicentibus quod in ejusmodi scriptis saepe naufragatur veritas et suffragatur falsitas, quippe praesentium mala periculose, bona plausibiliter dicuntur. Eo fit, inquiunt, ut, quia modo omnia magis ad pejus quam ad melius sint proclivia, scriptor obvia mala propter metum praetereat, et, bona si non sunt, propter plausum confingat.' However, at the urging of his friends, William decided to write about recent times. He continues, ibid., ii, 358: 'Quocirca illorum, qui mihi timent ut aut odiar aut mentiar, benevolentiae gratus, ita sub ope Christi satisfaciam, ut nee falsarius nee odiosus inveniar.' See also the end of the prologue to Book I of the Gesta Regum (cited above, p. 69 n. 101), and Gesta Regum, ii, 283 (prologue to Book III): William says he will treat William the Conqueror impartially 'bene gesta, quantum cognoscere potui, sine fuco palam efferam; perperam acta, quantum sufficieat scientiae, leviter et quasi transeunter attingam; ut nee mendax culpetur historia, nee ilium nota inuram censoria cujus cuncta pene, etsi non laudari, excusari certe possum opera'. 131 Gesta Regum. i, xxxiii et seqq., xlvii; Gesta Pontificum, pp. xv-xvii; Hugh Farmer, William of Malmebsury's life and work', Journal of Ecclesiastical History, xiii (1962), 45-6; Gransden, Historical Writing, i, 181-2 and nn. 13041.
Prologues in the Historiography of Twelfth-Century England
145
Examples of the practice may be given. The first suggests that he used the post hoc, ergo propter hoc method. Thus, he states that God punished Leofstan, abbot of Bury St Edmunds (1045-1065), for tampering with St Edmund's body, by inflicting him with gout of the hands, and that Baldwin, a well known physician, who later succeeded Leofstan in the abbacy, first visited Bury in order to cure Leofstan's gout.132 It seems that William apparently constructed this narrative by giving casual connections to disparate facts, all recorded by a respected early authority: Abbot Leofstan tampered with St Edmund's body;133 he contracted gout in his hands;134 Baldwin was a well known physician;135 he succeeded Leofstan.136 William also almost certainly used analogical arguments. For instance, there was no early authority for the pre-Conquest origins of the abbeys of Malmesbury or Sherborne. But William knew from the Anglo-Saxon chronicle that in 964 King Edgar expelled the clerks from the Old and New Minsters at Winchester, and from Milton and Chertsey, and introduced monks. He also knew that St Ethelwold's biographers, yElfric and Wulfstan, represented Ethel wold as the moving force in the transformation of the clerical communities at Winchester into monastic ones.137 It seems that William used these cases as analogues of what happened at Malmesbury and Sherborne ; he states that St Aldhelm drove the clerks from Malmesbury, and St Wulfsigedrove them from Sherborne. (Other twelfth-century historians seem to have used the same topoi for the account of the pre-Conquest origins of their houses).138 A final example of William's apparent use of a fallacious argument is of him drawing a general conclusion from, so far as we know7, an isolated case. It appears that he used Eadmer's strictures on the pre-Conquest monks of Christ Church, Canterbury, 139 as evidence for the decadence of all monasteries in the late Anglo-Saxon period.140 If so, his deduction could reflect unwarranted bias; the Anglo-Norman historians denigrated the Anglo-Saxon church in the interest of the new regime, which justified its 132 Gesta Pontificum, pp. 155-6. 133 Hermann, 'De Miraculis Sancti Eadmundi' in Memorials of St Edmund's Abbey, ed. Thomas Arnold (Rolls Series, 1890-6, 3 vols), i, 52-4. 134 Ibid., i, 56. 135 Ibid., i, 56, 63. 136 Ibid., i, 56. 137 Three Lives of English Saints, ed. Michael Winterbottom (Toronto Medieval Latin Texts, 1972), pp. 23, 45 respectively. 138 I.e. the historians of Worcester, Ely and Bury St Edmunds. See A. Gransden, 'Traditionalism and continuity during the last century of Anglo-Saxon monasticism,' Part II, Journal of Ecclesiastical History, above, pp. 76-8. 139 Eadmer's 'Life of St Dunstan' in Memorials ofSt Dunstan, ed. W. Stubbs (Rolls Series, 1874), pp. 237-8. 140 Gesta Pontificum, pp. 70-1. Cf. R. W. Southern, StAnselm and his Biographer (Cambridge, 1963), p. 247 and n. 1.
146
Legends, Traditions and History in Medieval England
existence partly on grounds of its programme of ecclesiastical reform. It is very likely that William intended to support this point of view. He was certainly not free of partisan bias. He did not write contemporary history until late in life, when he wrote the Historia Novella for Robert earl of Gloucester.141 Earl Robert was a patron sufficiently powerful to protect William and his abbey, and the Historia is strongly slanted in favour of him and his niece, the empress Matilda. There remains to discuss the light prologues throw on the methods which an historian might adopt to structure his work. A work's structure is sometimes explained in the prologue. The twelfth century was remarkable for the elaborate ways used to construct histories. Traditionally historiography in England was annalistic, but particularly in the twelfth century a few writers broke away from the strictly chronological arrangement. Instead they gave their works unity by other kinds of structure and by overriding themes. The most ambitious authors in this respect explained their methods in their prologues. Thus Eadmer explains how his two books on St Anselm, the Historia Novorum and the Vita Sancti Anselmi complement each other. In the prologue to the Vita he points out that the Historia dealt with the relations of the kings of England with St Anselm, but neglected his private life, character and miracles. This omission, at the insistence of his friends, he now tends to remedy. 'I have tried to adopt a plan to form a complete narrative in each case, so that neither work needs each other.' Nevertheless, in order to understand St Anselm properly, the reader should read both books.142 Eadmer inherited the idea of dealing separately with different aspects of a man's life from the Anglo-Saxon tradition of secular and sacred biography (a tradition derived from Einhard's Life of Charlemagne, and thus indirectly from Suetonius' Lives of the Caesars). However, Eadmer developed that tradition. The pre-Conquest examples are of single biographies divided into sections, each dealing with a particular aspect: typically, a man's public life; his private life; and, if he were a saint, his miracles. Eadmer, by producing two distinct volumes, made the physical division absolute.143 William of Malmesbury, like Eadmer (who may well have influenced him in this matter, as he did in others),144 saw his two principal works, the Gesta Regum and the Gesta Pontificum as complementary. In the prologue to the latter he writes: 'having formerly dealt [in the Gesta Regum] with the kings of England, it is not unreasonable that I should 141 Historia Novella, ed. Potter, p. xiii. 142 Vita Sancti Anselmi, ed. Southern, p. 2. 143 For the Anglo-Saxon tradition of the divided biography see Gransden, Historical Writing, i, 51-2 and nn. 60, 132, 136. 144 See ibid., i, 169, 170.
Prologues in the Historiography of Twelfth-Century England
147
145
now turn to its pontiffs. Under Bede's influence, he arranged the Gesta Regum in five books, according to the old kingdoms of Anglo-Saxon England. He adopted the same order for the Gesta Pontificum, substituting 'episcopal provinces' for kingdoms.146 But he deliberately alters the order. Both start with Kent, but whereas the Gesta Regum has Wessex second and East Anglia and Essex last, the latter two being, William says, 'of little interest',147 the Gesta Pontificum has East Anglia and Essex after Kent because of their distinguished and still continuing series of bishops.148 Moreover, Book V of the Gesta Pontificum is devoted to the patron saint of Malmesbury abbey, St Aldhelm. 'The reader like a pilgrim at last returns home.'149 And William arranges his material about St Aldhelm's life, as he explains,150 in four parts, according to subject matter, in the traditional Anglo-Saxon manner. Similarly, Henry of Huntingdon explains in the prologues to Books I and V of the Historia Anglorum why he divided his work into five parts. His motive was thematic rather than purely structural. He made the structure reinforce his work's overall edificatory theme. Divine vengeance, he argues, had inflicted five scourges on Britain from the beginning of its history until Henry's own time. First, the Romans conquered Britain; secondly, thePicts and Scots sorely harassed it; thirdly, the Anglo-Saxons came and conquered; fourthly, the Danes wreaked havoc; fifthly, the Normans subdued the land and still rule it.151 Henry describes the fourth scourge, England's sufferings from the Danes, particularly graphically. The Danes had descended 'like a swarm of wasps', 'sparing neither age nor sex', and decimated the country for 230 years, from the accession of King Ethelwulf (in 837) until the Norman Conquest,152 which, by imposing law and order, gave the people 'life and liberty'. 153 They were instruments of divine vengeance elicited by religious decadence and the corruption of the AngloSaxon church, which had shone with such lustre in its early days.154 Henry's periodisation was in the mainline of the ancient tradition of Christian historiography, which saw history as the expression of God's will on earth. St Augustine had familiarised the Middle Ages with the 145 Gesta Pontificum, p. 4. 146 Ibid., p. 4. 147 Gesta Regum, i, 2: 'Quod profecto fiet expeditius, si regna Orientalium Anglorum et Orientalium Saxonum post aliorum tergum posuero, quae et nostra cura et posterorum memoria putamus indigna.' 148 This is fully explained in Gesta Pontificum, p. 139.
149 150 151 152 153 154
Ibid., p. 330. Ibid., p. 331. Hen. Hunt., pp. 137-8 (prologue to Book V). Ibid., p. 139. Ibid., p. 138. Ibid., p. 139.
148
Legends, Traditions and History in Medieval England
concept of the Seven Ages, symbolising the seven days of the Creation.155 Henry's version of providential periodisation was different. It was based on the Old Testament view that history illustrates God's power to punish the sinful; in his scheme God appears not as Creator, but as Judge. The immediate source of Henry's thematic structure was probably Bede, to whom he owed his nostalgic vision of the Golden Age of Northumbria.156 Bede, following Gildas, attributed a plague and the Anglo-Saxon invasion, which successively afflicted the ancient Britons, to God's anger with the latter for their 'drunkenness, hatred, quarrelling, strife, envy, and other similar crimes'.157 Moreover, others before Henry had seen the Viking incursions and Norman Conquest as divine vengeance on the AngloSaxons. King Alfred, in the preface to the Pastoral Care, ascribed the ninthcentury Viking incursions to God's wrath at their neglect of learning.158 Eadmer, whose Historia Novorum or some common source was used by Henry, regarded the Viking attacks of the eleventh century as divine punishment for King Ethelred's murder of his brother.159 The AngloSaxon Chronicle, which Henry certainly used, states that God did not help the Anglo-Saxons against the Normans at Hastings 'because of [their] sins'.160 Yet again, Eadmer attributes God's anger to Harold's perjury.161 Henry, therefore, had plenty of models encouraging him to see English history as a series of divine punishments, though the causes given for them varied. Possibly Henry derived the idea of the Danish scourge lasting 230 years from William of Malmesbury. William asserted 223 years' historiographical 'sleep' followed Bede's death.162 This 'sleep' was the result of intellectual torpor and, since 'idleness is the enemy of the soul', it was part of the Anglo-Saxons' moral decadence. It was finally ended by a dramatic revival of religious life after the Norman Conquest. Emphasis is laid by both Norman and Anglo-Norman historians on this reformation, with the implication that it was one justification for the Conquest.163 To highlight 155 The Six Ages of the World, corresponding to the six days of the Creation, and representing epochs in history, gained currency through St Augustine's De Civitate Dei, xxii, 30. See Auguste Luneau, L'Histoire du Salut chez les Peres de I'Eglise (Paris, 1964), pp. 285 et seqq., 352 et seqq. 156 Hen. Hunt., p. 139. 157 Hist. Eccl, i, 14 (ed. Plummer, i. 30). Cf. Gildas, De Excidio Britanniae, cc. 22, 24. 158 Simon Keynes and Michael Lapidge, Alfred the Great, Asser's Life of King Alfred and other contemporary sources (Harmondsworth, 1983), p. 125. 159 Historia Novorum, p. 3. 160 The Anglo-Saxon Chronicle, ed. D. Whitelock with D. C. Douglas and S. I. Tucker (London, 1961), p. 143 (D, s.a. 1066). 161 Historia Novorum, p. 9. 162 Gesta Regum, i, 2. 163 See e.g.: Guillaume de Poitiers, Histoire de Guillaume le Conquerant, ed. with
Prologues in the Historiography of Twelfth-Century England
149
the reformation it was necessary to paint a black picture of the preceding period. Indeed, the idea of the 'sleep' could have originated as part of Norman propaganda. This suggestion is perhaps supported by the fact that there is no hint of it in either Alfred of Beverley's or William of Newburgh's survey of English historiography: both say that Bede had many successors, of praiseworthy assiduity though none of comparable talent.164 Possibly William of Newburgh was here following Alfred of Beverley, who wrote in a fairly remote place and may be supposed, therefore, to have been little, if at all, influenced by Norman propaganda. Moreover, the idea of a historiographical 'sleep' after Bede's death occur sin the work of at least one eleventh-century French historian.165 Henry might well have thought that the period of decadence would have more or less coincided with the Danish scourge. Ralph Diceto likewise explains the structure of his work in the prologue. He too adopted a religious periodisation. He divided his narrative into three parts, before the Law, under the Law, and of Grace, < denoting the very ancient, the ancient and the modern. And he explains the problems each period presents to the would-be historian.166 Diceto's prologue also contains much other material. Although put together in haphazard fashion, one fact clearly emerges: he had the convenience of readers much in mind. He adopted two expedients to help them. First, he lists twelve different subjects (persecutions of the church, schisms, ecclesiastical councils, royal coronations, and so on), and draws a sign after each. Any reference in the narrative to one of these subjects is indicated, as he explains, by the appropriate sign in the margin.167 (This system of signa was to be
a French translation Raymonde Foreville (Les Classiques de 1'Histoire de France au Moyen Age, xxiii, Paris, 1952), p. 127; Eadmer, HistoriaNovorum, pp. 12-15 passim', William of Malmesbury, Gesta Regum, ii, 326; Orderic, ii, 208 et seqq. 164 Alfred of Beverley, Annales, p. 3: 'Similiter et post Bedam plures per Anglorum ecclesias regum tempora diligencius perscrutantes, ipsorum gesta sollerti indagine annotare curaverunt, de quibus non nulla studiosius investigata huic opusculo sunt inserta.' William ot Newburgh in Chrons. Stephen, Henry II, and Richard I, i, 18: 'Sane post Bedam non defuere, qui ab ipso seriem temporum atque eventuum nostrae insulae ad nostram usque memoriam ducerent; illi quidem minime comparandi; pro religiosa tamen opera et fideli, quanquam minus diserta narratione, laudandi.' 165 In 1044 the French historian Raoul Glaber had remarked that no one wrote history for 200 years after Bede and Paul the Deacon; Raoul Glaber, Les cinq livres des ses histoire(900-1044), ed. M. Prou (Collection de textes pour servit a Tetude..de 1'histoire, Paris, 1886), p. 1 (cited Guenee, 'Les premiers pas', p. 136). 166 Diceto, i, 18-20. 167 Ibid., i, 3-4.
150
Legends, Traditions and History in Medieval England
adopted and elaborated by Matthew Paris.168) Diceto's second aid to readers was to make compendia and extracts. For example, he compiled his list of historians and supplied extracts from their prologues, so that readers who had no access to their books could learn about their works.169 To conclude, we have to consider whether the two generalisations of Bernard Guenee mentioned above, about the twelfth-century historical prologues of western Europe, are exactly applicable to the English examples. First, was Hugh of St Victor an influence of primary importance? Certainly he had some influence on John of Salisbury, and a very strong one on Ralph Diceto. But Guenee naturally writes from a European point of view7. In discussing English twelfth-century historiography, while not forgetting the native tradition,170 he puts the emphasis on continental influence.171 This tends to obscure the profound influence of Bede, which the English prologues make very evident. Bede was by far the most highly regarded English historian, and his Historia Ecclesiastica was - as it still is- an unrivalled source for England's early history. Therefore, it is not surprising that this influence was powerful and pervasive. As has been seen, William of Malmesbury and Henry of Huntingdon were Bede's self-declared continuators, and the structure of William's Gesta Regum and Gesta Pontificum was partly modelled on that of the Historia Ecclesiastica. The growing precision with which the prologues describe the authors' sources no doubt owed much, as Guenee points out, to Hugh of St Victor. But English historians already had the precedent of Bede's prologue to the Historia Ecclesiastica. Although it seems that Bede's influence waned in southern England in the last half of the twelfth century, in the north it remained as strong as ever; this is most vividly illustrated by William of Newburgh's prologue, which sets Bede's authority against that of Geoffrey of Monmouth. The second question concerns one of Guenee's central themes. He argues that in the course of the twelfth century the dominance of rhetoric over historiography declined; instead scholarship gained supremacy.172 Historians became less concerned with literary style and structure and more with learning- with the search for, and evaluation and even criticism of, the works of their predecessors. Rather than write history in literary mode, they began to produce compilations of extracts from standard authorities, 168 169 170 171 172
See Gransden, Historical Writing, i, 364 and n. 56. Diceto, i, 19-20. Cf. Gransden, op. cit., i, 231 and n. 102, 234, 235 and n. 130. See Guenee, 'L'Histoire entre 1'eloquence et la science', pp. 361-2, 363. Ibid., 357-70 passim, Guenee, 'Les premiers pas', pp. 136-52 passim. Idem, 'L'Histoire entre 1'eloquence et la science, pp. 357-70 passim, idem, 'L'Historien et la compilation au xiiie siecle', Journal de Savants (1985), pp. 119-35.
Prologues in the Historiography of Twelfth- Century England
51
assembled and sorted in scholarly fashion, and presented in strictly chronological sequence. Thus twelfth-century historiography paved the way for that of the thirteenth century; then the 'monastic chronicle', which begins as a compilation and is throughout composed of annals, had a monopoly. Can this trend be detected in the English prologues? In general the English evidence concurs with Guenee's conclusion. For example, the prologues of Eadmer and William of Malmesbury make clear their preoccupation with literary style and structure. Diceto's prologue, on the other hand, shows less concern with these matters, but is remarkable for the number of extracts it contains. Symptomatic of this trend is the prologue to Roger of Howden's chronicle. Roger did not compose it himself at all, but simply copied from Symeon of Durham.173 Nevertheless, like all generalisations, this one of Guenee's can only be accepted with reservations. The prologues highlight static factors in twelfth-century English historiography. The influence of rhetoric by no means died; Henry of Huntingdon and Richard of Devizes amply prove this. Nor was learning a prerogative of the last half of the century. (Respect for 'authority' was perennial in the Middle Ages; this in itself always encouraged historians to study standard history books.) William of Malmesbury, indeed, was one of the most learned historians England has ever produced. And he wras perfectly capable, when he wished, of effectively criticising his sources. William of Newburgh's critical acumen would have been exceptional in any age. An historiographical generalisation, like any other, can only be roughly right. Deviations inevitably occur because of the peculiar character and talents, background and immediate circumstances of individual authors.
173 Hoveden, i, 3-4.
This page intentionally left blank
6
The Growth of the Glastonbury Traditions and Legends in the Twelfth Century 1
A
recent scholar has written of 'the widespread belief, by no means the monopoly of high Anglicans only, that it was to Glastonbury, first of all places in these islands of ours, that the Christian faith first came, and that Glastonbury is therefore the fountain and headspring of Christianity in England, and indeed in the whole of the British Isles'.2 Glastonbury claimed in the Middle Ages to be one of the most hallowed spots in Britain not only because of its ancient tradition, but also because of the numerous relics it housed, of saints and of king Arthur and queen Guinevere. This reputation originated in the AngloSaxon period and grew in the four centuries following the Norman Conquest. It was mainly the result of work by the abbey's chroniclers and by others whom the monks commissioned to write. In its development the twelfth century was crucial: during this period the monks 'proved' Glastonbury's antiquity and holy associations, sowing the seeds of the more baroque manifestations of its later medieval legends. The reasons are obvious why the monks wanted to increase the abbey's reputation by the use of literature to prove its long and holy tradition. The earlier the date of an abbey's foundation the higher its status. In particular this would give an abbot precedence at an ecclesiastical council,3 but in a more general way antiquity gave prestige. Relics, too, contributed to prestige, and an abbey with the relics of a famous founder was especially fortunate.4 Prestige armed a religious community against its 1
This paper is based on one delivered to the Glastonbury Antiquarian Society in St. John's church, under the chairmanship of my father, Mr. Stephen C. Morland, F.S.A. I am indebted to him and to my colleague Dr. B. F. Hamilton for help on specific points. 2 R. F. Treharne, The Glastonbury Legends, London 1967; reprinted, without the index, Abacus Books 1975, 4. 3 See J. Armitage Robinson, Two Glastonbury Legends: King Arthur and St. Joseph of Arimathea, Cambridge 1926, 40-1 and n.i. 4 See Jonathan Sumption, Pilgrimage, an Image of Medieval Religion, London 1975, 165-7.
154 Legends, Traditions and History in Medieval England enemies, which enabled it to defend its privileges and property more effectively. It also increased its wealth by attracting pilgrims. The crowds who came to worship, perhaps to be cured, perhaps only to look at the shrines of the saints, brought offerings which could be an important source of monastic income.1 At the time of the Norman Conquest Glastonbury abbey, although recognised as an ancient Christian site of outstanding holiness, needed to substantiate its traditions and legends. The circumstances of its foundation were veiled in the mists of time and lacked the picturesque details which did so much to convince the pious multitude. Moreover, although it had relics of a number of saints, many were Northumbrian or Celtic, most of whom were of limited reputation in the south. Nor did it have a famous patron saint of its own. Here it was at a disadvantage compared with other great Benedictine houses, some of which had had such patrons from their foundations: Durham had St. Cuthbert, St. Albans claimed St. Alban, Bury St. Edmunds had St. Edmund, king and martyr, and Worcester St. Oswald, to whom was added St. Wulfstan after the Norman Conquest. Westminster, too, acquired a famous saint, Edward the Confessor, in the post-Conquest period. Canterbury itself had a host of early archbishops, starting with St. Augustine, and then in 1170 gained the biggest pilgrim attraction of all, Thomas Becket. Glastonbury had no equivalent. It is true that it had some claim to St. Duns tan because he had been abbot of Glastonbury before becoming archbishop of Canterbury. But Canterbury's claim was stronger since, as was generally recognised, he was buried in the cathedral. Moreover, the monks asserted that St. Patrick the apostle of Ireland was buried in the abbey, but this claim too presented problems.2 The period of transition following the Norman Conquest was especially difficult at Glastonbury. The Norman abbot, Thurstan, who was appointed in about 1077, had no respect for the abbey's customs and his intransigence culminated in bloodshed in the church—at least two monks were killed and many more wounded.3 In addition Glastonbury had difficulties similar to those faced by other pre-Conquest houses. Its estates, already impoverished by Danish raids and the dilapidations of the last two Anglo-Saxon abbots, had suffered from the Norman settlement,4 and its prosperity was further threatened by the attrition of spiritual prestige. The new ecclesiastical hierarchy tended to regard the Anglo-Saxon saints 1 For the lucrativeness of the pilgrim trade see C. R. Cheney 'Church-building in the Middle Ages', Bulletin of the John Rylands Library, xxxiv (1951-2), 29—32 (reprinted in Cheney, Medieval Texts and Studies, Oxford 1973, 346—63), and Sumption, op. cit., 158—65. 2 For Glastonbury's claim to the relics of St. Patrick and St. Dunstan see below, 159, 162, 163-5 passim. 3 See the Anglo-Saxon Chronicle, s.a. 1083; The Anglo-Saxon Chronicle, a revised translation ed. Dorothy Whitelock with D. C. Douglas and S. I. Tucker, London 1961, 160, and Adami de Domerham Historia de Rebus gestis Glastoniensibus, ed. Thomas Hearne, Oxford 1727, i. 113—16. 4 Adam de Domerha?n, i. 110—13.
The Growth of the Glastonbury Traditions and Legends
155
with scepticism. As archbishop Lanfranc said: 'When I turn over in my mind the accounts of these saints, I cannot help doubting the quality of their sanctity';1 and he struck some from the liturgical calendar of Christ Church, Canterbury.2 Nor was Glastonbury's need of prestige and money particularly acute only in the Anglo-Norman period. In 1184 a catastrophic disaster struck the abbey; the church and monastic building were burnt to the ground. The place, lamented its chronicler, 'was reduced to a pile of ashes, its relics to confusion'.3 The cost of rebuilding was immense and could only be met with the help of those impressed by its fame. The abbey's troubles lasted well into the thirteenth century. It needed all its prestige to defeat the attempt by Savaric bishop of Bath (i 192-1205) and his successors first to move the see of Bath to Glastonbury and then to assert the right to overlordship of the abbey.4 The monks' financial resources were severely strained by litigation in the royal courts and papal curia. Monks habitually responded to such challenges by propaganda, in order to augment their houses' reputations. They wrote local histories and saints' 'Lives'. Both were probably read mainly by the monks themselves, but through them the propaganda was transmitted to neighbours and pilgrims; it was hoped that the well-informed public would give more generously. The Norman Conquest provided a strong incentive to this kind of writing: the monks of Durham and Evesham, for example, produced local histories to prove that their foundations had long and glorious traditions, unbroken by the Conquest;5 and many houses wrote Lives of their patron saints—at Canterbury Osbern6 and Eadmer7 wrote in defence of the Canterbury saints, which was doubtless one reason why some were restored to the liturgical calendar.8 Glastonbury conformed to this general trend, but besides lacking a detailed foundation story and famous relics, the monks were apparently short of literary talent to make the best of what they had. In the twelfth 1 The Life of St. Anselm Archbishop of Canterbury by Eadmer, ed., with an English translation, R. W. Southern (Nelson's Medieval Texts, 1962), 50. Cf. A. Gransden, Historical Writing in England c. 550 to c. 1307, London 1974, 105. 2 See F. A. Gasquet and E. Bishop, The Bosworth Psalter, London 1908, 27—39. 3 Adam de Domerham, ii. 334. 4 For this dispute, a full account of which is given by Adam of Domerham (Adam de Domerham, ii. 352f. and see below, 340), see M. D. Knowles, 'Essays in Monastic History: V. The Cathedral Monasteries', Downside Review, Ii (1933), 94—6, and J. Armitage Robinson, 'The First Deans of Wells', Somerset Historical Essays, London 1921, 68-70. The monks appealed to the antiquity of Glastonbury in the dispute; they ask in one of their articles against Jocelin bishop or Bath (i 206—42) 'An omnibus ecclesiis Angliae Glastoniensis ecclesia sit antiquior et vetustior, saltern per famam?': Adam de Domerham, ii. 453. 5 Printed in Symeonis Monachi Opera Omnia, ed. Thomas Arnold (Rolls Series, 1882, 1885, 2 vols.), i. 3-135, and Chronicon Abbatiae de Evesham, ed. W. D. Macray (Rolls Series, 1863). Cf. Gransden, op. cit., 111-21. 6 For Osbern and his hagiographies see R. W. Southern, Saint Anselm and his Biographer, Cambridge 1963, 248—52. 7 For Eadmer's liagiographies, see ibid., 277—87. 8 See Gasquet and Bishop, op. cit., 33-4.
156
Legends, Traditions and History in Medieval England
century the abbey produced no one capable of writing in its defence. Later it was to be more fortunate. In the thirteenth century Adam of Domerham wrote its history to 1291, with the general intention of showing how the abbey's property and privileges, formerly so great but now sadly diminished, could be preserved and increased, but in particular to record its victory over the claims of the bishop of Bath.1 And early in the fifteenth century a monk of Glastonbury called John abbreviated Adam's work, adding various embellishments, and continued it until about 1400. (A further brief continuation was added which goes down to the late fifteenth century.) 2 Because of the paucity of literary talent at Glastonbury itself in the twelfth century, the monks turned to outsiders to record and amplify the abbey's history. They commissioned at least one well-known historian, William of Malmesbury, and, as will be suggested below, probably employed two other writers, Caradoc of Llancarfan and Gerald of Wales. It is certain that William and Gerald visited Glastonbury before writing, and Caradoc may have done so too. The use of outsiders as propagandists was by no means unusual, and in most instances they seem to have visited the place for which they were to write. Hagiography, indeed, became such an industry in the AngloNorman period that a class of professional hagiographers began to develop. Many of these were foreigners. There was, for example, Goscelin from the abbey of St. Benin's at St. Omer in Flanders, who came to England and travelled from monastery to monastery writing Lives of the patron saints. 3 Such professionals from abroad were sufficiently numerous to arouse Eadmer's anger; he wrote in disgust that they would tell any lie if paid enough.4 (However, Eadmer was being too severe because many of the 'Lives' by foreigners, like those by the AngloNormans themselves, are of considerable value for local history.) The practice of commissioning outsiders to write propaganda remained common throughout the twelfth century. Thus Ailred of Rievaulx, having attended the translation of relics at Hexham in 1155 at the invitation of the canons, subsequently wrote for them an account of the Hexham saints.5 Similarly, in 1163, the abbot of Westminster invited him to the translation of Edward the Confessor and commissioned him to write on the saint's life and miracles.6 1
For the printed edition of Adam's chronicle see above, 154 n. 3. John's chronicle and the continuation are printed mjohannis Glastoniensis Chronica sive Historia de Rebus Glastoniensibus, ed. Thomas Hearne, Oxford 1726, 2 vols. 3 For Goscelin and other foreign hagiographers see Gransden, op. cit., 107-11. 4 Eadmer says this in his letter to the monks of Glastonbury countering their claim to have the relics of St. Dunstan (cf. below 163 & n. 4); Memorials of St. Dunstan, ed. William Stubbs (Rolls Series, 1874), 415. 5 His work, De Saudis Ecdesiae Haugustaldensis, et eorum Miraculis Libellus, is printed in The Priory oj Hexham, its Chroniclers, Endowments and Annals, ed. James Raine (Surtees Society, xliv, xlvi, 1864, 1865, 2 vols.), i. 172-203. 6 Printed in Migne, Patrologia Latina, cxcv, coll. 737-90, from Roger Twysden, Histonae Anglic an ae Script ores X, London 1652, i, 369. Cf. The Life of Ailred of Rievaulx by Walter Daniel, ed., with an English translation, F. M. Powicke (Nelson's Medieval Texts, 1950). xlvii— xlviii. 2
The Growth of the Glastonbury Traditions and Legends
157
In order to build up the prestige of Glastonbury abbey, the twelfth century propagandists made use of two well-established literary traditions, hagiography and romance literature. Hagiography, although centred on the lives of saints, involved an interest in local history because nearly every saint was the patron of a place, usually where he was buried.1 The place attracted the hagiographer's attention from the time of the saint's first association with it, which in the writer's view might mark its origin; thus hagiographers tended to be interested in origins. The romance tradition, which was influenced by classical literature, belonged, in contrast to hagiography, to the secular world. Although typified by the chansons de geste, features of it invaded other types of writing, for example the Latin chronicle. Its values were popularised in England by Geoffrey of Monmouth's Historia Regum Britanniae which was finished in ns6. 2 The hero of romance, at all times obedient to the chivalric code of behaviour, might be an historic figure, but his alleged deeds were largely imaginary. To give verisimilitude to the tale, his career was often linked to recorded events. (Brutus, for example, Geoffrey of Monmouth's eponymous hero of Britain, was an exile from the Fall of Troy.) This led to a taste for historical parallels, to place the hero in the sequence of world history; and if he were a founder, this resulted in an interest in origins. Hagiography and romance were not, of course, totally distinct literary forms; each influenced the other. Both tended to be interested in origins, and the heroes of each were not always dissimilar. On the one hand, a saint might have the attributes of a warrior—he might even have been one earlier in life. (Alternatively a hero of romance might impinge on his career.)3 On the other hand, a warrior hero might be represented as a Christian fighting for God, and his tomb might become the centre of a cult. Generally speaking hagiography was the principal influence on Glastonbury's literary productions up to 1184, the year of the fire; thereafter romance was the most important factor. The influence of hagiography will be considered first. In this context the principal figure is William of Malmesbury whom the Glastonbury monks commissioned in about 1129 to write the abbey's history4 and the life of St. Dunstan. 5 Besides his proximity, there were good reasons for 1
For the value of hagiography for local history see Gransden, op. cit., 69, 71-3, 85—6, 89-91, 106-114 passim. 2 Printed in The Historia Regum Britanniae of Geoffrey of Monmouth, ed. Acton Griscom, London 1929. 3 For an example, see below, 169. 4 The text of the De Antiquitate Glastoniensis Ecclesiae as it survives today (see below, 158 and n. 5) is dedicated to Henry of Blois, bishop of Winchester (1129-1171), who also ruled Glastonbury as abbot. But originally William addressed the work to the monks of Glastonbury; see Adam de Domerham, i. 121-2. Cf. Armitage Robinson, 'William of Malmesbury 'On the Antiquity of Glastonbury',' Somerset Historical Essays, 4. 5 For William's two Lives of St. Dunstan see below, 159. The surviving Life is printed in Memorials of St. Dunstan, ed. Stubbs, 250-324. For the dedication to the monks of Glastonbury, see ibid., 250.
1 58
Legends, Traditions and History in Medieval England
employing him. He was a writer of ability and a recognised authority on history: he was, as it were, a specialist on the subject of continuity as well as a hagiographer. In 1125 he published his major work, the Gesta Pontificum,1 which stressed that the church of his time was a continuation of the Anglo-Saxon one. His declared intention was to fill the gap between Bede and his own day, paying particular attention to the lives of the preConquest saints, which, he said, the Anglo-Saxons had neglected to record.2 Already early in the twelfth century the monks of Worcester had commissioned him to translate from English into Latin the Life of Wulfstan, written by one of their number, Coleman, soon after the saint's death in io95.3 The Glastonbury monks wanted William both to glorify the abbey in general, and also to counter a particular attack which had recently been made on its antiquity. Osbern in his Life of St. Dunstan had asserted that Dunstan was the first abbot of Glastonbury.4 The monks, of course, knew better: the abbey had a far longer history and they wanted this proved beyond dispute. William of Malmesbury's De Antiquitate Glastoniensis Ecclesiae sought finally to establish that the abbey had a very long and very venerable history. Unfortunately we do not have William's original text. All that survives is the copy made by Adam of Domerham to form the early part of his own chronicle. The version of the De Antiquitate which Adam copied was not the text as William left it. It had numerous additions, mainly of legendary material, interpolated by the Glastonbury monks to enhance the abbey's reputation. Some idea of what William actually wrote can be gained by comparing the extant text of the De Antiquitate with the extracts William inserted in the third edition of his Gesta Regum: whatever is not in the Gesta Regum is suspect. Nevertheless, it is impossible to be certain of William's original text, or of the date of all the interpolations, although many were clearly added shortly after the fire of 1184'5 Besides the De Antiquitate William wrote saints' 'Lives' for the 1 Printed Willelmi Malmesbiriensis Monachi de Gestis Pontificum Anglorum, ed. N.E.S.A. Hamilton (Rolls Series, 1870). 2 See ibid., 4, and Willelmi Malmesbiriensis Monachi de Gestis Regum, ed. William Stubbs (Rolls Series, 1887, 1889, 2 vols.), i. 2. 3 Coleman's work does not survive but William of Malmesbury's Latin translation is extant and is printed in The Vita Wulfstani of William of Malmesbury, ed. R. R. Darlington (Camden Society, third series, xl, 1928). 4 Memorials of St. Dunstan, ed. Stubbs, 92. See William of Malmesbury's comment in his De Antiquitate Glastoniensis Ecclesiae; Adam de Domerham, i. 71. William makes the same criticism of Osbern in his own Life of St. Dunstan (for which see below, 159); Stubbs, op. cit., 251. 5 The interpolation of the De Antiquitate is fully discussed by W. W. Newell, 'William of Malmesbury on the Antiquity of Glastonbury', Publications of the Modern Language Association of America, xviii (1903), 459-512, and, independently, by Armitage Robinson, 'William of Malmesbury 'On the Antiquity of Glastonbury', Somerset Historical Essays, 1—25. For specific references to the fire in two of the interpolations see Adam de Domerham, i. 23, 37. Adam of Domerham's version of the De Antiquitate and his own chronicle (Adam de Domerham, i. 1-122, ii. 303-596, respectively) are in Trinity College, Cambridge,
The Growth of the Glastonbury Traditions and Legends
159
Glastonbury monks: he wrote two 'Lives' of St. Dunstan, one of which survives and is of some value for the history of the abbey,1 and 'Lives', now lost, of St. Patrick, St. Indract and St. Benignus.2 It is clear from William's Life of St. Dunstan that he fully appreciated the value of the earliest Life of St. Dunstan.* This was by an unknown writer, probably a foreigner, whose name began with 'B' and who wrote the 'Life' shortly after Dunstan's death partly on the information of members of the archbishop's intimate circle.4 Although he was not a monk of Glastonbury, 'B' may have visited the abbey. William's De Antiquitate may be regarded as a gloss, so to speak, on two passages by 'B'. One passage, following the statement that Dunstan's father took him as a boy to visit Glastonbury, describes the place itself: 'There was within the realm of king Athelstan a certain royal island known locally from ancient times as Glastonbury. It spread wide with numerous inlets, surrounded by lakes full of fish and by rivers, suitable for human use and, what is more important, endowed by God with sacred gifts. In that place at God's command the first neophites of the catholic law discovered an ancient church, built by no human skill though prepared by heaven for the salvation of mankind. This church was consecrated to Christ and the holy Mary his mother, as God himself, the architect of heaven, demonstrated by many miracles and wonderful mysteries. To this church they added another, an oratory built of stone, which they dedicated to Christ and to St. Peter. Henceforth crowds of the faithful came from all around to worship and humbly dwelt in that precious place on the island'.5 The other relevant passage in 'B' says that 'Irish pilgrims, as well as other crowds of the faithful, had a great veneration for Glastonbury, particularly on account of the blessed Patrick the younger, who was said most happily to rest in the Lord there'.6 From these passages it could reasonably be assumed that Glastonbury was extremely ancient as a Christian site, and the burial place of St. Patrick the apostle of Ireland. But the narrative was too bald and brief to stir enthusiasm. It needed amplification, and this is what William of 1
For the printed edition see above, 157 n. 5. William mentions that he wrote lives of SS. Patrick, Benignus and Indract, besides the two lives of St. Dunstan, in his prologue to the De Antiquitate; Adam de Domerham, i. 3; cf. ibid., 24, 113. For St. Indract and his connexion with Glastonbury see below, 349 n. i. 3 See Memorials of St. Dunstan, ed. Stubbs, 252 and n. 4, 258 and n. i, 263 and n. i, 265 and n. i. 4 For 'B' and suggestions as to his identity see Stubbs, op. cit., xi-xxvi. 5 Ibid., 6-7. 6 Ibid., 10-11. 2
MS.R.5.33, fols. i—i8 V , 2i~73 v , respectively. The De Antiquitate and Adam's chronicle are in the same good thirteenth century charter hand to fol. 51*; thereafter the handwriting of Adam's chronicle becomes progressively rougher. Both De Ant. and Adam's chronicle have long marginal additions in at least three hands of the thirteenth, fourteenth and fifteenth centuries. The MS. is described in M. R. James Catalogue of Manuscripts in the Library of Trinity College, Cambridge. Cambridge 1900-4, ii. 199. For two other references to the MS. see below, 166 nn. 4 , 6 .
160
Legends, Traditions and History in Medieval England
Malmesbury undertook. To amplify he used written evidence and the observations of his own eyes, and no doubt hearsay. William knew from the first passage cited that in the remote past missionaries came to Glastonbury. To discover who these were he turned to Bede's Ecclesiastical History. There he read that at the request of king Lucius, pope Eleutherus sent missionaries to Britain in A.D. 166.* These were the first missionaries known to have come to Britain and so corresponded with 'B's 'first neophites of the catholic law'. However, 'B's statement that the missionaries found a pre-existing church 'built by no human skill' gave William more trouble. In itself it contradicted 'B's description of the missionaries as the 'first neophites'. Here William apparently abandoned 'B' and used another now unidentified source. This authority stated that the old church was built by disciples of Christ.2 William comments that this was not impossible, for, as Freculfus said, St. Philip preached to the Gauls, and he could have sent missionaries over the channel. 'But', William writes, 'I will leave such disputable matters and stick to solid facts'.3 To fill out the early history William made remarkable use of visual evidence. He gave an account of the churches at Glastonbury from his own observation and much of what he says has been confirmed by archaeology. He gives the exact location of the earliest church, which he says was of wattle but had been covered with wooden planks and roofed with lead by Paulinus archbishop of York (625-633).4 This church was destroyed by the 1184 fire and no trace of it remains. However archaeology has revealed within the precincts a chapel of wattle,5 a type of construction favoured by the Celts in Britain. William also records that king Ine's church, which was next to the wattle one, was enlarged by St. Dunstan by the addition of side aisles, thus changing its shape from a rectangle to a square.6 Recent excavations have confirmed this.7 Most remarkable are William's observations on the two 'pyramids' or standard crosses in the cemetery.8 He records that one was twenty-six feet 1 Historia Ecclesiastica, Bk. I, cap. iv; Bk. iv, cap. xxiv; Venerabilis Baedae Historia Ecelestas tica, ed. Charles Plummer, Oxford 1896, i. 16, 352. The mission is also noticed in the Anglo-Saxon chronicle, s.a. 167; ed. Whitelock, Douglas and Tucker, 8. 2 See Armitage Robinson, Somerset Historical Essays, 10—11. 3 Gesta Regum, ed. Stubbs, i.-23-4. 4 Adam de Domerham, i. 17-18, 28, 53-4. A parallel to this method of renovating a wattle church is in Bede who states that bishop Finan (651-661) built the church at Lindisfarne of hewn oak and thatched it with reeds. Bishop Eadbert (688-698) removed the reeds and covered the whole with plates of lead: Historia Ecclesiastica, Bk. HI, cap. xxv; ed. Plummer, i. 181 (d. Joan and Harold Taylor, 'Pre-Norman Churches of the Border', Celt and Saxon: Studies in the Early British Border, ed. N. K. Chadwick, Cambridge 1963, 254 and n. i.). Similarly Wilfrid re-roofed the church at York with lead: The Life of Bishop Wilfrid by Eddius Stephanas, ed., with an English translation, Bertram Colgrave, Cambridge 1927, 34 (cap. xvi; cf. Plummer, op. cit., ii. 188). 5 See Taylor and Taylor, op. cit., 256. 6 Adam de Domerham, i. 54, and Memorials of St. Dunstan, 271. 7 Taylor and Taylor, op. cit., 256, and the same authors' Anglo-Saxon Architecture, Cambridge 1965, i. 251. 8 Adam de Domerham, i. 44—5.
The Growth of the Glastonbury Traditions and Legends
161
high and had five panels on which were carvings of people with their names transcribed beneath. The other cross was eighteen feet tall and had four similar panels. Although the names were, as William says, almost illegible with age, he managed to transcribe most of them. He identified them as abbots of Glastonbury, thus disproving Osbern's assertion that Dunstan was the first of the line. Modern scholarship does not suggest that these were in fact the names of abbots,1 but clearly William was right to describe the 'pyramids' as 'extremely ancient'; they must have belonged to the same period, the seventh century, as the Northumbrian crosses at Ruthwell and Bewcastle. In order to give what account he could of the Anglo-Saxon abbots from the seventh century onwards, William used the evidence of early charters which enabled him to record the abbots' acquisition of lands.2 He also testified to the burial at Glastonbury of the relics of a number of Irish, Welsh and Northumbrian saints. Here he must have derived his information from the shrines he saw in the abbey and perhaps also from the liturgical calendar used by the monks. There is, in fact, tenth century evidence supporting his claim with regard to some of the saints.3 1 See Aelred Watkin, 'The Glastonbury 'Pyramids' and St. Patrick's 'Companions' ', Downside Review, Ixiii (1945), 30—41. However, Dom Aelred Watkin's identifications are, perhaps, more definite than the evidence warrants. 2 William of Malmesbury's information, brief though it is, on the Anglo-Saxon charters of Glastonbury, which date back to the seventh century, is of great value because many of the texts are lost. See H. P. R. Finberg, The Early Charters of Wessex, Leicester 1964, 15 (Professor Finberg includes William's particulars in his hand-list; op. cit., 109 f.). 3 For William on Glastonbury's relics see Adam de Domerham, i. 18—30 passim (cf. Armitage Robinson, Somerset Historical Essays, 17-21, and the same author's The Times of St. Dunstan, Oxford 1923, 98-103). William includes SS. Ceolfrid, Aidan, Benedict Biscop, Gildas (see below, 163 and n. i), and Patrick (see below and the next note). Evidence on the association with Glastonbury of a number of Northumbrian and Celtic saints in the tenth century is provided by the calendar of the Bosworth Psalter from Christ Church, Canterbury, which is based on a Glastonbury calendar almost certainly taken to Canterbury by Dunstan when he became archbishop, in the late tenth century Glastonbury calendar in the Leofric Missal, and in a late tenth or early eleventh century tract on the resting places of saints: printed respectively in Bosworth Psalter, ed. Gasquet and Bishop, 76-118, cf. P. M. Korhammer, 'The Origins of the Bosworth Psalter', Anglo-Saxon England, ed. Peter Clemoes, Cambridge 1972-5, ii. 175—80; Leofric Missal, ed. F. E. Warren, Oxford 1883, 23-34 (for the association of the calendar in the Leofric Missal with Glastonbury, see ibid., liv); Die Heiligen Englands, ed. F. Liebermann, Hannover 1889. The Bosworth Psalter and Leofric Missal both state that abbot Ceolfrid was buried at Glastonbury: Gasquet and Bishop, op. cit., 21, 106; Warren, op. cit., 31. The Leofric Missal and the tract on the resting places of saints state that Aidan was buried there: Warren, op. cit., 30; Liebermann, op. cit., 17. Moreover, the calendars show that besides the feasts of Ceolfrid and Aidan, those of Benedict Biscop (12 January), Gildas (29 January) and Bridget (i February) were kept at Glastonbury. Relics of the Northumbrian saints may have reached Glastonbury between c. 900 and c. 970: see Christopher Hohler, 'Some service books', Tenth Century Studies: essays in commemoration of the millenium of the Council of Winchester and the Regularis Concordia, ed. David Parsons, London and Chichester 1975, 69-7 i (cf. Adam de Domerham, i. 29). It should also be noted that the calendars in the Bosworth Psalter and Leofric Missal show that the feast of Paulinus (10 October) was observed at Glastonbury, which confirms that he had traditionally some connection with
162
Legends, Traditions and History in Medieval England
It was particularly necessary to emphasise that St. Patrick was buried in the abbey.1 The second passage quoted above from *B' states explicitly that it was St. Patrick the younger, that is the bishop, and the apostle of Ireland. This positive assertion indicates that Glastonbury's claim had already been questioned, as well it might, because it was far more likely that he was buried in Ireland, as the Irish contended. Probably the suggestion had been made that it was 'the elder', not his famous namesake, who rested at Glastonbury. Generally speaking William of Malmesbury's treatment of the antiquity and relics of Glastonbury was restrained and scholarly. The same cannot be said of those who succeeded him as the abbey's historians. As the twelfth century proceeded the monks' interest by no means declined but study became more disreputable; they stopped at neither fantasy nor forgery. Soon after William of Malmesbury wrote the De Antiquitate, a Welshman, Caradoc of Llancarfan, composed a Life of St. Gildas for the Glastonbury monks.2 Little is known for certain about Caradoc—only a colophon in one MS. of the Life attributing the work to him,3 and a brief notice by Geoffrey of Monmouth: Geoffrey speaks of him as a contemporary who wrote a history of the later Welsh kings, from the point where his own history ended.4 This work has not survived, but Geoffrey's statement suggests that Caradoc was a writer of some reputation. It has been argued on the grounds of the pro-Glastonbury bias in the Life of St. Gildas that Caradoc, having been a canon of Llancarfan, became a monk of Glastonbury.5 However, it seems equally likely that the Glastonbury monks commissioned him to write. His Life amplifies a brief statement by William of Malmesbury that Gildas spent many years at Glastonbury, 1
William of Malmesbury had stated that St. Patrick the apostle of Ireland came to Glastonbury; ibid., 18-19 (cf. Armitage Robinson, Somerset Historical Essays, 12—17). Besides the passage cited from 'B', the two calendars cited in the previous note show that the feast of Patrick 'the bishop', i.e. the apostle of Ireland (17 March), was kept at Glastonbury. The tenth-eleventh century list of saints' resting places states simply that 'St. Patrick' was buried there (Liebermann, op. cit., 17). The two calendars also show that the feast of Patrick 'the elder' (24 August) was kept at Glastonbury, and the Bosworth calendar adds that he rested there; see Gasquet and Bishop, op. cit., 21. For the problem as to which St. Patrick was buried in the abbey see Memorials of St. Dunstan, ed. Stubbs, Ixxviii n. 3, 10 and n. 8; C. H. Slover, 'William of Malmesbury and the Irish', Speculum, ii (1927), 271—3; H. P. R. Finberg, 'St. Patrick at Glastonbury', Irish Ecclesiastical Record, cvii (1967), 345—54 passim; the latter article is reprinted in the same author's West-Country Historical Studies, Newton Abbot 1969, 70-88. Cf. below, 165. 2 Printed in T. Mommsen, Monumenta Germaniae Historica, Auctorum Antiquissimorum, xiii, Chronica Minora, Berlin 1898, iii. 107-10. 3 Ibid., 3, 110. Cf. J. S. P. Tatlock, 'Caradoc of Llancarfan', Speculum, xiii (1938), 140. 4 Hisloria Regum Britanniae, Bk. XH, cap. xx; ed. Acton Griscom, 124 and n. 2, 125, 536. 5 Tatlock, op. cit., 141. the abbey (see above, 344 and n. 3). I am indebted to Mr. Christopher Hohler and Mr. C. P. Wormald for help with the subject matter of this note.
163
The Growth of the Glastonbury Traditions and Legends 1
died there and was buried in front of the altar in the wattle church. To this outline Caradoc added graphic details of Gildas's flight from his hermitage near the Bristol channel and his stay at Glastonbury, ending with his urgent request on his deathbed to be buried in the abbey, and his honourable burial 'in medio pavimento ecclesiae sanctae Mariae'.2 The fire of 1184 was a stimulus to further invention; undoubtedly most of the interpolations in the De Antiquitate were made soon afterwards. One of them, the so-called charter of St. Patrick, clearly shows concern for the pilgrim trade. This document is dated A.D. 430 but is in fact a forgery probably made in the post-fire period. It reinforces Patrick's connexion with Glastonbury by giving full details of his arrival and career as the first abbot. It describes how he climbed the Tor, fighting his way through the undergrowth, and discovered St. Michael's oratory built by the two first missionaries. And here the Charter makes specific concessions to pilgrims—all who visit the oratory will have thirty days indulgence, and those clearing the approaches 'with axes and hatchets, to make access easier for pious Christians' are to have a hundred days indulgence.3 The monks not only tried to strengthen their claim to St. Patrick's relics, they also staked a claim to St. Dunstan's. Already in William of Malmesbury's time a rumour was current that St. Duns tan was buried at Glastonbury, and not in Canterbury cathedral. His body, it was asserted, had been stolen and brought to Glastonbury a hundred years earlier. Eadmer wrote a long, angry letter (the only known evidence for the rumour) abusing the monks for their mendacity: had he not as a boy seen the elevation of St. Dunstan's body in the cathedral before a great multitude ? Did they not remember the annual pilgrimage from Canterbury to Glastonbury in honour of the saint?4 The story with full elaboration and with its denouement after the 1184 fire is narrated in one of the interpolations to the De Antiquitate.5 It tells of the removal of Dunstan's body to Glastonbury in the early eleventh century and its 'rediscovery' in the abbey church after the fire. This far-fetched tale culminates in an account of the actual 'rediscovery'; which may well describe an event actually staged by the monks. The whole is worth closer examination because it foreshadows the 'discovery' of the bodies of king Arthur and queen Guinevere a few years later. The story has numerous circumstantial details to make it sound convincing. It alleges that shortly after Canterbury was burnt to the ground 1 Adam de Domerham, i. 18. For a tenth-century tradition connecting Gildas with Glastonbury see above, 161, n. 3. 2 For references to king Arthur in Caradoc's Life see below, 353. 3 Adam de Domerham, i. 19—22. The charter is printed in translation and discussed in Armitage Robinson, Somerset Historical Essays, 12-16. Cf. Finberg, 'St. Patrick at Glastonbury', 345-64 Eadmer's letter, which provides the evidence for Glastonbury's claim, is printed in Memorials of St. Dunstan, 412-22. Cf. ibid., cxv-cxvi, and Southern, St. Anselm and his Biographer, 282, 285. 5 Adam de Domerham, i. 35-8; cf. ibid., ii. 335-6.
164
Legends, Traditions and History in Medieval England
by the Danes in 1012, king Edmund happened to visit Glastonbury. He told the abbot about events at Canterbury. The abbot was grief-stricken, and, telling the king of St. Dunstan's virtues, asked for help in bringing his relics to Glastonbury where he had 'first imbibed the milk of religion'. The king agreed to a plan and four monks were sent to Canterbury. Their names are given and the author alleges that they had been chaplains to St. Dunstan at Canterbury and had attended his funeral. (This should reassure readers that they knew the exact location of his tomb.) Subsequently they had served his successor St. Elphege until his martyrdom. When they arrived at Canterbury they found Dunstan's tomb, his bones and ring lying on gold and topaz exactly as they remembered. On their return to Glastonbury with the relics there was much discussion about how to hide them safely in case, when peace returned, the archbishop wanted them back. Eventually it was decided that their burial place should be known only to two wise old monks, who each on his deathbed should each pass on the information to another wise old monk. The box in which the relics were placed is then carefully described. (To leave the reader in no doubt that it was the same which was later 'rediscovered'.) The story proceeds to tell how the secret of Dunstan's burial place was at last revealed after the 1184 fire. Previously a gifted young monk called John de Watelege acquired influence over his master, John Canan, at that time the possessor of the secret. Eventually after much asking, Canan gave Watelege a cryptic indication of the relics' whereabouts. But his words, though they were common knowledge after his death, baffled the monks. It was only after the fire that light dawned on two particularly assiduous brethren, Richard of Taunton and Ralph Toe. They went to the place, removed a stone and there found the box. They called the prior and convent. (The reader needed to know that there were witnesses to what followed.) The box was unmistakably the same as that previously buried. Nor could there be any doubt whose relics it contained because one end was inscribed 'S' and the other 'D'. The box was opened and inside lay Dunstan's bones and ring. The relics were carefully removed and placed with due reverence in a reliquary already containing an arm and leg of St. Oswald. It is hardly necessary to point out that this story is full of inconsistencies. Why did the abbot hear of the sack of Canterbury from king Edmund when apparently the four monks of Glastonbury who had been serving archbishop Elphege returned at that time? Why did only one monk, John Canan, know of the hiding place, when according to the original arrangement two wise old monks should have known it? Why was it only after the 1184 fire that his cryptic words were understood? Clearly the story was concocted to add the relics of an important saint to Glastonbury's store, and incidentally to deprive Canterbury of them. At the same time the monks were at pains to stress that the fire had not destroyed the abbey's other relics. Adam of Domerham's chronicle specifically mentions the salvaging of the relics of St. Patrick, St. Indract
The Growth of the Glastonbury Traditions and Legends
165
1
and St. Gildas. Glastonbury was, indeed, rich with relics; it would, an interpolator of the De Antiquitate writes, take an immense volume to enumerate them all. He continues: 'In addition to the relics of the blessed Patrick and of those saints we have mentioned, the church of Glastonbury houses many more. No place you can walk lacks relics of the blessed—the paving stones, the areas beside and above the high altar, the very altar itself, are full of relics. No wonder the resting place of so many saints is called a celestial sanctuary on earth! How happy are those, dear Lord, whose lives are changed for the better by this holy place! None can fail of salvation who has the favour and intercession of such saints! I do no injustice to religion if I believe that wherever a stone triangle or square is deliberately placed in the paving, carefully set and bonded with lead, there lies some hidden, holy relic. The place's antiquity and the abundance of saints command reverence; by night no one presumes to loiter in the church, and during the day no one would dare spit—one shudders all over at such a disgusting thought!'2 Nevertheless, despite the labours of William of Malmesbury and his successors, the abbey still towards the end of the twelfth century lacked one famous saint exclusively its own. Its claim to St. Dunstan was more than dubious. Nor was its claim to St. Patrick generally accepted. One reason why John of Glastonbury wrote his chronicle was to disprove a statement of Ranulf Higden in his very popular work written early in the fourteenth century, the Polychronicon* Higden asserted that the St. Patrick buried at Glastonbury could not be the apostle of the Irish because he was buried, as was well known, in Ireland. It was in fact St. Patrick the abbot who lived in about 850. John angrily denied this, using various specious arguments.4 What the Glastonbury monks needed above all after the 1184 fire was an outstanding patron saint. At this point, their study of hagiography having failed, they turned to another great literary tradition, that of romance. Surely this would yield more spectacular results. And indeed it did, for in 1191 they exhumed the bodies of king Arthur and queen Guinevere. The story of the exhumation is so well known that it is not necessary to dwell on it. Briefly, in 1191 the abbot Henry de Soilli ordered the monks 1 Ibid. ii. 335. St. Indract was according to tradition the son of an Irish king who followed St. Patrick to Glastonbury and was murdered in the vicinity with his companions by robbers. William of Malmesbury wrote his Life, now lost (see above, 159 and n. 2), and asserts that king Ine translated his relics to Glastonbury and placed them in a pyramid on the left of the high altar (Adam de Domerham, i. 35—6), which may be true; see G. H. Doble, 'Saint Indract and St. Dominic', Somerset Record Society, Ivii (194.2) Collectanea, iii, 3, 21. There is evidence suggesting that he was venerated at Glastonbury at least by the early eleventh century; ibid., i. 2 Ibid., i. 27. 3 Polychronicon Ranulphi Higden Monachi Cestrensis, ed. Churchill Babington and J. R. Lumby (Rolls Series, 1865-86), v. 304, 306. 4 Johannes Glastoniemis, i. 7-8.
166
Legends, Traditions and History in Medieval England
to dig between the two pyramids in the cemetary. They, surrounded by a crowd but screened from direct view by curtains, dug down over sixteen feet. Eventually they found an oak coffin containing the gigantic bones of a man, with a huge skull bearing the marks of ten wounds, and the skeleton of a woman with a lock of beautiful golden hair: one monk, jumped impetuously into the hole to seize the hair which turned to dust at his touch. In the grave was a lead cross, inverted under a stone, and on it was a Latin inscription: 'Here lies buried the renowned king Arthur on the Isle of Avalon'. The monks then reburied the bones in a mausoleum in their church.1 (The presence of witnesses and the inscription identifying the bones recalled the 'rediscovery' of St. Dunstan's bones.) The narrative of the exhumation has to be reconstructed from a number of sources, all of which have some information not in the others. Only one, the latest in date, is by a monk of Glastonbury, Adam of Domerham.The earliest known accounts are by Gerald of Wales. He wrote two: one is in his De Principis Instructione which he began before 1192;* the other is in his Speculum Ecclesiae, written in about 1217.* Adam of Domerham's chronicle also has two accounts, both shorter than Gerald's and one a marginal addition.4 The latter is derived almost word for word from the Speculum Ecclesiae or from a common source.5 The other, although expressed in different words,6 is substantially the same as Gerald's account with the exception of one or two additional statements (notably the mention of the curtains). It seems most likely that 1
What is meant by the monks' church baffled Dr. Nitze, who argues that the Lady Chapel is intended. This, however, leads to problems with the texts which are unnecessary if it is assumed that the new tomb was put in the abbey church even though rebuilding was in progress. See W. A. Nitze, 'The Exhumation of King Arthur at Glastonbury', Speculum, ix (1934), 360—1. 2 Giraldi Cambrensis Opera, ed. J. S. Brewer and others (Rolls Series, 1861—91, 8 vols.), viii. 126—9. For the date of De Principis Instructione see ibid., xiv—xviii. 3 Ibid., iv. 47-51. For the late date of the Speculum Ecclesiae see The Autobiography of Giraldus Cambrensis, ed. and translated H. E. Butler, with an introductory chapter by C. H. Williams, London 1937, 351. 4 Both accounts are on fol. 26 V in Trinity College, Cambridge, MS.1^.5.33: printed pardy in two columns in Adam de Domerham, ii. 341—3 (the added account is in the right hand column on page 341 and in the left on page 342, and continues on page 343 which has no columns). 5 For an addition to the account in the Speculum see below, 170 n. 4. 6 One short passage (Adam de Domerham, ii. 342) resembles the Speculum Ecclesiae. It reads 'Dehinc tumbam reginae, Arturo consepultae, aperientes, tricam muliebrem flavam et fbrmosam, miroque artificio consertam, inveniunt'. The Speculum (Giraldus Cambrensis, iv. 47) reads 'inventa fuit in eodem sepulchro trica muliebris, flava et formosa, miroque artificio conserta et contricata, uxoris scilicet Arthuri, viro ibidem consepultae'. However, this does not prove that the author of the account in Adam was here copying the actual words of the Speculum because the passage is apparently the work of a reviser, being mainly on an erasure and with one word in the margin. Another passage in Adam (Adam de Domerham, ii. 342) resembles one in Gerald's De Principis Instructione. It reads 'os unius tibiae a terra usque ad medium cruris et amplius in magno viro attingeret'. The passage in De Princ. Instr. (Giraldus Ca?nbrensis, viii. 129) reads 'Os enim tibiae ipsius appositum Itibiaej longissimi/viri loci, quern nobis abbas ostendit et juxta pedem illius terrae affixurn, large tribus digitis trans genu ipsius se porrexit'.
The Growth of the Glastonbury Traditions and Legends
167
Gerald's and Adam's narratives descend from some exemplar which contained all the details now divided between the two authors.* None of the above accounts give the date of the exhumation. This appears in two Cistercian chronicles, that of Ralph de Coggeshall (written c. 1223)* and that of Margam abbey in Glamorgan (of c. 1234).3 Comparison of the entries makes it clear that Coggeshall's is a shortened version of that in the Margam chronicle.4 Since direct borrowing is most unlikely, it must be supposed that there was a common exemplar. This narrative, despite one important variant,5 in general follows Gerald of 6 Wales. The postulated exemplar from which the extant accounts by Gerald of Wales and Adam of Domerham descend was almost certainly by Gerald himself. This is suggested by the fact that he seems to have attended the exhumation and written from his own observation.7 The graphic details must surely have been written by an eyewitness: there are the mention of 1 The evidence does not warrant the firm opinion expressed by Dr. Nitze (op. cit., 360) that Adam's account is independent of Gerald's. Professor Treharne and Professor Alcock also treat Gerald and Adam as independent sources: Treharne, The Glastonbury Legends, 102—3, and Leslie Alcock, Arthur's Britain, London 1971, 74, respectively. 2 Radulphi de Coggeshall Chronicon Anglicanum, ed. Joseph Stevenson (Rolls Series, 1875).
36.3
The chronicle of Margam in Annales Monastici, ed. H. R. Luard (Rolls Series, 1864-9), i. 21-2. 4 Nitze's contention (op. cit., 359) that the entry in the Margam chronicle is dependent on Coggeshall cannot be sustained. Coggeshall follows almost verbatim the Margam text tor three quarters of the account and then ends abruptly. 5 In addition to the important variant discussed below (172) the version of the inscription on the lead cross differs in both Margam and Coggeshall from that given by Gerald of Wales, and agrees with that given by Adam of Domerham: Gerald alone adds 'cum Wenneuereia vxore sua secunda'. As this reference to Guinevere does not appear in the copy of the inscription given by Camden (see below, 352 n. 7), it seems likely that it was Gerald's invention. Professor Alcock argues that Coggeshall's account differs from Gerald's in another particular, i.e. in the description of the coffin: Coggeshall states that the bones lay 'in quodam vetustissimo sarcophago', while Gerald places them 'in terra quercu concava': Coggeshall, 36; Giraldus Cambrensis, viii. 127; Alcock, Arthur's Britain, 74. However, there is no conflict here because in the medieval period the word sarcophagus does not necessarily mean, as it did in Roman times, a stone coffin; it merely signifies a coffin: see R. E. Latham, Revised Medieval Latin Word-List, London 1965, 419, under sarcographia. Indeed, Adam of Domerham describes the coffin in question as sarcophagum ligneum mirae magnitudinis: Adam de Domerham, ii. 341. 6 Although there is general similarity in content, verbal echoes are hard to find because the Margam account is so much shorter than Gerald's narratives. Professor Treharne and Professor Alcock both treat Coggeshall as an independent authority for the exhumation: Treharne, op. cit., 93, and Alcock, op. cit., 74. As they overlook the Margam chronicle they fail to recognise that it and Coggeshall's account derive from an exemplar which may well have been of Glastonbury provenance (see below, 172 and n. 2). 7 Treharne (Glastonbury Legends, 97) expresses the opinion that Gerald wrote his account after visiting Glastonbury in 1192 or 1193. The evidence for this date is the fact that Gerald saw the new tomb already prepared, but while Henry de Soilli was still abbot (he was consecrated bishop of Worcester in December 1193). However, it seems almost certain that he was an eyewitness; he could have seen the tomb on a later occasion, or heard about it.
168
Legends, Traditions and History in Medieval England 1
the curtains and the description of the monk jumping into the hole to seize the lock of hair.2 Gerald's extant accounts refer to the author's presence. Abbot Henry 'showed us' the tibia of one of Arthur's legs, demonstrating its size by holding it against the leg of the tallest man present—it reached three inches above the knee.3 Similarly, having described the exact position of the lead cross, Gerald writes that 'it was removed from under the stone and handed round by abbot Henry so that we could examine it and read the inscription'.4 There can be no reasonable doubt that the exhumation of king Arthur and queen Guinevere was bogus, a spectacle put on for the credulous public.5 The very existence of king Arthur is not established beyond dispute6 and certainly Guinevere was a figment of the imagination. The conclusion cannot be avoided that the Glastonbury monks deliberately buried two skeletons, complete with inscribed cross, and then staged the 'discovery'. The circumstances are hard to explain in any other way. The 'discovery' was so very opportune in view of the abbey's financial plight (the inscribed cross itself presents serious problems).7 And why the curtains if not to conceal the details of what was going on? The reasons why the monks decided to adopt king Arthur as their patron, which made it necessary to find his body so that his tomb could become the centre of a cult, are not immediately obvious. Probably a number of factors attracted the monks to the idea. The fame of king Arthur was widespread, particularly among the nearby Welsh. The oral 1
Adam de Domerham, ii, 341. Giraldus Cambrensis, iv. 47—8; viii. 127. Ibid., viii. 129; cf. above, 166 n. 6. 4 Ibid., iv. 50. 5 Professor Alcock argues in favour of the authenticity of the exhumation. He asserts that though the forgery of documents was well known in the Middle Ages, 'the Glastonbury exhumation, if it was a fake, was of a quite different character from the general run of monkish forgeries' (op. cit., 76). His argument neglects the probability that the monks themselves had recently faked the 'discovery' of St. Dunstan's relics. Professor Alcock also states that he would find it easier to accept that the exhumation was bogus if 'we believed that Glastonbury already had a traditional Arthurian connection' (op. cit., 80; cf. 75—6). Here he overlooks the fact that Glastonbury had already been associated with king Arthur in Caradoc of Llancarfan's Life of St. Gildas (a work not referred to by Professor Alcock); see below, 169. 6 For a summary of the evidence on king Arthur's historicity see J. S. P. Tatlock, The Legendary History of Britain (University of California Press 1950), 178-229. See also Thomas Jones, 'The Early Evolution of the Legend of Arthur', Nottingham Mediaeval Studies, viii (1964), 3—21, (this article was first published in Welsh in Bulletin of the Board of Celtic Studies, xvii (1958), 237-52, and was translated by Gerald Morgan), and K. H.Jackson, The Arthur of History', Arthurian Literature in the Middle Ages, ed. R. S. Loomis, Oxford 1959, 1-11. Professor Alcock argues in favour of the authenticity of the Arthurian entries in the Annales Cambriae the extant text is mid-tenth century although it used earlier material; Alcock, op. cit., 45-55. 7 William Camden gave the inscription but did not reproduce the cross itself in his Britannia, London 1586, 104. In the sixth edition (London 1607), 166, he gives a picture of the cross with the inscription on it. All discussion of the cross and inscription depends on accepting that Camden's picture of the cross and his epigraphy of the inscription is correct (it could of course be the result of his antiquarianism). Professor Alcock's opinions (op. 2
3
The Growth of the Glastonbury Traditions and Legends
169
legends had received definition and elaboration in Geoffrey of Monmouth's Historia, a best seller,1 which had popularised Arthur's reputation as never before. In one respect Arthur was a very suitable patron for Glastonbury. Geoffrey, following Nennius,2 said that Arthur fought with the image of the Virgin Mary on his shield 'which forced him to think perpetually of her' :3 the abbey church, then in process of rebuilding, was dedicated to St. Mary. However, Geoffrey of Monmouth did not associate king Arthur with Glastonbury. This connexion was apparently first made in writing by Caradoc of Llancarfan in his Life of St. Gildas. Caradoc wrote under the influence of romance literature as well as hagiography, and brought king Arthur to Glastonbury at the time when Gildas was living there. Arthur 'the tyrant' came with an army to snatch Guinevere back from her abductor, the wicked king Melvas. A reconciliation between the kings was achieved by the abbot and Gildas, and in return both kings granted lands to the abbey, promising reverence and obedience, and never to injure it.4 Caradoc says nothing of Arthur's burial at Glastonbury. For this the monks had to rely on the passage in Geoffrey's Historia which stated that Arthur when mortally wounded at the battle of Camlan was carried to the Isle of Avalon.5 All that was necessary was to identify Avalon with 1
For the popularity of the Historia see Gransden, Historical Writing in England c. 550—c. 1307, 201—2 and nn. 2 See Historia Brittonum, cap. Ivi; ed. T. Mommsen, M.G.H., Auctorum Antiquissimorum, xiii, Chronica Minora, Hi. 199. 3 Historia Regum Britanniae, Bk. ix, cap. iv; ed. Acton Griscom, 438. 4 Mommsen, op. cit., 109—10. 5 Historia Regum Britanniae, Bk. xi, cap. ii; ed. Acton Griscom, 501. cit., 78—80), which coincide with those of C. A. Ralegh Radford (see The Quest for Arthur's Britain, ed. Geoffrey Ashe, London 1968, 126-38 passim), deserve attention. Accepting Camden's reproduction as accurate, he dates the epigraphy to the tenth or eleventh century. He suggests that the monks of St. Dunstan's time copied a sixth century inscription, perhaps reading 'Hie sepultus iacit Arturius,' which marked the burial place of king Arthur which was, it is postulated, next to a mausoleum in the old cemetery. They copied it to mark Arthur's grave when the mausoleum was demolished, prefixing 'Rex inclitus' and appending 'in insula Avalonia'. It should be noted that Professor K. H.Jackson dates the epigraphy of the inscription to the sixth century (see his review of Professor Alcock's book in Antiquity, xlvii (1973), 81); this suggests the possibility that the cross was the actual one used to mark Arthur's original burial. However, it seems much more likely that the monks faked the cross in 1191 and deliberately used archaic letter forms. It is most improbable that the monks of St. Dunstan's day (and even less of the sixth century) would have added 'in Avalonia', because the identification of Glastonbury with Avalon was first made, at least in writing, by Gerald of Wales (below, 170 & n. 1) . And they might well have imitated the lettering of an old inscription, perhaps one in the cemetery; the ancient letters on the two pyramids had certainly attracted attention (see above, 345, and Giraldus Cambrensis, viii. 127). In this context an example of visual antiquarianism at Wells may be noted. Early in the thirteenth century the canons commissioned a series of effigies of the pre-Conquest bishops of Wells, to increase their prestige in their struggle with the monks of Bath over the right to elect the bishops. Two of these wear the low, rounded mitres and simple dress of the Anglo-Saxon period ('clear signs of genuine antiquarianism'): Lawrence Stone, Sculpture in Britain in the Middle Ages, Penguin Books 1955, 106—7. I am indebted in writing this foot-note to a discussion with Mr. P. W. Dixon.
170
Legends, Traditions and History in Medieval England
Glastonbury. This identification first appears in Gerald of Wales.1 It may well have been thought that people could easily be persuaded to accept it because Glastonbury was surrounded by lakes and marshes, and the Tor has a mysterious appearance. There is, however, no proof that the actual suggestion originally came from the monks. Possibly it came from the royal court, from Henry n himself. Gerald of Wales asserts rather vaguely that the king 'told the monks everything', having 'heard about it from an ancient Welsh bard knowledgeable on history', that is, exactly where and how deep to dig and what they would find.2 He also advised them to lay the bones in a marble tomb.3 The addition to Adam of Domerham asserts that abbot Henry de Soilli, who was elected in September 1189, was a relative and former friend of king Henry.4 Gerald may well have been right in attributing the suggestion to Henry ii. Romance literature and the cult of king Arthur flourished at court. King Henry had been brought up partly by his uncle, Robert of Gloucester, to whom Geoffrey of Monmouth dedicated the first issue of his Historia,5 and Henry himself commissioned a romance history, the Roman de Rou.6 In addition Henry was very concerned over the abbey's financial troubles. The abbacy was vacant from 1180, when abbot Robert died, until Henry de Soilli's succession, and after the fire Henry n allowed what revenues remained after the monks had been provided with their necessities, to contribute to the rebuilding rather than go to the royal exchequer.7 He may have realised that a flourishing Arthurian cult would help matters. It would also be useful to him politically because it would publicise that king Arthur was actually dead: in current folklore it was believed that he merely slept on the Isle of Avalon and would one day awaken and again lead the Britons to victory. Such a view was damaging to a king trying to subdue the Welsh.8 1
Giraldus Cambrensis, iv. 49; viii. 128. 'Rex Angliae Henricus secundus, sicut ab historico cantore Britone audierat antique, totum monachis indicavit, quod profunde, scilicet in terra per xvi. pedes ad minus, corpus invenirent, et non in lapideo tumulo sed in quercu cavato : ibid., viii. 128. 'Dixerat enim ei [i.e. to Abbot Henry] pluries, sicut ex gestis Britonum et eorum cantoribus historiis rex audierat, quod inter pyramides duas, quae postmodum erectae fuerant in sacro coemeterio, sepultus fuit Arthurus . . .': ibid., iv. 49. 3 'Dictus autem abbas corpore reperto, monitis quoque dicti regis Henrici, marmoreum ei sepulchrum fieri Fecit egregium . . .': ibid., iv. 51. 4 'Conditus fuit Rex Arthurus (sicut per regem Henricum abbas Henricus didicerat, cujus consanguineus et dudum familiaris extiterat: qui eciam Rex hoc ex gestis Britonum, et eorum cantoribus historicis, frequenter audierat . . .': Adam de Domerham, ii. 341. 5 See J. S. P. Tatlock, 'Geoffrey and King Arthur in Normannicus Draco', Modern Philology, xxxi (1933), 124-5. 6 Maistre Wace's Roman de Rou et des Dues de Normandie, ed. H. Andresen, Heilbronn 1877, 1879, and Chronique des Dues de Normandie par Benoit, ed. Carlin Fahlin, Lund 1951-67- See Tatlock, Legendary History of Britain, 465, 467, and M. D. Legge, 'The influence of Patronage on Form in Medieval French Literature', Stil- und Formprobleme in der Literatur, ed. Paul Bockmann, Heidelberg 1959; Vortrage des VII Kongresses der Internationalen Vereinigung fur moderne Sprachen und Literaturen in Heidelberg, 139. 7 Adam de Domerham, ii. 334-5. 8 Tatlock, 'Geoffrey and King Arthur in Normannicus Draco 122—3. 2
The Growth of the Glastonbury Traditions and Legends
171
Court interest in the Arthurian cult at Glastonbury did not end with Henry n. In 1278 Edward i, his queen Eleanor, and their retinues, visited Glastonbury. The tomb of king Arthur and queen Guinevere was solemnly opened in their presence. Adam of Domerham describes the occasion in detail.1 The next day the king placed Arthur's bones in one casket and the queen placed Guinevere's in another; each sealed the respective casket with the royal seal, and the caskets were put back in the tomb, which had been moved, at Edward's command, to a place in front of the high altar. Edward's motives for the disinterment were similar to those of Henry n: he was a devotee of the cult of king Arthur ;2 he wanted to help the monks raise money (in this instance partly to finance abbot John of Taunton's building activities);3 and, because of his attempts to impose English rule on the Welsh following the treaty of Conway (November 1277), he needed it to be known that king Arthur was dead.4 An objection to attributing the monks' decision to 'discover' king Arthur to the influence of Henry n is the time factor: Henry died in July 1189 and the monks did not dig until 1191. But this does not present an insuperable difficulty. Henry's advice would not necessarily have been taken at once, and after his death the need for money became even more acute: in the words of Adam of Domerham, king Richard 'turned his mind to matters of war and took no interest in our new church, and so work on the building came to a standstill because there was no one to pay the labourers'.5 Once the monks had made the decision, maximum publicity was desirable. It seems likely, therefore, that lacking a writer of their own, they invited Gerald of Wales to attend the exhumation and commissioned him to write it up afterwards (in the same way as the canons of Hexham and the abbot of Westminster had commissioned Ailred of Rievaulx).6 They would have considered Gerald a suitable person: it is even possible that Henry n himself suggested him, because in about 1184 Gerald had entered royal service.7 Gerald was accustomed to recording his experiences and observations: he wrote on the Irish campaign of 1187 on which he accompanied John at king Henry's command;8 and he wrote on 1
Adam de Domerham, ii. 588-9. For Edward f s interest in king Arthur (with a reference to his visit to Glastonbury) see R. S. Loomis, 'Edward I, Arthurian Enthusiast', Speculum, xxviii (1953), 114—27. 3 See VCH, Somerset, ii, 91. For a list of buildings put up by John of Taunton see Adam de Domerham, ii. 573. 4 See F. M. Powicke, King Henry III and the Lord Edward, Oxford 1947, ii. 724. 5 Adam de Domerham, ii. 341. Treharne (Glastonbury Legends, 106) does not accept that Henry n could have originated the idea of the search for king Arthur's bones. 6 Above, 156. 7 The Autobiography of Giraldus Cambrensis, ed. Butler, 81 and n. i. 8 Ibid., 86 ff. After the campaign Gerald wrote the Expugnatio Hibernka and the Topographia Hibernica (printed in Giraldus Cambrensis, v); he dedicated the Topographia to Henry n (ibid., 20) and the Expugnatio, first to king Richard and then to king John (ibid., 222, 405). 2
172
Legends, Traditions and History in Medieval England
archbishop Baldwin's tour of Wales in 1188, when he helped to preach the crusade.1 The attempts of the monks of Glastonbury to publicise king Arthur's exhumation probably extended further than the commissioning of Gerald. It is likely that they distributed propaganda pamphlets to other religious houses,2 which probably had at least one important variant from the original narrative. As has been mentioned3 both Coggeshall and the Margam chronicler, who surely derived their accounts from some such broadsheet, differ in one particular from Gerald and Adam. Instead of attributing the discovery to deliberate excavation undertaken on Henry n's advice, they ascribe it to chance: a monk had expressed a strong desire to be buried between the pyramids and the gravediggers unexpectedly came upon the bones of Arthur and Guinevere.4 Presumably the monks of Glastonbury suppressed the part played by Henry n because they considered that the story sounded less contrived without it. They may also have feared that mention of king Henry's involvement would alienate readers because he was not popular with the Church. Although Gerald of Wales gave a full record of king Arthur's exhumation, his work as a piece of propaganda had one weak link: unless the public were sure that Glastonbury had once been Avalon, it was impossible to be certain that the bones discovered were really those of Arthur and Guinevere, or that the inscription on the lead cross was not a fake. Gerald had simply stated that Glastonbury was the former Avalon, giving independent etymologies of both names.5 Glastonbury meant the glassy borough in the English language, and Avalon was the island of apples in Celtic. But there was nothing to connect the one with the other. In order to secure the identification the monks adopted and elaborated a pre-Conquest eponymous foundation story. Their version of the eponym is one of the interpolations in the De Antiquitate.6 It alleges that Glastonbury took its name from a swineherd called Glasteing. He lost his sow, a remarkable creature with eight legs, and pursued her for a long distance. Eventually, taking the track known as 'Sugewege' (in Celtic 'the marshy way') from Wells, he reached Glastonbury. There he found his sow suckling her piglets under an apple tree. Delighted with the fertility of the place, he made it his home with his family. Thus Glastonbury took its name from Glasteing. By adopting this 1 Ibid., 99 ff. After the tour Gerald wrote the Itinerarium Kambriae and the Descriptio Kambriae (printed in ibid., vi.). 2 A copy may have reached Christ Church, Canterbury, in the early thirteenth century: this is suggested by the fact that Gervase of Canterbury knew of the identification of Avalon, as Arthur's burial place, with Glastonbury. See The Historical Works of Gervase of Canterbury, ed. William Stubbs (Rolls Series, 1879, 1880), ii. 19. 3 Above, 167. The Margam chronicle also alleges that Mordred was buried with Arthur and Guinevere. 4 Radulfus de Coggeshall, 36; Annales Monastici, i. 21. 5 Giraldus Cambrensis, iv. 49; viii. 128. 6 Adam de Domerham, i. 16—17.
The Growth of the Glastonbury Traditions and Legends
173
etymology the monks were abandoning Gerald of Wales's. But they kept his explanation of Avalon, as the island of apples, adding, however, that it was probably so called because Glasteing found his sow under one of the apple trees. And so the story proved that Glastonbury and Avalon were one and the same place.1 The Celtic affiliations of the eponym are well known.2 There was an early tale that St. Patrick had resurrected and baptised a swineherd called Cass mac Glaiss. In the course of the ninth century the name became Glass mac Caiss and the story was localised at Glastonbury. Here were the seeds of the later eponym—but nothing to identify Glastonbury with Avalon. Concentration on the Celtic aspects of the story has tended to overshadow the influence of romance literature on its final form. Eponyms were a feature of romance and had their roots in classical antiquity. Virgil had, for instance, asserted that Rome took its name from Romulus. Moreover, an animal often played an essential part in ancient foundation myths. There was the wolf in the Romulus and Remus story, and even more relevant to our purpose the sow in the story of the foundations of Rome's ancestor cities, Lavinium and Alba Longa. Virgil mentions the white sow and her thirty piglets in this context,3 while the Greek historian Dionysius relates that the sow was pursued by the Trojans over a distance of twenty-four stades and was found giving birth to her litter on the site of the future city.4 However, the direct influence on the Glastonbury monks was Geoffrey of Monmouth. The writer's account of the source of his story is itself reminiscent of Geoffrey: he states that it was taken 'from ancient books of the Welsh' ;5 Geoffrey had claimed that the source of the Historia was 'an ancient book in the Welsh tongue'.6 Moreover, the form of the story may well have been borrowed from Geoffrey. He was particularly fond of eponyms, alleging, for example, that Britain took its name from the first settler Brutus, London from king Lud, and Colchester from king Cole. It should be noted that the Glastonbury eponym incorporates a minieponym, as it were: the sow is the eponymous heroine of 'Sugewege' which the monks by a false etymology now called Sows' Way.7 (To this day a farm off the Wells Road is called Southway.) The monks of Glastonbury had, therefore, by the end of the twelfth 1
Previous scholars do not seem to have appreciated the use of the Glasteing eponym in linking Glastonbury with Avalon. 2 See Slover, 'William of Malmesbury and the Irish', 276-80 and nn. for further references, and Finberg, 'St. Patrick at Glastonbury', 354-5 and n. 36. 3 Aeneid, Bk. in, lines 389-93; Bk. vin lines 42-5. 4 See Virgil, Aeneid Book VIII, ed. K. W. Gransden, Cambridge 1976, 188. 5 'Haec de antiquis Britonum libris sunt': ibid., 17. 6 Geoffrey states that his source was a book lent him by Walter, archdeacon of Oxford— 'quendam britannici sermonis librum vetustissimum': Historia Regum Britanniae, Bk. i, cap. i; ed. Acton Griscom, 219. 7 The false etymology resulted from the confusion of sugga, a marsh, with sugu, a sow; see Finberg, op. cit., 354.
174
Legends, Traditions and History in Medieval England
century at last acquired in king Arthur a much needed patron, exclusively theirs. But a famous patron constituted only half, so to speak, of an abbey's prestige; the other half was its antiquity.1 The story of Glasteing only gave an explanation of Glastonbury's origin as an inhabited place, not as a Christian centre. And for propaganda purposes the monks needed to trace their community's Christian origins. To do this they resorted after the 1184 fire to both the hagiographical and to the romance literary traditions. The results of their researches are to be found in the so-called Charter of St. Patrick. By reference to Geoffrey of Monmouth they were able to enlarge on William of Malmesbury's statement that, as far as could be known with any certainty, the founders of the abbey were two missionaries sent by pope Eleutherus in A.D. i66.2 Geoffrey supplied the missionaries' names (although he made no mention of them coming to Glastonbury); they were called Pagan and Deruvian.3 William of Malmesbury had hinted at even earlier origins, tentatively suggesting that the church ('built by no human skill') which the two missionaries found was the work of disciples of St. Philip.4 Now the monks were more specific: the charter of St. Patrick states categorically that both St. Philip and St. James had sent twelve disciples who, as instructed by the archangel Gabriel, had built the church at Glastonbury.5 The date of the conversion of Britain was, therefore, almost contemporary with that of France and Spain, an idea sure to appeal to the incipient patriotism of the late twelfth century. The Glastonbury monks had only to find a leader for the twelve founding disciples and supply an exact date. This they did in the mid-thirteenth century. They set St. Joseph of Arimathea,* whose cult was growing in popularity, at the head of the band, and dated the arrival at Glastonbury to A.D. 6s.6 Thus Glastonbury abbey acquired the earliest foundation date of any in England, and a patron whose fame survives even to this day. The monks dove-tailed the legend of St. Joseph into that of king Arthur by asserting that St. Joseph was an ancestor of king Arthur—Adam of Domerham gives the genealogy in true romance style.7 Henceforth the legends of both developed side by side; but their later evolution, which can be read in the pages of John of Glastonbury, belongs to literature rather than historiography. 1
See above, 153. 160. Historia Regum Britanniae, Bk. iv, cap. xix, ed. Acton Griscom, 328-9; Adam de Domerham, i. 19—21. 4 Above, 160. 5 Adam de Domerham, i. 20. 6 See Armitage Robinson, Two Glastonbury Legends, 28. 1 Johannes Glastoniensis, i. 56-7. 2 Above, 3
Note: For revisory notes to this chapter see below pp. 329—30.
7
Realistic Observation in Twelfth-Century England T. D. KENDRICK has already commented on 'nascent medieval topography' (which he describes as 'rather casual') in England, and cites examples of topographical descriptions from chronicles.1 It is proposed here to examine in more detail the ability of medieval writers in twelfth-century England to see and describe the world around them. Besides topographical observation, I shall include observation of small objects (such as goldsmiths' work and books), of mankind itself (people's physical appearance, character and behaviour both individually and corporately as social beings), and of animals and birds. Writers had various motives for descriptive writing. Admiration of the beautiful and wonder at the extraordinary were constant motives throughout history. Moreover, a writer often had a commemorative intention: he might, for example, want to preserve for posterity the appearance of a great man, or of a work of art in order to commemorate the artist.2 But to some extent visual sense is an individual gift. Therefore some writers (such as William of Malmesbury), with a developed visual sense, will always turn more readily than others to descriptive writing. Two intellectual trends could also encourage realistic description. The first was Christian piety. A writer's spiritual love, centred on God and his saints, stirred his interest in any person or object with a holy association. His earthly loyalty and affection were centred on his own home — usually a monastery. Therefore he was likely to observe carefully anything contributing to its reputation — for example, any building of outstanding beauty, or one which proved the place's great antiquity, or a relic of the patron saint, which demonstrated its holiness. Visual perception became particularly acute if the house's business interests were at stake. Monks began to study palaeography and diplomatic because they needed to defend their houses' property and privileges against enemies armed with forged charters. The second intellectual trend encouraging descriptive writing was the study of the classics. The influence of classical antiquity took various forms. Admiration for classical civilization sometimes led to an interest in archaeological remains of Roman Britain. Moreover, study of some classical authors provided models for descriptive writing. Thus Suetonius' Lives of the Caesars had descriptions of the Caesars' physical appearance. (William of Malmesbury adopted the Suetonian mode for his descriptions of the Anglo-Norman kings). Probably more important for our purpose was the attitude of mind engendered by classical studies. The 1 T. D. Kendrick, British Antiquity (London, 1950), p. 184. For a large collection of extracts from printed editions of early documents and literary sources relating to art and architecture, including some of the passages cited below, see O. Lehmann-Brockhaus, Lateinische SchriftqueUen zur Kunst in England, Wales und Schottland vom Jahre 901 Us zum Jahre 1307 (Miinchen, 1955-1960, 5 vols.; Verqffentiichungen des ZentraUnstituts fur Kunstgeschichte in Miinchen, no. 1 etc.). 1 This motive is explicitly stated by Matthew Paris; see below p. 17 8 & n. 15.
176
Legends 3 Traditions and History in Medieval England
idea of Rome as a city of art treasures and monumental buildings provoked the competitive spirit: England too had treasures and fine cities. Furthermore, such classical authors as Sallust fostered an interest in the trivialities of everyday life. It must, of course, be remembered that the hagiographical tradition and classical studies could both discourage realistic descriptive writing. For example, the hagiographer and the author influenced by classical works usually aimed at giving the general rather than the particular truth about their hero, to show how such a saint or king would have behaved and what he would have looked like, not his actual behaviour and appearance. Similarly descriptions of inanimate objects and of the outside world were sometimes based on classical models. Therefore the reader must be on his guard against idealization and literary imitation.3 However, the examples given below which are surely authentic, suggest that it is easy to overestimate the extent to which mediaeval writers were bound by these literary conventions. The twelfth century was, until the literary developments of the fourteenth and fifteenth centuries, pre-eminent for descriptive writing. A number of writers described what they saw in detail and a few drew rational deductions from their observations. The use of observations as evidence amounted to rudimentary historical research. (Perhaps, therefore, Professor V. H. Galbraith's statement that 'It would be a brave man who sought to widen the list of medieval achievements by the inclusion of historical research'4 needs slight modification). Both the religious motive and the incentive provided by classical studies gained momentum in the twelfth century. Generally speaking, the former was strongest early in the twelfth century and the latter towards the end. The Norman Conquest and settlement presented the Anglo-Saxon church with a challenge and put its saints on trial. (Archbishop Lanfranc questioned the right of some Anglo-Saxon saints to liturgical commemoration).5 The reputation of the Anglo-Saxon church as a whole, and the prestige and prosperity of each monastery in particular, depended on having an unbroken and glorious history. Therefore hagiographers and historians turned with enthusiasm to the study of ecclesiastical history, determined to prove that the Anglo-Saxon church had a creditable past. The influence of classical authors is already apparent early in the century (notably in the works of William of Malmesbury), but it is particularly marked in the works of such late twelfth-century 'humanists' as Gerald of Wales and Walter Map, and resulted occasionally in objective curiosity. 8 For the influence of classical literature on mediaeval representations of heroes and rulers, and of landscape, see E. R. Curtius, Europaische Literatur und lateinisches Mittelalter (Bern, 1948), chapters ix, x. 4 V. H. Galbraith, Historical Research in Medieval England (London, 1951), p. 7. 6 For Lanfranc's attitude to Anglo-Saxon saints see The Life of St. Anselm Archbishop of Canterbury by Eadmert ed., with an English translation, R. W. Southern (Nelson's Medieval Texts, 1962), pp. 50-54. For the changes made by Lanfranc in the liturgical calendar of Christ Church, Canterbury, see F. A. Gasquet and E. Bishop, The Bosworth Psalter (London, 1908), pp. 37-39. For the test of fire to which Walter, the Anglo-Norman abbot of Evesham (1077-1086) submitted the relics at Evesham see Chronicon Abbatiae de Evesham. ed. W. D. Macray (Rolls Series, 1863), p. 323.
Realistic Observation in Twelfth-Century England
177
Although the twelfth century was most remarkable for descriptive writing, the preceding period and the thirteenth century also produced examples. Gildas, Bede andNennius were all capable of using archaeology as historical evidence.6 But here they were exceptional. Realistic descriptions by other writers were the result of admiration, wonder and piety towards saints. The description of Romano-British buildings in the Old English poem the Ruin was obviously written because of admiration and wonder.7 Eddius Stephanus must have intended to praise Wilfrid by describing his churches and the treasurers at Hexham and Ripon.8 (Presumably the biographer of Edward the Confessor had the same motive for describing the abbey church at Westminster).9 And the reference to ancient remains in Bede's Life of St. Cuthbert10 and in Felix's Life of St. Guthlac11 are incidental to the hagiographical theme. Such examples of realistic descriptions are 'casual' and isolated; clearly as yet authors had no compelling motive to write them. Again in the thirteenth century examples of realistic observations are sporadic, 6 For Gildas on the Roman wall see De Excidio Britanniae et Conquestu, cc. 15-18, ed. T. Mommsen in Monumenta Germaniae Historica, Auctorum Antiquissimorum, xin, Chronica Minora Saeculi iv-vii, in (Berlin, 1898), 33-35; Gildas' interpretation of the evidence is discussed by C. E. Stevens, 'Gildas Sapiens' in EHR, LVI (1941), 356-360. For Bede's derivation of the name Horsa apparently from a Roman tombstone see HE, i. 15; Venerabilis Baedae Historia Ecclesiastica Gentis Anglorum, ed. C. Plummer (Oxford, 1896, 2 vols.), i. 31 (cf. H. M. Chadwick, The Origin of the English Nation [Cambridge, 1924], pp. 42-43). For Nennius' interpretation of archaeological evidence at Cair Segeint (Segontium) see Historia Brittonum, c. 25, ed. Mommsen, op. cit. m, 166. Cf. F. Lot, Nennius et VHistoria Brittonum (Paris, 1934), pp. 59-60. For hoards of coins found at Segontium see R. E. M. Wheeler, 'Segontium and the Roman Occupation of Wales' in Y Cymmrodor, xxxm (1923), 111 sq. Alcuin mentions the Roman walls at York; see De Pontificibus et Sanctis Ecclesiae Eboracensis Carmen, lines 19-37, printed in Monumenta Alcuiniana praeparata a Philipp Jajfe, ed. W. Wattenbach and E. Dummler (Aalen, 1964), p. 82. For a description of Bugga's church by Aldhelm see Carmen in Ecclesia Mariae a Bugge exstructa, printed in Aldhelmi Opera, ed. R. Ehwald in Mon. Germ. Hist. Auctorum Antiquissimorum, xv (Berlin, 1919), 14-18. Brief descriptions of the topography of Lindisfarne and of the author's own monastery are in ^Ethelwulf's De Abbatibus, ed., with an English translation, A. Campbell (Oxford, 1967), pp. 10, 12. 7 For internal evidence suggesting that The Ruin refers to Bath, see Three Old English Elegies, The Wife's Lament, The Husband's Message, The Ruin, ed. R. F. Leslie (Manchester, 1961), pp. 23-26. For the poet's talent for vivid description see ibid. pp. 28-29. 8 For Eddius' description of York, Ripon and Hexham see The Life of Bishop Wilfrid by Eddius Stephanus, ed., with an English translation, Bertram Colgrave (Cambridge, 1927), pp. 35, 37, 47; cf. H. M. Taylor and J. Taylor, Anglo-Saxon Architecture (Cambridge, 1965, 2 vols.), i, 301; n, 516-518, 700-709. For Eddius' description of the Gospels in gold letters on purple parchment which Wilfrid gave to Ripon see Colgrave, op. cit. p. 37; for such gospel-books, executed from the sixth to the twelfth centuries, see E. Maunde Thompson, An Introduction to Greek and Latin Palaeography (Oxford, 1912), pp. 32-33. 9 The Life of King Edward, ed., with an English translation, Frank Barlow (Nelson's Medieval Texts, 1962), pp. 45-46. 10 St Cuthbert was looking at the Roman walls of Carlisle, and a Roman well, when he had his vision of King Ecgfrith's defeat; Two Lives of St. Cuthbert, ed., with an English translation, Bertram Colgrave (Cambridge, 1940), p. 122. 11 For a description of the prehistoric or Roman barrow inhabited by Guthlac see Felix's Life of St. Guthlac, ed., with an English translation, Bertram Colgrave (Cambridge, 1956), pp. 92-95, 182184.
178
Legends, Traditions and History in Medieval England
the result of individual interest rather than a general trend. Matthew Paris, himself an artist, had a developed visual sense. He wrote the well-known description of Henry Ill's elephant,12 and also described the buffalo, specimens of which were owned by Richard of Cornwall.13 Especially notable are his descriptions of works of art (for example of the wash-bowl presented by Queen Margaret of France to Henry III).14 His aesthetic sense was augmented by a desire to commemorate the artists who had worked at St. Albans.15 (The Gesta Abbatum is partly a history of the abbey's art and architecture). The monks of Bury St. Edmunds had a considerable antiquarian interest in their house16 and also left a graphic description of the abbey's twelfth-century illuminated Bible (now Corpus Christi College, Cambridge, MS 2).17 But only at the beginning of the century and at the very end (perhaps in the early fourteenth century) do we find lifelike descriptions of people's appearances and characters comparable to those written in the late twelfth century. Jocelin of Brakelond's description of Samson abbot of Bury St. Edmunds (1182/3-1211)18 can be regarded as an extension of the twelfth-century mode, while Nicholas Trevet's descriptions of Henry III and Edward I were the product of another humanist revival.19 Realistic observation in the twelfth century will be discussed in two parts. 12 See Brit. Mus. MS. Cotton Nero D i, f. 168V and attached slip, and the flyleaves of Matthew's autograph of the Chronica Major a in Corpus Christi College, Cambridge, MS. 16. See also Chronica Majora, ed. H. R. Luard (Rolls Series, 1872-1884, 7 vols.), v. 489; F. Madden, 'On the Knowledge possessed by Europeans of the elephant in the thirteenth Century' in The Graphic and Historical Illustrator, ed. E. W. Brayley (London, 1834), pp. 835-336, 352; and R. Vaughan, Matthew Paris (Cambridge, 1958), pp. 256-257. 13 Chronica Majora, v. 275. 14 Ibid. v. 489. Another product of Matthew's visual sense was his work on heraldry; for the coats of arms he inserted in his historical works see Rolls of Arms: Henry HI, ed. T. D. Tremlett (Harleian Society cxni-iv, 1961-1962). 16 Recording the artistic achievement of St. Albans, Matthew writes: 'Haec idcirco scripturae immortal!, ac memoriae, duximus commendanda, ut penes nos, haud ingratos, eorum vigeat cum benedictionibus recordatio, qui studioso labore suo opera ecclesiae nostrae adornativa post se reliquerunt'; Gesta Abbatum Monasterii Sancti Albani, ed. H. T. Riley (Rolls Series, 1867-1869, 3 vols.), i. 233. 16 This appears, for example in: the thirteenth-century tract on the dedication of the altars, chapels and churches at Bury St Edmunds, in Brit. Mus. MS. Harley 1005, ff. 217V-218V (extracts are printed in M. R. James, On the Abbey of St. Edmund at Bury (Cambridge Antiquarian Society, octavo series, xxvm, [1895], pp. 161-162); the account of the uncovering of the walls of an ancient round church at Bury in 1275; The Chronicle of Bury St. Edmunds 1212-1301, ed., with an English translation, Antonia Gransden (Nelson's Medieval Texts, 1964), p. 58; and in the thirteenth century Gesta Sacristarum, printed in Memorials of St. Edmund's Abbey, ed. T. Arnold (Rolls Series, 1890-1896, 3 vols.), n. 289-296. 17 Ibid. n. 290. For this passage concerning the Bury Bible in the Gesta Sacristarum see E. G. Millar, English Illuminated Manuscripts from the tenth to the thirteenth Century (Paris-Brussels, 1926), pp. 30-32. 18 The Chronicle of Jocelin of Brakelond, ed., with an English translation, H. E. Butler (Nelson's Medieval Classics, 1949), pp. 39-40. 19 Nicholai Triveti Annales, ed. T. Hog (English Historical Society, 1845), pp. 279-283. Cf. F. M. Powicke, King Henry III and the Lord Edward (Oxford, 1947, 2 vols.), II. 686-687. Trevet also gives a good topographical description of Winchelsea; Annales, p. 167.
Realistic Observation in Twelfth-Century England
179
First, it will be considered as exemplifying the Christian motive for writing. Second, it will be considered as a response to classical studies.
The desire to establish continuity with the Anglo-Saxon past is a marked feature of the history of Durham written in the early twelfth century, probably by the monk Symeon of* Durham.20 The author's obsession with continuity permeates the work, and caused him to describe some material objects. It also accounts for some statements which are of dubious credibility. The author wanted to give venerable associations even to flaws in Durham's treasures. Thus he asserts that the crack in the stone cross (which, he alleges, was the cross of Ethelwold, the bishop of Lindisfarne who died in 740), standing in St Cuthbert's cemetery, was acquired when the Danes raided Lindisfarne.21 Similarly he tells a story apparently to provide a creditable explanation for a flaw, a slight water stain at the top of some pages, in the precious Gospel-book then owned by Durham cathedral (now the Lindisfarne Gospels in the British Museum). Having briefly described the book which, he states, was written by Eadfrith, and bound with wonderful decoration in gold and gems, by the hermit Billfrith at the order of Bishop Ethelwold,22 he relates how, when the book was borne from Lindisfarne with St Cuthbert's body and carried to the mouth of the Derwent for embarkation (which was divinely prevented) to Ireland, it accidentally fell in the sea, to be miraculously recovered three days later from the beach at Whithorn.23 This story can hardly be true, because Whithorn is thirty miles from the mouth of the Derwent. However, the legend that the Gospels accompanied St Cuthbert's body is itself quite plausible.24 In the south of England the man to contribute most to vindicate the reputation of the early Anglo-Saxon church and to establish continuity between past and present was William of Malmesbury. He accused the Anglo-Saxons of neglecting 20
Printed Symeonis Monachi Opera Omnia, ed. T. Arnold (Rolls Series, 1882-5, 2 vols.), I. For post-Conquest interest in Bedan Northumbria see ibid. i. 108 sq., and M. D. Knowles, The Monastic Order in England (Cambridge, 1949), pp. 165-171. Already in the tenth century St. Oswald had searched for relics of the Northumbrian saints and tried to refound Ripon (Historians of the Church of York, ed. James Raine, [jn.] Rolls Series, 1879-1894, 3 vols., i. 462), and in the early eleventh century Alured, sacrist of Durham (great-grandfather of Ailred of Rievaulx) went relic-hunting in Northumbria; see Symeon, i. 88-89, and The Priory of Hexham, its Chronicler'St Endowments and Annals, ed. James Raine (Surtees Society XLIV, 1864, 2 vols.), i. liii. 21 Symeon, i. 39. For a reference to two marvellously carved stone crosses at the head and foot of Acca's tomb (one of which is probably to be identified with the cross now hi Durham cathedral) see ibid. ii. 33, and Taylor and Taylor, op. cit. i. 305. 22 Symeon, i. 67-68. This passage is fully discussed by T. J. Brown in Codex Lindisfarnensis, ed. T. D. Kendrick and others (Oltun and Lausanne, 1960, 2 vols.), I. 5-11. 23 Symeon, i. 64-68. This story is fully discussed by T. J. Brown, loc. cit. i. 21-23. 24 See ibid. i. 23-24. The Stonyhurst Gospel, which had a close connection with St Cuthbert, were found in St Cuthbert's tomb when it was opened in 1827; see R. A. B. Mynors, 'The Stoneyhurst Gospel' in The Relics of Saint Cuthbert, ed. C. F. Battiscombe (Oxford, 1956), pp. 357-358.
180
Legends, Traditions and History in Medieval England
their past.26 No one since Bede had tried to record English history and few had troubled to write the lives of saints. To remedy the deficiency William wrote the Gesta Regum and the Gesta Pontificum in 1125, and also hagiographies and his book on the antiquity of Glastonbury, the De Antiquitate Glastoniensis Ecclesiae. His corpus of works concerns both Anglo-Saxon history in general, and also the history of his own abbey, Malmesbury, and of Glastonbury abbey. Besides recording in the Gesta Pontificum all he could discover about the lives of the Anglo-Saxon archbishops, bishops and saints, William included what is virtually a gazetteer of ecclesiastical England. He divided the work into bishoprics. He begins each section with a description of the episcopal see, its site and buildings (the careers of the successive bishops follow), and ends with an account of the religious houses in the diocese, complete with topographical descriptions. It is likely that most (perhaps all) of William's descriptions were based on personal observation, not hearsay. He was, therefore, the first Englishman to travel to get his evidence — to write history from information collected on the spot. His notices of the generous hospitality of some monasteries (such as Reading and Tewkesbury)26 may represent formal thanks for a visit. The surviving autograph copy of the Gesta Pontificum is a small volume which could have fitted into his pocket.27 The Gesta Pontificum has descriptions of numerous churches and their treasures. Thus William describes St Wulfstan's tomb at Worcester,28 and mentions the splendid glass in Rochester cathedral,29 both of which have long since vanished. Of particular value to the architectural historian today is his account of the round church built at Athelney by King Alfred (of which no trace now remains), still served in William's day by a community of ascetic monks. It was built, William records, in a novel style: 'Four posts fixed in the earth support the whole structure, and it has four aisles with round ends arranged around the spherical building.'30 26
William of Malmesbury, Gesta Regum, ed. William Stubbs (Rolls Series, 1887-1889, 2 vols.), i. 2; Gesta Pontificum, ed. N. E. S. A. Hamilton (Rolls Series, 1870), p. 4. 26 Gesta Pontificum, ed. Hamilton, pp. 193, 295. 27 Magdalen College, Oxford, MS. 172. Five pages of this MS are reproduced in facsimile by Hamilton, op. cit. who gives in the footnotes William's revisions of his text. 28 Gesta Pontificum, p. 288. 29 Ibid. p. 138. 30 '[Rex Elf red us] fecitque ecclesiam, situ quidem pro angustia spatii modicam, sed novo edificendi modo compactam. Quattuor enim postes solo infixi totam suspendunt machinam, quattuor cancellis opere sperico in circuitu ductis'; ibid. p. 199. For the value of William's description see G. Baldwin Brown, The Arts in Early England (London 1903-1937, 6 vols.; new edition of vol. 2, 'Anglo-Saxon Architecture', London 1925), n. 196. For King Alfred's foundation of the monastery see Asser's Life of King Alfred c. 92, ed. W. H. Stevenson (Oxford, 1904), pp. 79-80. William of Malmesbury also mentions the Anglo-Saxon church at Bradford-upon-Avon: Gesta Pontificum, p. 346; cf. Baldwin Brown, op. cit. n. pp. 17, 160, 297, 302-303. William mentions an Anglo-Saxon church, now lost, at St. Albans: Gesta Pontificum, p. 316; cf. Baldwin Brown, op. cit. u. 187. William's description of Hexham (Gesta Pontificum, p. 255) is mainly based on that by Eddius Stephanus (see above p. 31 and n.8 and Baldwin Brown, op. cit. u. 175-176), and is of less interest than the detailed description written by Richard of Hexham soon after 1138, which is printed in The Priory of Hexham, ed. Raine, 1.10-14,
Realistic Observation in Twelfth-Century England
181
The fifth book of the Gesta Pontificum is devoted to the life of St Aldhelm, patron and co-founder of Malmesbury abbey.31 William included in it information about local history. Like Symeon of Durham, he shows a penchant for giving local antiquities religious associations, in order to enhance the reputation of his house. He connected material objects with St Aldhelm. There was preserved at Malmesbury a beautiful old cope. It was, William records, of fine scarlet silk and had peacocks enclosed in black roundels embroidered on it. William asserts that this cope had belonged to St Aldhelm.32 He also states that the series of free standing stones near Malmesbury were erected to mark the places where the bearer of St Aldhelm's body rested on the way from Doulting, where St Aldhelm died, to Malmesbury, where he was buried.33 And William claims that a crack in the altar at Bruton, a gift, according to William, from Pope Sergius to St Aldhelm, was caused by the fall of the animal (perhaps a camel) which carried it over the Alps.34 William of Malmesbury*s most ambitious and intelligent excursion into local history was in the De Antiquitate Glastoniensis Ecclesiae, written sometime between 1129 and 1139.35 The reasons, perhaps biographical,36 for William's close interest in Glastonbury, are obscure. But he certainly wrote to please Henry of Blois, bishop of Winchester, who ruled Glastonbury abbey during his episcoPATE (1129-1171),37 AND TO COUNTER THE ASSERTION BY oBSERN, THE cANTERBURY hagiographer, in his Life of St Dunstan, that Dunstan was the first abbot of Glastonbury, which impugned Glastonbury's claim to a long and venerable Christian past.38 William sought to prove Glastonbury's antiquity as an abbey and its even greater antiquity as a holy place. He used, besides documentary evidence, extant antiquities to prove his thesis. He particularly called attention to two 'pyramids,' or burial crosses, in the cemetery.39 One, he states, was and fully discussed in Baldwin Brown, op. cit. n. 149-184 passim. Cf. Taylor and Taylor, op. cit. i. 297-312 passim and A. W. Clapham, English Romanesque Architecture (Oxford, 1930-1934, 2 vols.), i. 44-45, 72-73. 31 The Gesta Pontincum is divided into five books; the last is a Life of St. Aldhelm, and includes much historical material concerning Malmesbury abbey. 32 Gesta Pontificum, p. 365. The peacock motif was popular from the seventh century, or earlier, until the thirteenth century: I owe this information to Mr. Donald King, Deputy Keeper of Textiles in the Victoria and Albeit Museum. 33 Gesta Pontificum, pp. 383-384. 34 Ibid. p. 373. However Hugh Farmer, 'William of Malmesbury's Life and Work' in Journal of Ecclesiastical History, xin (1962), p. 40, accepts William's statements concerning the cope and the Bruton altar. 35 Printed in Adami de Domerham Historia de Rebus Gestis Glastoniensibus, ed. Thomas Hearne (Oxford, 1727, 2 vols.), i. 1-122. 36 For criticism of William's work see Gesta Regum, n. 357. For the possibility that he was passed over for the abbacy in 1120 see ibid. i. xxxviii-ix. 37 See J. Armitage Robinson, 'William of Malmesbury 'On the Antiquity of Glastonbury' ' in his Somerset Historical Essays (Oxford, 1921), pp. 3, 4. 88 See Domerham, p. 71, and Armitage Robinson, loc. cit. pp. 3, 23. 89 Domerham, pp. 44-45. This passage is printed in translation by Aelred Watkin, 'The Glastonbury 'Pyramids' and St. Patrick's 'Companions' ' in Downside Review, LXIII (1945), pp. 30-31. The
182
Legends, Traditions and History in Medieval England
twenty-six feet high and had five storeys, on four of which ancient names could be deciphered, though the stone was weathered. William copies the names. He also copies the names from the other 'pyramid' which had four storeys and was eighteen feet high. William states that these were the names of the abbots before Dunstan, thus proving, on his own evidence, the abbey's great antiquity: William may well have been partly right, because some of the names probably did commemorate the earliest abbots of Glastonbury, of the late seventh and eighth centuries.40 He was certainly right in his general view that Glastonbury had been a holy place from the early Christian period in Britain.41 And his obviously eye-witness description of the 'pyramids' is particularly valuable to the historian today because no trace of them now remains.42 Besides William of Malmesbury, Osbern and Eadmer, both monks of Christ Church, Canterbury, made intelligent use of antiquities. Their hagiographies contributed much to re-establishing the reputations of the Anglo-Saxon saints connected with Canterbury after the Conquest. Osbern, an Englishman who found it difficult to adapt to Norman rule (Lanfranc sent him for two years to Normandy to help him readjust), visited Glastonbury when writing his Life of St Dunstan. He saw Dunstan's cell which he described precisely, using it to demonstrate the saint's austerity. This den, 'more like a tomb for the dead than a habitation for the living,' was so small ('not more than five feet long and two and a half feet wide') that Dunstan could never have lain down to sleep, and the door, with a small window in it, constituted one wall.43 Eadmer, best known as biographer of Anselm, archbishop of Canterbury (1093-1109), resembled Osbern in his concern for preserving the memory of the Anglo-Saxon saints. Besides the threat to continuity offered by the Norman Conquest, he responded to an additional challenge. Canterbury cathedral was burnt to the ground in 1067. The cathedral's reputation could have suffered serious damage if contemporaries had believed that the relics it housed had been lost in the conflagration. Therefore Eadmer wrote a work, De Reliquiis Sancti Audoeni et quorundam aliorum sanctorum quae Cantuariae in aecclesiae domini Salvatoris habentur, to prove that the relics from the old choir had been safely placed in the new one.44 He described the exact location of the relics in the old rendering in Robinson, loc. cit. p. 21, is less precise than that by Watkin (see Watkin, loc. cit. p. 30 n.5). 40 See the suggested identifications of the names given by Watkin, loc. cit. pp. 35-40. 41 For what is known today of the Anglo-Saxon churches at Glastonbury, with references to William of Malmesbury's information, see Taylor and Taylor, Anglo-Saxon Architecture, I. 250-257. See also the same authors' 'Pre-Norman Churches of the Border' in Celt and Saxon, Studies in the Early British Border, ed. N. K Chadwick (Cambridge, 1963), pp. 256-257. 42 The 'pyramids' remained standing until the eighteenth century, when they were dragged away. They were remarked on by other writers from the late twelfth century onwards, but by then their inscriptions were indecipherable; see Watkin, loc. cit. pp. 81-35. 48 Memorials of St. Dunstan, ed. William Stubbs (Rolls Series, 1874), pp. 183-184. 44 Printed A. Wilmart, 'Edmeri Cantuariensis cantoris nova opuscula de sanctorum veneratione et obsecratione' in Revue des Sciences religieuses, xv (1935), 362-370. The architectural description in this work is also printed with an English translation (together with other references in the works of
Realistic Observation in Twelfth-Century England
183
choir, and this caused him to describe its architecture. Therefore he wrote what was in fact the first architectural history to be composed in England — a precedent which, as will be seen below, was not neglected at Christ Church. A characteristic passage reads: 'Beyond the middle of the length of the nave there were two towers which projected above the aisles of the church. The south tower had an altar in the middle of it which was dedicated in honour of the blessed pope Gregory, and at the side was the principal door, which formerly was called by the English — and still is — the Suthdure.'45 The Norman Conquest resulted in the most serious,* but not the only, threat to the continuity of the monastic tradition in mediaeval England. The anarchy of King Stephen's reign (1135-1154) and Henry II's judicial reforms also made monks fear for their ancient rights and privileges. Again monks showed a heightened interest in their houses' past. In the first half of Henry II's reign chronicles were written in numerous monasteries, for example at Ely, Peterborough, Ramsey and Battle, which were principally intended to record the houses' history, authenticated with copies of documents, as far back as possible. The Ely chronicler begins with a Life of the patron saint, St Etheldreda, and uses visual evidence to connect the abbey's treasures with the Anglo-Saxon past. He describes a precious altar cloth; it was of blood-coloured satin on top, and had green sides decorated with gold thread and gems and edged with gold fringe — it was, he said, the work of, and a gift from Queen Emma.46 The Ramsey chronicler was particularly interested in remains from the Anglo-Saxon period. He gives a valuable description of the earliest stone church at Ramsey, the tower of which fell in St Oswald's time, because of faulty foundations. Nothing now remains of this church which was in the Carolingian style: Two towers overshadowed the roofs; the smaller one, at the west end, at the front of the church, presented a beautiful spectacle from a long way off to those entering the island; the larger one, in the centre of the quadrifid structure, was supported by four columns, one attached to the next by spreading arches to prevent them falling apart. According to the architecture of that distant age, this was a fine enough building.47
The Ramsey chronicler also mentions the ruins of an old crypt in the cemetery, which, he stated, were 'witness and evidence' of a nunnery which King Canute and Bishop ^Ethelric had intended to found there. But God in His mercy had prevented them !48 Eadmer and other early writers throwing light on the architectural history of Christ Church) in R. Willis, The Architectural History of Canterbury Cathedral (London, 1845), pp. 9-13. See also R. W. Southern, Saint Anselm and his Biographer (Cambridge, 1963), pp. 262 n. 2, 370-371. 45 'Dein sub medio longitudinis aulae ipsius duae turres erant, prominentes ultra aecclesiae alas. Quarum una, quae in austro erat, sub honore beati Gregorii papae altare in medio sui dedicatum habebat, et in latere principale octium aecclesiae, quod antiquitus ab Anglis et nunc usque Suthdure dicitur;' Wilmart, loc. cit. p. 365. 46 Liber Eliensis, ed. E. O. Blake (Camden Soc. third series, xcn, 1962), p. 149. For a description of the altar at Ramsey see Chronicon Abbatiae Rameseiensis, ed. W. Dunn Macray (Rolls Series, 1886), p. 90. 47 Ibid. p. 41. See Baldwin Brown, op. cit. n. 268-269; Clapham, op. cit. I. 90. See Historians of the Church of York* ed. Raine, i. 434 for an earlier reference to this church. 48 Chron. Abb. Ram. p. 126.
184
Legends, Traditions and History in Medieval England
But the Ramsey chronicler had more than a nostalgic, antiquarian interest in the Anglo-Saxon past. Of all the twelfth-century chroniclers, he took the most methodical interest in charters. He explicitly states that the anarchy of Stephen's reign provided the incentive: he wanted to provide written evidence of the abbey's titles to land and privilege in case these were again challenged in a period of chaos.49 Therefore the Ramsey chronicler copied numerous charters into his chronicle. He concentrated particularly on pre-Conquest charters, thus becoming one of the earliest Anglo-Saxon scholars. He examined 'very ancient schedules of charters and cyrographs,' which, as they were 'nearly all in English', he translated into Latin. He comments on the poor condition of some of the documents, 'disintegrating with age.'50 The Ramsey chronicler was also interested in seals. He appreciated the additional authority given to a charter by the royal seal, and remarks on the fact that in early times only the king and great men, not people of lesser importance had seals.51 The same point is made in the contemporary chronicle of Battle abbey which records that Richard de Lucy, the justiciar, when defending in court the authenticity of an unsealed charter of Battle abbey, pointed out that in the old days 'not every little knight had his own seal, but only kings and really important people.'52 Interest in the physical appearance of devices for the authentication of documents is vividly demonstrated in 'Benedict of Peterborough': it has a drawing of the rota of King William II of Sicily, at the end of a copy of William's charter granting dower to his betrothed, Henry II's daughter Joan 53 (PL 4). Forgery, a weapon against attacks on rights to property and privilege, provided a powerful incentive to palaeographical and diplomatic study, because suspect charter were examined attentively.54 Thus Gervase of Canterbury, writing between about 1188 and 1199, records how a forgery was detected in 1181. In order to strengthen their claim to exemption from obedience to the archbishop of Canterbury, the monks of St Augustine's produced what purported to be a papal diploma granted to St Augustine of Canterbury. But, Gervase records, though the parchment (actually a reused piece) was old, the handwriting, style and leaden bull were not.55 49 60
Ibid. p. 4.
Ibid. pp. 65, 176-177. 61 Ibid. p. 65. 52 Chronicon Monasterii de Bello, ed. J. S. Brewer (Anglia Christiana, London, 1846), p. 108. 63 See also Gesta Regis Henrici Secundi Benedicti Abbatis, ed. William Stubbs (Rolls Series, 1867, 2 vols.), i. 172 n. 4. A picture of the same rota is also in Roger of Howden's chronicle; see Chronica Magistri Rogeri de Houedene, ed. William Stubbs (Rolls Series, 1868-1871, 4 vols.), n. 98. (For evidence suggesting that Roger of Howden wrote 'Benedict' see D. M. Stenton, 'Roger of Howden and 'Benedict' ' in EHR, LXVIII (1953), 574-582 passim). For a reference to this rota, on a diploma dated at Palermo in 1182, see Arthur Engel, Recherches sur la Numismatique et la Sigillographie des Normands de Sidle et d'ltahe (Paris, 1882), p. 87. 54 For examples of the criticism of documents at the papal curia in the twelfth century see R. L. Poole, Lectures on the History of the Papal Chancery (Cambridge, 1915), pp. 143-162, 65 The Historical Works of Gervase of Canterbury, ed. William Stubbs (Rolls Series, 1879, 1880, 2 vols.), i. 296-297. For this case see M. D. Knowles, 'The Growth of Exemption' in Downside Re-
4 'Benedict of Peterborough' (rota of William II of Sicily). (London, British Library, MS Cotton Vitellius E xvii,fo. 28)
5 A cripple from Gerald of Wales, Topographia Hibernica, dist. iii, cap. xxxv. (London, British Library, MS Royal 13 B viii,fo. 30v) 6 Irish hornblowers from Gerald of Wales, Topographia Hibernica, dist. iii, cap. xxxiv. (London, British Library, MS Royal 13 B viii,fo. 30)
7 Plan of waterworks of Christ Church, Canterbury, c. 1165. (Cambridge, Trinity College, MS R.17.1,fos. 284v-285)
Realistic Observation in Twelfth-Century England
185
Gervase of Canterbury was the last representative in the twelfth century of the monastic antiquarian tradition exemplified by William of Malmesbury. Like William he was interested in the monastic geography of England. His Mappa Mundi^ compiled to preserve the memory of the England of his own day, i nearly complete list of the monasteries of England (he gives four hundred and thirty eight houses), grouped in counties, arranged in three columns: in the first column is written, as relevant, 'archbishopric,' 'bishopric,' 'abbey' or 'priory'; in the second is the name of the place and the patron saint; and in the third column is the order to which the house belonged. But Gervase's antiquarian interest was mainly centred on Christ Church. Like Eadmer, he faced a challenge to the continuity of the cathedral's spiritual tradition. In 1174 the choir of the cathedral was again burnt to the ground, to be rebuilt in the next decade by William of Sens and his successor. Once again it was necessary to prove that no relics had been lost, and that the tombs of the archbishops, especially that of Thomas Becket, had been safely disposed in the new choir. Gervase begins his Chronicle with an account of the fire and of the rebuilding, including an architectural history of the cathedral from AngloSaxon times and a vivid description of the new choir.57 His purpose was, he states, to prevent the two previous churches, both destroyed by fire, from ever being forgotten. He probably had a personal reason for strong interest in and detailed knowledge of the new cathedral's architecture. It is likely that he was the monk appointed by William of Sens to oversee the workmen after William had been incapacitated by a fall from the scaffold.58 Gervase's interest in architectural history was stimulated, like Eadmer's, by preoccupation with relics. 'It is impossible,' he writes, 'to show clearly the resting-places of the saints, which are in various parts of the church, without first describing the building itself where they are housed.'59 He copies Eadmer's account of the location of the relics in the pre-Conquest church, and describes exactly where they and the archbishops' tombs were placed in the new choir — the climax is the account of the translation of Thomas Becket.60 Gervase traced the cathedral's architectural history in great detail. He relied on Eadmer for his account of the pre-Conquest church, but described the view, L (1932), 411-415. For a similar case, of 1221, involving Dunstable priory see Annales Monastici, ed. H. R. Luard (Rolls Series, 1864-1869, 5 vols.), in. 66, and V. H. Galbraith, Studies in the Public Records (London, 1948), pp. 48-52. 56 Printed in Gervase of Canterbury, n. 414-449. It is fully discussed by M. D. Knowles, 'The Mappa Mundi of Gervase of Canterbury' in Downside Review, XLVIII (1930), 237-247. 57 Gervase, i. 7-29, passim. For an English translation see R. Willis, The Architectural History of Canterbury Cathedral, pp. 32-62 passim. 58 Gervase, i. 20, records that William of Sens after his accident, 'veruntamen quia hiems instabat, et fornicem superiorem consummari oportebat, cuidam monacho industrio et ingenioso qui cementariis praefuit opus consummandum commendavit, unde multa invidia et exercitatio malitiae habita est, eo quod ipse, cum esset juvenis, potentioribus et ditioribus prudentior videretur'. The omission of the young monk's name would be surprising if he were not Gervase himself. 69 Ibid. i. 12. «° Ibid. i. 22-3.
186
Legends, Traditions and History in Medieval England
nave of Lanfranc's church, which had survived the 1174 fire, from his own observation. (He admits that he cannot describe Lanfranc's choir because it was pulled down by Prior Conrad in Archbishop Anselm's time in order to build the new one). He describes Conrad's choir from memory, and compares it with the new choir in order to highlight the latter's architectural novelties. He points out that the new columns were nearly twelve feet higher than the old ones, and unlike the latter had carved capitals. He states that the new rib-vaults replaced plain stone vaults and painted wooden roofs, and notices that the new choir had two triforiums instead of one, was higher and lighter, and was richer in marble and carving. He ends by explaining why the new choir narrowed at the end. Gervase concludes with an appreciation of the value of visual evidence: 'All these things can be more clearly and pleasurably seen with the eye than learnt about from the written or spoken word.'61 Gervase' strong interest in the architecture of Christ Church was undoubtedly aroused by the fire of 1174. However it should be noted that about ten years earlier another event had stimulated the monks' interest in the architecture of their monastery. Sometime between 1153 and 1167 they had installed a new system of water distribution and drainage within the precincts. This gave rise to the two earliest known surveys made in England, in order to show where the pipes lay. One survey is a plan of the monastery, showing the cathedral and principal buildings, which though not correct in all details is in general accurate (PL 7). The other is a plan of the extra-mural waterworks.61*
I turn now to descriptive writing encouraged by study of the classics. William of Malmesbury himself was much influenced by the classics. His admiration for classical antiquity is well demonstrated by his inclusion in the Gesta Pontificum, a work devoted to Christian history, of a description of the ruins of an impressive Roman hall in Carlisle, and a transcription of the inscription on it.62 But the most important result of William's study of the classics, especially of Suetonius' Lives of the Caesars, was on his perception of people. Suetonius provided William of Malmesbury with models for the physical description of people and taught him a technique for illustrating facets of character. William, the first writer in England to write more than one pen-portrait,63 used the Suetonian method of short, 61
Ibid. i. 28. Both surveys are discussed and reproduced in The Canterbury Psalter ed. M. R. James (London, 1935) pp. 53-56, and end of volume. 62 Gesta Pontificum. p. 208. This inscription is in R. G. Collingwood and R. P. Wright, The Roman Inscriptions of Britain (vol. 1 only published, Inscriptions on Stone, Oxford, 1965), i. 316-317 (no. 950). 63 A few pre-Conquest writers described men's physical appearance. Bede briefly but graphically described the appearance of Bishop Paulinus (HE, ii. 16). Alcuin, writing abroad, described Willibrord, in the Suetonian style (quoted HE, ed. Plummer, n. 293). There is a description of Edward the Confessor in the anonymous Life, apparently based on a description of St. Omer in a contemporary hagiography; The Life of King Edward, ed. F. Barlow, p. 12 and n. 1. The earliest description of an 61a
Realistic Observation in Twelfth-Century England
187
snappy descriptions.64 How far his descriptions are realistic is hard to say. As they have verbal echoes of Suetonius and it is unlikely that William had met the men he describes, they could be literary rhetoric, without historical basis. This is probably true of his portrait of St Wulfstan, bishop of Worcester (10621095) which has no distinctive features —- Wulfstan was 'of medium height, overtopped by the very tall, but exceeding the very short, and all his limbs were well-porportioned.5>65 On the other hand, William's pen-portraits of the Anglo-Norman kings are sufficiently concrete to justify credence. He could have obtained his information from people who had seen the kings. For example, he writes that William Rufus 'was squarely built, had a florid complexion and yellow hair, an open countenance and multi-coloured eyes, varied with glittering specks; he was of astonishing strength, though not very tall, and had a protruding stomach.'66 William also adopted the Suetonian technique of illustrating character, both good and bad traits, by recounting anecdotes and trivial details. A good example of William's use of anecdote is a story he tells to show William Rufus' extravagance. Rufus reprimanded his chamberlain for buying him a cheap pair of boots, exclaiming angrily, 'Only three shillings!? you son of a whore, since when has a king worn such cheap boots? Go and bring me a pair worth a silver mark!'67 Moreover William probably borrowed from Suetonius his method of writing passages exclusively devoted to sketching characters in some detail. (He treats the Anglo-Norman Kings in this way).68 The classics and romance literature (itself indebted to classical studies) fostered interest in secular life. William of Malmesbury was the first writer to divide secular from ecclesiastical history explicitly, dealing with the former in the Gesta Regum and the latter in the Gesta Pontificum. And the first pen-portrait of a layman other than a king was written in the mid-twelfth century: Ailred of Rievaulx gives a vivid description in the Suetonian style of Walter Espec, lay patron of Rievaulx abbey —- a very tall man, with black hair, a bold, lined Englishman is that of Pelagius by Jerome, who knew him well ('a huge, fat highland dog' who 'walked like a tortoise'); see J. N. L. Myres, 'Pelagius and the End of Roman Rule in Britain' in Journal of Roman Studies, L (1960), 24. 64 See e.g. Suetonius, Duodedm Caesares, ed. Maurice Rat, with a French translation (Paris, 1981, 2 vols.), i. 51; ii. 65, 127 (Caesar, c. XLV; Caligula, c. 1; Claudius, c. xxx). For the influence of Suetonius on William of Malmesbury treated generally see M. Schiitt, 'The Literary Form of William of Malmesbury's 'Gesta Regum' ' in EHR, XLVI (1981), 255-260. 66 The Vita Wulfstani of William of Malmesbury, ed. R. R. Darlington (Camden Society, third series, XL, 1928), p. 46. For a similar description of Thomas archbishop of York (1070-1100) see Gesta Pontificum, p. 257. 66 Gesta Regum, n. 374. For descriptions of William I and Henry I see ibid. n. 885, 488. For a short reference to King Athelstan's appearance, based on observations made when his tomb at Malmesbury was opened, see ibid. i. 148. w Ibid. n. 368. 68 See ibid. n. 335-336, 366-371, 488. A character-sketch of William the Conqueror occurs in the Anglo-Saxon chronicle, s.a. 1087; The Anglo-Saxon Chronicle, a revised translation ed. D. Whitelock with D. C. Douglas and S. I. Tucker (London, 1961), pp. 163-164.
188
Legends, Traditions and History in Medieval England
face, with a voice like a trumpet, 'uniting eloquence with a certain majesty of sound.'69 The twelfth century is notable for numerous topographical descriptions of castles. These reflect contemporary interest in warfare, encouraged both by romance literature, popularized by Geoffrey of Monmouth's Historia Regum Britanniae (published ca. 1136), and by conditions during the anarchy of Stephen's reign. They also reflect men's response to the impressive castles which played a crucial role during the anarchy. Geoffrey of Monmouth himself briefly describes the Tower of London.70 And so many topographies of castles (of Exeter, Bristol, Bath, Oxford, Cricklade and Faringdon)71 occur in the Gesta Stephani as to suggest that the author had some particular reason for interest in their defensive potential. For example, he writes of Oxford: The city is very securely protected, inaccessible because of the very deep water that washes it all round, most carefully encircled by the palisade of an outwork on one side, and on another finely and very strongly fortified by an impregnable castle and a tower of great height.
A similar topography of Scarborough castle is in the late twelfth-century chronicle of William of Newburgh, an Augustinian canon: A huge rock, almost inaccessible on account of precipices on all sides, drives back the sea which surrounds it, except for a narrow ascent on the west. On its summit is a beautiful grassy plain, more than sixty acres in area, with a spring of fresh water issuing from a rock. At the entrance, which is difficult of access, there is a royal castle, and below the incline begins the town which spreads to the south and north but faces west, defended on this side by its own wall, on the east by the castle rock, while bcth sides are washed by the sea.72
Most remarkable for realistic descriptive writing were a group of writers working in the last decade of Henry II's reign and in the reign of Richard I. Their common features were, in varying degrees, an interest in man as an individual and an objective curiosity about man's environment. All were influenced, to a greater or lesser extent, by classical studies, but none imitated exactly any particular classical author. Foremost were Gerald of Wales, Walter Map, and William Fitz Stephen. Other writers sharing some of their characteristics were Adam of Eynsham, chaplain and biographer of St Hugh of Lincoln, Ralph Diceto, dean of St Paul's (1180/1-1202), Richard of Devizes, a monk of St Swithuns, Winchester, and Lucian, a monk of St WerburghV, Chester. All these men, except apparently Richard of Devizes and Lucian, had contact, with one or more of the famous schools of the day. Gerald of Wales, Walter Map and 69
Ailred, Relatio de Standardo, in Chronicles of the Reigns of Stephen, Henry II, and Richard I, ed. Richard Hewlett (Rolls Series, 1884-1889, 4 vols.), in. 183. 70 Geoffrey of Monmouth, Historia Regum Britanniae, c. m. 10; ed. Acton Griscom (London, 1929), p. 291. Cf. ibid. c. i. 17; c. m. 20 (pp. 252, 301, respectively). 71 Gesta Stephani, ed., with an English translation, K. R. Potter (Nelson's Medieval Texts, 1955), pp. 22, 37-38, 38-39, 92, 113, 120, respectively. For a detailed account of the Isle of Ely's natural defences and a brief comment on Bedford see ibid. pp. 66, 155. 72 William of Newburgh, Historia Rerum Anglicarum, in Chrons. Stephen, Henry II, and Richard /, ed. Hewlett, i. 104.
Realistic Observation in Twelfth-Century England
189
Adam of Eynsham had connections with Oxford73 and Lincoln cathedral74— Gerald mentions the schools at Lincoln.75 Fitz Stephen, who describes the schools in London,76 and Diceto were Londoners, and both Gerald and Walter Map (who held a prebend at St Paul's)77 often lived in the capital. Gerald78 and Walter Map,79 and probably Diceto had studied at Paris university,80 and all three had close connections with the Angevin court, a notable cultural center. Gerald of Wales and Walter Map both give pen-portraits of Henry II, whom they knew personally. (Walter Map entered royal service soon after 116081 and Gerald of Wales in about 1184).82 Gerald's description, as the most graphic of the two, may be quoted. Henry was: a man of reddish, freckled complexion with a large round head, grey eyes which glowed fiercely and grew bloodshot in anger, a fiery countenance and a harsh, cracked voice. His neck was poked forward slightly from his shoulders, his chest was broad and square, his arms strong and powerful. His frame was stocky, with a pronounced tendency to corpulance, due to nature rather than indulgence, which he tempered by exercise.83 73 Gerald read one of his works to the students of Oxford in about 1185, and spent two years in Oxford from 1193 to 1194; see Giraldi Cambrensis Opera, ed. J. S. Brewer and others (Rolls Series, 1861-1891, 8 vols.), i. 72, 294. Cf. The Autobiography of Giraldus Cambrensis, ed. H. E. Butler (London, 1937), pp. 97,139 and n. 5. Walter Map became archdeacon of Oxford in 1196; J. Le Neve, Fasti Ecclesiae Anglicanae (Oxford, 1854, 3 vols.), n. 64. Adam, a monk of Eynsham (about five miles from Oxford), was of an Oxford burgher family; see The Life of St. Hugh of Lincoln, ed. D. L. Douie and H. Farmer (Nelson's Medieval Texts, 1961-1962, 2 vols.), i. viii-ix. 74 Gerald of Wales retired to Lincoln to write and study in 1196; Giraldus, i. 93, and Butler, op. cit. p. 127 and n. 2. Map was precentor of Lincoln; J. Le Neve, op. cit. n. 82. Adam lived partly at Lincoln as St. Hugh's chaplain from 1197 to Hugh's death in 1200; Life of St. Hugh, i. x-xi. 76 See Giraldus, i. 93. 76 See Fitz Stephen's description of London in his Life of Saint Thomas in Materials for the History of Thomas Becket, ed. J. C. Robertson and J. B. Sheppard (Rolls Series, 1875-1885, 7 vols.), in. 4-5. 77 J. Le Neve, Fasti Ecclesiae Anglicanae 1066-1300, i, compiled by D. E. Greenway (London, 1968), p. 60. 78 Giraldus, i. 23, 45 sq.; vni. 292. 79 For Map's stay in Paris see Walter Map, De Nugis Curialium, ii. 7; cf. v. 5: ed. M. R. James (Anecdota Oxoniensia, 1914), pp. 69, 225. sq. 80 Diceto studied in Paris, probably at the university and had other close contacts with France and the Angevin empire; see The Historical Works of Master Ralph de Diceto, ed. William Stubbs (Rolls Series, 1876, 2 vols.), i. xvii sq., xxxi sq. 81 See De Nugis, ii. 3; v. 6: ed. James, pp. 65-66, 246, Cf. T. Wright's introduction to his edition of De Nugis (Camden Society L, 1850), p. vi. 82 See Butler, op. cit. p. 81 and n.l. 83 Expugnatio Hibernica, in Giraldus, v. 302; the pen-portrait and character-sketch (discussed below) is printed, in an English translation, in English Historical Documents 1046-1089, ed. D. C. Douglas and G. W. Greenaway (London, 1953), pp. 386-388. Walter Map's pen-portrait of Henry II is in his character-sketch of the king; see Walter Map, De Nugis Curialium, v. 6, ed. James, p. 237 and the next note. For another contemporary pen-portrait and character-sketch of Henry II see Peter of Blois' Epistola 66 (written in 1177), printed in Materials for the History of Thomas Becket, ed. Robertson and Sheppard, vn. 571-575 (and in Petri Blesensis Bathoniensis archidiaconi Opera omnia, ed. J. A. Giles, Oxford 1846-1847, 4 vols.). Peter was a prolific writer whose career resembled that of Gerald of Wales though much of it was on the continent. He was born at Blois, took orders, studied at Bologna and taught at Paris. He was successively tutor of William II of Sicily, and secretary to Rotrou archbishop of Rouen, to Baldwin archbishop of Canterbury, and to Queen Eleanor,
190
Legends, Traditions and History in Medieval England
Moreover, Gerald of Wales and Walter Map give detailed character-sketches of Henry II. They treat the subject en bloc, besides telling numerous illustrative anecdotes in Suetonian fashion. Map's description of Henry mainly concerns his abilities as a ruler (he had 'discretion in the making of laws and the ordering of all government, and was a clever deviser of decisions in difficult and dark cases'). But Map also gives some more personal touches. He ascribes Henry's restless energy, which made him a tireless traveller 'tolerant of the discomforts of dust and mud,' and exhausted his household, to his fear of growing too fat. And he describes his good temper and good manners, but criticises him for a tendency to withdraw from company when away from home.84 Equally vivid is Gerald of Wales' description of Henry's character.85 Like Map, he comments on his extraordinary energy; after a day of strenuous physical activity he would sit neither before, or after supper, so that 'by such great and wearisome exertion he would wear out the whole court by continually standing.' Adam of Eynsham in the Life of St Hugh illustrates character with anecdotes and trivial details, even mentioning facets of St Hugh's character which were not entirely creditable —• he relates that Hugh was sometimes irritable when he presided over the chapter at Lincoln.86 He recounts anecdotes about the kings which show Henry II's quick temper and sense of humour,87 Richard I's forgiving nature,88 and King John's lack of religious feeling, meanness, and worldly ambition. (He illustrates John's unpleasant character partly with a story of how St Hugh showed him a carving of the Last Judgment on the tympanum at Fontevrault, hoping that the sight of the torments of the damned would frighten him into repentance: but John turned away to look at some carvings of proud kings, saying it was them whom he would emulate.)89 Adam also had an eye for everyday life. He describes a baby's response to the saint's presence: The tiny mouth and face relaxed in continuous chuckles. .. It then bent and stretched out its little arms, as if trying to fly, and moved its head to and fro. . . . Next it took St. Hugh's hand in both of its small ones, and using all its strength raised it to his face, immediately licking rather than kissing it.90
A feature of a number of late twelfth-century writers was an interest in towns, and became archdeacon of Bath in about 1175 and of London in about 1192. He knew Henry II well. Short pen-portraits and character-sketches of Thomas Becket occur in the Life by William Fitz Stephen (Materials, in. 17) and in the Icelandic Life (Thbmas Saga Erkibyskups, ed. E. Magnusson, Rolls Series, 1875, 1883, 2 vols., I. 29). 84 De Nugis, v. 6; ed. James, pp. 237-242 passim. Map records that he once crossed the channel with Henry II (ibid. p. 242). His character-sketch of Henry is printed, in an English translation, in Douglas and Greenaway, op. cit. pp. 389-390. Map also gives character-sketches of King Canute and Henry I; De Nugis, v. 4; v. 5: ed. James, pp. 211, 218-220. 85 Giraldus, v. 301-306. Gerald also gives character-sketches of Henry II's sons and of William Longchamp; Giraldus, v. 193-201; iv. 399 sq.t respectively. 86 See The Life of St. Hugh of Lincoln, ed. Douie and Farmer i. 124. 87 Ibid. i. 115-119. 88 Ibid. n. 101. Cf. ibid. i. xlii-iv. 89 Ibid. n. 140-141. 90 Ibid. i. 129-130.
Realistic Observation in Twelfth-Century England
191
both in their topography and in the customs of their citizens. This interest developed throughout the twelfth century (writers in earlier periods rarely refer to towns)91 as a result of a number of factors. Towns were playing an increasingly important part in the economic and social life of England. Moreover, the idea of the city was fostered by trends in European history. The reform of the papacy attracted attention to Rome. William of Malmesbury copied a poem in praise of Rome, and gives a topography of the Eternal City.92 Interest in cities was further encouraged by the crusades — cities of Byzantium and in the crusading states were famous and prosperous. William of Malmesbury gives topographical accounts of Constantinople, Antioch and Jerusalem,93 and the author of the Itinerary of Richard I describes Jerusalem and its holy places.94 And early in the thirteenth century the Cistercian chronicler Ralph of Coggeshall gave a detailed, though rather inaccurate account of the topography of Constantinople.95 Study of the classics also fostered an interest in cities, the centres of ancient civilization. Geoffrey of Monmouth derived his interest in the derivation of the names of towns from the classics. Virgil explained the derivation of the name Rome with an eponym (the story of Romulus and Remus): Geoffrey used eponyms to explain the names of London,96 Gloucester,97 and Leicester.98 And by calling London 'New Troy,' Geoffrey shows that he regarded London as a reincarnation, so to speak, of that ancient city. Although the descriptions of cities written in England in the late twelfth century are not modelled exactly on any classical work, they, like those by classical authors, divide into two categories, the laudatory99 and the satirical.100 William Fitz Stephen's description of London, Lucian's description of Chester, and Ralph Diceto's description of 91 Gildas realized the importance of cities in Britain; De Excidio, c. 3; ed. Mommsen, p. 28. Alcuin praised York as a center of commerce; De Pontificibus . . . Carmen, lines 19-37; ed. Wattenbach and Dummler, p. 83. Thellth-century hagiographer Goscelin mentioned London's commercial prosperity; C. H. Talbot, 'The Liber Confortatorius of Goscelin of Saint Bertin' in Analecta Monastica III, xxxvm; Rome, 1955), p. 49. For a similar passage see The Life of King Edward, ed. Barlow, p. 44. 92 Gesta Regum, u. 402-408. For the poem and the topography (which dates from the late seventh or early eighth century) see ibid. 11. cxxi-ii, and J. K. Hyde, 'Medieval Descriptions of Cities' in Bulletin of the John Rylands Library, XLVIII (1966), 321. 93 Gesta Regum, u. 411-412, 415-416, 422, respectively. For William's sources for these descriptions see ibid. n. cxxii-raxiii. 94 Itinerarium Peregrinorum el Gesta Regis Ricardi in Chronicles and Memorials of the Reign of Richard I, ed. William Stubbs (Rolls Series, 1864-1865, 2 vols.), i. 435. 96 Radulphi de Coggeshall Chronicon Anglicanum, ed. Joseph Stevenson (Rolls Series, 1875), pp. 149-150. 96 Historia Regum Britanniae, iii. 20; ed. Acton Griscom, p. 301. See also Geoffrey's derivations of Billingsgate and Ludgate, ibid. hi. 10, 20; ed. Acton Griscom, pp. 291, 302. 97 Ibid. iv. 15; ed. Acton Griscom, p. 324. 98 Ibid. ii. 11; ed. Acton Griscom, p. 262. 99 The laudatory attitude to cities in classical literature is well represented by Aelius Aristides* Roman Oration; printed in translation by J. H. Oliver, The Ruling Power (Transactions of the American Philosophical Society, new series, XLIII, pt. 4,1953), pp. 895-907. Virgil praises Rome throughout the JEneid. 100 An obvious classical prototype is Juvenal's Satire III, on Rome.
192
Legends, Traditions and History in Medieval England
Angers, are laudatory, while Richard of Devizes' accounts of English towns are satirical. Richard of Devizes, who wrote his chronicle between 1192 and 1198, frequently cites from such authors as Virgil, Horace, Juvenal, Lucan and Ovid.101 He includes satirical references to a number of English towns, dissociating himself from his remarks by a literary conceit: he puts them into the mouth of a French Jew advising a Gentile youth where to settle in England.102 The young man should avoid London, a sink of iniquity, crowded with pimps, gamblers and parasites ('actors, smooth-skinned boys, belly dancers' and the like). The youth should also avoid Canterbury, Rochester, Chichester, Oxford, Exeter, Worcester, Chester, Hereford, York, Durham, Norwich, Lincoln and Bristol — Richard gives a brief reason for the undesirability of each. For example, York 'is full of Scotsmen, filthy and treacherous creatures scarcely men,' Bath 'is placed or rather dumped down in the midst of valleys in an exceedingly heavy air and in sulphurous vapour, at the gates of Hell,'102a and at Bristol 'there is no-one who is not or has not been a soap maker, and every Frenchman loves soapmakers as he does a dung-heap.' The Jew concludes by advising the youth to settle in Winchester, 'the city of cities, the mother of all and better than all others,' the only city tolerant of Jews, where men can study and the citizens are generous and courteous. But Richard adds a satirical remark even here — Winchester is full of gossip and rumours. The account of London written sometime between 1173 and 1175 by William Fitz Stephen, a clerk who had been in the service of Thomas Becket, is the most detailed and realistic description of a city written in mediaeval England.103 It has many allusions to and citations from classical authors — for example Virgil, Horace, Persius and Ovid. Fitz Stephen wrote primarily because of the circumstances of his times. His love of and loyalty to London were roused by the attention paid to Canterbury as a result of the martyrdom of Thomas Becket in the cathedral. Fitz Stephen prefixed the description of London to his Life of Saint Thomas, the pretext being that Becket, like Fitz Stephen, was a Londoner — and just as Sallust had described Africa in his history of the Carthaginian war, so Fitz Stephen would describe London. London, he writes, was once a metropolitan see (a reference to Gregory the Great's arrangements for the ecclesiastical organisation of England), and would, 'it is thought,' be so again, unless the martyr Thomas' fame perpetuates the honour for Canterbury where his body lies. But London could justly dispute Canterbury's claim to close association with Becket because, though he died in Canterbury, he was born in London.104 101
See, for example, The Chronicle of Richard of Devizes, ed., with an English translation, J. T. Appleby (Nelson's Medieval Texts, 1963), pp. 63-64 and n. 3. 102 Ibid. pp. 65-67. io2a A more favourable description of Bath, which mentions the popularity of its health-giving hot springs, is in the Gesta Siephani, ed. K. R. Potter, pp. 38-39. 103 The text is in Materials, in. 2-13 (see above note 76). For an English translation, by H. E. Butler, see F. M. Stenton, Norman London (Historical Association Leaflets nos. 93, 94, 1934). See also Hyde, loc. cit. pp. 324-325. 104 Materials, HI. 2-3.
Realistic Observation in Twelfth-Century England
193
Fitz Stephen describes the topography of London, the Tower, the walls with their towers and gates, the Thames ('teeming with fish'), the suburban houses with wells and spacious gardens, the meadows and forests beyond. Even more remarkable is his long account of the social customs and pastimes of the Londoners. He describes the horse fair at Smithfield, and the cook-shops on the banks of the Thames, which catered for rich and poor and were especially handy if guests arrived unexpectedly. He mentions the disputations of students studying in London, thus providing evidence for the existence of schools in the capital. Then he describes the games of young Londoners, the ball-games, horse-races and dangerous ice sports played on the frozen Thames. Like Rome in ancient and in Christian times, London was notable both for the usefulness and hard work of its citizens, and also for its pleasures and entertainments.105 Following Geoffrey of Monmouth, Fitz Stephen asserts that London ('New Troy') produced 'not a few men who subdued many nations and the Roman empire to their sway' in the pre-Christian era. He ends his account of famous men with Thomas Becket ' 'than whom she bore no whiter soul nor one more dear' (Horace, Sat. L v. 41-2) to all good men in the Latin world.' The two other laudatory descriptions of cities written at this period have less detail than Fitz Stephen's. Lucian, in his work in praise of Chester, De Laude Cestrie,106 written in about 1195, gives numerous references to classical authors, especially to Horace, Virgil, Ovid and Seneca, besides citing the Bible and the Fathers107 — and Geoffrey of Monmouth.108 The influence of Rome appears in the comparison of the church of St Peter at Chester with St Peter's, Rome.109 Apparently the work originated as a series of sermons intended to please and edify the citizens of Chester,110 for it has long, homiletic passages, and many allegories (for example, the two Roman streets meeting in the city center symbolize the cross, the four gates angelic guardians).111 Nevertheless, Lucian describes the plan of the city and its geographical position.112 He also describes the city walls, the river Dee with its daily tide, wide sands and busy harbour, frequented by merchants from Spain, Ireland, Aquitaine and Germany, and by fishermen.113 The other laudatory description of a city, the account of Angers in Ralph Diceto's Ymagines Historiarum,lu is more straightforward. It is uncertain whether Diceto was himself the author or whether he copied it from some now lost Angevin source.115 The description of the bridge at Angers is particularly 105
Ibid. m. 8-9. Printed Liber Luciani de Laude Cestrie, ed. M. V. Taylor (Lancashire and Cheshire Record Society, LXIV, 1912). For the date of the De Laude see ibid. pp. 8-10. 107 See ibid. p. 15. 108 Ibid. p. 64. 109 Ibid. p. 52. 110 See Ibid. pp. 19-20. 111 Ibid. pp. 46-47. 112 Ibid. pp. 44, 45, 58. 113 Ibid. p. 46. 114 Diceto, Works, ed. Stubbs, n. 291-292. 116 The description of Angers also occurs in the Historia Comitum Andegavensium; see Chronlques 106
8 A crane, grus, from Gerald of Wales, Topographia Hibernica, dist. i, cap. xiv. (London, British Library, MS Royal 13 B viiijo. 9) 9 A swan from a bestiary of c. 1200. (Oxford, Bodleian Library, MS Ashmole 1511, fo. 71)
10 Pigeons from a bestiary of c. 1120-30. (Oxford, Bodleian Library, MS Laud Misc. 247, fo. 161v)
11 Hedgehogs from a bestiary of c. 1200. (Cambridge, University Library, MS Ii.4.26,fo. 28v]
12 A bramble from an early twelfth-century herbal. (Oxford, Bodleian Library, MS Bodley 130, fo. 26)
13 A milk thistle from an early-twelfth century herbal. (Oxford, Bodleian Library, MS Bodley 130Jo. 37)
194
Legends, Traditions and History in Medieval England
vivid: it had 'workshops in little houses (of earth, wood and stones) . . . placed opposite each other and arranged under one almost uniform roof so making the bridge (which is mainly wood in the middle) like a real street, always open to passers-bye but sheltered from the sun.'116 Diceto also has a passage (probably copied from a lost continental source), on the social customs of the people of Aquitaine, including a description of their cullinary methods. It states, for example, that 'the men of Poitou love beef for daily fare. When the pepper and garlic have been mixed together in a mortar, the fresh meat needs as a condiment either the juice of wild apples or that of young vine shoots, or grapes.'117 This interest in social customs, reminiscent of Fitz Stephen's account of the Londoners, is particularly characteristic of Gerald of Wales. But while Diceto and Fitz Stephen concentrated on civilized man, Gerald turned his attention to primitive peoples, describing the customs of the Irish and Welsh in his Topographia Hibernica11* and in the Descriptio Kambriae.11* Ostensibly he wrote about the Irish and the Welsh because he thought writers had neglected them. But he seems also to have wanted to point out the shortcomings of civilized life by implicitly contrasting it with the primitive condition.120 His view of both peoples as barbarians (with consequent vices and virtues) probably owed something to classical literature. Gerald of Wales collected his information for the Topographia Hibernica when he visited Ireland with Prince John in 1185, and for the Descriptio Kambriae when he toured Wales with Archbishop Baldwin, who was preaching the crusade, in 1188. Much of what he says about the Irish is based on folk-lore, not fact, but he records some vivid details which are surely authentic. Most of his information about the Welsh seems to be based on objective observation. He says that both des Comtes d'Anjou, ed. P. Marchegay and A. Salmon, with an introduction by E. Mabille (Societe de 1'IIistoire de France, 1856, 1871), pp. 336-338. Stubbs suggested that Diceto was the author of the Historia; Diceto, Works, n. xxiv-xxix. If so, it is likely that Diceto wrote the description. Stubbs also suggested that the description of the customs of the people of Aquitaine (see below), which is not in the Historia, was by the same author, whether Diceto or not, as the description of Angers. See ibid. n. 293 n.l. lie por the bridge see C. Port, Dictionnaire Historique Geographique et Bibliographique de Maine-etLoire (Angers, 1874-1878, 3 vols.), i. 105-106. 117 Diceto. i. 293-294. Unlike the descriptions of food cited by Curtius, op. cit. pp. 189-190, this passage appears to be realistic. 118 Printed Giraldus, v. 3-204. 119 Printed ibid. vi. 155-227. Gerald's Itinerarium Kambriae, which also has a few first-hand observations, is printed in ibid, vi, 3-152. 120 Such a contrast seems to be implied, for example, in the account of the natural growth of the Irish from their infancy: '. . . fere cuncta naturae relinquuntur. Non in cunabulis aptantur; non fasciis alligantur; non frequentibus in balneis tenera membra vel foventur, vel artis juvamine componuritur. Non enim obstetrices aquae calentis beneficio vel nares erigunt, vel faciem deprimunt, vel tibias extendunt. Sola natura quos edidit artus, praeter artis cujuslibet adminicula, pro sui arbitrio et componit et disponit.' And the Irish grow up fine, handsome men. Ibid. v. 150. The same criticism of civilized man seems implicit for example in the account of the hospitality of the Welsh. The Welsh feed guests lavishly, but 'non ferculis rnultis, non saporibus et gularum irritamentis coquina gravatur; non mensis, non inappis, non manutergiis, domus ornatur. Naturae magis student quam nitori.' Ibid. vi. 183.
Realistic Observation in Twelfth-Century England
195
peoples neglected agriculture, trade, and all productive labour.121 He describes the primitive clothes of the Irish in realistic terms: they wore close fitting hoods hanging over the shoulders, 'of parti-coloured strips sewn together,' under which they had rugs instead of cloaks, and breeches often with socks attached to them.122 The Welsh slept all together, in their day-clothes, a thin cloak or tunic, on a coarse blanket spread on the floor along the side of the room, keeping each other warm, and before they turned over warming themselves by the fire.123 Gerald describes how the Irish rode, bareback, without boots or spurs, but with a sort of riding crop, and reins which served both as bridle and bit.124 He appreciated the virtues of these hardy peoples (especially those of the Welsh). The Irish had by nature fine physiques —• they needed no human skills to help their growth.125 They loved leisure and liberty above all else, and were gifted musicians.126 The Welsh too were very musical,127 and they were also brave soldiers128 and most generous to guests (Gerald describes their customs at table in detail) — and they kept their teeth the cleanest in the world (they made them 'gleam like ivory by constant rubbing with green hazel and by wiping them with a woollen cloth').129 Moreover, the Topographia Hibernica has lively illustrations in the margins, in pen and colour, of scenes from Irish life and other subjects (Rs. 5, 6, 8). As these pictures are in the earliest known manuscript of the work (from St Augustine's, Canterbury) it seems likely that the originals were executed by Gerald himself, or at least under his supervision. Gerald of Wales was a remarkably acute observer of other aspects of the outside world besides people and social anthropology. He also noticed and vividly described antiquities and natural history. His principal motives were objective curiosity and aesthetic appreciation. The former motive probably elicited his description of Stonehenge. Having accepted Geoffrey of Monmouth's statement that Stonehenge had been magically transported from Kildare in Ireland, he proceeds with an eye-witness description. 'It is wonderful' he writes 'how stones of such huge size were ever collected together and erected in one place, and how equally large stones were so skilfully placed on top of such immensely tall stones; thus they seem to hang suspended, as if in mid-air, appearing to be supported more by the artifice of craftsmen than by the tops of the upright stones.'130 121
Ibid. v. 151-152; vi. 180. Ibid. v. 150. 123 Ibid. vi. 184. 124 Ibid. v. 150. 125 See above p. 48 n. 120. 126 Ibid. v. 152 sq. 127 Ibid. vi. 186 sq. 128 Ibid. vi. 180-181. 129 Ibid. vi. 185. 130 Ibid. v. 100-101. Gerald says that the stones were transported from Kildare, where they were known as the Giants' Ring, and that originally they came from Africa. Here Gerald was following Geoffrey of Monmouth, who had also apparently seen Stonehenge; see Historia Regum Britanniae, viii. 10-12; ed. Acton Griscom, pp. 409-414. Cf. J. S. P. Tatlock, The Legendary History of Britain, Geoffrey of Monmouth9s Historia Regum Britanniae and its early vernacular versions (California, 1959), pp. 40-43. Geoffrey's view that the stones had been brought to Stonehenge from afar is of course 122
196
Legends, Traditions and History in Medieval England
Moreover Gerald was interested in Roman remains. He comments on the charred walls of the ancient fort at Carmarthan ('the magnificent walls, still partly standing, situated on the noble river Towy'),131 and describes Roman Caerleon, 'where you can still see many remains of its former grandeur; . . . the spacious palaces (which formerly had guilded roofs in the Roman fashion. . .), an enormous tower, remarkable baths, the ruins of temples, and a theatre all enclosed with a fine wall, parts of which still stand.' Gerald also remarks on the subterranean structures and the acquaducts, and particularly on the hypocaust system 'constructed with wonderful art.'132 His aesthetic motive is well illustrated by his graphic description of an illuminated manuscript he saw in Ireland, probably the Book of Kells.133 Moreover, to this period belong the earliest realistic descriptions of animals and birds,134 for which the ever-curious Gerald of Wales was mainly responsible. He describes, for example, the Irish hare and the habits of the bear,136 but most remarkable is his descriptions of St Hugh of Lincoln's pet swan: It was about as much larger than a swan as a swan is than a goose, but in everything else, especially in its colour and whiteness, it closely resembled a swan, except that in addition to its size it did not have the usual swelling and black streak on its beak. Instead that part of its beak was flat and bright yellow in colour, as were also its head and the upper part of its neck.136
It is possible to identify Hugh's swan from this description as a whooper swan. Gerald accompanies this description with details about its habits, how 'it would fly over the surface of the river, beating the water with its wings, and giving vent to loud cries.' Adam of Eynsham, who copied Gerald's account, added some details, notably concerning the swan's affection for St Hugh.137 Finally, it should be noted that the occurrence of realistic descriptive writing in twelfth-century England has a counterpart in art. Despite the dominance of Romanesque, a heavily stylized and conventionalized art form, there are examples of realistic pictures. Gerald of Wales, in the Topographia Hibernica, was probably responsible not only for the vivid scenes from Irish life, but also for correct (they probably came from the Prescelley mountains in Pembrokeshire); see Stuart Piggott, 'The Stonehenge Story' in Antiquity, xv (1941), 305-319. 131 Giraldus, vi. 80. 132 Ibid. vi. 55-56. 133 Ibid. v. 123-124. For the identification of the book described by Gerald with the Book of Kells see The Book of Kells, ed. E. II. Alton and P. Meyer (Bern, 1950-1951, 3 vols.), in. 14-16. 134 YiJliam Fitz Stephen's description of the action of horses is probably an example of classical influence and not of realistic observation. He describes the *rocking-horse' action and the 'lateral trot* action, both of which occur frequently in classical sculpture and numismatics; Materials for the History of Thomas Becket, ed. Robertson and Sheppard, m. 6. 136 Giraldus, v. 57 (cf. ibid. v. Ixxii-lxxiii); vi. 114-117, respectively. Gerald's knowledge of natural history is fully discussed by U. T. Holmes, 'Gerald the Naturalist' in SPECULUM xi (1936), 110-121. 136 Giraldus, vii. 74. 187 Life ofSt Hugh, ed. Douie and Farmer, i. 105-109.
Realistic Observation in Twelfth-Century England
197
illustrations of animals, birds and fishes in Ireland. Particularly remarkable is the fairly accurate coloured drawing of a crane (Fig. 5).138 Even the Bestiaries and Herbals of the period have a few realistic pictures. Usually an artist faithfully copied the traditional, often fanciful, picture in his exemplar, especially if it was of a creature or plant (such as an elephant or a delphinium139) which he had never seen. However, if he knew from personal ob servation what his subject looked like, he might draw it realistically. Thus there are, for example, realistic pictures of a swan, pigeons, hedgehogs, a bramble an thistle (Pis. 9-13).14° These pictures (with the exception of the two plants which could have been drawn from life) were probably drawn from memory and are not accurate in every detail: they are slightly stylized and show a tendency to conflate one species with another. Nevertheless, they demonstrate that some artists, like some writers, had closely observed the world around them.141 138
Illustrating Gerald of Wales' Topographic Hibernica, dist.i, cap. xiv. The pronounced tail resembles that of a crane, although the general shape is most like a heron (whose favourite food is the eel), and the colouring (black with white breast and belly, red bill and legs) is that of a black stork. Holmes, loc. cit. p. 117, unequivocally identifies the drawing as a black stork. I owe the comments on the birds in this (pi. 8) and the next two plates (9 and 10) to my colleague in the University of Nottingham, Dr. A. K. Kent. 139 See The Herbal ofApuleius Barbaras from the early twelfth century Manuscript formerly in the Abbey of Bury St Edmunds, ed. R. T. Gunther (Roxburghe Club, 1925), pp. xxiv-xxv. 140 Plate 9, a swan, from a Bestiary of ca. 1200 (Oxford, Bodleian Library, MS. Ashmole 1511, f. 71) is a reasonably life-like representation of a mute-swan. However, the beak is not the correct shape or colour and is closer to that of a goose than a swan. Plate 10, pigeons, from a Bestiary of ca. 1120-1130 (Oxford, Bodleian Library, MS, Laud Misc. 247, f.!61v); the beak and legs are not correct, but the preening-action of the second from the left, and the landing-action of the second from the right, are very lifelike. Plate 11, hedgehogs, from a Bestiary of ca. 1200 (Cambridge, University Library, MS. Ii.3.26,f.28v); reproduced The Bestiary, being a Reproduction of the Manuscript Ii.4.26 in the University Library, Cambridge, ed. M. R. James (Roxburghe Club, 1928). Plate 12, a bramble, from the early twelfth-century Herbal reproduced by Gunther, op. cit. (Oxford, Bodleian Library, MS. Bodley 130, f.26). Plate 13, a milk thistle, from the same Herbal (f.37). This milk thistle, Carduus Mariannus, does not illustrate the text which describes snakeweed: see Gunther, op. cit. p. 113. 141 Brunsdon Yapp, having studied pictures of birds in medieval manuscripts, came to this conclusion: 'Not until the mid-13th century, in Gothic manuscripts, does there appear any attempt at the kind of accurate drawing that might satisfy a modern illustrator . . . There is then a sudden flowering of birds in decoration, which reaches its climax in England in the early years of the 14th century, and soon afterwards ceases even more abruptly than it began.' W. Brunsdon Yapp, Birds in Medieval Manuscripts (British Library, 1981), p. 71. This indicates that the twelfth-century examples which I mention are rare. Brunsdon Yapp also points out that the vast majority of the realistic pictures of birds in 13th-early 14th century manuscripts are not found in bestiaries at all, but in other kinds of book, such as bibles and psalters. The artists of pictures in bestiaries were almost always dominated by the traditional bestiary illustrations, much as the authors of the texts relied generally on the authority of the Physiologus (probably of the second century A.D.), Pliny, Isidore, the bible, Fathers of the church and the like, and not on personal observation. See ibid. 114 and most recently: Xenia Muratova, 'Workshop Methods in English Late Twelfth-Century Illumination and the Production of Luxury Bestiaries' in Beasts and Birds of the Middle Ages, ed. W. B. Clark and M. T. McMunn (University of Pennsylvania Press, Philadelphia, 1989), pp. 56—7, and J. B. Friedman, Teacocks and Preachers: Analytic Technique in Marcus of Orvieto's Liber de Moralitatibus, Vatican lat. MS 5935' in ibid. 185-6.
This page intentionally left blank
8 The Chronicles of Medieval England and Scotland
Introduction The aim of the following study is to classify and characterise the chronicles of medieval England and, to a lesser extent, of Scotland. Attention is especially paid to their later sections, those dealing with the authors' own times. The subject will be dealt with in seven parts: 1. 2. 3. 4. 5. 6.
Definition of the term 'chronicle', and the usage adopted here A brief survey of chronicle writing The provenance of chronicles Motives for writing chronicles The government's use of, and influence on, chronicles Suggested reasons for the decline of the chronicle as an historiographical genre 7. The composition of the 'contemporaneous' sections of chronicles. 1. Definition of the term 'chronicle3
The first step in any attempt to characterise the chronicles of medieval England and Scotland must be to classify them, that is to distinguish them from other kinds of historical writing. This is necessary because the term 'chronicle' has been used so loosely in medieval and modern times that it has lost any precise meaning. Here it will be used only of general, serious historical writings. Therefore, local histories and biographies, such as the 'chronicle of Battle Abbey' and the 'chronicle ofjocelin of Brakelond', will be excluded. Also to be excluded are histories in the romance style, most of which are in French, for example the Chronique of Jordan Fantosme. Because their primary purpose was to entertain, they do not share a number of the features typical of chronicles as characterised below. Chronicles, then, by our definition fall into the category of general, serious history. However, further classification is necessary, because the category itself is composite. It includes both 'histories' and 'chronicles'. The distinction between these two genres was already recognised in
200
Legends, Traditions and History in Medieval England
classical times,1 and was well expressed in the late twelfth century by Gervase of Canterbury: It is characteristic of history to tell the truth, to persuade those who read or hear it with soft words and elegant phrases and to inform them about the deeds, ways and lives of anyone it truthfully describes; it is an essentially rational study. A chronicle, on the other hand, reckons the years, months and Kalends from the Incarnation of our Lord, briefly tells of the deeds of kings and princes which happened at those times, besides recording any portents, miracles or other events.2 Thus histories were in literary form, and, although their contents were expected to be 'true', they were selected in order to suit a theme. They observed chronology, but were not structured by it. Chronicles, on the other hand, were dominated by their chronological structure; a chronicler entered the events of each year under the appropriate year-date. Examples of histories are Eusebius3 Ecclesiastical History* and Bede's Ecclesiastical History of the English People* Examples of chronicles are Eusebius' Chronicle,5 and the two chronicles which Bede appended to his two books on time.6 However, the two genres of histories and chronicles tended to overlap. A chronicler might write eloquently and at length, revealing his own opinions, so that his work acquired features of a history. Gervase of Canterbury deplored this tendency. Quoting Matthew 23.5, he said: 'They make broad their phylacteries, and enlarge the borders of their garments.'7 Examples of chroniclers who thus exceeded the limitations of
1 Cicero had distinguished between the early annalists and the later rhetorical historians: De Oratore, II, 51-64; cf. Latin Historians, ed. T.A. Dorey (London, 1966), pp. xiixiii. 2 The Historical Works of Gervase of Canterbury, ed. William Stubbs (Rolls Series, 1879-80, 2 vols), i. 87; cited V.H. Galbraith, Historical Research in Medieval England (London, 1951), repr. in idem, Kings and Chroniclers (Hambledon Press, London, 1982), p. 2. 3 The Ecclesiastical History, ed., with an English translation, K. Lake and J.E.L. Oulton (Loeb Classical Library, London, 1949-57, 2 vols). 4 The standard edition is Venerabilis Baedae Historia Ecclesiastica gentis Anglorum . . ., ed. Charles Plummer (Oxford, 1896, 2 vols). A more recent edition, with an English translation, is by Bertram Colgrave and R.A.B. Mynors (Oxford Medieval Texts, 1969). 5 For St Jerome's Latin translation of Eusebius' Chronicle see Die Chronik des Hieronymus: Hieronymi chronicon, ed. Rudolph Helm (Eusebius Werke, vii, Die Griechischen Christlichen SchriftstellerdesErstenJahrhunderts, xlvii, 24 and 34, Berlin, 1956). 6 Bede's chronicles, one from the Creation to 703 and the other from the Creation to 725, form the last part of his De Temporibus and De Temporum Ratione respectively; printed parallel Monumenta Germaniae Historica, Auctores Antiquissimi, xiii, Chronica Minora Saeculi ivvii, vol. 3, ed. T. Mommsen (Berlin, 1898), 247-317, 247-327. 7 Gervase of Canterbury, ed. Stubbs, i. 87-8.
The Chronicles of Medieval England and Scotland
201
their genre are Henry of Huntingdon8 and Matthew Paris.9 Chronicles, therefore, must be differentiated from histories. A further distinction has to be made, between chronicles and annals. The term 'annals' itself is confusing, since a chronicle consists of individual 'annals', an 'annal' being the entry for each year. Used generically, the term 'annals' is usually understood to mean a chronicle with very short yearly entries. The earliest annals written in the Christian West were notes of a line or two made in the margins of Easter Tables, opposite the appropriate years. By process of evolution annals came to be written independently of Easter Tables.10 The entries might be copied from an earlier work or works, or composed by the author himself fairly soon after the events they record happened. However, the distinction between chronicles and annals defined in this way is blurred. A work might, and often does, start with brief annals and then become prolix, or, which is rarer, vice versa. Bernard Guenee adopts another distinction between them.1 * In his view a chronicle is a compilation from various written sources, mainly standard works, while annals are a more or less contemporaneous account of events. But this definition, like the other, has a drawback, because very often a work begins as a compilation, and continues as a contemporary account. It is with the last kind of work, combined chronicle and annals according to Guenee's definition, that this article is concerned, and for convenience they will all be called chronicles.12 2. Survey of Chronicle Writing in England The chronicle tradition in England goes back at least to the eighth century, when somewhere in Northumbria Latin annals were added to the
8 Henrici Archidiaconi Huntendunensis Historia Anglorum, ed. Thomas Arnold (Rolls Series, 1879). 9 Matthaei Parisiensis, Monachi S. Albani, Chronica Majora, ed. H.R. Luard (Rolls Series, 1872-83, 7 vols). 10 See R.L. Poole, Chronicles and Annals (Oxford, 1926), pp. 5 et seqq. P. Grosjean, 'La date du Colloque de Whitby', Appendix 'Sur les annales anglaises du VIP siecle jointes a des tables pascales', Analecta Bollandiana, Ixxviii (1960), 255-60; The Anglo-Saxon Chronicle, ed. G.N. Garmonsway (Everyman's Libary, London, 1953, revised ed. 1954), pp. xx-xxii and n. 2 and plate. 11 Bernard Guenee, Histoire et Culture Historique dans VOccident Medieval (Paris, 1980), pp. 203-4. 12 This usage is narrower than that adopted by John Taylor, The Use of Medieval Chronicles (Helps for Students of History, no. 70, Historical Association, 1965). For example, he includes in his survey the St Albans' Gesta Abbatum (p. 12), a local history, Higden's Polychronicon (p. 4), largely an encyclopaedia, and the works of Langtoft (p. 9) and Froissart (p. 17), both in the romance style. None of these fall within our category.
202
Legends, Traditions and History in Medieval England
chronological summary of Bede's Ecclesiastical History.1^ But it owed its extraordinary vitality to the Anglo-Saxon Chronicle which, although based up to the early ninth century on now lost Latin annals, was first composed (to 890) in Wessex in the reign of King Alfred.14 Copies were sent to important churches where they were continued. At this period its history is obscure, but, by the late tenth century, copies were at the Old Minster Winchester, Abingdon, Ripon and either at Evesham, Worcester or York.15 In the eleventh century Christ Church and St Augustine's, Canterbury, and Peterborough all had copies. Other copies are lost.16 A few bridged the Norman Conquest; the Peterborough one continued until the accession of Henry II.17 Although the Norman Conquest doomed the Anglo-Saxon Chronicle, since English was no longer the language of educated people, its influence survived. Once translated into Latin it could continue in serviceable use in the many monasteries where it had taken root. The task of translation had begun almost at the time of its inception. Asser translated a substantial portion for his Life of King Alfred,13 and Aethelweard the ealdorman did the same for the chronicle which he composed in about 1000.19 However, the first translation of the whole was apparently made shortly after the Norman Conquest, in the late eleventh century, at Christ
13 See: Historia Ecclesiastica, ed. Plummer, i. 361-3, ii. 345 note; Chronica MagistriRogeride Houedene, ed. William Stubbs (Rolls Series, 1868-71, 4 vols), i. xxviii-xxx; P.H. Blair, 'Some observations on the 'Historia Regum' attributed to Symeon of Durham', Celt and Saxon, ed. N.K. Chadwick (Cambridge, 1963), pp. 86-99. See also D.N. Dumville, 'A new chronicle fragment of early British history', English Historical Review, Ixxxviii (1973), 31214. 14 See: The Anglo-Saxon Chronicle, ed. D.N. Dumville and Simon Keynes, iv, MS. B, ed. Simon Taylor (D.S. Brewer, Cambridge, 1986), p. xi;J. B. Bately, 'Manuscript layout and the Anglo-Saxon Chronicle', Bulletin of the John Rjlands Library, Ixx (1988), 23 and nn. 15 For the claims of Evesham, Worcester and York see respectively: Two of the Saxon Chronicles Parallel, ed. Charles Plummer, on the basis of an edition by J. Earle (Oxford, 1892-9, 2 vols, repr. 1952 with two notes by D. Whitelock), ii. Ixxv-lxxvii; Anglo-Saxon Chronicle, ed. Garmonsway, p. xxxvii; F.M. Stenton, Anglo-Saxon England (3rd edn, Oxford, 1971), p. 681, supported by The Peterborough Chronicle, ed. D. Whitelock (Early English Manuscripts in facsimile, iv, Copenhagen, 1954), pp. 29-30. For the continuations in general see Plummer and Earle, op. cit., ii. xxxvii et seqq. 16 Plummer and Earle, op. cit., ii. cxxv-cxxvii and Anglo-Saxon Chronicle, ed. Whitelock, pp. xvii-xxi. 17 The Peterborough Chronicle, 1070-1154, ed. Cecily Clark (2nd edn, Oxford, 1970). 18 Asser's Life of King Alfred . . ., ed. W.H. Stevenson (Oxford, 1904; new impression with a contribution by D. Whitelock, 1959), pp. Ixxxii-lxxviii, xcix-c, and Simon Keynes and Michael Lapidge, Alfred the Great: Asser's Life of King Alfred. . . (London, 1982), pp. 556 (for further references see ibid., index under 'Asser: learning of). 19 The Chronicle of JEthelweard, ed., with an English translation, Alistair Campbell (Nelson's Medieval Texts, 1962), pp. xvii-xviii, xxi-xlivpassim.
The Chronicles of Medieval England and Scotland
203
Church, Canterbury (BL Cotton MS Domitian A VIII). 20 The Anglo-Saxon Chronicle bequeathed two things to the postConquest period: a fund of information about Anglo-Saxon times, and a historiographical model. The first chronicler to make extensive use of it was apparently the monk of Worcester commonly referred to as Florence,21 who started the Chronicon ex Chronicis in the late eleventh or early twelfth century. For Anglo-Saxon history up to the eighth century Bede was, of course, the Worcester monk's principal authority, but thereafter he relied mainly on the Chronicle; in fact much of his material is little more than a Latin translation. Subsequent chroniclers followed suit in recognising that the Chronicle was a source of basic importance. Moreover, it provided an historiographical model for the writing of annals shortly after the events to be recorded took place, and also provided a starting point for continuation. No doubt its existence encouraged the Worcester monk to compile his chronicle in the first place. (There was also another model, one brought from the continent by Robert Losinga, bishop of Hereford (1079-95), that is the Universal Chronicle by the German, Marianus Scotus.) At Worcester, therefore, the annalistic tradition was adapted to the needs of the post-Conquest period. Worcester itself became a centre of historiography. Copies of the Chronicon ex Chronicis, which was continued at Worcester ultimately to 1140 by a monk called John, soon reached other monasteries. A copy was at Durham probably shortly after 1119,22 and almost immediately John had completed the annal for 1130 other monasteries received copies. By the end of the twelfth century copies were at, for example, Abingdon, Coventry, Gloucester, Bury St Edmunds and Peterborough.23 The influence of the Worcester chronicle continued to
20 See N.R. Ker, Catalogue of Manuscripts containing Anglo-Saxon (Oxford, 1957), no. 148. The Latin text is edited by F.P. Magoun, jr,' Annales Domitiani Latini: an Edition', Mediaeval Studies, ix (1947), pp. 235-95. For a detailed account of Canterbury as an historiographical centre in the late Anglo-Saxon period and under the Anglo-Normans see D.N. Dumville, 'Some aspects of annalistic writing at Canterbury in the eleventh and early twelfth centuries', Peritia, ii (1983), 23-57. 21 Florence's part in the composition of the Worcester chronicle is problematical. See: Vita Wulfstani, ed. R.R. Darlington (Camden Society, third series, xl, 1928), pp. xi-xviii; idem, Anglo-Norman Historians (London, 1947), p. 14; Martin Brett, 'John of Worcester and his contemporaries', in The Writing of History in the Middle Ages, Essays Presented to R.W. Southern, ed. R.H.C. Davis and J.M. Wallace-Hadrill (Oxford, 1981), p. 104 (and n. 3 for further references); above pp. 116-20, 123. 'Florence' is printed Florentii Wigorniensis Monachi Chronicon ex Chronicis, ed. Benjamin Thorpe (English Historical Society, 1848-9, 2 vols, repr. Krau, Vaduz, 1964). Cf. above p. 123 Note. 22 Hunter Blair, 'Some observations on the Historia Regum attributed to Symeon of Durham', pp. 107-10. 23 See: The Chronicle of John of Worcester 1118-1140, ed. J.R.H. Weaver (Anecdota Oxoniensia, 1908), pp. 4-9; Brett, loc. cit., pp. 106-10.
204
Legends, Traditions and History in Medieval England
spread in the thirteenth century.24 By that time the most important centre of historical writing was St Albans. There, in about 1259, Matthew Paris wrote an abbreviated version of his Chronica Major a, the Flores Historiarum which was a chronicle from the Creation to 1258. It was continued by another monk of St Albans to 1265. A copy with the continuation went to Westminster abbey where it was further continued to 1307 and then to 1326.25 From Westminster copies with all or part of the continuation to 1307 soon reached, for example, St Augustine's Canterbury, St Swithun's (as the Old Minster was now called) Winchester, Rochester, Norwich and St Benet of Hulme.26 Nor was its popularity confined to the Benedictines; the Augustinians of Merton (in Surrey),27 the Cistercians of Tintern, and the secular canons of St Paul's and also those of St Mary's Southwark, all had copies, which their chroniclers plundered and continued.28 Moreover, the citizen of London who wrote the Annales Londonienses (1194-1316) extracted from the Flores to 1306 and continued his chronicle to 1330;29 the Flores thus contributed to the evolution of a new species of chronicle, the London chronicle. Another historiographical centre in the thirteenth century was Bury St Edmunds. The first recension of its chronicle, by the monk John Taxter, covered the period from the Creation to 1265; it was later continued to 1296 and then to 1301.30 The recension to 1296 was influential especially
24
A copy of the Worcester chronicle (or extracts from it) was available at St David's in Wales, in the thirteenth century; see Kathleen Hughes, 'The Welsh Latin chronicles: Annales Cambriae and related texts', Proceedings of the British Academy, lix (1973), 246. 25 Matthew Paris's Flores and its St Albans and Westminster continuations are printed Flores Historiarum, ed. H.R. Luard (Rolls Series, 1890), 3 vols). 26 Bodleian Library Laud MS 572. See Flores, ed. Luard, i. xxxi-xxxiii, xvii-xviii (cf. N.R. Ker, Medieval Libraries of Great Britain (2nd ed., London, 1964), p. 201), xxvi-xxvii, xxix, xxii-xxiv, respectively. Variants from the standard Westminster text of the Flores in the Rochester version are printed in Flores, ed. Luard, iii. 327-8. 27 For the Merton copy, which apparently preserved an otherwise lost Westminster version of the Flores, see below pp. 247-8. 28 Ibid., i. xxii-xxiv, xxvii-xviii, xxix, respectively. The variants from the standard Westminster text of the Flores in the Tintern version are printed in ibid., iii. 328-48. 29 Chronicles in the Reigns of Edward I and Edward II, ed. W. Stubbs (Rolls Series, 1882-3, 2 vols), i. xvii-xviii, and J. Catto, 'Andrew Horn: law and history in fourteenth-century England' The Writing of History in the Middle Ages, ed. Davis and Wallace-Hadrill, pp. 367, 375. 30 The Bury chronicle is printed from 1212, The Chronicle of Bury St Edmunds, ed., with an English translation, A. Gransden (Nelson's Medieval Texts, 1964). For discussion of how the contemporary annals were composed see below pp. 226-38 passim.
The Chronicles of Medieval England and Scotland
205
in East Anglia;31 it was plundered by the chroniclers of Norwich,32 St Benet of Hulme,33 Peterborough34 and Spalding.35 In the early fourteenth century much of it was copied by the chronicler of St John's, Colchester,36 and by the middle of that century an abstract was with the Grey Friars of Babwell, near Bury St Edmunds, which later passed to the Grey Friars of Lynn.37 Although only exceptionally did the influence of the Bury chronicle spread outside eastern England, it was known to a chronicler in Wales, probably a monk of the Cistercian abbey of Whitland, in the late thirteenth century,38 and in the late fourteenth century a copy of'Taxter' was in the Augustinian friary in York.39 The thirteenth century was the heyday of the monastic chronicle, when, as will be seen, chronicles fulfilled an important record function. Most monasteries, besides those already mentioned, kept chronicles; for example, the Benedictine houses of Burton,40 Tewkesbury,41 Battle,42 St Swithun's and Hyde abbey in Winchester,43 and Worcester,44 and the
31
For the dissemination of the Bury chronicle see below pp. 226-35 passim, 241-3. A similar centre of chronicle writing was St Swithun's Winchester. See N. Denholm-Young, 'The Winchester-Hyde chronicle', English Historical Review, xlix (1934, repr. in idem, Collected Papers on Medieval Subjects (Oxford, 1946)), 85-93, and see below n. 43, pp. 211 and n. 64, 225 and n. 122, 235 and n. 184. 32 See Bartholomaei de Cotton Monachi Norwicensis Historia Anglicana 449-1298, ed. H.R. Luard (Rolls Series, 1859), pp. lii-lviii. 33 Chronicajohnannis de Oxenedes, ed. Henry Ellis (Rolls Series, 1859). 34 The Peterborough version (Corpus Christi College, Cambridge, MS 92) of the Bury chronicle, which covers the years from 1152-1294 and is a continuation of the chronicle of 'Florence' and John of Worcester, is printed in Chronicon ex Chronicis, ed. Thorpe ii. Cf. below pp. 228, 237, 238. 35 prmteci in Joseph Sparke, Historiae Anglicanae Scriptores Varii (London, 1723, 2 vols), pp. 1-137, and reprinted as Chronicon Angliae Petriburgense, ed. J.A. Giles (Caxton Society, 1845). Cf. Felix Liebermann, 'Ueber Ostenglische Geschichtsquellen', Neues Archiv, xviii (1892), 235-45. 36 The Colchester annals are described and printed in Felix Liebermann, Ungedruckte Anglo-Normannische Geschichtsquellen (Strasburg, 1879), pp. 158-65. 37 See below p. 279-80. 38 Hughes, loc. cit., pp. 247-9. 39 Bury Chronicle, ed. Gransden, pp. xxxvii-xxxviii. 40 A.D. 1004-1262, printed in Annales Monastici, ed. H.R. Luard (Rolls Series, 1864-9, 5 vols),i. 183-500. 41 A.D. 1066-1262, printed in ibid., i. 43-170. For another chronicle, 1258-1263, copied at Tewkesbury but not necessarily composed there, see below p. 216 n. 83. 42 Brutus to 1264 (continued to 1286), mainly unpublished. Described and annals 1258-65 printed in Charles Bemont, Simon de Montfort Comte de Leicester (Paris, 1884, repr. Geneva, 1976), pp. xiv-xv, 373-80. 43 The chronicles of these two houses are closely related. A version, A.D. 519-1277, is printed in Ann. Mon., ii. 3-125. Cf. Denholm-Young, 'Winchester-Hyde chronicle', pp. 8693. 44 Incarnation-1308; printed in Ann. Mon., iv. 355-560. Cf. Denholm-Young, loc. cit., p. 88, and below p. 208 and n. 55.
206
Legends, Traditions and History in Medieval England
Cistercian abbeys of Margam and Waverley, and Melrose in southern Scotland.45 Similarly, chronicles were kept by the Dominican priory of Dunstable and the Augustinian abbey of Osney.46 These chronicles, with few exceptions, end late in Henry Ill's reign or early in Edward Fs. In the fourteenth century the monks' near monopoly of chronicle writing was increasingly challenged by secular clerks and then by laymen. Nearly all the chroniclers belonging to these classes were Londoners; men such as Adam of Murimuth, a prebendary of St Paul's,47 and Robert of Avesbury, registrar at the court of the archbishop of Canterbury at Lambeth,48 besides, of course, the authors of the London chronicles proper.49 The most popular of all fourteenth-century chronicles were the Brut chronicles, which apparently originated in the reign of Edward I; probably they too were by secular clerks, and written in London.50 The monastic tradition of chronicle writing enjoyed its last period of vitality in Richard II's reign. Indeed, at St Albans the revival lasted well into the fifteenth century. There Thomas Walsingham, the last great St Albans' chronicler, brought Matthew Paris' Chronica Majora up to date.51 However, in the fifteenth century the writing of chronicles became primarily an activity of secular clerks and laymen. They managed to keep the chronicle tradition alive because they accommodated it to the rise of 45 The Margam chronicle (1066-1232) and the Waverley chronicle (Incarnation-1291) are printed in Ann. Mon., i, 3-40, ii. 129-411. The Melrose chronicle, A.D. 731-1275, is edited in facsimile by A.O. and M.O. Anderson (limited edn, London, 1936); see below Plate V. 46 The Dunstable chronicle (Incarnation-1297) and Osney chronicle (A.D. 601 -1297) are printed in Ann. Man., iii. 3-408, iv. 3-352. For the Dunstable chronicle see below pp. 226, 237. 47 Printed in Adae Murimuth Continuatio Chronicarum. Robertus de Avesbury De Gestis MirabilibusRegisEdwardi Tertii, ed. E. Maunde Thompson (Rolls Series, 1889), pp. 3-219. 48 Printed in ibid., pp. 279-471. For what is known about Avesbury's life see: ibid., pp. xxii-xxiii; C.L. Lethbridge, 'Robert de Avesbury', English Historical Review, xxii (1907), 292. 49 The earliest London chronicle in print is the Cronica Maiorum et Vicecomitum Londoniarum, probably by Arnold Fitz Thedmar, an alderman; printed in De Antiquis Legibus Liber, ed. Thomas Stapleton (Camden Society, original series, xxxiv 1846), pp. 177. The author of the Annales Londonienses may well have been Andrew Horn, fishmonger, alderman and chamberlain of London (see above, p. 204 and n. 29). See also, with further references, A. Gransden, Historical Writing in England, [i], c. 500-c. 1307 (London, 1974), pp. 509-17 and nn., and, for the fifteenth-century versions, below p. 207 n. 52. 50 Printed The Brut or the Chronicles of England, ed. F.W.D. Brie (Early English Text Society, cxxxi, cxxxvi, 1906, 1908, 2 vols, repr. 1960); cf. idem, Geschichte und Quellen der mittelenglischen Prosachronik the Brute of England oder the Chronicles of England (Marburg, 1905), and John Taylor, English Historical Literature in the Fourteenth Century (Oxford, 1987), Chapter 6. 51 For Thomas Walsingham and the St Albans' tradition of historiography see: The St Albans Chronicle 1406-1420, ed. V.H. Galbraith (Oxford, 1937), pp. xxvii-lxxi; Gransden, Historical Writing, [i], Chapter 16; idem, Historical Writing in England, ii, c. 1307 to the Early Sixteenth Century (London, 1982), ii, Chapter 5; Taylor, English Historical Literature, pp. 5978.
The Chronicles of Medieval England and Scotland
207
the vernacular. In the same way as the tradition survived the Norman Conquest because the Anglo-Saxon Chronicle was translated into Latin, so now it survived because chroniclers wrote in English. The earliest versions of the Brut chronicle are in French, but by the late fourteenth century they were being composed in English. So too were the closely related London chronicles, which proliferated in the fifteenth century and were particularly favoured by the merchant class.52 The longest of the Brut chronicles ends in 1475, and the longest London ones in the early sixteenth century. Nevertheless, the influence of the London chronicles survived in the Tudor period. Some of them supplied Robert Fabyan with information for his New Chronicles of England and France (completed in 1504), which preserved many of their characteristics, notably, besides the chronological structure, the interest in, and knowledge of, the City of London.53 Furthermore, the chronicle of the Grey Friars of Newgate, and its continuation from the friary's dissolution in 1538 to the accession of Queen Mary, are virtually a London chronicle.54 The influence of the London chronicles is very evident in the works of the Tudor historians Edward Hall, John Stow and Raphael Holinshed. They used the London chronicles as sources, and retained the chronological structure. John Stow's Chronicles of England (1580) was in fact a re-edition of the old London chronicles; Stow borrowed from them, and when they came to an end he added new annals up to his own day. The subsequent fate of the chronicle tradition will be discussed in Part 6 below. 3. The Provenance of Chronicles Monasteries, especially Benedictine ones, were the homes par excellence of chronicle writing. Their chronicles are more numerous, and on average of better quality, than those of other institutions or of individual nonmonastic authors. They offered chroniclers from generation to generation a combination of facilities. Their libraries contained standard histories and chronicles, which could serve as historiographical models. Such books also provided information necessary for the compilation of early history. Although monastic archives consisted principally of the house's charters and other business documents, a few 'public records' were 52
For the fifteenth-century London chronicles see: C.L. Kingsford, Chronicles of London (Oxford, 1905); The Great Chronicle of London, ed. A.H. Thomas and I.D. Thornley (London, privately printed, 1938); with further references, Gransden, Historical Writing in England, ii 227-48 and nn. 53 The New Chronicles of England and France by Robert Fabyan . . ., ed. Henry Ellis (London, 1811). 34 A.D. 1189-1556, printed in Monumenta Franciscana, ed. J.S. Brewer and Richard Hewlett (Rolls Series, 1858, 1882, 2 vols), ii. 143-260.
208
Legends, Traditions and History in Medieval England
usually stored with them. The archives themselves impressed on monks the virtue of record-keeping, besides supplying material for the history of the not-so-distant past, and for recent and contemporary history. Moreover, the monastic chronicler was well placed to hear news by word of mouth. For local affairs he had the benefit of his house's position as a landowner, since this entailed close contact with tenants and neighbours. For national affairs he could question visitors to the guest house. Once a tradition of chronicle writing was established in a monastery it tended to be self-perpetuating; one chronicle provided a model, and also the foundation stone, or at least a quarry, for another. The tradition at Worcester, for example, may well have begun as early as the late tenth century, and lasted until towards the end of the thirteenth,55 while that at St Albans lasted from the early thirteenth century until the early fifteenth. Other institutions, notably secular cathedrals and the City of London, could supply chroniclers with facilities similar to those enjoyed by monastic chroniclers, but to a lesser extent. In fact, no secular cathedral except St Paul's in London established a flourishing chronicle tradition.56 Although London chronicles proliferated in the late Middle Ages, nearly all were apparently individual compositions — few can be proved to have been officially sponsored by the Guildhall.57 They lacked, therefore, stable corporate backing, and, indeed, their quality does not compare with that of the best monastic chronicles. 4. Motives for Writing Chronicles In a few cases there is evidence suggesting that a chronicle was written as a result of a direct commission, but usually the reasons are more obscure and less simple. One or other of a variety of reasons, or several, might 55
For Worcester's claim to have been the home of one version (D) of the Anglo-Saxon Chronicle see above, p. 202. and n. 15. For versions of the Anglo-Saxon Chronicle which were at Worcester after the Norman Conquest but are now lost see Anglo-Saxon Chronicle, ed. Whitelock, p. xx. For post-Conquest chronicles written at Worcester see above pp. 203 and nn. 21, 23, 24, 205 and n. 44. 56 Two chronicles survive from St Paul's, written in the last half of the thirteenth century and in the first half of the fourteenth (other St Pauls' chronicles may well be lost): Annales Sancti Pauli (1064-1274), described and extracts printed by F. Liebermann, Monumenta Germaniae Historica, Scriptores, xxviii (Hannover, 1888), pp. 548-51; Annales Paulini (to 1341, printed in Chronicles oj'the Reigns of'Edward 1and EdwardII, ed. William Stubbs (Rolls Series, 1882-3, 2 vols). Murimuth (d.? 1347) was a prebendary of St Paul's (above p. 206). Earlier, the prolific chronicler, Ralph Diceto, was dean of St Paul's (1180/1-1201); see Radulphi de Diceto Lundoniensis Opera Historica, ed. W. Stubbs (Rolls Series, 1876, 2 vols). 57 Some of the London chronicles were quasi-official productions, that is, they were by authors who held office in the City and used City records. But there is no evidence that any was actually commissioned by the mayor and corporation; see: D.C. Cox, 'The French chronicle of London', Medium Aevum, xlv (1976), 207; Gransden, Historical Writing, [i], 51114;^zW.,ii, 71,230-3, 242.
The Chronicles of Medieval England and Scotland
209
encourage or cause someone to write a chronicle. Therefore, for the sake of clarity, the subject will be discussed from a number of angles in two main sections, each with sub-sections, first, 'internal' incentives, and, secondly, 'external' ones. i. 'Internal'incentives
These comprise the chronicler's spontaneous motives, not those he might adopt on account of external circumstances and pressures. However, these motives were not peculiar to the chronicler as an individual, but as a member of the community to which he belonged, whether a monastery, the City of London, or whatever. (a) RELIGIOUS PURPOSE
It is commonly held that chronicles were intended to be edifying and serve a religious purpose, to strengthen Christian faith and improve behaviour.58 It is true that any account of past events could be seen as demonstrating the workings of God's providence on earth. However, the edificatory element in chronicles has been exaggerated, perhaps as a result of applying to 'chronicles' a characteristic feature of'histories' and even more of hagiographies. In general only those chroniclers who 'broaden their phylacteries', writers such as Henry of Huntingdon and Matthew Paris, point out and dwell upon examples of divine retribution and the like. Bede had adopted for his two World Chronicles the periodisation of the Seven Ages, symbolising the seven days of the Creation and the seven stages of a man's life.59 Matthew Paris adopted these divisions for his Chronica Majora, but he was exceptional in this respect (as in many others). Most chroniclers were satisfied with the simple annalistic arrangement. Although Robert Fabyan divided his chronicle into seven books, these symbolised the Seven Joys of the Virgin Mary, not the days of the Creation.60 And although Ranulph Higden used the divisions of the Seven Ages for his Polychronicon?1 the latter is not a 'chronicle' by our definition. (b)
INFORMATORYPURPOSE
There can be no doubt that one of a chronicler's main motives was to satisfy man's natural curiosity about the past and his desire for contemporary news; a chronicle was both a history book and a newspaper. 58
See e.g. R.G. Collingwood, The Idea of History (Oxford, 1946), pp. 52-6 passim, and J.R. Hale, The Evolution of British Historiography (London, 1967), pp. 9-10. 59 The concept of the Seven Ages derived from St Augustine, De Civ. Dei, xxii, 30. See: Auguste Luneau, L'Histoire du salut chez les Peres de I'finglise (Paris, 1964), pp. 285 et seqq., 352 etseqq. C.A. Patrides, The Grand Design of God (London, 1972), pp. 18-22 and nn. 60 See Gransden, Historical Writing in England, ii. 246 and n. 175. 61 See John Taylor, The Universal Chronicle of Ranulf Higden (Oxford, 1966), p. 39.
210
Legends, Traditions and History in Medieval England
It is noticeable that newsworthy events, such as the Viking invasions during the Anglo-Saxon period, the Barons' War in Henry Ill's reign, the Peasants' Revolt of 1381 and the deposition of Richard II, were all accompanied by a spate of chronicle writing, and may in some instances have provided the initial incentive to composition.62 The chronicler's immediate audience, his own community, might well have encouraged him to write simply because he presented information in an attractive form; his chronological, episodic narrative suited medieval taste. Whatever sentiments and opinions a chronicler expressed were bound to please his fellow monks, since in normal circumstances they would have been their own. Although undoubtedly monks of a community might be bitterly divided concerning internal matters, it is unlikely that they often were about national politics and other external affairs. Their strong sense of corporate identity and their communal self-interest must have militated against divergence in attitude, and (as should become apparent in the next section) a chronicler was his community's mouthpiece. The fact that nearly all the thirteenth-century chronicles share a common outlook, makes it fair to conclude that the predominant element in the total monastic population also did so; one result of this would have been that the information purveyed in the chronicle of one house would have been congenial to monks of another. (c) CHRONICLES AS RECORDS
Besides serving a general, informatory purpose, chronicles had particular, local interest and utility. A chronicler's concern for his own monastery, or its equivalent, tended to override all other considerations. He fitted the history of his own institution into the framework of general history, which gave the chronicle additional interest to his audience. Moreover, it was useful to have local history in such a context. For example, a monastery's comparative antiquity might be demonstrated in this way, which would enhance its prestige. More specifically, a chronicle with notices of a community's acquisition of property and privileges and of its important lawsuits, perhaps including the citation in extenso of relevant documents, could serve as a business record. There is no doubt that often an important motive for composition was to provide a useful record. The record element is clear in many monastic chronicles and also in the fifteenth century London ones. It is especially characteristic of the monastic chronicles of the thirteenth century. As records, chronicles had an advantage over cartularies or registers; information and documents in them were in strictly chronological order. This was no small advantage especially in the thirteenth century when the 62
For the suggestion that the baronial opposition to Henry III, the Peasants' Revolt and Richard IPs deposition prompted the writing of chronicles see: Gransden, Historical Writing, [i]. 407; ibid., ii. 142, 162-3.
The Chronicles of Medieval England and Scotland
211
monasteries were proliferating and accumulating innumerable documents, but when the organisation of archives was in its infancy.63 It was becoming increasingly difficult to find a specific document; but if its date were known, there was at least a chance of finding it in a chronicle. Although the thirteenth-century monastic chronicle, like other chronicles, was a literary production and usually seems to have been kept in the monastic library, it is best regarded as a product of the recordkeeping mentality. It was apparently an 'official' production in the same way as cartularies and registers were. This is implied by the passage in the prologue of the Winchester/Worcester chronicle which describes how contemporaneous annals should be written.64 Having specified that current news should be noted in plummet on a schedule of loose leaves attached to the end of the chronicle, the author proceeds: At the end of each year let he who is ordered, not just anyone who likes, record for posterity what he thinks best, entering it briefly at the end of the book. This shows that in this instance at least a monastic superior commissioned the chronicler. It also shows that a chronicle might be by more than one monk (the same passage later mentions that a chronicle 'was composed by various people'). Thus chronicles were corporate productions. Most are anonymous; only the more sophisticated ones have names attached. Their anonymity, however, is of little importance since the resources which a chronicler used and, in normal circumstances, the views he expressed (as I have argued above) must surely have been those of his community, at least so far as they related to national affairs. ii. 'External'incentives
Influences from outside might impinge on any chronicler, whether monk, secular clerk or layman, and cause or encourage him to write. He might receive a commission from the government or from an individual, or simply intend to please a patron, or he might be inspired by patriotism. These possibilities will be considered in turn.
63 SeeM.T.Clanchy, From Memory to Written Record (London, 1979), pp. 122-47 (esp. pp. 123, 139, 143, 147). 64 The passage is printed from the Worcester chronicle, which copied it from a lost Winchester chronicle (Denholm-Young, op. tit., p. 88), in Annales Monastici, ed. Luard, iv. 355. It is cited by V.H. Galbraith (who ascribes it to the Worcester chronicler), 'The St. Edmundsbury chronicle, 1296-1301', English Historical Review, Iviii (1943), 53; cf. C.R. Cheney, 'The making of the Dunstable Annals, A.D. 1233 to 1242', in idem, Medieval Texts and Studies (Oxford, 1973, first publ. in Essays in Medieval History presented to Bertie Wilkinson, ed. T.A. Sandquist and M.R. Powicke (Toronto, 1969)), p. 228. For the light which the passage throws on the compilation of the contemporary annals of a chronicle see below p. 235 and n. 184 (for the text itself).
212
Legends, Traditions and History in Medieval England
(a) ROYAL AND GOVERNMENT
SPONSORSHIP
There was no tradition of official, in the sense of government sponsored, historiography in medieval England. Not one chronicle explicitly acknowledges commission by the king. However, the content and tone of some chronicles provide evidence, convincing in some cases, that they were official. A case can be made for regarding the Anglo-Saxon Chronicle to 890 as an official history. If it was, it would be the earliest surviving English example.65 It may well have owed its inception to someone at the court of King Alfred, if not to the king himself. Since it is in Old English, it could well have been a product of Alfred's policy of promoting the vernacular as a written language. His intention was to raise the standard of education, primarily of the clergy, but also of thegns. The Chronicle demonstrates the rise of the West Saxon dynasty and of Alfred's achievements, and at the same time relates the history of the English people as a whole. Copies of the Chronicle were sent to important churches for preservation and continuation, but the various local versions seem to have retained some connections with the royal court. Apparently, indeed, the Chronicle was continued at Edward the Elder's court; a version survives which is strongly biased in Edward's favour.66 Even those chroniclers composing continuations locally received common instalments from some centre, probably one connected with court of the kings of Wessex at least for the earlier periods, for the years 892-924, 925-975 and 983 1018.67 After the Norman Conquest very few historical works were written which can be regarded as official. Nor do any of the most obvious examples, all of them fifteenth-century, fall within our definition of
65
Plummer and Earle, Two of the Saxon Chronicles Parallel, ii.civ, take the view that King Alfred himsef wrote or commissioned the Chronicle, R.H.C. Davis, 'Alfred the Great: propaganda and truth', History, Ivi (1971), 173-7, considered that its purpose was to rally support for Alfred in his struggle against the Vikings, even by distorting facts in Alfred's favour. Dorothy Whitelock, however, argues strongly against this opinion; D. Whitelock, The Importance of the Battle ofEdington, A.D. 878 (lecture to the Annual meeting of the Friends of Edington Priory Church, 1977, Westbury, Wilts.), pp. 1-4. Janet Bately supports her view; J.B. Bately, 'The compilation of the Anglo-Saxon chronicle, 60 BC to AD 890', pp. 127 and n. 3, 128. For a balanced judgement see Keynes and Lapidge, Alfred the Great, pp. 39-41. 66 For the Edwardian version of the Anglo-Saxon Chronicle see The Chronicle of jEthelweard, ed., with an English translation, Alistair Campbell (Nelson's Medieval Texts, 1962), pp. xxviii-xxix. 67 See Plummer and Earle, op. cit., ii. cxiv-cxxii, and The Battle ofBrunanburh, ed. Alistair Campbell, (London, 1938), pp. 1-7, 34-6. For the idiosyncratic character of the annals for Ethelred IFs reign (979-1016) in the Chronicle, which were compiled en bloc after his death, see Simon Keynes, 'The declining reputation of King Aethelred the Unready', in Ethelred the Unready: Papers from the Millenary Conference, ed. David Hill (British Archaeological Report, British Series, lix, 1978), pp. 229-36.
The Chronicles of Medieval England and Scotland
213
'chronicle'.68 They are the Gesta Henrici Quinti,69 John Capgrave's Liber de Illustribus Henricis™ and the two propaganda tracts of Edward TV's reign, the Chronicle of the Rebellion in Lincolnshire and the Historic of the Arrivall of Edward IV in England and the Finall Recovery e of his Kingdomes from Henry VI, 1471.71 Four other works approach nearer to our 'chronicle' classification. They are all arranged in chronological order, in annals, but on the other hand their last annals were not apparently composed contemporaneously with the events recorded; rather they seem to have been written en bloc retrospectively, though soon after the date of the concluding annal. The two earliest, the so-called 'Merton' Flores Historiarum (to 1306) and its continuation (to 1326) by Robert of Reading, were written in Westminster abbey.72 It appears, indeed, that in the early fourteenth century the monks of Westminster were writing official history for the kings of England. If so, for this brief period they can be compared with the monks of St Denis, in so far as the latter throughout the Middle Ages composed the official history of the kings of France. Westminster stood in similar relation to the kings of England as St Denis to the kings of France; the king was the abbey's patron, and the abbey was the scene of coronations, a royal mausoleum and situated close to the seat of government. The fulsome terms used by the 'Merton' Flores to praise Edward I are best explained by assuming that it was written to please king and court. Perhaps it was intended for Edward I himself. However, the fact that the account of each reign, from King John's to Edward I's, is prefaced by a picture of the respective king in full regalia sitting on the coronation throne, and that the series ends with a similar but rather larger picture of 68
Earlier, in Richard I's reign, a history in monograph form was written in favour of the king's opponents, the Appellants; Historia sive Narracio de Modo et Forma Mirabilis Parliamenti apud Westmonasterium Anno Domini Millesimo CCCLXXXCI . . . Thomas Favent Clericum indictata, ed. May McKisack (Camden Miscellany, xiv, 1926), pp vi-viii, 1-27. 69 Printed Gesta Henrici Quinti, ed. F. Taylor and J.S. Roskell (Oxford, 1975). For the Gesta's purpose of promoting the French war see ibid., pp. xxiii-xxviii. 70 Printed Johannis Capgrave Liber de Illustribus Henricis, ed. F.C. Hingeston (Rolls Series, 1858). 71 Printed Chronicle of the Rebellion in Lincolnshire, 1470, ed. J.G. Nichols (Camden Miscellany, i, 1847), and Historie of the Arrivall of Edward IV in England and the Finall Recoverye of his Kingdomes from Henry VI, A.D. M.CCCC.LXXI, ed. John Bruce (Camden Society, original series, i, 1838). For these works as official histories see Gransden, Historical Writing, ii. 261-5, and idem, 'Politics and historiography during the Wars of the Roses', Medieval Historical Writing in the Christian and Islamic Worlds, ed. D.O. Morgan (School of Oriental and African Studies, University of London, 1982), pp. 128-31. 72 These continuations of the Flores are discussed below pp. 245-65. The version of the Flores eulogizing Edward I is preserved in a copy made for Merton priory; the manuscript is described and the variants from the standard version are printed in Flores Historiarum, ed. H.R. Luard (Rolls Series, 1890, 3 vols), i. xv-xvii, iii. 239-327. Robert of Reading's chronicle is printed in ibid., iii. 137-235.
214
Legends, Traditions and History in Medieval England
Edward II,73 suggests the possibility that the 'Merton' Flores was written for presentation to Edward II, perhaps on the occasion of his coronation. Robert of Reading, writing about Edward II's reign, also produced a strongly biased narrative in highly coloured prose. In his case the bias was not in favour of the king, but against him. His consistent and eloquent vilification of Edward II indicates that he wrote for Queen Isabella and Roger Mortimer, to justify their seizure of power and Edward's deposition. The two other historical works which may be official are Robert of Avesbury's De Gestis Mirabilibus Edwardi Tertii and Thomas Walsingham's Ypodigma Neustriae. Avesbury eulogises Edward III, especially as a hero of chivalry,74 and is at pains to justify the French war and Edward's claim to overlordship of Scotland.75 His subject-matter, therefore, suggests the possibility that he wrote for the king. It is not inconceivable that the king commissioned him to write, since Avesbury was a well-educated Londoner (the fact that he gives details about the riot in Oxford in 1355 suggests that he had studied there) ,76 Walsingham explicitly dedicates the Ypodigma to Henry V in the prologue, and states that his intention is to provide the king with a history of Normandy, to justify the Norman campaigns.77 Again, it is not impossible that Henry V commissioned the work. Walsingham lived within fifteen miles of London and was a chronicler of wide reputation. Moreover, after Henry Bolingbroke's coup of 1399 he wrote in the Lancastrian interest, including in his narrative propagandist material issued by the new regime.78 Nevertheless, too much weight should not be attached to the dedication as evidence that Henry V commissioned the Ypodigma or even that it was intended for his eyes. Such a dedication was a literary commonplace. The practice of dedicating a work to a famous person began in ancient times, and flourished throughout the Middle Ages.79 The suggestion that the 73 The picture of the coronation of Henry III and that of Edward I are reproduced Pis 28, 29. 74
See e.g. Avesbury, ed. Thompson (above p. 206 n. 48), pp. 285, 390-1, 409-10, 413, 451,
455.
75
Ibid., pp. 286-97, 302-3, 339. Avesbury, pp. 421-3, noticed H. Rashdall, The Universities of Europe in the Middle Ages, ed. F.M. Powicke and A.B. Emden (Oxford, 1936, 3 vols), iii. 98 n. 2. For the riot see The History of the University of Oxford, ed. J.I. Catto (Oxford, 1984), p. 167. 77 Ypodigma Neustriae a Thoma Walsingham, ed. H.T. Riley (Rolls Series, 1876), pp. x, 3-5. 78 See: Gransden, Historical Writing, ii. 139, 140 and nn. 159, 161, 141; G.B. Stow, jr, 'Richard II in Thomas Walsingham's chronicles', Speculum, lix (1984), 91-8. 79 See: Tore Janson, Latin Prose Prefaces (Acta Universitatis Stockholmiensis. Studia Latina Stockholmiensia XIII, Stockholm etc., 1964), pp. 95, 116-45passim; Gertrud Simon, 'Untersuchungen zur Topik der Widmungsbriefe Mittelalterlicher Geschichtsschreiber bis zum Ende des 12. Jahrhunderts', ArchivfurDiplomatik, iv (1958), 54-63passim; ibid., v-vi (1959-60), 136-46; D.W.T.C. Vessey, 'William of Tyre and the art of historiography', Mediaeval Studies, xxxv (1973), 436-8. See above pp. 125, 127, 129. 76
The Chronicles of Medieval England and Scotland
215
dedication may be no more than a conventional flourish, receives some support from the fact that the Ypodigma soon abandons the history of Normandy, and concentrates on that of England. (b) PATRONAGE
A few chronicles were written to please a patron, perhaps even in response to a commission. Some monasteries had individual patrons, and all had benefactors. The monks were grateful to them for past benefits and hoped for future ones. They also needed their protection, and that of any powerful figure, especially in troubled times. Patronage was even more important for secular clerks and laymen than for monks. The same considerations weighed with them, and in addition they needed the help of patrons for advancement in their careers. Historiography provided a means of expressing gratitude and currying favour. Any historical work which touched on the activities of the patron was biased towards him and tended to reflect his viewpoint; the same applied to its treatment of his relatives. (Often, of course, authors writing for patrons did not adopt the chronicle mode; they preferred to write history in verse, in the romance style.)80 As in the case of official chronicles, explicit evidence of commission by individual patrons is lacking. Again we have to rely on the content and tone of the work. It is impossible to be certain whether the bias in a chronicle implies that it was actually commissioned by a patron, or merely reflects the chronicler's respect for him, accompanied perhaps by the desire to cultivate his good graces. A few examples may be cited of chronicles which were at least strongly influenced by patronage. During the anarchy of Stephen's reign Malmesbury abbey fell within Robert of Gloucester's orbit. William of Malmesbury dedicated the Historia Novella to Earl Robert;81 it is slanted in his favour and includes much eulogy of him.82 None of the thirteenth-century chronicles shows convincing signs 80 Examples of verse histories for patrons are: L'Histoire de Guillaume le Marechal, ed., with an abridged translation into modern French, Paul Meyer (Societe de 1'Histoire de France, 1891-1901, 3 vols), commissioned by the Marshal's son, William, fifth earl of Pembroke (1219-31); The Metrical Chronicle of Robert of Gloucester, ed. W.A. Wright (Rolls Series, 1887, 2 vols) (part of the text 1262-4 is also in Early Middle English Verse and Prose, ed. J.A.W. Bennett and G.V. Smithers (2nd edn, Oxford, 1968), pp. 158-64, 394-53 and see below), perhaps written for Sir Warin of Bassingbourn (see, with citations from Wright, rendered in modern English, of passages relating to Sir Warin, Warin of Bassingbourn fortified his Manor, no author or date, P1982, Cam. d. 982.5 in Cambridge University Library, and Gransden, Historical Writing, [i], 436); The Chronicle of John Hardyng, ed. Henry Ellis (London, 1812, repr. 1974), written to please various patrons (see C.L. Kingsford, 'The first version of Hardyng's chronicle', English Historical Review, xxvii (1912), 465 etseqq., and Gransden, op. cit., ii. 27'4-82 passim. 81 See The Historia Novella by William of Malmesbury, ed., with an English translation, K.R. Potter (Nelson's Medieval Texts, 1955). 82 Ibid., passim (esp. pp. 64-7).
216
Legends, Traditions and History in Medieval England
that it was written as a result of patronage.83 However, examples survive from the fourteenth century. It is likely that the secular clerk who wrote the French Brut to 1333 had some connection with the house of Lancaster; this would account for his Lancastrian bias, which appears especially in the fulsome praise of Thomas of Lancaster.84 The principal patron of Geoffrey le Baker (fl. 1350) was Sir Thomas de la More, but Baker's sympathetic interest in, and knowledge of, Humphrey, earl of Hereford, and other evidence indicate that he also had a Bohun patron.85 In Richard II's reign, Henry Knighton's hostility to king and government was no doubt partly because the earls and dukes of Lancaster were patrons of his abbey, St Mary's in the Meadows, Leicester.86 Adam of Usk (fl. 1352-c. 1421) was another Lancastrian partisan, partly no doubt because he wrote under the patronage of the earl of March and Archbishop Thomas Arundel.87 Conversely, it may well be that the Westminster chronicler gives a relatively favourable portrait of Richard II during the last decade of his rule because the king, the abbey's patron, was on good terms with the monks at that time.88 (c) PATRIOTISM A number of chronicles have passages with a patriotic ring. National 83
The chronicle from 1258 to 1263 preserved in a manuscript copied in Tewkesbury abbey (printed Ann. Mon., ed. Luard, i. 174-80). It is so pro-baronial in viewpont that it might well have been composed for a member of the house of Clare. The fact that a copy of the chronicle was made at Tewkesbury may be explained by the fact that the Clare earls of Gloucester were the abbey's patrons. (For the burial of the earls in Tewkesbury abbey see Michael Altschul, A Baronial Family in Medieval England: the Clares, 1217-1314 (Baltimore, 1965), pp. 34, 60, 92, 155, 164.) However, these two facts are insufficient to prove that the chronicle was actually composed there. Indeed, it seems unlikely that it was, since it makes no mention of Tewkesbury. 84 The version of the Brut to 1333, printed The Brut or the Chronicles oj'England, ed. Brie, i. For its Lancastrian bias see Gransden, op. cit., ii. 74-5. John Taylor, however, regards the Brut's judgment on Thomas of Lancaster as an expression of popular opinion; Taylor, English Historical Literature, p. 113. 85 For Baker's patrons see: Chronicon Galfridi le Baker de Swynebroke, ed. E. Maunde Thompson (Oxford, 1889), pp. x-xvi; Taylor, op. cit., p. 13. 86 Chronicon Henrici Knighton vel Cnitthon Monachi Leycestrensis, ed. J.R. Lumby (Rolls Series, 1889-95, 2 vols). See: V.H. Galbraith, 'The chronicle of Henry Knighton', Fritz Saxl 1890-1948. A volume of Memorial Essays from his friends in England., ed. D.J. Gordon (London, 1957), p. 136; Gransden, op. cit., ii. 178-81; Taylor, op. cit., pp. 13, 18. 87 For Adam of Usk's career and patrons see: Chronicon Adae de Usk de A.D. 1377-1421, ed., with an English translation, E. Maude Thompson (2nd edn., London, 1904), pp. xixxix; Taylor op. cit., pp. 11-12. For the bias in his chronicle see Gransden, op. cit., ii. 183 et seqq. 88 See The Westminster Chronicle 1381-1394, ed., with an English translation, L.C. Hector and B.F. Harvey (Oxford Medieval Texts, 1982), pp. Ixxiv-lxxv, for a different interpretation. But cf. Nottingham Medieval Studies, xxviii (1984), pp. 102-3.
The Chronicles of Medieval England and Scotland
217
triumphs are duly celebrated and national disasters bewailed. Such patriotic manifestations were not necessarily the result of a chronicler's spontaneous feelings. If he was an 'official' chronicler (as probably the author of the Alfredian Anglo-Saxon Chronicle and Robert of Avesbury were), or was influenced by government propaganda (a subject to be discussed in the next section), his patriotic tone might reflect the government's point of view rather than his own. On the other hand, the opinions and sentiments of chronicler and government might coincide. In any case, only rarely can patriotism have been strong enough to provide the initial motive for composition. Nevertheless, in some circumstances it must have encouraged chroniclers to write. A marked characteristic of the chronicle tradition in England is the propensity for criticising king and government. To suppose that such criticism necessarily reveals a chronicler's independence of mind would be an over-simplification. As has been seen, some chroniclers were influenced by patrons, and a patron might be a magnate in opposition to the king. As will be shown (in the next section) some of the chroniclers writing about Richard II's reign succumbed to Lancastrian propaganda which was spread under Henry IV to discredit Richard. These reservations aside, critical passages in general suggest that on occasion a chronicler's patriotism could override both his obsession with his own locality and his loyalty to the king. Then his criticisms were the result of genuine concern for the English people, a concern which made him take a detached view. Good examples of this phenomenon occurred in the late Anglo-Saxon period and in the thirteenth century. A sense of national identity distinct from the monarchy appears in the eleventh-century annals of the Anglo-Saxon Chronicle. Having noted, s.a. 1009, that King Ethelred allowed the large fleet, which he had built and assembled at Sandwich, to disperse, the chronicler comments: '[thus] the toil of all the nation [came] lightly to naught; and no better than this was the victory which all the English people had expected.'89 The threat of civil war under Edward the Confessor again aroused the chronicler's patriotism.90 When the host of Earl Godwin and his sons faced the king's forces in 1051, he writes: 'some thought it would be a great piece of folly if they joined battle, for in the two hosts there was most of what was noblest in England.' Similarly, he states uner 1052, on recording Godwin's return from Flanders with a strong force, that 'it was hateful to almost all [of his soldiers] to fight against men of their own race, for there was little else that was worth anything apart from Englishmen on either side.' This incipient nationalism was accompanied by xenophobia. King Edgar, the chronicler alleges under 959, 'indulged excessively in one bad 89 90
Anglo-Saxon Chronicle, ed. Whitelock, p. 89. Ibid., pp. 118, 123-4.
218
Legends, Traditions and History in Medieval England
practice: he loved evil foreign customs and was too persistent in bringing heathen manners into this land; and he attracted hither foreigners and enticed harmful people to this country.'91 On the death of Hardacnut, the chronicler welcomes the restoration of the English dynasty: 'all the people then received Edward the Confessor as was his natural right'.92 But later it deplores the favour Edward showed to 'Frenchmen', who 'promoted injustices and passed unjust judgment and [gave] bad counsel in this country', and caused discord between the king and Earl Godwin.93 Patriotism is, therefore, a characteristic of the Anglo-Saxon Chronicle. It may have been one reason why a version survived the Conquest by almost a century; it survived at Peterborough, in a pocket of English resistance. The thirteenth-century chroniclers almost without exception supported the barons against King John and Henry III. They expended some of their most eloquent prose on the Barons' War; the barons, in their eyes, were acting for the benefit of the realm, upholding the ancient laws and customs of England and the liberty of the Church, in face of encroachment and abuse by the central authority.94 The St Albans' Flores Historiarum (s.a. 1259/60) claims that the baronial reforms revived 'the spirit of justice'.95 Many chroniclers appealed to the right of the community of the realm to act against bad government, and to the barons' privilege of advising the king as his 'natural counsellors'.96 As in the tenth and eleventh centuries, the chroniclers hated foreigners. They attacked the queen's Savoyard relatives because, in their opinion, they usurped the barons' rightful influence over the king and became excessively rich; the Oseney chronicler feared they would 'overthrow the native inhabitants'.97
91
Ibid, p. 75. Ibid., p. 106. 93 Ibid., pp. 124, 125-6. 94 See Gransden, Historical Writing, [i], Chapter IS passim (esp. pp. 407, 414-17, 421-3 and nn. for further references). For the extreme pro-baronial attitude of a chronicle from Tewkesbury abbey see above p. 216 and note 83. For Matthew Paris's partisanship see: Richard Vaughan, Matthew Paris (Cambridge, 1958), p. 137-43; V.H. Galbraith, Roger Wendover and Matthew Paris (Glasgow, 1944, repr. 1970 and in idem, Kings and Chroniclers), p. 20; Gransden, op. cit., [i]. 368 et seqq. 95 Flores Historiarum, ed. Luard, ii. 439. 96 For references, variously expressed, to the community of the realm etc., see: Matthew Paris, Chronica Majora, ed. H.R. Luard (Rolls Series, 1872-83, 7 vols), iii. 495, v. 21; Flores Historarium, ed. Luard, ii. 426-7; Ann. Mon., ed. Luard, i. 471. For reference to the king's reliance on bad counsellors, neglect of the barons' advice, etc., see: Matthew Paris, Chron. Maj., iii, 395; Flores Historiarum, ed. Luard, ii. 464,466,470; Ann. Mon., ed. Luard, ii. 351. 97 Ann. Mon., ed. Luard, iv. 150-1. For other anti-alien passages see: Matthew Paris, Chron. Maj., ed. Luard, v. 229, 316-17; Flores Historiarum, ed. Luard, ii. 463, 479; Ann. Mon., ed. Luard, ii, 349, 355. 92
The Chronicles of Medieval England and Scotland
219
5. The Government's Use of, and Influence on, Chronicles The scarcity of official chronicles in medieval England could be understood to mean that in general the kings did not consider chronicles very useful. This would be a false conclusion. Although the kings commissioned few themselves, they recognised the value of other people's chronicles, as sources of information, records and potential vehicles of propaganda. It was pointed out above that in the thirteenth century the growth of monastic archives had outstripped the ability to organise them.98 The consequent problem of retrieving information from archives in a monastery was but a microcosm of those suffered by the central government. Edward I gave spectacular recognition to the value of chronicles, which stored information in chronological order, in 1291. When he claimed the right to try the case of the disputed succession to the Scottish throne, he turned for evidence of his overlordship not only to the public records but also to the monks, with a request that they consult their chronicles for precedents.' Their findings helped furnish the historical proofs of the English king's overlordship which were cited during the trial.100 Further examples of royal appeals to chronicle evidence survive from the fourteenth century. Such an appeal was made some time early in the 1340s, though no details are known about it: a letter survives from the abbot of Nutley to the Chancellor saying that he cannot provide a transcript of a chronicle because it is on loan to the dean of Wells.101 Another instance is known to have occurred in 1352, but again there are no details: Edward III summoned Ranulph Higden to come 'with all your chronicles and those in your charge to speak and treat with the council concerning matters to be explained to you on our behalf.102 In 1399 Henry Bolingbroke used chronicles to justify Richard IPs deposition, and in 1400, as Henry IV, to substantiate his claim to overlordship of Scotland.103 98
Above p. 210-11. See J. Bain, Calendar of Documents relating to Scotland (Edinburgh, 1881-8, 4 vols), ii. 122, nos 503, 504; F. Palgrave, Documents and Records illustrating the History of Scotland (Record Commision, 1837), pp. xcvi-xcvii and n. Cf. E.L.G. Stones, 'The appeal to history in Anglo-Scottish relations between 1291 and 1401', Archives, ix (1969), 11-21 passim. Cf. below p. 232, 237. 100 E.L.G. Stones and G.G. Simpson, Edward I and the Throne of Scotland 1290-1296: An Edition of the Record Sources for the Great Cause (Oxford, 1978, 2 vols), i. 137-62, ii. 296-309. 101 PRO, Ancient Correspondence (S.C.I), vol. xxxix, no. 161 (c. 1341-3). I owe this reference to Professor M.C.E.Jones. 102 J.G. Edwards, 'Ranulf, monk of Chester', English Historical Review, xlvii (1932), 94. 103 See Chronicon Adae de Usk, ed. Thompson, pp. 29-31, and Stones, 'The appeal to history in Anglo-Scottish relations', pp. 80-3. 99
220
Legends, Traditions and History in Medieval England
Kings appreciated that chronicles, besides providing useful information, were suitable places to preserve copies of documents of propagandist value. In 1291 Edward I sent copies of the letters of submission to his authority of the competitors for the Scottish throne,104 and in 1399 the new Lancastrian government distributed copies of the official 'record and process' of Richard IPs deposition to important monasteries for inclusion in their chronicles.105 Since on occasion chronicles attracted the attention of the central government, the question arises to what extent their authors enjoyed freedom of expression. The fact that many severely criticised king and government suggest that they wrote what they liked. Presumably they might feel particularly confident if they lived in a remote place, beyond the reach of retribution. An even stronger safeguard would have been a sympathetic, protective audience. This most monastic chroniclers had, since they wrote primarily for their own communities and also perhaps for those of neighbouring religious houses. However, the monks themselves were apparently not the only possible readers of monastic chronicles. It is likely that visitors to a religious house might read, or at least learn about the contents of, its chronicle. This is indicated by the fact that occasionally the author of some fairly ordinary chronicle admits that he considers it risky to express himself freely. Thus the Osney chronicle has an early note in the margin after the notice of the battle of Lewes (1264) which reads: because of the malice of the times it is not safe to tell the whole truth, so at present we will be silent concerning the names of the magnates who fled, and who were captured in the battle; and we have omitted from this history much else that was done in these days, for the sake of readers' peace, because what
104
See: the Canterbury chronicle in The Historical Works of Gervase of Canterbury, ed. Stubbs, ii. 297-9; Bury Chronicle, ed. Gransden, pp. 100-3; Bartholomaei de Cotton . . . Historia Anglicana, ed. Luard, pp. 180-2; the Peterborough chronicle in Florentii Wigorniensis . . . Chronicon ex Chronicis, ed. Thorpe, ii. 247-50 (however, the copy of the letter in the Peterborough chronicle may be a doctored version of the copy in the Bury chronicle; see below p. 237); the chronicles of Merton and Rochester in Flores Historiarum, ed. Luard, i. xvi, iii. 72 n. 1; and the chronicles of Dunstable, Waverley and Worcester in Ann. Mon., ed. Luard, ii. 411, iii. 368-9, iv. 507-9. For the letters of submission in the chronicles of St Benet of Hulme see below p. 236. 105 Seejohannis de Trokelowe . . . Chronica et Annales, ed. H.T. Riley (Rolls Series, 1866), pp. 252-86; Eulogium Historiarum, ed. F.S. Haydon (Rolls Series, 1858-63, 3 vols), iii. 382-4; Historia Vitae et Regni Ricardi Secundi, ed. G.B. Stow, jr (Pennsylvania, 1977), pp. 157-61; Chronicles of London, ed. Kingsford, pp. 16-47. Cf. M.V. Clarke, and V.H. Galbraith, 'The deposition of Richard IF, Bulletin of the John Rylands Library, xiv (1930, repr. in Fourteenth Century Studies by M.V. Clarke, ed. L.S. Sutherland and May McKisack (Oxford, 1937)), 44-6passim', H.G. Richardson, 'Richard IFs last Parliament', English Historical Review, Iii (1937), 41.
The Chronicles of Medieval England and Scotland
221
might please the king's men would displease those who favoured the barons.106
The chronicler of Furness in Lancashire also reveals unease. He notes that after the battle of Evesham (1265) people dared not speak of the miracles worked by Simon de Montfort's body for fear of the king.107 The Dictum of Kenilworth (1266), Chapter 8, prohibited such tales. Perhaps it was in response to this decree that mention of such miracles was erased from the chronicle of Bury St Edmunds.108 It is well known that two chroniclers, Matthew Paris and Thomas Walsingham, carried out drastic revision of their chronicles to reduce the anti-government element. Matthew Paris revised his chronicles to suppress or modify his abuse of the king, the queen's French relatives and other members of the 'establishment',109 and Thomas Walsingham did the same to his scurrilous attacks on John of Gaunt.110 Although William of Malmesbury's Gesta Regum and Gesta Pontificum are not strictly speaking 'chronicles', they deserve mention in this context because they provide an early example of similar revision. Late in life William modified his censure of William the Conqueror, William Rufus, and their ministers and officials.111 Fame had deprived Matthew Paris and Thomas Walsingham (as it had also William of Malmesbury) of the relative immunity which protected more obscure monastic chroniclers; it had brought them and their chronicles to the notice of important people. Such attention, facilitated by St Albans' proximity to London, must have made them circumspect, since it was necessary to keep on good terms with the great. At the same time it exposed them to the full brunt of government propaganda. Caution and the influence of propaganda were probably Walsingham's principal 106
'Et quia propter temporum maliciam non est tutum omnia vera loqui, ideo ad praesens tacemus nomina magnatum, [qui] fugerant et qui capti fuerant in bello predicto. Multa etiam alia quae sunt facta hiis diebus omittimus huic historiae interserere propter pacem legentium, quia forte quod placeret regalibus displiceret baronum fautoribus'; Ann. Mon., ed. Luard, iv. 148-9. Exactly who read, or listened to, monastic chronicles remains problematical. 107 The chronicle of Furness in Chronicles of the Reigns of Stephen, Henry II, and Richard 1, ed. Richard Hewlett (Rolls Series, 1884-9, 4 vols), ii. 548. 108 Chronicle of Bury St Edmunds, ed. Gransden, p; 33 and n. a. Cf. below p. 229 and Plate 15. 109 See: Vaughan, Matthew Paris, pp. 64-5, 117-24. 110 V.H. Galbraith, 'Thomas Walsingham and the Saint Albans chronicle, 1272-1422', English Historical Review, xlvii (1932), 21-5; The St Albans Chronicle 1406-1420, ed. idem (Oxford, 1937), p. li. 111 Willelmi Malmesbiriensis Monachi de Gestis Regum Anglorum Libri Quinque, ed. William Stubbs (Rolls Series, 1887-9, 2 vols), i. xxxiii et seqq., xlvii; Willelmi Malmesbiriensis Monachi de Gestis Pontificum Anglorum Libri Quinque, ed. N.E.S.A. Hamilton (Rolls Series, 1870), pp. xv-xvii; Hugh Farmer, 'William of Malmesbury's life and work', Journal of Ecclesiastical History, xiii (1962), 45-6.
222
Legends, Traditions and History in Medieval England
reasons for revision. He revised his chronicles after the 1399 revolution to make them conform with the Lancastrian viewpoint. Besides rehabilitating John of Gaunt, Henry IV's father, he underlined criticism of Richard II and his government (partly by extracting material from the 'record and process', which included a damning account of Richard's reign). His revision, therefore, not only improved the image of the Lancastrian dynasty, but in addition helped justify Henry's usurpation. Caution and the influence of propaganda may not have been Thomas Walsingham's only reasons for revision. Improved knowledge of public affairs and public figures could well have resulted in a more sympathetic attitude. This may have been so in Thomas Walsingham's case; it was almost certainly true in Matthew Paris's. Matthew came to know Henry III personally (he records that in 1247 and 1257 Henry gave him information for inclusion in his chronicle),112 and he composed and illustrated Saints' Lives in French verse for Queen Eleanor and for at least two noble ladies.113 It seems very likely that such acquaintance caused a genuine change of mind on Matthew's part. 6. Suggested Reasons for the Decline of the Chronicle
Why the chronicle tradition was moribund by the end of the fifteenth century is uncertain. A few possible reasons may be suggested. As has been seen, the strength of the tradition had lain in the monasteries. Most monasteries seem to have stopped producing chronicles in a routine way as part of their archives early in the fourteenth century. The Indian Summer of the monastic chronicle in Richard IPs reign and in the early fifteenth century was transient. Apparently monks no longer felt the urge to record national events. Perhaps they found other means of learning news, for example by newsletters, more satisfactory,114 or perhaps they had become more 'parochial' in their outlook. This last suggestion receives support from the number and generally high quality of the domestic histories of religious houses composed in the first half of the fifteenth 112 113
nn.
114
Chron. Maj., ed. Luard, iv. 644-5, v. 233-4, 617-18. Cf. Vaughan, Matthew Paris, p. 3. See Vaughan, op. at., pp. 170, 173, 178; Gransden, Historical Writing, [i]. 358-9 and
Newsletters were a source of information from the twelfth century onwards, and their value was fully recognised by chroniclers, who used them for information. See, with further references, E.L.G. Stones, 'The surrender of King John of Scotland to Edward I in 1296: some new evidence', Bulletin of the Institute of Historical Research, xlviii (1975), 103 and nn. 3, 4; A.E. Prince, 'A letter of Edward the Black Prince describing the Battle of Najera', English Historical Review, xli (1926), 417; C.A.J. Armstrong, 'Some examples of the distribution and speed of news in England at the time of the Wars of the Roses', Studies in Medieval History presented to F.M. Powicke, ed. R.W. Hunt, W.A. Pantin and R.W. Southern (Oxford, 1948, repr. 1969), p. 432; Gransden, Historical Writing, ii (see index under 'newsletters'); Taylor, English Historical Literature, pp. 140, 229, 270, 320.
The Chronicles of Medieval England and Scotland
223
century.115 One object of a 'house' history was to satisfy the monks' curiosity about their monasteries' origins and history. Another purpose was to improve esprit de corps by dwelling on (and often exaggerating) an individual monastery's antiquity and glorious past. This aspect of a 'house' history was also useful to the monastery in its external relations. It increased its prestige and recorded its rights and privileges, which strengthened its defences in an increasingly less sympathetic environment. These histories were both a manifestation of, and an incentive to, the growth of monastic localism. Another external circumstance may have discouraged the writing of chronicles. As the fifteenth century progressed writers' desire to keep or court the king's favour, or at least not to risk his displeasure, grew. It is known that such feelings influenced authors of histories which do not fall within our definition of 'chronicles'. This is proved by the fact that a number were turncoats: John Hardyng, John Capgrave and William Worcester switched sides from the Yorkist to the Lancastrian 'party', and John Rous, having supported the Yorkists, became an ardent partisan of Henry Tudor.116 Clearly these writers (like Sir John Fortescue)117 considered it expedient to support whichever party was in power. Monastic chroniclers had shown themselves vulnerable to government pressure from Edward I's reign onwards; perhaps during the Wars of the Roses and thereafter, in those increasingly difficult and insecure times, the religious preferred not to risk recording national events. The London chronicles flourished until towards the end of the fifteenth century. Their authors were in a different position from the monks. Since they wrote under the patronage of a strong, close-knit community, the London oligarchy, they had a comparatively safe environment. They consistently present the merchants' political (pro-Yorkist) viewpoint. Although the last of the actual London chronicles came to an end in the early sixteenth century, their influence continued. As has been seen, the chronicle of the Grey Friars of Newgate, which continued to Queen Mary's accession, is more or less indistinguishable from a London chronicle, and John Stow's Chronicles of England not only used the London chronicles as sources, but also, when they ended, continued to his own day with additional annals. Nevertheless, the tradition of adding annals year by year to a chronicle was dying even in London. Its demise was probably mainly owing to the introduction of printing, a decline in demand for news in the form of annals, and a change in historiographical taste. William Caxton printed the Brut in 1480 and the Polychronicon in 1482, and by the end of the 115
See Gransden, Historical Writing, ii, Chapters 12 and 13 passim. See Gransden, 'Politics and historiography during the Wars of the Roses', pp. 133-6. 117 See P. E. Gill, 'Politics and propaganda in fifteenth century England: the polemical writings of Sir John Fortescue', Speculum, xlvi (1971), 333-47. 116
224
Legends, Traditions and History in Medieval England
sixteenth century several of the most important chronicles were in print.118 The multiplication of copies meant that a chronicle once in print was easier to borrow and cheaper to buy. At the same time the text lost its 'organic5 quality: it was now fixed, not fluid. In addition, the demand for news in annalistic form probably declined. Such a 'news-medium' was slow and tended to be sketchy. Londoners can have had little difficulty in learning about current events; the City was the centre of national life, and a hub of news, from royal proclamations, public ceremonies and celebrations, verbal exchange and the like. Meanwhile the chronicle mode had become old-fashioned. Under the Tudors the factual record of events characteristic of the medieval chronicle lost favour; taste turned to histories, which, though still in chronological order, had a strong theme (usually in support of the Tudor dynasty).119 7. The Composition of the 'Contemporaneous' Sections of Chronicles The author wrote his book and then, like William of Malmesbury, invited his readers to send him, or like Matthew Paris or Thomas Walsingham, himself loaded the margins with, notes suggested by further reading and reflection. This process occupied many years and at any stage a copy might be bespoken by another religious house or a lay patron. In this way endless recensions arose, and not until the work became a classic, like Bede's Ecclesiastical History, was the text in any way fixed.120
V.H. Galbraith wrote this passage in 1959 to introduce his exposition of the evolution of Higden's Polychronicon. Although the latter is an encyclopaedia rather than a chronicle by strict definition, Galbraith's generalisation is certainly appropriate to the composition of a typical monastic chronicle except that nearly all, far from being potential 'classics', were second- or third-rate pieces of work. In these cases it was not 'classical' status which fixed their texts, but printing. As long as a chronicle remained in manuscript it had an 'organic' existence; only if and when printed did its text become static — but even then only barring new editions. The early part of a chronicle, which begins well before the author's own time, for example at the Creation, Incarnation or Norman Conquest, was, of course, a compilation from usually well-known authorities. Later, when the author wrote of the period within living memory, he used in addition oral information. Editors and other scholars have in general taken pains to 118
For the first appearance of chronicles in print see FJ. Levy, Tudor Historical Thought (Huntington Library, San Marino, California, 1967), pp. 120, 133-4. 119 For historical writing in the Tudor period see Levy, op. cit., and May McKisack, Medieval History in the Tudor Age (Oxford, 1971). 120 V.H. Galbraith, 'An Autograph MS of Ranulph Higden's Polychronicon The Huntington Library Quarterly, xxiii (1959-60), p. 3.
The Chronicles of Medieval England and Scotland
225
identify chroniclers' sources. Most recently L.C. Hector and Barbara Harvey in their edition of the Westminster chronicle have given a thorough analysis of that chronicle's sources, and D.E.R. Watt has done the same in his edition for the last two Books of the Scotichronicon.121 Such analyses include the specific problem of the relationship of one monastic chronicle with others composed at about the same time. This interrelationship of chronicles has been the subject of a few specialist studies: for example, N. Denholm-Young tried to unravel the intricate relationship of the Winchester, Hyde, Waverley and Worcester chronicles.122 A chronicler's greatest problem was how to compose the contemporary part of his work. Perhaps he was in touch with the chronicler of a nearby religious house. If so, he might be sent intalments about recent events by that chronicler, as the latter added them to his own work. He was, therefore, borrowing from a chronicle which itself was no more than a draft, still in a very unfixed state. Apart from such a possible chronicle source, a chronicler had little or no control over what or when news reached him, and had to impose chronological order and narrative coherence on a miscellaneous collection of items arriving sporadically, sometimes after he had composed the relevant annal. Galbraith did much to illuminate the composition of the contemporaneous section (13921422) of the St Albans chronicle, when Thomas Walsingham was writing within two or three years of the events which he recorded. Galbraith described the process of composition: it was 'one of ceaseless experiment and change'; 'Closely connected with this fluid character of the text, inevitable in an age when there was nothing equivalent to the publication of a book, is the annalistic nature of Walsingham's . . . chronicle'; 'Such piecemeal composition inevitably involved a good deal of recasting from time to time on grounds of style, or because fresh information came to hand, or because the writer's feelings changed.'123 Apart from Galbraith, scholars do not seem to have appreciated the problem of how the contemporary sections of chronicles were compiled. But the St Albans chronicle was a monastic chronicle par excellence. Matthew Paris and Thomas Walsingham wrote with a degree of detail and eloquence unmatched by other monastic chroniclers, and the number of surviving copies of their works is exceptionally great. It is the intention in what follows to examine the methods apparent in more typical 121
The Westminster Chronicle, 1381-1394, ed. Hector Harvey, pp. xliii-liv. For Professor Watt's edition of the Scotichronicon see below, p. 226 n. 133. 122 The chronicle of St Swithun's, Winchester, was used by the chroniclers of Hyde, Waverley and Worcester; see N. Denholm-Young, 'The Winchester-Hyde Chronicle', EHR, xliv (1934), 85-93 (repr. idem, Collected Papers on Medieval Subjects (Oxford, 1946, new edn. Cardiff, 1969), pp. 86-95. 123 The St. Albans Chronicle, 1406-1420, ed. Galbraith, p. Ixiii.
226
Legends, Traditions and History in Medieval England
monastic chronicles, of how their authors wrote contemporaneously. The related problem of the dissemination of a chronicle, from its place of origin to other religious houses, will also be considered. The argument will be centred on the chronicle of Bury St Edmunds,124 a normal monastic chronicle and a product of the thirteenth century when the chronicle genre had its heyday (as explained above, pp. 205-6, 210-11). In its final form the Bury chronicle covered the period from the Creation to 1301. Its first recension, to 1265, was by John de Taxter, a monk of Bury. Another (anonymous) Bury monk wrote a second recension. He revised Taxter's chronicle and continued to 1296. A second continuator (also anonymous) then added annals to 1301. The Bury chronicle's strictly annalistic structure and the nature of its contents resemble those of many other monastic chronicles. For purposes of comparison a selection of these will be referred to below: the chronicle of John of Worcester (Creation to c. 1140);125 the Winchcombe 'Annals' (1182-1232);126 the chronicle of Melrose (Roxburghshire, 731-1275);127 the chronicle of St Swithun's, Winchester (519-1277);128 the Norwich chronicle (Incarnation to 1291);129 the chronicle of St Benet of Hulme (Norfolk), which goes under the name of John of Oxenedes' (Creation to 1292); 13° The Dunstable 'Annals' (Incarnation to 1297);131 and the Westminster chronicle (1381-94);132 Walter Bower's Scotichronicon (Noah to 1437)133 will also be mentioned, although it is not a typical monastic chronicle. (It has many features of a history in literary form, but nevertheless has a fair number of annalistic entries which are of use in the 124
The Chronicle of Bury St Edmunds 1212-1301, ed. Gransden. The Chronicle of John of Worcester, 1118-1140, ed. Weaver. 126 'The Winchcombe annals 1049-1181', ed. R.R. Darlington, in A Medieval Miscellany for Doris Mary Stenton, ed. P.M. Barnes and F.C. Slade (Publication of the Pipe Roll Society, new series, xxxvi, 1960). See also Eric John, 'A critical study of the sources of the annals of Winchcombe, Faustina B I, ff.21-29b (1182-1232)', M.A. thesis, University of Manachester (1949). 127 The Chronicle of Melrose, a facsimile edition from Cotton MS. Faustina B ix, ed. A.O. Anderson and M.O. Anderson (London, 1936). 128 printeci m Annales Monastici, ed. Luard, ii. 3-125. 129 This Norwich chronicle was incorporated by Bartholomew Cotton in his chronicle; Bartholomaei de Cotton, Historia Anglicana, ed. Luard, pp. 47-182. Gf. ibid., pp. xxiii-xxiv. 130 Chronica Johannis de Oxenedes, ed. Ellis. 131 prmted in Annales Monastici, ed. Luard, iii. 3-408. 132 For the standard edition see above p. 216 n. 88. 133 To 1383 Bower incorporated most of John Fordun's Chronicon Gentis Scotorum. Hi own chronicle is virtually a continuation of Fordun's. The whole is ed. Walter Goodall (Edinburgh, 1759, 2 vols). For an up-to-date edition of four of the books see Scotichronicon by Walter Bower in Latin and English, general editor D.E.R. Watt, ii, Books III and IV, ed. John and Winifred MacQueen (Aberdeen, 1989), viii, Books XV and XVI, ed. D.E.R. Watt (Aberdeen, 1987). The remaining books in this edition are planned to appear within the next decade. 125
The Chronicles of Medieval England and Scotland
227
study of how chronicles were composed.) In general, the length of individual annals in a chronicle increases as the chronicler nears his own day. Starting with brief, piecemeal entries, the prose tends to become fluent when it concerns events more or less contemporary with the author. The Bury chronicle conforms to this model. From 1264 Taxter was writing fairly soon after the events he recorded, and gives a detailed account of the Barons' War. His first continuator wrote more or less contemporaneously at least from the mid1280s, and probably did so throughout. The near contemporaneity of a substantial part of the Bury chronicle make it a suitable focus of attention for our study. The Bury chronicle has in addition other advantages which few other chronicles possess in combination. Of those listed above, excluding Bower because he is untypical, only John of Worcester has them all, and he does not belong to the thirteenth century, from which most of our examples are drawn. Part of the Bury chronicle survives in an early draft. It is true that the later annals of quite a few of the chronicles listed (John of Worcester, and the Winchcombe, Melrose, Dunstable and Westminster chronicles) survive in early drafts,134 but, with the exception of John of Worcester and the Dunstable chronicle, there is only one manuscript of each. The Bury chronicle, on the other hand, survives in a number of manuscripts and, moreover, was plundered by the chroniclers of quite a few other religious houses. None of the chronicles listed enjoyed comparable popularity except the Winchester chronicle and, again, John of Worcester.135 The best text of Taxter's chronicle is British Library Cotton MS Julius A 1, ff. 3-43v.136 It is a fair, but early, copy in one hand throughout. There is one other copy of Taxter, of the last half of the fourteenth century, College of Arms Arundel MS 6, ff. 109-24. Although not copied directly from Julius A 1, it is a descendant of it.137 The best text of the second recension of the Bury chronicle, and the only one of the second continuation, is College of Arms Arundel MS 30, ff. 97-205. To 1285 it is a fair copy, with only one change of hand, in the middle of the annal for 1268. From 1270 there are signs of early revision (see e.g. below Plate 14). Most significantly, from 1286 to 1296 there are many changes of ink and apparently of hand. Clearly this part of the manuscript is a draft, not a fair 134 por descriptions of the manuscripts of these chronicles respectively see: Weaver, op. cit., pp. 4-10; Darlington, loc. cit., pp. 111-12, and John, loc. cit., pp. 1-2; Melrose, ut. cit., pp. x et seqq; Cheney, 'The making of the Dunstable Annals, A.D. 1233 to 1242', pp. 210-13, 219; Westminster Chronicle, ed. Hector and Harvey, pp. xiv-xxi. 135 See above p. 203 and n. 23. 136 por Descriptions of the manuscripts of the Bury chronicle see Bury Chronicle, ed. Gransden, pp. xxxv-xliv, and for more detail idem, A Critical Edition of the Bury St Edmund's Chronicle in Arundel MS 30 (College of Arms), Ph.D. thesis, University of London (1956), pp. 21-68. 137 But see below p. 229.
14 The Chronicle of Bury St Edmunds; the first continuation (London, College of Arms, MS Arundel 30, fo. 167v). The end of the annal for 1279 and the beginning of that for 1280. An early reviser made corrections to the text in plummet in the margin, and these were incorporated in the text in ink. A passage (a notice of an eclipse, and a notice of William de Hoo's succession to the office of sacrist of Bury St Edmunds) is marked vacat in the margin in plummet.
15 The Chronicle of Bury St Edmunds; the first recension by John Taxter (London, British Library, MS Cotton, Julius A 1, fo. 43v). The end of the last annal, for 1265, with the passage alleging that Simon de Montfort's corpse worked miracles erased.
16 The Chronicle of Bury St Edmunds (London, College of Arms, MS Arundel 30, fo. 150). Annals from 1260 to 1263, with spaces between the annals for 1260 and 1262. A note of the birth of John de Hastings to Joanna, wife of Henry de Hastings, has been added at the end of the annal for 1262 (the Hastings were hereditary stewards of St Edmund's Liberty).
17 The Chronicle of Bury St Edmunds (Bury St Edmunds, Moyses Hall, MS, fo. 103v). The end of the annal for 1282. A space has been left after the notice of the birth of a daughter to Queen Eleanor for her name. The text in the College of Arms calls her Elizabeth, but Bartholomew Cotton has 'Walkiniana'. 'Oxenedes' has 'Elysabeth' written on an erasure.
18 The Chronicle of Melrose Abbey (London, British Library, MS Cotton, Faustina B IX, fo. 28v). Annals from 1208 to 1210, with spaces between the annals. A note recording the death of Richard abbot of Coupar-Angus and the succession of Abbot Alexander has been added at the end of the annal for 1209.
19 Scotichronicon by Walter Bower (Cambridge, Corpus Christ! College, MS 171, fo. 343). Book 16, the end of Chapter 15 and the beginning of Chapter 16, 1428-30. A space of about nine lines has been left after Chapter 15. The last sentence, 'Reliqui ..donantur', was added in a similar, but smaller hand and in darker ink. The original scribe made a marginal addition.
20 The Chronicle of Bury St Edmunds, the first continuation (London, College of Arms, MS Arundel 30, fo. 179). Part of the annal for 1291. By the copy of Edward Fs letter to the abbot and convent of Bury St Edmunds ordering them to copy into their chronicle the letters of submission to his judgement of the competitors for the Scottish throne, a scribe has written Sancti Petri de Burg' in plummet; clearly this was in preparation for a copy being made for Peterborough. Edward Fs and the competitors' letters must have been copied into the chronicle before the historical matter introducing them; since the scribe did not leave enough space for the latter, the final words were relegated to the margin.
21 The Chronicle of Bury St Edmunds, the first continuation (London, College of Arms, MS Arundel 30, fo. 190). Part of the (penultimate) annal for 1295. The scribe left a space after the notice of the dismissal of William March from the office of Treasurer. An early reviser recorded the succession of Walter Langton in plummet in the margin, and inserted it in ink in the space. Three passages are marked vacat. The first concerns the treason of Thomas de Turbeville; the second records March's disgrace and Langton's succession as Treasurer; and the third names the tax collectors in Suffolk.
22 The Westminster Chronicle (Cambridge, Corpus Christi College, MS 197A, p. 197). Annal for 1390, including a notice of the consecration of Brother Alexander [Bache] as bishop of St Asaph. The scribe left a blank space for the surname which was later inserted.
23 The Chronicle of Bury St Edmunds; the first recension by John Taxter (London, British Library, MS Cotton, Julius A 1, fo. 23v). A marginal addition (extracted from Ralph de Diceto's Abbreviationes Chronicorum, s.a. 773) to the annal for 772.
24 The Chronicle of Winchcombe Priory (London, British Library, MS Cotton, Faustina B I, fo. 13). The annals from 1074 to 1088. The chronicler wrote an annal, abridged from a Tewkesbury chronicle, after the year date, and then added material in blocks around it.
25 The Chronicle of Bury St Edmunds (London, College of Arms, MS Arundel 30, fo. 188v). Part of the annal for 1294. The succession of William de Hothum to the archbishopric of Dublin has been added in the margin.
26 The Chronicle of St Benet of Hulme commonly known as 'John de Oxenedes' (London, British Library, MS Cotton, Nero D.II, fo. 235v (new foliation)). Part of the annal for 1282. The name of Queen Eleanor's new born baby, 'Elysabeth', is written by a different scribe on an erasure.
27 The St Albans copy of Ralph Diceto's Tmagines Historiarum (London, British Library, MS Royal 13 E VI, fo. 134v). The annal for 1198, 'edited' for the scribe making an abbreviated version for Dunstable priory; the passages to be copied are marked with a diagonal stroke in the left-hand margin in reddish brown pencil. Compare the Dunstable chronicle, s.a. 1198 (Luard 1864-9:3.27).
228
Legends, Traditions and History in Medieval England
copy. Another manuscript of the second recension, also written at Bury, is now in the Moyses Hall Museum at Bury St Edmunds. It is in one neat book hand, apparently the same as that which wrote the annals from 1268 to 1285 in Arundel 30, but only goes down to 1283. From 1269 its variants from the text in Arundel 30 become important.138 It is a very early copy, made shortly after the date of its last annal.139 To these manuscripts must be added the chronicles of the principal borrowers from the Bury chronicle. In fact it was used by a number of chroniclers in various religious houses, mostly in East Anglia and adjoining regions, in the thirteenth and fourteenth centuries.140 However, only those chronicles which reproduce substantial parts of the text, that is the chronicles of Norwich, St Benet of Hulme and Peterborough, are helpful for the present inquiry. The Norwich chronicle copies Taxter for the annals from 1258 to 1263, and the second recension from 1279 to 1284.141 The Hulme chronicle uses the second recension for the annals from 1020 to 1169, and 1258 to 1292.142 The Peterborough chronicle also copies the second recension (probably Arundel 30 itself), from 1152 to 1295.143 Light is thrown on the Bury chronicle's evolution and dissemination by various kinds of evidence: study of the contents of the chronicle, and of the early manuscripts; collation of the various manuscripts of it,144 and of the relevant sections in the partly derivative chronicles; and study of the derivatives themselves. As stated above, Taxter wrote fairly contemporaneously from 1264. This conclusion is based partly on the immediacy of the narrative. Moreover, it conforms with evidence provided by Taxter himself. He states at the end of the annal for 1244: 'hoc anno scriptor presentis voluminis habitum suscepit monachicum dictus I. de Taxter5.145 Therefore, if he wrote in the 1260s, he would have been a monk for upwards of twenty years, which is perfectly feasible. This 138
Gransden, A Critical Edition, p. 38. 139 The annals for 1267 and 1268 are also preserved in the early fourteenth-century Bury customary, BL Harley MS 3977, ff. 55v-9. This text has small variants from both Arundel 30 and the Moyses Hall MS, but will not be considered here. 140 See above pp. 204-5. 141 See Cotton, ed. Luard, pp. lii-liv, Iv-lviii. 142 Another chronicle mostly compiled at Hulme, the so-called 'Cronica Buriensis', also contains citations from the Bury chronicle, probably from the version used by 'Oxendes'. The 'Cronica Buriensis' is a history of St Edmund's abbey from 1020 to 1346. It is mainly a compilation and there are changes of authorship at least in 1327 and 1335. Certainly to 1327 and probably to 1335 it was compiled at Hulme, but thereafter was continued at Bury. See below pp. 239-44. 143 p rm t ec j in Florentii Wigorniensis Monachi Chronicon ex Chronicis, ed. Thorpe, ii. 136-279. Cf. Bury Chron., pp. xvii, xliv. 144 A full collation of the texts is in the footnotes in Gransden, A Critical Edition. Selected variant readings of the texts, from 1212 onwards, are in the footnotes in Bury Chron., ed. Gransden. 145 Bury Chron., p. 3n.b.
The Chronicles of Medieval England and Scotland
229
evidence from the chronicle's subject matter is corroborated by the earliest manuscript, Julius A 1. The handwriting could well belong to the 1260s. In addition, an erasure suggests the possibility that it was written before the Dictum of Kenilworth, 30 October 1266. The statement under 1265 that Simon de Montfort's body worked miracles has been carefully scraped out146 (see Plate 15). An erasure can, of course, be made at any time, but possibly this one was a result of the Dictum: Chapter 8 forbade men to speak of Simon's alleged miracles.147 If this were so, the copy of Taxter from which Arundel 6 descends and the earliest draft of the second recension of the chronicle must already have been made, since both have the original reading. Taxter's continuator started by revising his exemplar. His main alterations were to insert a number of passages about classical history to the early part, and to omit the entry under 1244 that Taxter took the habit.148 He also made one small revision to the contemporaneous section: to the notice under 1265 about de Montfort's miracles he added 'as many say' ('ut plurimis dicitur').149 Perhaps this was in response to the changed political climate after Henry Ill's resumption of power. His continuation, as mentioned above, was probably composed fairly close to events at least from the early 1280s. Explicit evidence of the near contemporaneity of one annal, that for 1287, occurs in the annal itself. It describes how in that year St Edmunds abbey lost its manors of Semer and Groton in a judicial duel. Lamenting this misfortune, the chronicler observes that there was no hope of recovering them.150 But we know from other records that negotiations for their recovery were crowned with success in 1290.151 The manuscripts indicate that from the 1260s the author of the second recension took into account the need to allow for the inclusion of late news. The scribe of Arundel 30 left three lines blank after the annal for 1260 and two after that for 1262 (Plate 16). From 1265 he regularly left such spaces after each annal. Similar, but longer, spaces, comprising up to nine blank lines, occur after each annal in the Moyses Hall MS from 1265 onwards (Plate 17). In one instance both Arundel 30 and the Moyses Hall MS have a space near the beginning of an annal, after the first two brief entries of the annal for 1279.152
146
/foW.,p.33.Cf.abovep.221. R.E. Treharne, Documents of the Baronial Movement of Reform and Rebellion 1258-1267 (Oxford Medeival Texts, 1973), p. 322. 148 See Bury Chron., p. xix. 149 Ibid., p. 33 and n.a. 150 Ibid., p. 89. 151 CCR, 1288-96, p. 126. See V.H. Galbraith, 'The death of a champion', in Studies in Medieval History presented to P.M. Powicke, ed. Hunt, Pantin and Southern (repr. in Galbraith, Kings and Chroniclers), p. 292. 152 Bury Chron., p. 67 n.c. 147
230
Legends, Traditions and History in Medieval England
These spaces were intended for additions. In fact, an addition has been inserted in Arundel 30, in another hand than that of the scribe of the rest of the text, at the end of the annal for 1262: it records the birth to Joanna, wife of Henry de Hastings, of a son, John, at Allesley (in Warwickshire) on 6 May153 (Plate 16). This was a matter of particular interest to a Bury chronicler, because the Hastings were hereditary stewards of St Edmunds's Liberty. John succeeded his father on the latter's death in 1268/9. It is tempting to conclude that the addition was made at about that time, when information about John would have been especially topical. In any case, it may well have been made before the Moyses Hall MS was copied, since the entry is in the text in the hand of the original scribe in that manuscript (and in the Peterborough version of the second recension. It is not in Taxters's chronicle).154 A number of other chronicles have blank lines between annals for additions. A good example is the thirteenth-century chronicle of Melrose.155 Only one manuscript survives of the Melrose chronicle, which was probably the only one of it ever written. Therefore, the many changes of hand, blank spaces, and insertions in them and in the margins reflect the actual process of composition (see Plate 18). Another example of the practice of leaving spaces is provided by the very different Scotichronicon composed by Walter Bower, prior of Inchcolm (1417-49). Bower borrowed from John Fordun's chronicle until that ends in 1483. Thereafter his chronicle derives from no other chronicle, and from the early fifteenth century until 1438, when it ends, it was increasingly based on Bower's own experiences and observations.156 It is a much more sophisticated work than the Melrose chronicle, and survives in a number of copies. It is not in all respects a typical monastic chronicle. One untypical feature is that the chronology of the annals tends to be confused, and its main divisions are into books (for the reigns of the kings of Scotland), and chapters each approximating to an annal. The earliest manuscript (Corpus Christi College, Cambridge, MS 171) was copied under Bower's supervision, and substantial spaces were left between chapters for additions. Entries were inserted into two of these at an early date157 (Plate 19). There is evidence indicating that the Moyses Hall MS of the Bury 153 That this was the birthday of John de Hastings, 1st baron Hastings 1290-1312/13, is confirmed by an inquistion 'post mortem'; Cal. Inq. Post Mortem, Henry III, v. 229, no. 719. For the Hastings' hereditary stewardship of St Edmund's Liberty see L J. Redstone, 'The Liberty of St. Edmund', in Proceedings of the Suffolk Institute of Archaeology and Natural History, xv (1915), pp. 207-9. 154 Bury Chron., p. 89 n. d. 155
156
157
See above p. 226 and n. 127. See Professor Watt's introduction to Scotichronicon, viii, ed. Watt, pp. xvi-xx.
Ibid., pp. xiv, 238 n. b, 260 n. w.
The Chronicles of Medieval England and Scotland
231
chronicle and the Corpus MS of the Scotichronicon were each copied from exemplars which also had spaces between their respective annals or chapters, and that additions had been inserted into some of them. Collation shows that the Moyses Hall MS has three sentences, two under 1274 and one under 1277, which are not in Arundel 30.158 They record respectively that: Edward I forfeited the Limousin and Provence in the court of the king of France, and the countess of Provence was awarded them instead; the king of Spain opposed the election of the king of Germany as emperor of the Romans, with the agreement of the Genoese and the marquess of Montferrat; King Edward forfeited the castle of Fronsac (Gironde) in the king of France's court. The first of these additions occurs one sentence before the end of the annal for 1274, the second at the end, and the third is fairly near the beginning of the annal for 1277. It seems likely that at least the first two entered the chronicle as additions in a space at the end of the annal for 1274 in some now lost draft. The same may also be true of the third addition; as noted above, the annal for 1279 in Arundel 30 and the Moyses Hall MS has a space near the beginning. It is noteworthy that all three additions concern foreign affairs, about which one would expect news to arrive late. In the case of the Scotichronicon the significant feature is that a number of chapters end with short, inconsequential entries; these could well have originated as additions made in blank spaces left in the copyist's exempler.159 Thus, it can be seen that spaces between annals were intended, and sometimes used, for additions. They were particularly useful for information which came to hand after the annal for the relevant year had been composed. Spaces left elsewhere in a manuscript were similarly used. When the scribe of the Corpus MS of Bower's chronicle listed the chapter heads at the beginning of each book of the Scotichronicon, he left the verso blank. Two of these blank pages provided room for the insertion of four important letters: a letter of Pope Gregory XI and one of David II were inserted at the beginning of Book XV; and two of Henry IV of England are on the blank page at the beginning of Book XVI, with a heading indicating that they should be inserted in Chapter 30 ('Hec sequentes littere deberent situari infra sequentem librum XVI tum ad 158 Bury Chron., pp. 57 nn. dj, 63 n. a. However, these additions could have originated as marginalia. For the use of margins for additions see below pp. 233-4. It is hard to interpret the fact that an early annotator of Arundel 30, whose notes in faint plummet appear frequently in the margins, has written the incipits of each of these three additions at the appropriate places. It seems most likely that he collated the text in Arundel 30 with the Moyses Hall MS or its exemplar, and noted these additions in this way. 159 See e.g. Scotichronicon, viii, ed. Watt, pp. 80-2, 86, 124, 254, 264,266, 276,292, 296-8. The spaces in Corpus MS 171 were little used (see n. 157 above); it is only to be expected that more use would have been made of those in an earlier draft. However, Professor Watt tells me that these short entries almost certainly came from some pre-existing St Andrews' chronicle. For spaces left between annals for additions see below p. 249.
232
Legends, Traditions and History in Medieval England
XXX capitulum') — where they duly occur in the later manuscripts of the Scotichronicon.160 Whether or not these pages had been left blank on purpose for late additions is not clear. Sometimes it seems that a document might be copied in first and the explanatory narrative introducing it inserted afterwards. The Bury chronicle in Arundel 30 has the letters of submission to Edward I's judgement of the competitors for the Scottish throne, 5 June 1291, copied in a different hand from the preceding account of the dispute; the last nine words of the latter are written in the margin, obviously for want of space161 (Plate 20). Edward I himself had ordered (9 July 1291) specific religious houses, including Bury, to have these letters copied into their chronicles.162 Presumably the royal mandate was the immediate cause of their inclusion here. It would seem that the chronicler obeyed Edward's command promptly, before composing the introductory matter; he, therefore, left a space, but it proved too short. To leave a small gap in a line of text for some snippet of information not yet to hand was a common practice. The scribe of the Bury chronicle in Arundel 30 left such a gap in the annal for 1295 after an entry about the dismissal of William March from the office of Treasurer. Another early hand has filled it in with: £cui successit Walterus de Langeton'163 (Plate 21). In fact March was dismissed on 16 August 1295, and Langton appointed on the following 28 September. The Moyses Hall MS of the Bury chronicle (copied in, or shortly after, 1283) leaves a gap under 1282 for the name of the daughter who, it records, was born to Queen Eleanor at Rhuddlan (actually August, 1282)164 (Plate 17). The Westminster chronicle, which was written more or less contemporaneously probably from the late 1380s, and survives in an early copy, Corpus Christi College, Cambridge, MS 197A, has a similar gap. Under 1390 it notes the consecration by the archbishop of Canterbury on 8 May of Brother Alexander as bishop of St Asaph; a blank is left for his surname (Bache)165 (Plate 22). In the same way Bower leaves small gaps for names he does not know under 1277 and 1433.166 In one place the Westminster chronicle leaves a gap for another kind of information: it records that in 1390 Richard II created William count of Ostrevantz, a knight off the Garter and conferred 500 marks on him - but the '500' is in darker ink than the 160
Corpus Christi College, Cambridge, MS 171, 336V. Cf. Scotichronicon, viii, ed. Watt, pp. xiv, 310-13. 161 BuryChron.,p. 100 n. a. 162 See Stones and Simpson, Edward I and the Throne of Scotland, i. 65, 160, 178; ii. 120 (no. D. 46). 163 BuryChron.,p. 129 n.b. 164 Ibid., p. 77 n. a. Cf. below p. 234. 165 Westminster Chronicle., ed. Hector and Harvey, p. 434. 166 Scotichronicon, viii, ed. Watt, pp. 98 and n. j, 286 and n. a.
The Chronicles of Medieval England and Scotland
233
rest of the text, which indicates that it was inserted later.167 Margins were, of course, useful places for additions. The scribe of Taxter's chronicle in Julius A 1 made marginal additions, especially to the early part (see Plate 23). The new material was derived from standard literary sources. These marginalia are incorporated in the text of the Arundel copy of Taxter, and in the second recension of the Bury chronicle.168 Bower used the margins of the Scotichronicon in the same way for additions, but even more extensively (see Plate 19); these too were incorporated in the texts of the later copies. But perhaps the most striking example of such marginal additions is provided by the Winchcombe chronicle. The holograph manuscript, British Library Cotton MS Faustina B I, ff. 21-29b, shows that the chronicle was little more than a skeleton fleshed with copious marginalia (see Plate 24). C.R. Cheney describes the chronicler's method: 'He set the entry for each year as an island in a little sea of parchment. Then he added new islets of text, spread all round the original entry: sometimes there are as many as six for one year.'169 The sources of the Winchcombe chronicler's additions were other monastic chronicles, but margins were especially useful for the insertion of late news, both oral and documentary. Various examples of this usage could be cited, but three from the Bury chronicle in Arundel 30 illustrate it well. The above mentioned addition (p. 232) in a gap in the text, recording the succession as Treasurer of Walter de Langton in 1295, was probably made in response to a note in plummet in the margin: 'cui successit Walterus de Langeton' AngP thes'.' The other two examples both occur under 1294. They are by one annotator writing soon after the scribe of the text, in the margin in ink. The first note is next to the chronicler's entry recording that Edward I took over the religious houses and their revenues, putting his officials in charge of them and ordering a fixed allowance for each monk still resident. The annotator added: 'that is, 18d. a week per monk' ('videlicet cuilibet monacho per ebdomados xviij. d' ').170 The other marginal addition occurs by the notice that John de Sanford, archbishop of Dublin, who had been on an embassy to the king of Germany, returned to Yarmouth and died a few days later. The annotator wrote: 'Brother William de Hothum of the Order of Preachers succeeded him' ('Cui successit frater Willelmus de Hodone de ordine 167 Westminster Chronicle, p. 452. For a similar example see Scotichronicon, viii. 262 and n. e, and below Plate 19. The scribe left a marginal note incomplete at the end, not adding the number comprising 'the whole membership of Clan Cameron'. The note remained incomplete, and when subsequent scribes incorporated it into the texts of their copies of the Scotichronicon they left a small gap for the still unknown number. 168 Bury Chron., pp. xxxvii-xxxviii. 169 Cheney, 'The making of the Dunstable Annals', p. 227. 170 Bury Chron., p. 123 and n. c.
234
Legends, Traditions and History in Medieval England
predicatorum') 171 (Plate 25). Since Sanford died on 2 October 1294, and Hotham was provided to Dublin on 24 April 1296, this addition (and presumably the other one) was made after the latter date. Study of manuscripts, therefore, can throw light on a chronicle's development. Moreover, if there are more than one manuscript of a chronicle, the collation of the texts may well be revealing. Again, the Bury chronicle provides good examples. Comparison of the texts of the chronicle itself and also of it with its derivatives, the chronicles of Norwich, St Benet of Hulme and Peterborough, produces significant results. Collation of Arundel 30 with the Moyses Hall MS shows the presence in the latter of the three extra sentences under 1274 and 1277, discussed above.172 It also shows that one of them (that recording Edward I's forfeiture of the Limousin and Provence) was copied into the Hulme chronicle.173 These variants indicate that both Arundel 30 and the Moyses Hall MS descend from an earlier draft which was in process of revision. It is noticeable that from the early 1280s the Norwich, Hulme and Peterborough chronicles differ increasingly from their Bury exemplar, again indicating the latter's progressive revision. It was observed above that the Moyses Hall MS leaves a gap for the name of Queen Eleanor's daughter born in 1282.174 But Arundel 30 has 'Elizabeth'.175 The Hulme chronicle similarly has 'Elysabeth', but on an erasure, in a darker ink, and apparently not in the hand of the original scribe, though in a nearly contemporary one176 (Plate 26). This suggests that new information occasioned revision. The Norwich chronicle calls the baby 'Walkiniana'.177 It would seem from these scraps of evidence that there was some delay before a name reached the chronicler at all, and then some confusion over what it was. From 1286 onwards the Hulme chronicle diverges more and more from the surviving text of the Bury chronicle. In some instances it seems that the latter was corrected after the Hulme chronicle was written. For example, the Hulme chronicle asserts wrongly that Edward I crossed to France in about the middle of Lent, 1286;178 the Bury chronicle states that he did so after Easter (14 April), in May, which is correct.179 Again, under 1290, the Hulme chronicle wrongly names the duke of Brabant Walter;180 171 172 173 174 175 176 177 178 179 180
Ibid., p. 124 and n. b. Cf. above p.232 and n. 163 and Plate 21. Above p. 231. Oxenedes, p. 246. Cf. Bury Chron., p. 57 n. d. Above p. 232 and n. 164. Bury Chron., p. 11. BL Cotton MS Nero D II, f. 235V (new foliation); Oxenedes, p. 260. BL Cotton MS Nero V, f. 206 (new foliation); Cotton, ed. Luard, p. 162. Oxenedes, p. 267. Bury Chron., p. 86. Oxenedes, p. 276.
The Chronicles of Medieval England and Scotland
235
the Bury chronicle rightly calls him John.181 In another instance both chronicles are wrong, but disagree nonetheless: under 1286 the Hulme chronicle states that Hugh Balsham, bishop of Ely, died on 5 Ides July (11 July);182 the Bury chronicle gives 5 Ides June (9 June);183 in fact 5 Kaljuly (16 July) was the date of his death. So far we have been concerned with methods of expanding annals already composed for specific years, in order to include extra information, and with their revision. The next problem is how a chronicle was continued from year to year, in such a way that incoming material was digested and a coherent annal composed. The Winchester chronicler recommends a method in his prologue: Let it be your responsibility that a sheaf of loose leaves be added to the end of the book, on which should be noted in plummet the deaths of illustrious men and anything memorable you hear concerning the state of the kingdom. At the end of each year let he who is ordered, not just anyone who likes, record for posterity what he thinks best, entering it briefly at the end of the book.184
The Winchester chronicler explains why he insisted that 'not just anyone who likes' should write an annal: it was because he deplored any annal in 'rough, unpolished Latin'. One such might easily occur because, 'since the book is added to every year and is, therefore, composed by various people, it may fall into the hands of someone who writes barbariously.'185 In fact, most monastic chronicles seem to have been the work of one monk and of one or more continuators, each covering several years. Nor by any means were they all kept up from year to year; two, three or even more yearly intervals were apparently not uncommon. Moreover, there is no positive evidence as to how widespread the practice was of appending a schedule of blank leaves to a chronicle. No manuscript of a chronicle petrified, as it were, in process of production, with a sheaf still attached, seems to survive. However, the argument of probability, and evidence provided by collation of texts, suggest that some such 181 182 183 184
Bury Chron., p. 95. Oxmedes,p. 267. Bury Chron. p. 86.
Annales Monastici, iv. 355. See above p. 211 and n. 64. The whole passage reads: Considerantes pro multis causis in religione chronicas esse necessarias, istas vobis de vetustis rotulis neglectisque scedulis excerpsimus; et quasi de sub mensa Domini fragmenta collegimus, ne perirent. Non enim debet vestras urbanas aures offendere rudis et inculta Latinitas, qui soletis in scripturis magis sensui quam verbis incumbere, fructui potius quam foliis inhaerere. Nee mirandum, si liber annuatim augmentatur, ac per hoc a diversis compositus, in alicujus forte manus incident, qui proloquens fecerit barbarismum. Vestri itaque studii erit; ut in libro jugiter scedula dependeat, in qua cum plumbo notentur obitus illustrium virorum et aliquod de regni statu memoriale, cum audiri contigerit. In fine vero anni non quicunque voluerit, sed cui injunctum fuerit, quod verius et melius censuerit ad posteritatis notitiam transmittendum, in corpore libri succincta brevitate describat; et tune veteri scedula subtracta, nova imponatur. 185 See previous note.
236
Legends, Traditions and History in Medieval England
method was used to compile the more ambitious kind of chronicle. Obviously a chronicler had to assemble his material, documentary and other, before writing an annal; to attach it to the end of the chronicle would have been a sensible way of keeping it safe and readily accessible. The practice of appending a collection of relevant documents to an historical work was not unknown. The best surviving example is, of course, Matthew Paris's Liber Additamentorum.im But a collection of this sort was supplementary to an historical work; it was not meant to provide material for composing the narrative. However, it is likely that the copies of documents which form the last part of Thomas Elmham's Speculum Augustinianum represent a collectanea made preparatory to composition.187 Elmham planned to write accounts of the sixty-one abbots of St Augustine's, up to his own time, the early fifteenth century. But he only completed sections on the first fourteen abbots, to 806. He then gives brief particulars to 1087 and finally a collection of documents to 1191. In this context the Bury chronicle and its derivatives are again suggestive. The divergencies in their texts, which have already been mentioned, become even more marked as the years pass. It would appear that the borrowers were not using the Bury chronicle itself, nor even a proper draft, but merely notes and documents acquired from Bury. This is most vividly illustrated by an example from the Hulme chronicle: it has a Latin version of the letters of submission of the competitors for the Scottish throne to Edward Fs judgement, while the Bury chronicle has a French one; in both chronicles Edward's covering letter is addressed to the abbot and convent of St Edmunds.188 Study of individual manuscripts and collation of texts illuminates some aspects of the dissemination of a chronicle from the monastery where it was composed to other religious houses. Some 'parent' manuscripts show traces of preparation for use as an exemplar by the scribe of a derivative 186 printecj m Chronica Majora, ed. Luard, vi. The Bridlington chronicler (temp. Edward II, early Edward III) often refers readers to his Incidentia Chronicorum, which must have been a register of documents relevant to his narrative, similar to Paris's Liber Additamentorum. See Chronicles of the Reigns of Edward I and Edward II, ed. Stubbs, ii. xxiv and n. 3, 40, 53, 78 etc., and V.H. Galbraith, Historical Research in Medieval England (London, 1951, repr. in idem, Kings and Chroniclers), p. 32 n. 1. Documents were thought to add authority to a text. It was for this reason that Eadmer crowded the last part of his Historia Novorum, mainly concerning the Canterbury/York controversy, with documents. See Eadmeri Historia Novorum in Anglia, ed. Martin Rule (Rolls Series, 1884), esp. pp. 198 etseqq. Alan of Tewkesbury regarded his collection of Becket letters as of primary importance, and his Life of St Thomas as introductory to it. See Materials for the History of Thomas Becket, ed. J.C. Robertson and J.B. Sheppard (Rolls Series, 1875-85, 7 vols), ii. 300, 351, and Anne Duggan, Thomas Becket: A Textual History of his Letters (Oxford, 1980), esp. Chapter 3. 187 Historia Monasterii S. Augustini Cantuariensis, ed. Charles Hardwick (Rolls Series, 1858), pp. 347-479. 188 Oxenedes, p. 136; Bury Chron., p. 100. Cf above p. 232.
The Chronicles of Medieval England and Scotland
237
chronicle. Thus Arundel 30 of the Bury chronicle has a note in plummet in the margin by Edward Fs letter ordering the abbot and convent of St Edmunds to insert the submissions of the Scottish competitors into their chronicles: it reads 'Sancti Petri de Burg'189 (Plate 20). This must have been an instruction to a scribe making a copy of the chronicle intended for Peterborough abbey: he was to substitute an address to the abbot and convent of Peterborough for the existing one to the abbot and convent of Bury. The Peterborough version of the chronicle has the revised address.190 The annals from 1268 to 1295, the end of the first continuation, in Arundel 30, have numerous passages marked 'vacat' in the margins, in plummet.191 A number of the passages relate to Bury; others concern foreign affairs, particularly Sicily and the papacy, and Edward Fs relations with Wales. A few are notices about the weather and other natural phenomena, while some seem purely miscellaneous. The 'vacat' passages were presumably to be omitted by some copyist. In view of the fact that a sizeable proportion concern Bury, it seems safe to conclude that the copy was intended for some other house. The fact that the 'vacat' hand could well be the same as the 'Sancti Petri de Burg' one suggests the possibility that the house in question was Peterborough. If so, the resultant chronicle cannot be the Peterborough chronicle preserved in the Corpus MS, since very few of the 'vacat' passages correspond with those it omits. The copy of Ralph Diceto's Ymagines Historiarum belonging to St Albans is a convincing example of a chronicle marked for a scribe making a copy for another religious house.192 It has passages marked by small diagonal strokes in plummet in the margins (see Plate 27). These were directions for a copyist, and in this case it is known for which house the copy was intended. It was Dunstable priory. The passages marked generally correspond with those extracted for the Dunstable 'Annals'.193 Adaptation of a parent chronicle can also be detected by collation. For example, the Hulme chronicler kept the passage in the Bury chronicle recording that in 1275 the monks of Bury pulled down the ancient round of St Edmund, but added 'apud Sanctum ^Edmundum', and substituted 'creditur' for 'credimus' in the sentence 'credimus illam [capellam] fuisse 189
BuryChron.,p. 100 n./>. Ibid., p. 100 n. b. and Florentii Wigorniensis Monachi Chronicon ex Chronicis, ed. Thorpe, ii. 247. 191 The passages marked 'vacat' in Arundel 30 are noted in Bury Chron., pp. 39-129 nn. passim. Some of the marks are almost, and a few completely, illegible. For Matthew Paris's use of'vacat' marks when revising his own chronicles, to alter their political tone etc., see Vaughan, Matthew Paris, pp. 64-5, 117-24. 192 BL Royal MS 13 E VI (C in Radulphi de Diceto Lundoniensis Opera Historica, ed. William Stubbs (Rolls Series, 1876, 2 vols)). 193 See Cheney, 'The making of the Dunstable annals', pp. 216-17. 190
238
Legends, Traditions and History in Medieval England
que ad opus sancti Eadmundi primo fuit constructa'.194 Similarly, collation shows that the chronicler of St Peter's, Gloucester, when in the mid-twelfth century he copied the chronicle ofjohn of Worcester, added a notice of the death of Roger de Berkeley, patron of St Peter's, but omitted the account of the sudden (and soon fatal) illness during mass of Uhtred, precentor of Worcester.195 The question arises whether a chronicle was lent to another house for copying or whether the copy was made in the house of its origin. Maybe more research, for example on the palaeography of the manuscripts concerned, would provide answers. No doubt practice varied. In the case of the Bury chronicle in Arundel 30 there is evidence suggesting that at least the Peterborough copy was made at Bury itself: the hand which wrote 'Sancti Petri de Burg' in plummet in the margin almost certainly wrote 'Nota' in the same way by a number of entries about Bury. Particularly convincing evidence that the hands are identical is provided by two longer 'nota' marginalia; that is, 'Nota de Capella Marie' and 'Nota de Mort' abb [a] t [is] Simon' by the relevant entries in the annals for 1275 and 1279 respectively.196 The references are to the notice in the text to Abbot Simon de Luton's construction of the Lady Chapel in St Edmund's conventual church, and to his death.197 The Bury chronicle, therefore, considered together with its derivatives, throws light on how the contemporary part of a chronicle might be written, and how its text might spread to other religious houses. It is a fairly safe basis for generalisation since it is typical of its genre, and, as has been seen, parallels for the methods of composition which its authors employed can be found in other chronicles.
194
Bury Chron., p. 58; Oxenedes, p. 246. Chronicle ofjohn of Worcester, pp. 25 n. 3, 36 and n. 4. 196 Arundel 30, ff. 162V, 166V. Bury Chron., p. 58 n. c; Gransden, A Critical Edition, p. 413 n. 1282. 197 A copy of Matthew Paris's Flores Historiarum was executed at St Albans c. 1250 probably especially for the monks of Westminster, to whom it was later transferred. See below p. 245. 195
9
The Cronica Buriensis and the Abbey of St Benet of Hulrne
THE Cronica Buriensis, so called by its editor, Thomas Arnold,1 is a history of the Benedictine abbey of Bury St. Edmund's from its refoundation by King Canute in 1020 to 1346. The only known text is a fair copy of about 1400 in a volume from the library at St. Edmund's abbey, now Cambridge University Library Additional MS. 850, fos. 25v~48v: it is incomplete at the end, a gathering of eight leaves having been lost.2 The Cronica is not a methodical record of events with entries for each year. It has the succession of the abbots of Bury and notices of important events in the abbey's history, omitting many years altogether. Arnold treated the Cronica as the work of one man. It is, however, almost certainly a composite work, compiled by more than one author and at more than one time. Exactly how many contributed is unknown but there is evidence suggesting changes of authorship in 1327 and 1335. The work must have been completed before about 1400, the approximate date of the manuscript. The annal of 1327 reads like a contemporary account of the events it records. 1
Memorials of St. Edmund's Abbey (3 vols., Rolls Ser., 1890—6), iii, pp. vii-xv, 1-73 (hereafter referred to as Arnold). 2 The medieval foliation jumps from fo. ccxxi (fo. 48 of the modern foliation) to fo. ccxxx (now fo. 49). The Cronica today ends with a complete sentence as the first word or two of the next sentence, at the end of the last line on fo. 48v, have been totally erased by scraping. Arnold does not note that the text is incomplete.
240
Legends, Traditions and History in Medieval England
The choice of events to be recorded may have been determined partly by the availability of material. To 1292 the text is mainly composed of extracts from other chronicles; and from 1301 (there are no entries for the years 1293 to 1300) of documents. There are a few passages which are neither citations from known literary works nor from documents. Two of them have information not to be found in other literary sources. One relates to the collection of an aid in 1212 from the town of Bury by a monk, instead of as was customary by the burgesses.1 The second is the account of the revolt of the town against the abbey in 1327; this is independent of the account written at Bury, the Depraedatio Abbatiae.2 Arnold's identification of citations from, and passages reminiscent of, other chronicles in the Cronica is incomplete and somewhat misleading. He printed citations in small type with the name of the work in which they occur in the margin. He also printed letters and other documents in small type. He does not, however, mention that the Cronica is also related to a number of other works. The account at the beginning of the Cronica of the refoundation of Bury by monks of St. Benet of Hulme and of Ely appears to be mainly a conflation of the account inserted in the midtwelfth-century copy of Florence of Worcester from St. Edmund's abbey (MS. Bodley 297, p. 35o)3, that in the registers of Hulme and Bury (Brit. Mus., Cotton MS. Galba E. ii, fo. 36v4, and Brit. Mus, Harley MS. 1005, fos. 35, 35v respectively), both of about 1300, and that in the volume of St. Edmund's life and miracles, written in the last half of the fourteenth century (MS. Bodley 240, fo. 638; the passage has the rubric 'ex cronicis de Hulmo').5 The details of the confraternity between Hulme and Bury which follow in the Cronica are almost verbatim in Cotton MS. Galba E. ii, fo. 36v and Harley MS. 1005, fos. 35, 35v.6 The succession, with short lives, of the Bury abbots recorded in the course of the Cronica is probably derived from a list like that in the fifteenth-century Lakynhethe register of Bury (Brit. Mus., Harley MS. 743, fos. 52-3)' and the notice of Abbot Baldwin's building activities resembles that in Hermann's De Arnold, iii. 9-10. Noticed by M. D. Lobel, The Borough of Bury St. Edmunds (Oxford, 1935), P- 1242 Printed Arnold, ii. 327-54. For the account in the Cronica see below 243-4. For the revolt see M. D. Lobel, ' A detailed account of the 1327 rising at Bury St. Edmund's and the subsequent trial*, Proc. Suffolk Inst. ArchaeoL, xxi (1933), 2I5-3L 3 Printed Arnold, i. 341-2. 4 Printed W. Dugdale, Monasticon Anglicanum, ed. J. Caley, H. Ellis and B. Bandinel (6 vols. in 8, 1817-30), iii. 135 (hereafter referred to as Mon. AngL). 5 Printed Arnold, i. 359. 6 The version in the Cronica ('Crescente autem ibidem religione . . . unus fratrum loci illius collocetur:' Arnold, iii. 2) differs from that in Cotton MS. Galba E. ii (printed Mon. Angl., iii. 135) and Harley MS. 1005 in omitting a sentence reading 'quod usque in eternum gratia Dei firmiter observabitur' after * omnibus aliis bonis Deo placitis' and adding two sentences at the end reading 'Abbates vero utriusque ecclesie . . . frequentius exhibetur'. Cf. above p. 97. 7 Printed Mon. Angl., iii. 155-6.
The Cronica Buriensis and the Abbey ofSt Benet ofHulme
241
1
Miraculis Sancti Eadmundi. There are citations which appear to be from the chronicle of John of Wallingford.2 Arnold recognized citations from, and passages reminiscent of, Jocelin's life of Abbot Samson,3 the Annales Sancti Edmundi (a chronicle of the world from the Incarnation to 1212 compiled at Bury),4 the Electio Hugonis (an account, with documentation, of the disputed election as abbot of Bury of Hugh de Northwold 1213-14),5 and the Chronica Maiora of Matthew Paris.6 He noticed that the Cronica contains citations from the chronicle, compiled at Bury in the last half of the thirteenth century, covering the period from the Creation to 1301, which has been attributed to John de Everisden. But, though he identified some citations from ' Everisden', he overlooked many more. He used the only printed text of 'Everisden', appended by Benjamin Thorpe to his edition of Florence of Worcester (English Historical Society, 1848/9). Thorpe printed from the version of 'Everisden' in Corpus Christi College, Cambridge, MS. 92, which was written for Peterborough abbey. It only covers the years from 1152 to 1295 and omits many passages relating to Bury which occur in the best text of the chronicle written at Bury.7 The Cronica cites many of these 1 Printed 2
Arnold, i. 26-92. Cf. ibid., iii. 4 and i. 85. Brit. Mus., Cotton MS. Julius D. vii, fos. 61-110, for which chronicle see R. Vaughan, 'The chronicle of John of Wallingford', Eng. Hist. Rev., Ixxiii (1958), 66-77, and the same, 'The chronicle attributed to John of Wallingford', Camden Miscellany, xxi (1958). Citations from it in the Cronica are: s.a. 1,210 (Arnold, iii. 9) 'cum filio suo capta . . . ibidem fame interiit'; s.a. 1214 (ibid., pp. 10-11) ' Hie legationis suae officium. . . His ita gestis ad propria remearunt'. 3 The account of St. Edmund's translation in 1198 in the Cronica (Arnold, pp. 7-9) is related to that in Jocelin (The Chronicle of Jocelin of Brakelond, ed. H. E. Butler (1949), with an English translation, pp. 112 sqq.). 4 The work is incomplete at the end owing to the loss of leaves. Extracts printed Arnold, ii. 3-25, and F. Liebermann, Ungedruckte Anglo-Normannische Geschichtsquellen (Strassburg, 1879), pp. 97-115. The account of Abbot Samson's death in the Cronica (Arnold, iii. 9) is a citation from it. 5 Printed Arnold, ii. 29-130. The account in the Cronica (ibid., iii, pp. vii, 11-26) has briefer narrative passages than the Electio but has transcripts of three letters not in the Electio, viz: the letter ratifying Hugh's election (ibid., p. 14), letter from the archbishop of Canterbury, Stephen Langton, to Pope Innocent III (ibid., pp. 15, 16), letter from Eustace, bishop of Ely, to Pope Innocent III (ibid., p. 16). 6 The notice of the death of Hugh de Northwold, bishop of Ely, formerly abbot of Bury, in 1248 (ibid., p. 29),' quandoque abbas sancti Edmundi.. ita et episcopus episcoporum coruscauit', is in Matthaei Parisiensis Chronica Majora, ed. H. R. Luard (7 vols., Rolls Ser., 1872-83), v. 454-5. 7 College of Arms, MS. Arundel 30. For the 'Everisden' chronicle and the manuscript texts see V. H. Galbraith, 'The St. Edmundsbury chronicle, 1296— 1301', Eng. Hist. Rev., Iviii (1943), 51 sqq. Since Professor Galbraith wrote his article another manuscript (to 1283) of 'Everisden' has come to light, which is now in the Moyses Hall Museum at Bury St. Edmunds. An edition of MS. Arundel 30 has been prepared by the present writer for future publication in Nelson's Medieval Texts.
242
Legends, Traditions and History in Medieval England
passages omitted in the Peterborough text: Arnold did not recognize them. Arnold indicated that the Cronica has citations from the chronicle of English history to 1292 composed at St. Benet of Hulme and attributed to John de Oxenedes.1 Yet the passages which he ascribed to 'Oxenedes' are citations in ' Oxenedes' from * Everisden'.2 Collation of the Cronica, ' Everisden' and ' Oxenedes' suggests that these passages in the Cronica are citations from * Everisden' and not' Oxenedes'3: only two passages (not identified by Arnold) in the Cronica are in 'Oxenedes' and not in 'Everisden'.4 Collation also indicates that the Cronica and 'Oxenedes' cite the same version of 'Everisden', but that this version differed from those surviving today. Thus s.a. 1071 both the Cronica and 'Oxenedes' omit the first line of the verse inscribed on the altar which Pope Alexander II gave to Baldwin abbot of Bury; the line is in all the known texts of 'Everisden'.5 Another variant suggests that the lost version was not written at Bury: s.a. 1275 the Cronica and 'Oxenedes' add the phrase ' apud sanctum Edmundum' to the statement in' Everisden' that the chapel ^Chronicajohannis de Qxenedes, ed. H. Ellis (Rolls Ser., 1859) (hereafter referred to as Ellis). 2 'Oxenedes' has citations from 'Everisden' for the years 1020 to 1169 and 1258 to 1292 (the St. Albans chronicles are the main sources for the intervening period). Ellis apparently only knew the text of 'Everisden' from Thorpe's edition of Florence of Worcester and so did not identify many of the citations from it in 'Oxenedes'. The connexion between 'Oxenedes' and 'Everisden' is noticed in Bartholomaei de Cotton, monachi Norwicensis, Historia Anglicana, ed. H. R. Luard (Rolls Ser., 1859), p. Ivii. 3 ' Oxenedes' has passages from ' Everisden' not in the Cronica, and the Cronica has some not in ' Oxenedes'. It is possible that the Cronica cites ' Oxenedes' for passages common to 'Oxenedes' and 'Everisden', and only cites 'Everisden' directly for passages not in 'Oxenedes'. However it is more likely that the Cronica throughout cites ' Everisden' directly, as ' Everisden' is a better authority for Bury history than 'Oxenedes'. Collation supports this view: some of the 'Everisden' citations in the Cronica are fuller than in 'Oxenedes' which omits short sentences (e.g. 'Oxenedes' omits from 'Everisden' citations s.a. 1279 'Dominica videlicet in albis, . . manerium suum,' s.a. 1282 'Fraternitas etiam Duodene . . . xii marcas fuit taxata'; both sentences are in the Cronica: see Ellis, PP- 253, 259; College of Arms, MS. Arundel 30, fos. i66v, 169; Arnold, iii. 33, 35). Also 'Oxenedes' has some readings in 'Everisden' citations different fr6m those in both 'Everisden' and the Cronica (e.g. s.a. 1275 'fuerunt' before 'apud sanctum Edmundum';' Everisden' and the Cronica read ' venerunt': Ellis, p. 246; College of Arms, MS. Arundel 30, fo. 162; Arnold, iii. 31). Similarly 'Oxenedes' does not derive 'Everisden' citations from the Cronica as it has phrases from 'Everisden' not in the Cronica (e.g. s.a. 1282 like 'Everisden' it has 'modo predicto' before 'cepit contributionem'; the Cronica omits these words: Ellis, p. 258; College of Arms, MS. Arundel 30, fo. 169; Arnold, iii. 34). 4 Details of the expulsion of the Jews, s.a. 1290, and of the king's visit to Bury in 1292 (Arnold, iii. 35-6) are in 'Oxenedes' (Ellis, pp. 277, 285) but not in 'Everisden'. Cf. the briefer entries in College of Arms, MS. Arundel 30, fos. I77v, i84v. 5 Cf. Arnold, iii. 3; Ellis, p. 34; College of Arms, MS. Arundel 30, fo. I33V.
The Cronica Buriensis and the Abbey ofSt Benet ofHulme
243
of St. Edmund was pulled down; in the same entry they read * creditur' for' credimus' at the beginning of the sentence ' illam [capellam] fuisse que ad opus sancti Edmundi primo fuit constructa'.1 Collation of * Oxenedes' with the extant texts of ' Everisden' suggests that the lost version ended in 1290. Though the annals for 1291 and 1292 in * Oxenedes' appear to be related to * Everisden', they contain such striking variants as to suggest that the author was using a draft of 1 Everisden' and some of the same documents as the Bury chronicler used, together with his own knowledge. For example, the copy of Edward I's letter of 1291 relating to the Scottish succession case is addressed to the abbot and convent at Bury, but the letters recited in it of submission of the competitors to Edward's judgment are in Latin, though in ' Everisden' they are in French. The probability, although unknown to Arnold, that the Cronica cites the same version of 'Everisden' as the Hulme chronicle attributed to Oxenedes, supports his view that the Cronica was compiled at Hulme. Arnold writes (p. vii) of the Cronica: 'this chronicle, as many indications go to show, was written by a monk of St. Benet Hulme'. Undoubtedly three indications led Arnold to his conclusion. The first was the opening paragraph concerning the part played by St. Benet of Hulme in the foundation of Bury and the close relationship between the houses. The second was the presence of twelve letters to the abbot of Hulme, mostly from the abbot, prior and others at Bury, and of two from him (all relating to Bury), dated or dateable 1301 to 1335 (the last letter cannot be dated exactly but was probably written after 1335 and certainly before 1346).2 Arnold's third probable reason for ascribing the Cronica to Hulme was the inclusion in the vivid and surely contemporary account of the revolt of the town of Bury, which broke out on 14 January 1326/7, of a description of the flight of the sacrist, William de Stowe, to Hulme.3 It relates that Stowe hardly escaped, climbing the town wall with a ladder, helped by a carpenter, in the middle of the night, and reached Hulme only after evading an ambush at Newmarket. The Cronica does not record the length of his stay: it reads' venit ad sanctum Benedictum die Dominica proxima sequente, ibique moratus est usque ad . . .', leaving a blank for the date of his departure. Presumably Stowe was still at Hulme at the time of writing. One of the letters, dated i February [1326/7], is from the abbot of Bury thanking the abbot of Hulme for his hospitality to the unfortunate Stowe and asking for its extension.4 The Cronica records that other Bury monks, who were on holiday in the country, took refuge at Hulme but returned to Bury where they were imprisoned. It is not x
Cf. Arnold, iii. 32; Ellis, pp. 246-7; College of Arms, MS. Arundel 30, fo.
162V. 2
Arnold, iii. 48. It is a letter from William de Stowe as prior of Bury to John abbot of Hulme (1325-46) asking him to send three or four monks to the feast of St. Edmund. It follows a letter dateable to 1335. *Ibtd., p. 39. *Ibzd.t p. 41.
244
Legends, Traditions and History in Medieval England
unlikely that the chronicler obtained his information from the refugees themselves. This evidence only suggests that the Cronica as far as the end of the fourteen 'Hulme' letters was composed at Hulme. It is likely that the rest of it was written at Bury, for its contents have nothing to do with Hulme. They are documents relating to the dispute between William de Bernham, abbot of Bury 1335 to 1361, and John, abbot of the Premonstratensian house of Langley, over the reception at Bury of a fugitive canon of Langley, and to the dispute, 1345 to 1346, between Abbot William and William Bateman, bishop of Norwich, over the abbot's spiritual jurisdiction. The reason why a history of Bury should have been written at Hulme is obscure. Possibly it was one result of the close relationship between Bury and Hulme. The Cronica shows that it was customary for the abbots to attend each other's election, installation and funeral.1 The deed of confraternity which must underlie part of the opening paragraph stipulated that the houses were to help each other in times of poverty or trouble (such as fire or war), if necessary harbouring half the inmates of the stricken house. This deed of confraternity seems to have belonged to the type which Professor Knowles suggests was ' a kind of insurance on the part of the communities'.2 The interest of the Hulme monks in Bury must have been stimulated by the town's attack on its privileges and the flight of monks to Hulme. Seven of the fourteen ' Hulme' letters and one other document3 in the Cronica relate to the revolt. One is a papal bull appointing the abbot of Hulme legate to announce the excommunication of the rioters. If the revolt was the cause of the writing of the Cronica to 1327, the possibility cannot be disregarded that one of the Bury refugees at Hulme had a hand in it. ADDITIONAL NOTE Page 241 n. 5. There is a new edition of the Electio Hugonis: The Chronicle of the Election of Hugh Abbot of Bury St. Edmunds and Later Bishop of Ely, ed., with an English translation, R. M. Thomson (Oxford Medieval Texts, 1974). — n.-j. My edition of the Bury Chronicle in Arundel MS 30 in the College of Arms is published: The Chronicle of Bury St. Edmunds 1212-1301, ed., with an English translation, A. Gransden (Nelson's Medieval Texts, 1964).
10 The Continuations of the Flores Historiarum from 1265 to 1327 ATTHEW Paris started the Flores Hutoriarurn, a general history based
M n his own works, at the Creation of the world, and ended in
1249.' A copy of this chronicle (now MS. 6712 in Chetham's Library, Manchester) was made at St. Albans under Matthew Paris's supervision; he himself wrote the text from 1241 to 12492 and helped his pupils draw the series of pictures of the coronations of English kings (Arthur, Edward the Confessor, and the kings from William the Conqueror to John).8 This manuscript was probably executed specially for the monks of Westminster in about 125O.4 However, it was not transferred to Westminster until 1265, after a continuation from 1250 to 1265 had been added at St. Albans.5 The Flores Historiarum was continued in Westminster Abbey from 1265 to 1327. The first continuation ends abruptly in February 1307, and the second continuation begins with a notice of Edward I's death, on 7 July, 1307, and ends with Edward Ill's accession, on 25 January, 1327. The problems raised by both continuations have been studied with especial reference to the manuscript evidence by a number of scholars, F. Mad1 Matthew Paris' Flores Historiarum and its continuations are printed in Flores Historiarum, ed. H. R. Luard, i (Rolls Series, 1890, 3 vols.). 2 See R. Vaughan 'The Handwriting of Matthew Paris,' Transactions of the Cambridge Bibliographical Society, (1953), 384, 390, and N. R. Ker, 'From 'Above Top Line' to 'Below Top Line',' Celtica, 5 (1960), 15-16. I am grateful to Miss Hilda Lofthouse, librarian of Chetham's Library, for allowing me facilities for examining the manuscript. 3 The pictures are described and reproduced by A. Hollaender, 'The pictorial work in the Flores Historiarum of the so-called Matthew of Westminster,' BJRL, 28 (1944), 361-381. The picture of Henry Ill's coronation is missing because of the loss of a leaf. The picture of Edward's coronation was executed at Westminster. 4 See Matthaei Parisiensis Monachi Sancti Albani, Historia Anglorum, ed. F. Madden (Rolls Series, 18661869, 2 vols.), I. 24 n. i, and V. H. Galbraith, Roger Wendover and Matthew Paris (Glasgow, 1944, reprinted 1970), p. 25 and n. 2. To this evidence may be added the fact that the picture of the coronation of Edward the Confessor in the Chetham MS. (f. H5V of the library's foliation) is twice as big as the other coronation pictures, covering half a page. 5 See Madden, op. cit. i. xxiii-xxiv.
246
Legends, Traditions and History in Medieval England
den,6 H. R. Luard, 7 W. Stubbs8 and T. F. Tout.9 Nevertheless, problems concerning the composition and authorship, and even the provenance of one version, remain. By reviewing the manuscript evidence again in close conjunction with the style and content of the continuations, it is possible to make new suggestions which could contribute to the solution of some of these problems. The first continuation, from 1265 to 1307, survives in a number of manuscripts and in two versions. Luard describes the Chetham MS. and fourteen other manuscripts containing the same version of the text, none of which is the author's/ authors' autograph.10 The Chetham MS., which also contains the text of the second continuation, to be discussed below, is itself a copy made by a number of scribes, mostly of the late thirteenth and /early fourteenth centuries. The annals from 1298 to 1302 are in a later fourteenth-century hand, which may also have written the last part of the annal for 1326 and that for 1327 in the second continuation:11 certainly this section was copied well after the events it records. Other manuscripts of this version, which for convenience may be called the 'Westminster' Flores, are known to have belonged in the middle ages to the monasteries of Norwich, Rochester and Tintern, and one was owned by St. Paul's cathedral. The other version of the 1265-1307 continuation of the Flores is based on the 'Westminster' Flores but has considerable variants, omissions and additions. The earliest known manuscript of this version was copied at or more probably for the Augustinian priory of Merton in Surrey, and is now MS. 123 in Eton College Library. 12 All the six copies of this version, which we will call the 'Merton' Flores, described by Luard apparently descend from the Merton manuscript.13 The evidence that the 6 Madden, op. tit. i. xxiv-xxvii and nn. 7 Flores, ed. Luard, i. xii-xvii, xl-xliii. 8 Chronicles of the Reigns of Edward I and Edward //, ed. William Stubbs (Rolls Series, 1882-1883, 2 vols.), ii. xii-xiii. 9 T. F. Tout, 'The Westminster Chronicle attributed to Robert of Reading,' EHR, 31 (1916), 450-464 (reprinted in The Collected Papers of T. F. Tout (Manchester, 1932-1934, 3 vols.), ii. 289-304). 10 Flores, i. xii-xxxiii. 11 See Flores, i. xiv, and Tout, op. tit. pp. 459-460. 12 I am indebted to Mr. Patrick Strong, librarian of Eton College, and to the College authorities, for depositing Eton College MS. 123 in die British Museum for me to study. The text is a fair copy in a book-hand, although from the end of 1284 there are occasional changes of ink and perhaps of handwriting. The manuscript ends incomplete in 1306 because of the loss of a leaf, but there is no reason to suppose that it did not once extend to the beginning of 1307 where the 'Westminster' Flores ends, as do the descendants of the Eton MS. 13 Luard, op. tit. i. xv-xvii, xxix-xxxiii. The so-called Tenison MS., formerly Phillipps MS. 15732, is now MS. 426 in the Beinecke Rare Book and Manuscript Library in Yale University Library. (See the catalogue of H. P. Kraus, New York, no. 117, 1967, item 29). A microfilm is in Nottingham University Library. The text is a conflation of the 'Merton' Flores and the 'Westminster' one, which
The Continuations of the Flores Historiarum
247
Eton College MS. was made for the canons of Merton is irrefutable: the manuscript has early notices concerning Merton priory in the margins,14 and the letter of Edward I containing the submissions of the competitors for the throne of Scotland to his judgement as overlord, of 1291, in the text is addressed to the prior and convent at Met ton.15 Therefore, it has generally been accepted that the 'Merton' Flores was actually composed at Merton.16 Nevertheless the evidence for this is inconclusive. The 'Merton' Flores only copies twro of the four entries in the 'Westminster' Flores relating to Merton priory,17 and shows no especial interest in the Augustinian order in general. (On the other hand, one of its long additions to the 'Westminster' Flores concerns the fate of a Benedictine abbey, Dunfermline, which was sacked by the Scots in 1303)-18 It seems more likely that the exemplar of the Eton College MS. was written at Westminster. The Eton College MS. itself provides some evidence supporting this view. It has a series of pictures of the coronations of the kings of England from William the Conqueror to Edward I, of better quality than the rather mediocre handwriting seems to justify, which were presumably copied from its exemplar. As has been seen Matthew Paris provided a similar series of pictures for the copy of the Flores which he apparently intended for the Westminster monks, who had of course a close interest in the coronation ceremony. Not only was the coronation performed in their church, but they also had custody of the regalia and of the books relating to the service.19 The artist of the exemplar of the Eton College MS. has one detail which shows accounts for the fact that its copy of Edward I's letter on the Scottish succession case (see below) is addressed to the abbot and convent of Westminster, not to the prior and convent of Merton. The annal for 1306 and the first part of that for 1307 derive from Nicholas Trevet's Annales. It ends with the appreciation of Edward I (for a small addition see below n. 82) and an account of events up to the deposition of Edward II, from Murimuth. It ends with two lines of verse: Carneruam natus, princeps Edwardus amatus; Ingratis gratus, est morte graui cruciatus. For references to other conflated texts see Flores, i. xx. 14 See Flores, iii. 250 n. i, 292, 302 n. 4 (Eton College MS. ff. 227, 251*, 254V). For an insertion in the text concerning Merton priory see ibid. iii. 84. 15 See Flores, i. xvi. 16 However, Richard Vaughan doubts the Merton provenance of the exemplar of the Eton College MS.; see his Matthew Paris (Cambridge, 1958), p. 101. 17 Flores, ii. 46, 51. For the entries in the 'Westminster* Flores concerning Merton priory omitted in the 'Merton' Flores see ibid. ii. 81 and n. 3, 88 and n. i. 18 Flores, iii. 311-312. 19 See P. E. Schramm, A History of the English Coronation, translated from the German by L. G. Wickham Legg (Oxford, 1937), pp. 40, 75, 80.
28 The 'Merton' Flores Historiarum: the coronation of Henry III. (Windsor, Eton College, MS 123Jo. 194)
29 The 'Merton' Flores Historiarum: the coronation of Edward I (who holds the rod with the dove). (Windsor, Eton College, MS 123Jo. 237)
248
Legends, Traditions and History in Medieval England
his knowledge of the regalia. He depicts Edward I holding the rod with the dove, carefully delineated, in his right hand.20 The rod with the dove also appears in the picture of King John's coronation, only less well drawn. This emblem is not generally represented in fourteenth and fifteenth century art, although it was in use in the coronation service at least from the time of Richard I. Usually the king is shown holding a rod with a floriated finial in his right hand (he held the sceptre with fleur-de-lis and orb in his left hand), as in the other coronation pictures in the Eton College MS. It is surely likely that the rod with the dove had attracted the especial attention of the artist of the exemplar of the Eton College MS. If he was working at Westminster his interest in the emblem could have been aroused at the time of Edward I's funeral, because the king was buried in full regalia, holding the rod with the dove. The actual rod, which was described in detail after Edward's tomb was opened in 1774, bore a close resemblance to the artist's representation.21 The artistic evidence could, therefore, be interpreted as indicating that the 'Merton' Flores was composed and illustrated at Westminster soon after Edward I's funeral. More evidence supporting this hypothesis will be adduced when discussing the author's outlook and argument.22 Neither version of the continuation of the Flores Historiarum from 1265 to 1307 has an ascription of authorship. On the other hand the continuation from 1307 to 1327 was according to its own evidence composed by a monk at Westminster called Robert of Reading up to the beginning of the annal for 1326. The relevant passage reads: Sicque frater Robertus de Redinge, quondam monachus ecclesiae bead Petri Apostoli Westmonasterii, cronicarum, vitae quoque suae, finem conclusit. Et in praemissis rnagister Adam Murimoth, olim canonicus ecclesiae sancti Pauli Apostoli Londoniarum, qui texuit ab anno regni regis Edwardi secundi post conquestum sexto usque ad annum regni regis Edwardi tertii filii ejusdem vicesimum, luculentius procedit.23
Tout has already discussed the problems inherent in this passage.24 It 20 Eton College MS. 123, f. 237. (The picture of Kingjohn's coronation is on f. 184). For the rod with the dove see L. G. Wickham Legg, English Coronation Records (Westminster, 1901), pp. lii-liii. 21 See Joseph Ayloffe, 'An Account of the Body of King Edward the First, as it appeared on opening his Tomb in the Year 1774,' Archaeologia, 3 (1775), 384. 22 Below pp.252ff. 23 Flores, iii, 232. 24 Tout, op. cit. pp. 450-455. For a payment by the prior of Westminster to 'brother Robert de Rading,' dated 5 Ocotober 1305, see Documents Illustrating the Rule of Walter de Wenlok, Abbot of Westminster, 1281-1307, ed. B. F. Harvey (Camden fourth series, ii, 1965), pp. 91-92.
The Continuations of the Flores Historiarum
249
is not impossible that Robert of Reading was the author, but if so, there must have been more than one monk of that name at Westminster during Edward II's reign, because a Robert of Reading died in 1317, too early for the chronicler. The ascription-passage itself does not have the authority of contemporaneity. The reference to Adam Murimuth formerly' canon of St. Paul's dates it to after 1347, the year of Murimuth's death.25 Moreover, as Tout has pointed out, the author of the remainder of the chronicle after the ascription-passage used not only Murimuth's chronicle but also that of Robert of Avesbury which was composed at the earliest after September 1356.26 Therefore, the passage was written well after the middle of the fourteenth century. The manuscript evidence agrees with this conclusion. The earliest text of the 1307-1327 continuation, as of the 'Westminster' Flores, is in the Chetham MS. (The only other medieval copy, in Cottonian MS. Cleopatra A xvi in the British Museum, is of the fifteenth century.)27 The change of hand at the beginning of the annal for 1326 noted above,28 occurs at the ascription-passage. The handwriting up to this point, though clearly not the author's autograph, is of the first half of the fourteenth century. (Its near contemporaneity with the events it records is demonstrated by the fact that substantial blank spaces, some of more than half a column, are left at the end of each annal for the addition of late news). We are mainly concerned with this part of the text preceding the ascriptionpassage, which will be called the 1307-1326 continuation. To turn now to the question of literary style. Tout argued on stylistic grounds, irrespective of Robert of Reading's authorship, that one author must have written not only the 1307-1326 continuation but also the annals of the 'Westminster' Flores from about i2g8. 29 The whole, as Tout pointed out, is in a distinctive style: the prose is florid and bombastic with rhetorical invocations and numerous biblical allusions and citations, and is punctuated with couplets of verse. In fact these stylistic features occur in the 'Westminster' Flores almost from the beginning in 1265,*° although they become much more noticeable after 1298: this 25 See Adae Murimuth Continuatio Chronicarum. Robertas de Avesbury De Gestis Mirabilibus Regis Edwardi Tertu, ed. E. M. Thompson (Rolls Series, 1889), p. ix. 26 Ibid., p. xxii; Tout, op. at., p. 454. 27 Cleopatra A xvi, ff. 67-193, contains the 'Westminster' Flores from 1298 to 1307 and then the 1307-1326 continuation. This is followed by the chronicles of Adam Murimuth and of John of Reading, consecutively, thus producing a continuous history from 1298 to 1367. 28 Above p. 246. 29 Tout, he. at. pp. 456-45730 See for examples of biblical citations (from 1266), Flores, iii. 8, 9, 13, 111, 115-120 passim, and of verse couplets (from 1272), ibid., iii. 29, 75.
250
Legends, Traditions and History in Medieval England
suggests that the original series of annals to about 1298, which were probably written fairly close to the events recorded, were refurbished and continued to 1307 in the florid style in the early fourteenth century. This style is particularly characteristic of the 'Merton' Flores (which apparently Tout did not distinguish from the 'Westminster' one). Two passages may be quoted, one from the 'Merton' Flores and one from the 1307-1326 continuation, to show the marked stylistic similarity between these works. The passage in the 'Merton' Flores describes the plight to which England was reduced by the Provisions of Oxford. It reads: Reliqua autem hujus provisionis, videlicet causam, exordium, et processum, regni vero divisionem et desolationem, contentiones et discordias, depraedationes et incendia, intrusiones ecclesiarum, persecutiones clericorum, obsidiones castrorum, tribulationes civitatum, exhaeredationes procerum, gemitus pauperum, extorsiones plebium, redemptiones captivorum, defectus senum, miserias orphanorum, deflorationes et suspiria virginum, fletus et lamenta viduarum, praelia et seditiones, et caetera dampna et gravamina, quae ab initio praedictae provisionis usque post bellum Eveshamiae illius occasione in regno Angliae acciderant et fiebant, quae vox, quae poterit lingua retexere?31 The 1307-1326 continuation has a similar lament on the state of England after the battle of Boroughbridge, reading: Planxerunt mulieres generosae maritorum privationem et natorum, luxerunt ingenui parentum amissionem, doluerunt cives urbium desolationem, deflevit Ecclesia propriam et regni subitam confusionem, quibus communitas undique vallata, liber et servus, dives et pauper, quisque recenter sumpsit lamentum, et qui videbantur laetitiam in vultu praeferre, genas suas occultis lacrimis lugentes regabant.32 Two examples may be given to illustrate the use in both works of biblical allusions and citations. The passage from the 'Merton' Flores is part of the account of Edward I's siege of Stirling castle in 1304. It has extracts from 2 Reg. xi. 20, 21 (here printed in italics), Joab's advice to David, woven into its text: Res stupenda nimis, ex turbine molae violentissimo, quasi ad icturn tonitrui, dextrarius regis resupinatis pedibus corruit, quasi asina Balaam dans locum exterminatori. Et accurrentes commilitones traxerunt regem per declivum mentis, increpantesque dixerunt, 'Domine, quare proprius
31 Flores, iii. 248. 32 Flores, iii. 213.
The Continuations of the Flores Historiarum
251
caeteris inermis ad murum acceditis, ut praeliamini? An ignoras quod multa desuper ex muro tela mittantur ? Quis percussit Abimeiechfilium Jerobaal f Nonne mulier misit super eum fragmentum molae de muro, et interfecit eum in Thebes f Manete amodo in papilionibus. Sive enim fugerimus, non magnopere de nobis ad eos pertinebit, sive media pars de nobis ceriderit, non satis curabunt, quia sola persona vestra pro decem milibus computatur. Melius est igitur ut sitis nobis aliorsum in praesidio'**
The comparable passage in the 1307-1326 continuation describes the escape of Roger Mortimer, the younger, from the Tower of London in 1323. It uses the account of St. Peter's escape from prison, in Act. xii. 4, 6-11, and reads: misit rex crudelis ministros detestabiles ad Turrim Londoniarum, ubi Rogerus et Rogerus de Mortuo Mari compedibus habebantur constricti, ut denuo majori artarentur supplicio dolentes, volens post paucos dies juniorem Rogerum producere populo et acerba rnorte dampnare. Cumque producturus eum esset rex, ecce in node sancti Petri ad Vincula Spiritus Domini astitit et gratia ejus affuit in habitaculo carceris, tactoque corde Rogeri excitavit eum dicens, 'Surge velociter et sequere me.Et egressus sequebatur eum, quia verum est quod fiebat per Christum; non ergo aestimabat se visum videre. Transiens autem primam et secundam custodiam pervenit ultra Thamense flumen ..34 Comparison of these four passages surely helps dissipate any remaining doubts as to the Westminster provenance of the 'Merton' Floret. Moreover, the isolation of this literary style links both versions of the Flores from 1265 to 1307, and the 1307-1326 continuation to two other works. The account in the 'Westminster' Flores of the burglary of the royal treasury at Westminster in 1303^ refers in its concluding words to its source: 'Subsequitur passio monachorum Westmonasteriensium secundum Johannem, cujus copiam non habentas quaerant et invenient et postulantes accipient.'
This tract on the burglary, the 'passion according to John,' is now apparently lost. On the evidence of the extract in the 'Westminster' Flores, which has a bombastic tirade, replete with biblical citations, against the imprisonment by the king of ten Westminster monks accused of complicity, the original tract must have been in the same style as the 'Mer33 Flores, iii. 318. 34 Flores, iii. 217. 35 Flores, iii. 115-117. For reference to this account of the burglary see T. F. Tout. 'A Mediaeval Burglary,' BJRL, 2 (1914-1915), 364. For the burglary in general see ibid. pp. 348-369, E. H. Pearce, Walter de Wenlok, Abbot of Westminster (London, 1920), pp. 146-166, and H. F. Westlake, Westminster Abbey (London, 1923, limited edition, 2 vols.), ii. 430-446.
252
Legends, Traditions and History in Medieval England
ton' Flores and the 1307-1326 continuation. The fourth work in this group is the Commendatio Lamentabilis in Transitu Magni Regis Edwardi, a lamentation on the death of Edward I addressed to his queen, Margaret, by John of London'.36 The Commendatio Lamentabilis starts with a eulogistic description of Edward's appearance and character (modelled on Peter of Blois' description of Henry II) and proceeds with panegyrics on him, spoken, as it were, by Margaret and various social groups (the bishops, the earls and barons, the knights, and the like). It has obvious stylistic affinities with the 'Westminster' and TVferton' Flores, with the 1307-1326 continuation and, judging from the surviving extract, with the tract on the burglary. The 'Merton' Flores and the 1307-1326 continuation both have passages in the threnetic mode (for example the laments on the condition of England quoted above),37 while the extant passage from the tract on the burglary is virtually a lament on the fate of the suspect monks. Furthermore, the Commendatio Lamentabilis is in the characteristic florid prose, with biblical allusions and rhetorical invocations. To illustrate the similarity a passage in the Commendatio Lamentabilis may be compared with one in the 'Merton' Flores. Addressing Edward I's knights the Commendatio Lamentabilis recalls the peace imposed by the king; it reads, with an allusion to Isai. ii. 4 ('Et conflabunt gladios suos in vomeres'): 'Gladii nostri conflabuntur in vomeres et in falces bellastices lanceae redigentur.' It then invokes England: 'O felix Anglia et vere beata terra cujus rex est nobilis .. Quo proficiscente in regione viventium et ejus regnante vocabulo pax vigescit, aemulus obmutescit, et tranquillitate ecclesia reflorescit.'38 The comparable passage in the 'Merton' Flores is in an invocation of England, with reference to her sufferings during the Barons' War. The author recalls her previous peace and prosperity: 'In te, gladiis conversis in vomeres, pax et religio viguerunt, ut caeteris omnibus regnis catholicis speculum sisteres et exemplum.'89 There is some, though inconclusive, evidence concerning the authorship of these works. The John' who wrote the burglary-account could of course be John of London,' the author of the Commendatio Lamentabilis, However, such an identification is only valid if this John of London was a monk of Westminster, which cannot, because the name was a 36 The Commendatio Lamentabilis is printed Chrons. Edward I and Edward //, ed. Stubbs, ii. 3-21. Its view of history is discussed in B. Smalley, English Friars and Antiquity in the early fourteenth Century (Oxford, 1960), pp. 9-12. 37 Above p. 250. 38 Chrons. Edward I and Edward If, ii. 16. 39 Flores, iii. 267.
The Continuations of the Flores Historiarum
253
common one, be proved. Nevertheless, there was a monk at Westminster called John of London in the early fourteenth century .S9a He is apparently to be identified with John Bever, a monk at Westminster.40 This identification provides corroborative evidence for his authorship of the 'Merton' Flores, because one fourteenth-century copy (an abbreviated text from St. Augustine's Canterbury), has a note contemporary with the manuscript attributing the work to John Bever.41 There is, therefore, the possibility that John of London alias John Bever, monk of Westminster, wrote the 'Merton' Flores, the burglary-account and the Commendatio Lamentabilis. He may also have been responsible for the 'Westminster' Flores in its present form: he may have revised a preexisting continuation of the St. Albans' Flores Historiarum to about 1298 and then continued it to 1307. (The reference to John's passio monachorum could be to the author's own work). On stylistic evidence he could also have written the 1307-1326 continuation. However, this conclusion would be rash because it would involve discarding the medieval attribution to Robert of Reading, who, as a product of the same literary school, might well have written a similar florid prose. It is now necessary to examine the light thrown on the continuations of the Flores Historiarum from 1265 to 1327 by their outlook and argument. Just as the continuations have a distinctive prose style, so 39a 'A John of London' appears among the monks of Westminster indicted by Edward I for complicity in the burglary of the royal treasury in 1303; CPR, 1301-1307, p. 195: For an identification of John of London, the monk of Westminster, with a former rector of Newland in the Forest of Dean (1264-1302), of the same name, see C. Fortescue-Brickdale, 'A Gloucestershire Rector's Lament for Edward I', Transactions of the Bristol and Gloucestershire Archaeological Society, 61 (1939), p. 195. The evidence there adduced is quite inconclusive. I owe this reference to Miss E. Danbury. 40 The identification of the author John of London with John Bever is suggested by T. D. Hardy, Descriptive Catalogue of Materials relating to the History of Great Britain and Ireland (Rolls Series, 1862-1871, 3 vols.), iii. 282. The evidence that the monk John of London was also called John Bever is a mandate preserved in Westminster Abbey from the commissary general of the archdeacon of London to the archdeacon of Westminster to publish the excommunication of 'brother John de London, dictum Le Bevere,' for contumacy in respect of a summons for a crime committed within the archdeaconry of London, dated 28 March 1310. This identification does not, however, prove that this John of London was the author of the Commendatio Lamentabilis. Nevertheless, Madden, Historia Anglorum, i. xxiv-xxv and n. i, tentatively accepts that he was the same man, and the possibility is discussed by Stubbs, Chrons. Edward I and Edward II, ii. xi-xiii. On the other hand, Miss Smalley maintains the work's anonymity; Smalley, op. cit. p. 9 n. i. 41 BM MS. Harley 641. At the foot of f. i is written in a fifteenth-century hand: 'Cronica de edicione domini Johannis died Bever' monachi Westmonasterii. De libraria monasterii Sancti Augustini Cantuariensis Distinct' T. Abbatis.' (A similar inscription is at the end of the chronicle, f. ii5v). The reference is to Thomas de Fyndone, abbot of St. Augustine's 1283-1310. The earlier part of the text is based on Geoffrey of Monmouth. Then at the Norman conquest it begins to follow the 'Merton' Flores which it slightly abbreviates, and also interpolates a few paragraphs; see Flores, i. xxxi-xxxiii. It is noteworthy that the five texts of the Commendatio Lamentabilis used by Stubbs all follow copies of the Flares', see Hardy, op. cit. iii. 309, and Chrons. Edward I and Edward //, ii. vii-xi.
254
Legends, Traditions and History in Medieval England
also do they have distinctive attitudes to politics. When the Flares left St. Albans for Westminster in 1265, it acquired a new political bias. Matthew Paris and his continuators at St. Albans (like the great majority of contemporary chroniclers) were harsh critics of the central government and ardent supporters of the baronial cause. At Westminster the Flares was continued in a royalist tone. Thus the 'Westminster' Flares calls Henry III an innocent, God-fearing man whose dearest wish was to end civil discord.42 Recording Henry's death, it writes: 'God and those who faithfully supported Henry knew what an innocent, patient man he was, and with what devotion he worshipped the Saviour. And, above all, the miracles which followed his death, show how God valued his life.43 The 'Westminster' Flores estimate of Edward I, despite occasional criticism, tends to be eulogistic. (In general the other chroniclers also favour Edward). It enthusiastically supports his Scottish policy. Edward was Scotland's rightful lord, the glorious victor,44 the most fortunate king, who 'trampled on the horns of the proud and silenced the roar of rebels.'45 The very walls of Stirling castle, captured in 1303, bore witness to his glory: the damage done to them by the besiegers' missiles were 'indelible tokens of the lasting victory and great triumph of this magnificent king.'46 This royalist bias is even more marked in the 'Merton' Flores. The author modified the few passages in the 'Westminster' Flores which were critical of the king. For example he rewrote the passage on Edward I's oppressive treatment of the clergy when they refused an aid in 1296, thus omitting a particularly biting piece of invective. ('The king turned to cruelty, and the clergy were so constantly afflicted that it was as if the madness of Nero was revived in England').47 Under 1303 he shortened the tirade against Edward for the imprisonment of the monks suspected of complicity in the robbery of the royal treasury, and rewote it in such a bombastic style that it is almost incomprehensible.48 Besides copying the passages in the 'Westminster' Flores which favour the king, the author of the 'Merton' Flores added more of his own. He inserted a
42 43 44 45 46 47 aulici 48
Flores, iii. 15. Flores, iii. 28. Flores, iii. 112. Flores, iii. 118. Flores, iii. 119-120. Flores, iii. 99. The author of the 'Merton' Flores diverted some of the blame on to thefalsi clerici, curiales; ibid. iii. 291. Flores, iii. 313.
The Continuations of the Flores Historiarum
255
resume of events culminating in the battles of Lewes and Evesham which justifies the monarchy and discredits the barons.49 He accuses the baronial leaders whom he calls 'princes of faction', of being moved by hate, ambition and greed rather than by patriotism,50 and of causing untold suffering by the civil war, which they engendered.51 He insists on the ultimate legitimacy of monarchy,52 defending the king's right to endow his foreign relatives53 and arguing that Henry could have annulled the Provisions of Oxford, which he refers to as the fwoditiones Oxoniae, by 'common consent' because they had been agreed by 'common consent'— only his scrupulousness made him seek papal absolution from his oath to observe them.54 Henry was a man of peace, and monarchical government was the custodian of peace. Henry's restoration to power in his 'perfect and spacious vineyard' represented the restoration of peace.55 The author gives a paean on peace, contrasted with horrors of civil war, in his most eloquent prose.56 The 'Merton' Flores praises Edward I. The author emphasises, in the account of events immediately preceding the battle of Evesham, the part played by the then Lord Edward in securing his father's victory over Simon de Montfort.57 Later, when Edward was king, he treats him as the architect of peace. He was especially interested in Edward's Scottish campaigns which tried to impose peace on the north. His most eulogistic passages are in his elaboration of the account in the 'Westminster' Flores of the siege of Stirling castle.58 Edward was brave and wise, his interference in Scotland was justified by historical precedents, and during the siege he was protected by the angel of the Lord from the blows of the devil. The biblical citations and allusions, discussed above,59 serve to embellish Edward's persona as the Lord's annointed. The strength of the political bias in the 1307-1326 continuation of the Flores rivals that in the 'Merton' Flores. However, the bias is in the opposite direction. No other chronicler attacks Edward II with such unremitting virulence. He is eloquent on Edward II's malefactions and 49 50 51 52 53 54 55 56 57 58 59
Flores, iii. 251-266 Flores, iii. 252-254. Flores, iii. 248. (See above Flores, iii. 254. Flores, iii. 252. Flores, iii. 255. Flores, iii. 266. Flores, iii. 266-268. Flores, iii. 264-265. Flores, iii. 315-321. Above pp. 250-1.
p. 250).
256
Legends, Traditions and History in Medieval England
accuses him of tyranny. The king, 'paralysed by sloth, won disgrace not fame.'60 The author dwells on the failure of his Scottish campaigns61 and his virtual loss of Gascony.62 Edward was a coward in battle, fleeing in terror from the Scots in 1322, 'spurring on his horse, trembling and defenceless/' and allowing the English to suffer all the miseries of defeat.63 And so he lost the glory and power won by his famous ancestors whose fame was extolled throughout the world.64 He broke his oath to observe the Ordinances65 and habitually 'forgot in the morning what he had promised in the evening.'66 His avarice was insatiable, with the result that the exactions of the royal justices reduced many counties to irremediable poverty,67 and he 'stretched forth his hand to vex' not 'certain of the church' (Acts xii.i) but every single prelate by his wicked ferocity!68 He followed evil counsel69 and made decisions 'in secret in his chambers with his intimates'70 and tried to circumvent opposition in parliament by delay, so that the lords, their patience and money exhausted, agreed to his demands.71 He preferred the company of the lowborn to that of the nobles72 whose counsel he neglected and whom he grievously oppressed, planning to humble and even to exterminate them. Under 1321 he writes: 'In his insane fury he hated all the magnates with such wicked hatred that he plotted to overthrow completely the great men of the realm together with the whole English nobility.'73 Conversely, the author of the 1307-1326 continuation praises the barons for their wisdom and magnanimity, their strength and courage in protecting England by war. 74 They were steadfast in their support of their leader, Thomas of Lancaster, preferring in 1321 to die in the faith
60 Flores, iii. 192. 61 Flares, iii. 159, 176, 190 and passim. 62 Flores, iii. 221-223. 63 Flores, iii. 210. 64 Flores, iii. 192-193. 65 Flores, iii. 201. 66 Flores, iii. 222. (Cf. ibid. iii. 228). 67 Flores, iii. 218. 68 Flores, iii. 218. 69 See e.g. Flores, iii. 143-144, 148, 192-193, 209, 230. 70 Flores, iii. 219. 71 Flores, iii. 220. 72 The chronicler describes how Edward amused himself 'relicto nobilium consortio .. orientalia stagnorum aequora in comitatu Cantebrugge cum simplicium personarum magna comitiva ..;' Flores, iii. 173. 73 Flores, iii. 200. 74 Flores, iii. 149, 188.
The Continuations of the Flores Historiarum
257
of Christ for the liberty of the church and realm, rather than to violate their oath to uphold the Ordinances.75 Thomas himself died a martyr to the cause, and 'the manifold goodness of this famous man, the generous alms he gave, his pious acts when alive and the merits of his posthumous miracles, worked by divine clemency, require a book on their own.76 Moreover, the author consistently supports Isabella and Mortimer (unlike Adam Murimuth and Geoffrey le Baker who are bitterly hostile to them). In the passage quoted above on Mortimer's escape from the Tower, he implies, by the use of biblical citations, a comparison of Mortimer with St. Peter.77 He is outspoken about Edward II's treatment of Isabella, who, he asserts, pleaded with the king in 1311 to turn from his intimates (confabulatores) 'to the excellence of the magnates, the wisdom of the clergy and the protection of the community,' for the sake of peace.78 Later, in 1324, he accuses the king of cruelty to the queen, by depriving her of her household, and exclaims: 'On the insane stupidity of the king of England, which must be condemned by God and man! He should not love his own infamy and illicit bed, full of sin, and should never have removed from his side his noble consort and her tender wifely embraces, in contempt of her noble birth!'79 The author's Francophilia appears in the assertion that Isabella's brother, Charles IV and the nobility of France, grieved at her ill-treatment (and impressed by the good looks and good manners of her son Edward, the heir-presumptive), will help her to invade England.80 We have, therefore, two chronicles, the 'Merton' Flores and the 13071326 continuation, of exceptionally strong political bias. Both stop short just before the end of the reign to which they relate, one in February 1307 and the other in February 1326. However, there is evidence suggesting the possibility that the text of the 'Merton' Flores, or at least drafts for it, did originally extend until the death of Edward I on 7 July, 1307, and that part of this continuation may be recovered from the chronicle of Adam Murimuth, who, as he himself states, used Westminster material.81 Murimuth's appreciation of Edward I's achieve75 Flores, iii. 204. 76 Flores, iii. 214. 77 Above, p. 478. 78 Flores, iii. 148. 79 Flores, iii. 2*9. 80 Flores, iii. 231. 81 Murimuth states that he found chronicles in Westminster Abbey up to 1305 and that from that year he wrote from his own knowledge; Murimuth, pp. 3-4. However, it is likely that Murimuth used historical material at Westminster to a later date (see the next note).
258
Legends, Traditions and History in Medieval England
rnents following the notice of his death coincides in tone with the account of the king in the 'Merton' Flores. It specifically attributes the defeat of Simon de Montfort and the barons to Edward's heroism, and recites his victories over England's enemies, especially the Scots, both of which subjects figure large in the 'Merton' Flores*2 Similarly, there is a possiblity that the 1307-1326 continuation originally went as far as the accession of Edward III. The ascriptionpassage, at the beginning of the annal for 1326 which states that Robert of Reading ended both his chronicle and his life, does not preclude such a conclusion. It could merely signify the point at which Robert stopped writing, not the date of his death. Perhaps he even left drafts for his continuator. If so, it is likely that this continuation is partly preserved by Murimuth. The account of Edward II's deposition in his chronicle sustains the theme of the 1307-1326 continuation of the Flores. It represents Edward as abdicating voluntarily, admitting his own unworthiness to rule, saying: 'I greatly lament that I have so utterly failed my people, but I could not be other than I am; I am pleased that my son who has been thus accepted by all the people should succeed me on the throne.' And so 'the whole community of the realm immediately received the young Edward as king.'83 It remains to examine possible causes for the extreme bias in favour of Henry III and Edward I in the 'Merton' Flores, and against Edward II in the 1307-1326 continuation. The problem is whether the affection of the Westminster monks for Henry III and Edward I, and their hatred of Edward II are sufficient explanation. Certainly it is not surprising that the 'Westminster' Flores adopted a pro-royalist attitude to Henry III. It supported the king partly no doubt from reasons of circumspection, because the governement was next door to the abbey (the royal treasury was in the abbey itself). But their support was surely sincere because Henry, as the abbey's patron, was a generous benefactor and
82 Murimuth, p. 10. Murimuth follows the notice of the funeral with two lines of verse celebrating the peace which Edward I imposed. The presence of the same lines in the Flores (iii. 128) suggests that Murimuth was still using Westminster material for his annal for 1307. The lines read: Dum viguit, rex, et valuit tua magna potestas, Fraus latuit, pax magna fuit, regnavit honestas. The so-called Tenison MS. (see above p. 246 n. 13) adds two more lines of verse: Scotos Edwardus, dum vixit, supptditavit, Terruit, afflixit, depressit, dilaniavit. 83 Flores, iii. 235. See below p. 262.
The Continuations of the Flores Historiarum
259
the builder of their new church. The 'Westminster' Flores describes how when Henry was ill in 1271 'the monks of Westminster, fearing to lose their patron, processed with bare feet, in the rain to the New Temple and there celebrated mass for him to the Blessed Virgin Mary.' 84 However, such good relations with Henry III expressed in the 'Westminster' Flores do not explain the rewriting for the 'Merton' Flores of the account of the Barons' War in an even more royalist tone, in the early fourteenth century. Moreover, the abbey's relations with Edward I, despite his gift of the regalia of the Scottish kings and the Stone of Scone to the abbey in 1297, 85 were not always sufficiently good to explain the revision of the 'Westminster' Flores to produce in the 'Merton' version an even more eulogistic account of him. Conversely, relations between Edward II and the abbey were hardly bad enough to justify the invective against him in the 1307-1326 continuation, although they were sometimes strained.86 (The chronicler expressed the anger of the Westminster monks at the removal in 1319 of the royal Exchequer and court of King's Bench from Westminster palace to York,87 and under 1320 he records that Edward occupied, 'not without sacrilege,' a cottage within the abbey precincts 'called Borgoyne, for (Edward) preferred to have the title 'of Borgoyne' rather than the titles used by his glorious predecessors.')88 Another explanation may be tentatively suggested for the political bias of the 'Merton' Flores and the 1307-1326 continuation. Perhaps they were official histories, written in response to royal command. Such an hypothesis would explain some of the problems raised by these works. With regard to the 'Merton' Flores, the following suggestions may be made. It was written by John of London, alias John Bever, a monk of Westminster, after the death of Edward I (7 July, 1307), at the command of Edward II, perhaps for presentation on the occasion of his coronation (24 February, 1308). John based the work on the 'Westminster' Flores, for the composition of which he himself was partly responsible. (The eulogy of Edward I in the account of the siege of Stirling castle in the 'Westminster' Flores suggests the possibility that 84 Flores, in. 22-23. 85 Flores, iii. 101. 86 The monks of Westminster clearly objected to Edward I's oppressive taxation of the clergy, and came into conflict with him over the indictment of some of their number for complicity in the burglary of the royal treasury; see above p. 251 & n. 35. 87 Flores, iii. 191. 88 Flores, iii. 193. For minor causes of conflict between Edward's II's household and Westminster Abbey see Tout, op. tit. p. 457 n. 2.
260
Legends, Traditions and History in Medieval England
John wrote this passage at the command of Edward I to celebrate his victory). There is evidence corroborating the hypothesis that the 'Merton' Flores was written for Edward II's coronation. The coronation pictures in the earliest text, the Eton College MS., show a preoccupation with the coronation ceremony, and the careful delineation of the rod with the dove could be interpreted as indicating that the pictures were executed soon after Edward's funeral when the emblem, may have attracted especial attention. 89 The theme of the 'Merton' Flore.s also supports our hypothesis. It commemorates the acts of Edward I, the new king's father, emphasising his military victories — his contribution, as the Lord Edward, to the defeat of Simon de Montfort, and his victories, as king, over the Scots. It is noteworthy that the new ordo compiled for Edward II's coronation included a prayer that the king should succeed against enemies and rebels.90 Its interest in the Barons' War, apparent in the long resume of events leading up to the battle of Evesham, and the eloquent paean on peace91 are explicable if the work was written at Edward II's accession, because the king was again under baronial pressure.92 The author must surely have been thinking of his own times when he wrote that even if Henry III did have shortcomings, they did not vitiate his heirs' monarchical rights, because 'both divine and human law provide that once the fault or defect ceases, the punishment is never transmitted to the heirs, lest the iniquity of the father should attend the son and the punishment exceed the crime.'93 (That the author wrote 'heirs' not 'heir' shows that he was thinking further ahead than Edward I). Moreover, the only shortcoming of Henry III which the author specifically mentions is his 'intolerable prodigality' which made the baronial take-over necessary:94 this can be regarded as a veiled admonition to Edward II lest the same fault on his part should lead to a similar result. Additional support for the suggestion that the 'Merton' Flore.s was started immediately after Edward I's funeral is provided by the Com89 See above pp. 247-8. 90 'Tribue ei, optimus Deus, .. ut sit fortissimus regum, triumphator hostium, ad opprimendas rebelles et paganas nationes; sitque suis inimicis satis terribilis, praemaxima fortitudine regalis potentiae. Foedera (Record Commission edition), ii. pt. i, 34. 91 See above pp. 254-5. 92 See Schramm, op. at. pp. 76, 207-208, and M. McKisack, The Fourteenth Century (Oxford, 1959), pp. 4 sq. 93 Flores, iii. 254. The coronation ordo of 1308 laid especial emphasis on Edward II's succession by hereditary right. 94 Flores, iii. 254.
The Continuations of the Flores Historiarum
261
mendatio Lamentabilis. This lament, described by Stubbs as a 'sort of mortuary Eloge or funeral sermon,' must have been written soon after the king's death. The close similarity between its point of view and that of the 'Merton' Flores suggests that the latter was written at about the same time. The Com mendatio Lamentabilis, like the 'Merton' Flores, emphasises that the Lord Edward helped Henry III defeat the barons: the knights say that they, with the great Edward, 'snatched England, from the lion's jaw when they freed Henry III, like Daniel, from that wild beast, Simon de Montfort.'95 And it too praises Edward when king for his wisdom, bravery and success in war which won for England the benefits of peace.96 With regard to the 1307-1326 continuation, it seems likely that it was written after the deposition of Edward II and before the fall of Isabella and Mortimer in October 1330. It may have been intended for presentation to Edward III on the occasion of his coronation, and is best understood as a friece justificative for Isabella's and Mortimer's coup. It reads as if it were intended to substantiate the accusations levelled against Edward II in the assembly held on 13 January, 1327, which deposed him. The chronicle of the Cistercian abbey of Pipewell gives the fullest account of the proceedings. It records : There, by common assent of all, the archbishop of Canterbury declared how the good King Edward when he died had left to his son the lands of England, Ireland, Wales, Gascony and Scotland in good peace; how Gascony and Scotland had been as good as lost by evil counsel and evil ward; how, further, by evil counsel the son had destroyed the greater part of the noble blood of the realm, to the dishonour and loss of himself, his realm and all the people; and how he had done many other marvels.97 It will be noticed that the charges made against Edward are precisely those enlarged on in the 1307-1326 continuation. 98 Moreover, assuming that Murimuth preserves the lost ending of the chronicle, its account of 95 Chrons. Edward I and Edward //, ii. 14. The Commendatio Lamentabilis compares Edward I with David and Solomon. It is noteworthy that prayers in the coronation ordo of Edward II ask God to endow the king with the qualities of various Old Testament figures such as Abraham, Moses, David and Solomon; see Foedera, ii, pt. i, 33-34. 96 Chrons. Edward I and Edward II, ii. 16 and passim. See above p. 254. 97 The translation is that in M. V. Clarke, 'Committees of Estates and the Deposition of Edward II,' Historical Essays in Honour of James Tail, ed.J. G. Edwards, V. H. Galbraith and E. F.Jacob (Manchester, 1933), p. 35. (The Anglo-Norman text is printed in ibid. p. 44 and in M. V. Clarke, Medieval Representation and Consent (London-New York-Toronto, 1935), pp. 194-195). The accusations against Edward II appear in an expanded form in Adam de Orlton's Apologia of 1334; Twysden, Decem Scriptores, c. 2765, and Foedera, ii. pt. i, 650. 98 See above p. 257.
262
Legends, Traditions and History in Medieval England
Edward II's abdication and Edward Ill's accession had a propagandist element, for clearly the chronicler was trying to make what was illegal seem legal. Its assertion that the abdication was Voluntary 1 reflects Isabella's and Mortimer's insistance on a formal resignation,' and Edward's alleged confession of his unworthiness to rule may well be a fiction suggested by the proceedings of the deposition assembly. Finally, the author calls the assembly itself a parliament: but it hardly deserved the appellation. 100 If our hypothesis that the 'Merton' Flores and the 1307-1326 continuation are official histories is correct, it provides a possible explanation of the fact that each stops just before the end of the reign which it describes. Perhaps only one complete copy of the original draft of each chronicle was made in Westminster Abbey. These complete copies were presented to the kings. Edward II's book could well have been lost cfuring the upheaval of the deposition, and meanwhile at Westminster no other complete copy had been made; perhaps the monks, satisfied with the 'Westminster' Flores, were not sufficiently interested. Therefore, in order to meet outside demand, it was necessary for scribes to use as exemplar the incomplete copy made for Merton priory. With regard to the 1307-1327 continuation, political considerations may explain the survival of only one early copy, itself incomplete. Edward Ill's copy might have been deliberately destroyed after the fall of Isabella, because its favourable attitude to them was no longer acceptable. The same factor could have resulted in the destruction at Westminster of other early copies and it is likely that it would have discouraged the making of more. Official histories were rare in medieval England and those which seem to deserve the title were by secular clerks.101 Thus, in the twelfth 99 See W. Stubbs, The Constitutional History of England (fifth edition, Oxford, 1891-1906, 3 vols.), ii. 379-380. 100 See B. Wilkinson, 'The Deposition of Richard II and the Accession of Henry IV,' EHR, 54 (1939), 223-230. 101 Before the Norman Conquest, the Alfredian Anglo-Saxon chronicle and the version written under Edward the Elder were almost certainly written to please Alfred and Edward and are biased accordingly: Plummer argues that the former was written under Alfred's 'direction and supervision'; see Two of the Saxon Chronicles Parallel, ed. C. Plummer, on the basis of an edition by J. Earle (Oxford, 1892, 1899, 2 vols., reprinted with two notes by D. Whitelock, 1952), ii. civ. For the 'official' element in the Edwardian chronicle see The Chronicle of Aethelweard, ed., with an English translation, Alistair Campbell (Nelson's Medieval Texts, 1962), pp. xxvii-xxxii. It should also be noted that the Norman accounts of William the Conqueror, especially that by William of Poitiers (Hutoire de Gillaume le Conquerant, ed., with a French translation, R. Foreville, Les classiques de I'histoire de France au moyen age, xxiii, Paris, 1952), are official histories in the sense that they praise the king and justify the Conquest. Moreover, the Norman work which formed the basis of L'estoire de la
The Continuations of the Flores Historiarum
263
century the chronicle of Roger of Howden, a clerk in the king's service, has features of an official history. 102 In the fourteenth century the De Gestis Mirabilibm Regis Edwardi Tertii by Robert of Avesbury, recorder of the archbishop of Canterbury, could, judging from its eulogistic tone, have been written to please Edward III, even at his command. And in the fifteenth century a chaplain of the royal household wrote the Gesta Henrici Quinti, 'an outstanding piece of propaganda, designed to justify the king's character and policy.' 103 Nevertheless, it is not unlikely that the monks of Westminster should have produced two official histories in the first third of the fourteenth century. Initially it is possible that Matthew Paris undertook to write the Flores Historiarum for Westminster Abbey as a result of royal encouragement. In October, 1247, when he attended the translation of Edward the Confessor at Westminster, Henry III told him to record the proceedings;104 the translation is duly noticed in the Flores which he completed soon atterwards. l05 *Such an undertaking in response to the king would accord with evidence supplied by his Chronica Majora and Historia Anglorum which shows that at about this time Matthew adopted a more favourable attitude to Henry. Both works in their original form are notable'for their strong anti-royalist bias. But in 1250 Matthew decided to modify his invective. He marked some passages in the Chronica Majora as offensive: many of these were omitted by the scribe making a new copy for St. Albans in 1250 or soon after. Matthew also erased some passages, substituting milder ones, and he revised the Historia Anglorum
guerre sainte (ed. Gaston Paris (Paris, 1897)) and of the Itinerarium Peregrinorum et Gesta Regis Ricardi (printed in Chronicles and Memorials of the Reign of Richard /, ed. W. Stubbs (Rolls Series, 1864, 1865, 2 vols.), i), eulogizes Richard I, in chivalric terms, and justified his behaviour during the crusade. Cf.J. G. Edwards, 'The Itinerarium Regis Ricardi and the Estoire de la Guerre Sainte' mHistorical Essays in Honour of James Tail, ed. Edwards, Galbraith and Jacob, pp. 59-77, and Das Itinerarium Peregrinorum, ed. H. E. Mayer (MGH, Schriften, xvii, Stuttgart, 1962), pp. 80-102. The Gesta Stephani (ed., with an English translation, K. R. Potter (Nelson's Medieval Texts, 1955)), which may also be of foreign authorship (see ibid. pp. xxx-xxxii) praises King Stephen as a chivalric hero. 102 Printed Chronica Magistri Rogeri de Hoedene, ed. W. Stubbs (Rolls Series, 1868-1871, 4 vols.). See also F. Barlow, 'Roger of Howden', EHR, 65 (1950), 352-360, and D. M. Stenton, 'Roger of Howden and Benedict', EHR, 68 (1953), 574-582. 103 J. S. Roskell and F. Taylor, 'The Authorship and Purpose of the Gesta Henrici Quinti: i,' BJRL, 53 (1970-1971), 428. The Gesta is printed Henrici Quinti, Angliae Regis, Gesta, ed. B. Williams (English Historical Society, 1850). 104 Matthew Paris, Chronica Majora, ed. H. R. Luard (Rolls Series, 1872-1883, 7 vols.), iv. 644-645. For a translation of the passage see Vaughan, Matthew Paris, p. 3. Henry's interest in Matthew Paris' historical work also appeared when he visited St. Albans in 1257, on which occasion he gave Matthew a list of the canonized kings of England and the names of two hundred and fifty English baronies. 105 See Vaughan, op. tit. pp. 108-109.
264
Legends, Traditions and History in Medieval England
even more drastically.1053 Perhaps the dedication of his Life of Edward the Confessor, in Anglo-Norman verse, which he wrote at about the same time, to Henry's queen, Eleanor of Provence, is also indicative of a rapprochment with the royal family.106 There is more evidence for Edward I's interest in history than there is for Henry Ill's. His appreciation of the value of chronicles as records of precedents, and of the uses of historiography for propaganda is wellknown. His clerks searched the chronicles for evidence relating to the case of the competitors for the Scottish throne in 1291, and he wrote to the monasteries ordering the monks to search their chronicles for similar material.107 The evidence collected was synthesized by the royal clerk, John of Caen, into the official account of the proceedings, the Processus Scotiae, which gave an account of Anglo-Scottish relations from 901 to 1252 strongly biased in favour of Edward's claims.108 Edward also ordered chroniclers throughout England to insert in their works copies of the competitors' letters of submission to Edward as overlord.109 Furthermore, there was a literary interest in history at court, fostered by the cult of King Arthur. The Dominican, Nicholas Trevet, wrote a chronicle in Anglo-Norman dedicated to Edward I's daughter Mary.110 And a number of books and works of art (including a coronation throne) were executed at Westminster for the king and members of the royal family.111 Moreover, the possibility of French influence cannot be discounted. In the late thirteenth and early fourteenth centuries cultural connections between England and France were close, reinforced by the marriage of Edward I with Margaret, daughter of Philip II of France (in September 1299), and by Edward II's marriage with Isabella, daughter of Philip the Fair (in January 1308). The official chronicle of the kings io5a Ibid. pp. 64-5, 117-124. 106 See La estoire de seint Aedward le ret, ed., in facsimile, M. R.James (Roxburghe Club, 1920), pp. 12, 17. For date of work see Vaughan op. cit. p. 168. 107 Edward I's appeal to historical precedent in the Scottish succession case is fully discussed by E. L. G. Stones, 'The Appeal to History in Anglo-Scottish Relations between 1291 and 1401: i,' Archives, 9 (1969), 11-21 passim. For Edward I's letters to the religious houses see ibid. p. 12 and n. 8, J. Bain, Calendar of Documents relating to Scotland (Edinburgh, 1881-1888, 4 vols.), ii. 110, no. 470, and F. Palgrave, Documents and Records illustrating the History of Scotland (Record Commission, 1837), pp. 6067. 108 Foedera, i. pt. ii, 762-784. See V. H. Galbraith//^/srtonoa/ Research in Medieval England (London, 1951), P- 36, ancf Stones, op. cit. pp. 17-19. Edward I again appealed to history to defend his claim to overlordship of Scotland against Boniface VIII; see ibid. pp. 19-21. 109 See Bain, op. cit. ii. 122, nos. 503-504, and Palgrave op. cit. pp. xcvi-xcvii and n. no This chronicle is unpublished; see M. D. Legge, Anglo-Norman Literature and its Background (Oxford, 1963), pp. 299-302. 111 See Peter Brieger, English Art 1216-1307 (Oxford, 1957), chapter xii passim.
The Continuations of the Flores Historiamm
265
of France, the Grandes chroniques, described by its modern editor as 'a monument raised to the glory of the French monarchy,'112 provides a continuous history from the origins of the monarchy in the sixth century, until, ultimately, the end of the fifteenth century. It was begun in the mid-thirteenth century and later continued by the monks of the royal foundation of St. Denis, near Paris. If we are correct in our hypothesis that the 'Merton' Flores and the 1307-1327 continuation are an official history by the monks of Westminster, the English monarchy had for a short period a comparable work produced by a comparable royal foundation.
112 Les Grandes chroniques de France, ed.J. Viard (Soc. de 1'histpire de France, 1920-1953, 10 vols.), i. viii. See also A. Molinier, Les sources de I'histoire de France (Paris, 1901-1904, 5 vots.), Hi, no. 2530, and the unpublished Ph. D thesis for Edinburgh University (1969) by Sarah Murphy Farley, French Historiography in the Later Middle Ages with special Reference to the 'Grandes chroniques de France.'
Note For an additional note to pp. 259-61 see below pp. 332-33.
This page intentionally left blank
11 The Alleged Rape by Edward III of the Countess of Salisbury1 THE story of the rape by Edward III of the countess of Salisbury in 1342 has five component episodes: (1) Edward III falls in love with the countess when he visits the castle of William Montagu, first earl of Salisbury (1337-44), during a campaign against the Scots; (2) in order to see the countess again Edward holds a grand tournament in London to which the earl of Salisbury is commanded to come with his wife; (3) Edward returns to the earl's castle and rapes the countess; (4) the countess confesses to her husband; (5) the earl takes consequent action. The full story of this villain cas is told by the Hainault chronicler, Jean le Bel, who wrote more than fifteen years after the alleged incident.2 The last three episodes are briefly noticed in the Chronographia ^.egum Francorum, a chronicle compiled in the early fifteenth century in the abbey of St. Denis in Paris, based on earlier historical collections made at St. Denis.3 The same three episodes are similarly noticed in other continental chronicles: these will not be cited in evidence here because they are closely related to the Chronographia* 1. I am indebted to my colleague Dr. M. C. E. Jones, and to my husband K. W. Gransden, for kindly reading this article in typescript and making useful comments and suggestions. I am also grateful to Professor E. L. G. Stones for help on specific points. However, I am responsible for any errors or omissions. 2. Chronique de Jean le Be/, ed. J. Viard and E. Deprez (Soc. de 1'histoire de France, 1904, 1905, 2 vols.), i. 290-4; ii. 1-4, 30-33. For the date when these parts of the chronicle were written see ibid. i. pp. xiv-xvi. 3. For the composition of the Chronographia see the printed edition, Chronographia Kegum Francorum, ed. H. Moranville (Soc. de 1'histoire de France, 1891-7, 3 vols.), i.pp. xliii-xlvii. 4. These episodes are in: Chronique normande du xive Siecle, ed. A. and fi. Molinier (Soc. de 1'histoire de France, 1882), pp. 54, 59-60; Istore et cronique de Flandres, ed. K. de Lettenhove (Brussels, 1879-80, 2 vols.), ii. 6, 9; the Chronique de Flandre in L. Delisle, Histoire du chateau et des sires de Saint-Sauveur-le-Vicomte suivie de pieces justificative! (Valognes, 1867), pp. 89-90. These chronicles are in French and the relevant passages in all three are almost identical. Scholars disagree concerning the relationship of the Chronographia to the Chronique normande and the Istore et cronique de Flandres. Moranville, op. cit. i. pp. v-xxvi passim argues that the Chronique normande and the Istore used the same St Denis source as the Chronographia. On the other hand H. Pirenne, *Les sources de la chronique de Flandre jusque'en 1342', in Etudes d* histoire du moyen age dediees a Gabriel Monod, ed. E. Lavisse (Paris, 1896), pp. 367-370, and his 'L'Ancienne chronique de Flandre et la Chronographia R.egum Francorum*, in Compte rendu de seances de la commission royale d^histoire ou recueil de ses bulletins, cinquieme serie, tome viii, ive bulletin (1898), pp. 199-208, and A. Molinier, in Revue historique, Ixvii (1898), 90-92, and his
268
Legends, Traditions and History in Medieval England
Episode (5) differs in substance in Le Bel from the account in the Chronographia.1 Froissart (the first and second redactions of whose chronicle follow Le Bel closely) explicitly condemned Le Bel's story. In the first redaction he suppressed the prediction of the rape at the end of episode (i) and the rape itself, episode (3). In the second redaction (the version in the Amiens MS.) he again suppressed the prediction of the rape, and added instead the famous description of the game of chess played by the king and the countess.2 Moreover, besides omitting episode (3), he inserted, s.a. 1345, a critique of its veracity. He writes: You have heard me speak above of Edward's love for the countess of Salisbury. The chronicle of Jean le Bel speaks of this love less properly than I must, for, please God, it would never enter my head to incriminate the king of England and the countess of Salisbury with such a vile accusation. If respectable men ask why I mention that love, they should know that Jean le Bel relates in his chronicle that the English king raped the countess of Salisbury. Now, I declare that I know England well, where I have lived for long periods mainly at the royal court and also with the great lords of that country. And I have never heard tell of this rape although I have asked people about it who must have known if it had ever happened. Moreover, I cannot believe and it is incredible that so great and valiant a man as the king of England would have allowed himself to dishonour one of the most noble ladies of his realm and one of his knights who had served him so loyally all his life.3 In the third redaction of his chronicle Froissart suppressed all reference to the countess of Salisbury at the grand tournament of 1342, and also the additions relating to the rape-story in the second redaction.4 Kervyn de Lettenhove, after research on the records and other chronicles of the period, exposed some errors in Le Bel's story Les sources de rhistoire de France (Paris, 1901-6,6 vols.), iv. 23-26, argue that the Chronique normande is the primary source of the other chronicles, However, if this is the case' it would not vitiate the probability that the author of the Chronique normande used some material from St. Denis in common with the author of the Chronographia. Certainly the latter cannot have copied his account of our episode (5) from the extant text of the Chronique normande, because the Chronographia has one passage, et ea bereditavit filiam suam quia masculinum'Jheredem non habebat (see below p. 337) which has no equivalent in the Chronique normande. It should be noted that the passages in question are not in the Grandes chroniques written at St. Denis. The latter copies many passages from the same source as the Chronographia until well into the fourteenth century. However, from 1340 to 1350 it has an original account apparently written by a monk of St. Denis contemporaneously with the events recorded, and was one source used by the Chronographia; see Grandes chroniques de France, ed. J. Viard (Soc. de Fhistoire de France, 1920-53, 10 vols.), viii. pp. xiii-iv; ix. pp. iv-vi, and below p. 339 n. 4. 1. See below pp. 273 ff. 2. See Chroniques de Jean Froissart, ed. S. Luce, G. Raynaud and L. and A. Mirot (Soc. de Thistoire de France, 1869-1966,14 vols., in progress), ii. 135, 340-2. 3. Ibid. iii. pp. xviii-xix, 293 (cf. i. pp. Ixiv-lxv). 4. Ibid. iii. pp. 198, 294 ff. (cf. i. p. Ixxiv).
The Alleged Rape by Edward HI of the Countess of Salisbury
269
1
and concluded that Froissart's disbelief was justified. Le Bel's editor, Jules Viard, could only point out in Le Bel's defence that part of the story is also in the Chronographia and two other continental chronicles: he did not mention that these chronicles have a common historiographical archetype.2 An examination of the story concerning the alleged rape shows that it is a mixture of fact, error and the unverifiable. The five episodes will be discussed in turn. (Le Bel and the Chronographia will be provisionally treated as primary authorities, postponing for now the consideration of the possibility that both relied on a common source.) Episode (i). Edward III went to relieve 'the castle of Salisbury', where the countess was staying, after the retreat of David Bruce. It was being held for the earl of Salisbury by 'a gentle esquire, the son of the earl's sister, who bore the name William Montagu after his uncle.'3 Le Bel places this episode in 1342. It could have happened in December 1341 or January 1342, when Edward was campaigning in the north.4 The castle in question must have been Wark-on-Tweed which Edward had granted to the earl of Salisbury in 1329 for life (and in 1333 in fee tail).5 There is no known evidence corroborating either that the countess of Salisbury was at Wark at this time, or that the castle was being held, in the absence of the earl (who was a prisoner in France from Easter 1340 until early in June I342),6 by the young William Montagu. Moreover, the passage almost certainly has one error. The young William Montagu who succeeded to the earldom of Salisbury in 1349 is generally accepted as the son of the first earl and his wife Katharine. Episode (2). The grand tournament in London to which King Edward summoned the earl of Salisbury. Here for the first time Le Bel names the countess of Salisbury. He calls her Alice.7 In fact the wife of William Montagu first earl of Salisbury was Katharine. Le 1. Lettenhove examines the story in Oeuvres de Froissart, Chroniques, ed. Kervyn de Lettenhove (Brussels, 1867-77, 25 vols.), iv. 458-62 (cf. iii. 517-24). 2. LeBel,ii.3on. 2; cf. above p. 333^4. 3. Ibid. i. 285-94 passim. Froissart follows Le Bel in stating that the young William Montagu was the first earl's nephew; Froissart, ed. Luce, ii. 129, 346. 4. The statement that Edward III visited Wark is accepted as probably true by C. J. Bates, The Border Holds of Northumberland, i (Newcastle, 1891), pp. 359-69, who dates the visit to sometime between 13 and 20 Dec., 1341 (ibid. p. 364). 5. C[alendarof]P[atent]R[olls],1327-1330,.,p. 392; CPR, 1330-1334, p. 462. Cf. K. H. Vickers in A History of Northumberland (issued under the direction of the Northumberland County History Committee, Newcastle, 1893-1926, 12 vols.), xi. 52. It is not possible to substantiate Lettenhove's assumption that Wark castle was being held by Edward Montagu in 1341/2. See Lettenhove, op. cit. iii. 516. 6. For the agreement between William Montagu, earl of Salisbury, and Philip VI, for the earl's release in exchange for the earl of Moray, dated 2 June 1342, see J. du Tillet, Recuei/ des guerres et traicte^ dtentre les roys de France et d*Angleterre (Paris, 1588), fo. 62V, and Lettenhove, op. cit. iii. 524-6. Apparently the earl of Salisbury was actually released on 4 June; ibid. iv. 459. 7. L,eBe/yii. 2.
270 Legends, Traditions and History in Medieval England Bel's error could have been the result of one of two simple confusions.1 On the one hand he may have confused Katharine with Alice, the aged widow of Thomas of Lancaster. This Alice, who died in 1348 was suo jure countess of Lincoln and Salisbury. If anyone had been asked the name of the countess of Salisbury whom King Edward had (allegedly) raped, Alice's name might well have come to mind, especially as her matrimonial career had won her some notoriety.2 On the other hand Le Bel could have confused the countess Katharine with the wife of Edward Montagu, brother of the first earl of Salisbury, who was called Alice.3 If this was the cause of the error, it would partly explain another probable mistake in Le Bel, the statement that the young William Montagu was the earl's nephew.4 He was of course the nephew of Edward Montagu and Alice. In conclusion, the mistakes made by Le Bel (which are reproduced by Froissart) concerning the earl of Salisbury's family connections could be the result of ignorance. Le Bel dates the grand tournament to the middle of August I342.5 There is no evidence to corroborate that a tournament was held at that time. Le Bel makes two statements about the tournament which suggest that he misdated a conflation of two tournaments held in the spring of 1342. He records the death in the jousts of John, the eldest son of Henry Beaumont. Adam Murimuth records John's death at the tournament at Northampton on 14 April.6 Le Bel also mentions the presence of William, count of Hainault; Murimuth states that a tournament was held in the count's honour at Eltham on 9 May.7 However, there is an objection to the hypothesis that Le Bel intended to refer to a grand tournament in the 1. For the suggestion that Le Bel may have confused the name deliberately see Margaret Galway, * Joan of Kent and the Order of the Garter', University of Birmingham Historical Journal, i (1947), 36-40, who argues that the countess of Salisbury in question was Joan of Kent who married William Montagu, second earl of Salisbury (1344-97) probably in 1340/1 and certainly before 1343; she was divorced in or before October 1349. See G. E. Cfockayne], Complete Peerage, xi. 389-90 and n.a. 2. See ibid. vii. 687-8. In 1317 Alice, who was born in 1281, was carried off by Sir Richard de St. Martin who claimed to have known her before her marriage to Thomas of Lancaster. In 1335/6 she married as her third husband Sir Hugo de Freyne (by whom she had previously been ravished), despite a vow of chastity. 3. See ibid. ix. 84-85. It should also be noted that William Montagu, second earl of Salisbury, contracted a marriage with Alice, younger daughter of Thomas of Brotherton before marrying Joan of Kent; ibid. xi. 389 n. k. 4. Above p. 269, and below 273 n. 3. 5. L,e Bel, ii. 2. Froissart borrows Le Bel's date in the first redaction of his chronicle, but alters it to Candlemas (2 Feb.) in the second redaction; Froissart, ed. Luce, iii. 2, 197. Tournaments on neither date are mentioned by D. Sandberger, Studien tiber das Kitterturn in England vornehmlich wahrenddes 14 Jahrhunderts (Berlin, 1937). 6. Adae Murimuth Continuatio Chronicamm. Robertus de Avesbury De Gestis Mirabilibus Regis Edwardi Tertii, ed. E. M. Thompson (Rolls Series, 1889), p. 124. The inquisition post mortem, dated 26 June 1342, does not give the date of the death of John Beaumont, son of the late Henry Beaumont earl of Buchan; Calendar of] Inquisitions] p[ost] mprtem viii, n. 381. (Cf. ibid. viii. n. 271 and G. E. C[ockayne,] op. cit. ii. 61). 7. LeBe/, ii. 3; Murimuth, p. 124. Cf. Sandberger, op. cit. p. 54.
The Alleged Rape by Edward HI of the Countess of Salisbury
271
spring of 1342. The earl of Salisbury could not have taken his wife to a tournament at that time because he was not released by the French until early in June I342.1 Therefore Le Bel must presumably have been referring to a summer tournament. Possibly the fact that he records an embassy to Edward III from Brittany immediately after the tournament provides a clue to the date.2 This embassy from John de Montfort's party was, Le Bel states, immediately followed by another one. (He incorrectly asserts that the second embassy was headed by the duchess de Montfort herself.)3 He records that as a result of these embassies which asked for help against Charles of Blois, Edward sent an expedition led by Robert of Artois and the earl of Northampton. It is known that representatives of John de Montfort's cause were active negotiating with Edward III in 1341 and until the summer of I342.4 On 23 July 1342 Edward at Windsor confirmed an agreement, of 21 February 1342, with de Montfort's party.5 The agreement stipulated the terms on which Edward would send military aid under Artois and Northampton. Possibly Le Bel had in mind the negotiations resulting in this indenture. If so, the grand tournament could be dated to sometime after early June (when the earl of Salisbury was released by the French), and before 23 July (the date of the indenture). It may be suggested that Le Bel's erroneous date, mid-August, for the grand tournament, arose from confusion with the date when the expeditionary force sailed for Brittany -14 August. Episode (3). The rape. Le Bel tells the story in highly-coloured dramatic prose. The Chronographia states briefly: 'The king of England violated the wife of the earl of Salisbury.' Both authorities place the alleged event in 1342, in the same chronological context; after the earl of Salisbury left with Robert of Artois on the Breton expedition, on 14 August. (Le Bel infers that Edward sent the earl purposely, so that he could pursue his intention with the duchess ),6 and before Edward himself sailed for Brittany, in mid-October (he landed at Brest on 19 October).7 It is not impossible that between i. Seeabovep.269andn. 6. z. LeBe/,11.4-6. 3. Ibid. ii. 7-8. Cf. Chron. norm. ed. Molinier, p. 259 n. i, and A. Le Moyne de la Borderie, Histoire de Bretagne (Rennes and Pads. 1896-1914, 6 vols.), iii. 447. The Chronographia also incorrectly states that the duchess de Montfort came on an embassy to Edward III in 1342; Chronographia, ii. 196. 4. See Murimuth, p. 121; Foedera (Record Commission edition), ii. pt. 2, 1189, 1205; Borderie, op. cit. iii. 447; J. Le Patourel, 'Edward III and the kingdom of France/ History, xliii (195 8), 187. 5. Pfublic] R[ecord] O[ffice], Duchy of Lancaster, Cartae Miscellaneae (D.L.36), i. no. 198.1 owe this reference to Dr. M. C. E. Jones, who also called my attention to a payment to Amaury de Clisson and Bernard de Guyngen, for expenses for a visit to England, dated 7 March 1342; P.R.O.,E 403/323, m. 31. 6. LeBe/,ii.$o. 7. M. C. E. Jones, DucalBrittany 1364-1399 (Oxford, 1970), p. 143, citing the chronicle of Richard Lescot. This date is corroborated by a warrant under the privy seal, dated at Brest, 19 Oct. 1342; P.R.O., C. 81, 287, no. 15132. However, Adam Murimuth dates Edward Ill's departure for Brittany 23 Oct.; Murimuth, p. 128. Perhaps Murimuth
272
Legends, Traditions and History in Medieval England
these dates Edward paid a fleeting visit to Wark, but there is no known evidence of it.1 Episode (4). The countess's confession. This is described in graphic detail by Le Bel and noticed more briefly in the Chronographia? Both authorities, which are uncorroborated by other evidence, date it to after the return of King Edward and the earl from Brittany: they landed at Weymouth on 2 March 1343.3 Episode (5). The action taken by the earl of Salisbury as a result of the confession. Here Le Bel and the Chronographia have, besides points of coincidence, both complementary and conflicting information. Le Bel states that the earl of Salisbury was grief-stricken at the countess's revelations, particularly because of the king's manifest ingratitude for his faithful service. He therefore undertook to give half his property to the countess and their son, who was twelve years old. He then went to London, confronted the king with his iniquity and handed back everything he held of the king, 'for the aid of his son.' He ended his harangue: 'Neither you nor anyone else in this land will ever set eyes on me again.' Thereupon he left to fight the Moors in Spain, and was killed at the siege of Algeciras.4 The Chronographia similarly represents the confrontation between the king and the broken-hearted, angry earl, but asserts that the earl took his friends with him. He disseized himself of all his lands in the presence of his peers and *ea hereditavit filiam suam quia masculinum heredem non habebat, sic quod uxor eius haberet doagium quamdiu viveret.' Next, he left the court, defied Edward and crossed to France. He went to Philip VI and handed him letters of confederation (litteras de confederations) of Oliver de Clisson and
derived his date from knowledge that governmental business under the privy seal continued to be conducted at Portsmouth until 23 Oct.: this is demonstrated by the fact that the extant privy seal warrants, except the one mentioned above, were issued at Portsmouth from 19 to 23 Oct.; P.R.O., C. 81. 287, nos. 15129-15131, 15133-15155. (There is then a gap in the dates of the privy seal warrants until i Nov., when they are dated at Plouguerneau and Brest. For problems in reconstructing the royal itinerary see the next note.) 1. Kervyn de Lettenhove (Froissart, iv. 459) concluded, on the evidence of the documents printed from the chancery rolls in Foedera, that Edward III did not go north in the period between 14 Aug. and 23 Oct. 1342. However, H. C. MaxwellLyte, Historical Notes on the Use of the Great Seal of England (London, 1926), pp. 71-72, 80, 236, 247-8, 251 ff., 405-9, has demonstrated that the chancery records do not provide infallible evidence for the king's itinerary at this period. See also P. Chaplais, English Royal Documents, King John to Henry VI, 1199-1461 (Oxford, 1971), pp. 26, 39-45. Nevertheless, a search through the privy seal warrants (P.R.O., C. 81/285-287) and the wardrobe accounts (P.R.O., E. 101/389/14, E. 101/390/1 and E 36/204) corroborates the evidence of the chancery records in this instance. I am indebted to Miss Elizabeth Danbury for making a preliminary search of E. 36/204. The expenses of Edward III at Newcastle-on-Tyne recorded in E. 101/389/40, were almost certainly all incurred during his Scottish campaign in the winter of 1341. 2. LeBe/, ii. 32-33; Chronographia, ii. 204. See below pp. 342-3. 3. Calendar of]C[lose] R[ol/s], 1345-1546, p. 97. 4. Le Bel, ii. 3 3-34.
The Alleged Rape by Edward III of the Countess of Salisbury
273
Godfrey de Harcourt with Edward III. Soon afterwards he left France 'and was never again seen by anyone'.1 There is nothing in the public records to confirm that Edward III and the earl of Salisbury had this bitter quarrel after their return from Brittany. Nor is there any evidence confirming the earl's alleged arrangements for his property: on the contrary, on his death Katharine had dower in his possessions, and subsequently, in 1349, William Montagu, second earl of Salisbury, had livery of the inheritance, in accordance with the normal procedure.2 The statement in the Chronographia that the earl had no male heir is contrary to the evidence that the second earl of Salisbury was son and heir to the first earl.3 Similarly, its assertion that the earl defied the king cannot be substantiated. Moreover, Le Bel's statement that the earl was killed at Algeciras is demonstrably untrue; he died of a wound inflicted in the tournament at Windsor in April 1344.4 However, episode (5) in both Le Bel and the Chronographia has some true facts. Le Bel correctly associated the earl with the war in Spain. By 30 August 1343 he was with the earl of Derby in Castile, negotiating on Edward Ill's behalf with King Alfonso, and he probably later fought the Moors.5 The precise date when he left England is apparently unknown, but he could have visited Paris before joining the earl of Derby (who left England before the middle of March I343)6 in Castile. It is, therefore, not impossible that he visited Paris on the way. If so, he could have betrayed Oliver de Clisson and Godfrey de Harcourt to King Philip.7 Oliver de Clisson, one of the principal Breton lords who had supported Charles of Blois' claim to the duchy of Brittany,8 and Godfrey de Harcourt, a powerful Norman lord, had both been captured by the 1. Chronographia, ii. 204-5. The Chronique normande, p. 59, omits the statement that the earl gave his inheritance to a daughter because he had no male heir. The Istore et croniques de Flandres, p. 9, and the Flemish chronicle in Delisle, op. cit. p. 90, omit the statement that the earl had no male heir. 2. See CCR, 1343-1346, p. 307; CCR, 1346-1349, p. 450; CCR, 1349-13^4, p. 107; Cal. Inq.p.m. ix. nos. 64, 310. 3. See Cal. Inq. p.m. viii. no. 532. For a suggestion that Le Bel's statement that the young William Montagu was the nephew of William Montagu first earl of Salisbury (above p. 26 9 and n. 3), was intended to cast doubt on the second earl's legitimacy, see Galway, loc. cit. pp. 39-40 (cf. ibid. pp. 24-25). 4. This is recorded in a version of Murimuth's chronicle; Murimuth, p. 232. 5. CCR, 1343-1346, p. 226 (a transcript of this entry on the Close Roll is printed in Foedera, ii. pt. 2,1232). Cf. G.E.C., Complete Peerage, xi. 387 and n.h. 6. See Kenneth Fowler, The King's Lieutenant, Henry of Grosmont, First Duke of Lancaster 1310-1361 (London, 1969), pp. 45-46. Cf. Froissart, ed. Lettenhove, iv. 459-60. 7. This possibility is mentioned by Raymond Gazelles, La socitte politique et la crise de la royaute sous Philippe de Valois (Paris, 1958), p. 153. The alleged incident is also mentioned by Simeon Luce, Histoire de Bertrand du Guesclin (Paris, 1876), p. 47 n. 2, who cites P. H. Morice, Histoire ecclesiastique et civile de Bretagne (Paris, 1750, 1756, 2 vols.), i. 268; however Morice was here citing the chronicle of Flanders, not an independent source (see above p. 267 and n. 4). 8. For Oliver de Clisson, brother of Amaury de Clisson (for whom see above p. 27 1 n. 5) see Gazelles, op. cit. pp. 143,147,15 3-4.
274
Legends, Traditions and History in Medieval England
English at Vannes in December 1342, and had apparently come to terms with their captors.1 Harcourt rebelled against King Philip in I343. 2 Philip carried out a purge of these and other traitors in May 1343; Clisson was executed in Paris, on 2 August 1343, and Harcourt was exiled. How Philip discovered Clisson's treason is unknown (the records of his and Harcourt's trial are lost),3 but the possibility of a revelation by the earl of Salisbury cannot be discounted. On the evidence available, allowing the chroniclers a due margin of error, it is surely impossible to decide how much truth there is in the story, with its complex of related incidents, of the rape of the countess of Salisbury. There is undoubtedly a possibility that Edward III raped the countess Katharine (or some other lady). The contrast between the ideals of chivalric society (which are reflected in Froissart's protestation concerning the rape-story), and the actual practice of those violent times is a truism. However, irrespective of how much truth the rape-story contains, it can be regarded from another angle. It can be considered as a piece of political propaganda. Propaganda was one by-product of the Hundred Years War. Probably there was a propagandist element in the anti-French bias evinced by the English chroniclers, especially by Adam Murimuth and Geoffrey le Baker. Murimuth argues at length the legal case for Edward Ill's claim to the crown of France, and stresses that the French were opposing the 'true heir' (John de Montfort) in Brittany.4 He consistently refers to Philip VI as Philip de Valois, not as king of France. Baker is more virulent, calling Philip a pseudo-king and tyrant, who planned to annihilate England, and was, like his subjects, inflated with pride.5 In France, the abbey of St. Denis, Paris, was the home of the official histories of the kings of France.6 During the Hundred Years War its historical productions 1. Chronographia, ii. 202-3 J Grandeschroniques, ed. Viard, ix. 241-2. 2. Gazelles, op. cit. pp. 152-3. Cf. Grandes chroniques, ed. Viard, ix. 243. 3. Gazelles, op. cit. p. 15 5 n. 6. 4. Murimuth, pp. i oo-i ,121. 5. Chronicon Galfridi le Baker de Swymbroke, ed. E. M. Thompson (Oxford, 1889), pp. 54,55,123,143. 6. The first recension of the Grandes chroniques, written at St. Denis and started at the request of Louis IX, presents an account of French history strongly biased in favour of the French monarchy to 1340. Its continuation, also written at St. Denis, to 1350 has an even more marked 'official* bias, as a result of increased royal interest in the work (John II was anxious to whitewash the failures of Philip VI, notably in the Hundred Years War). The second recension was commissioned by Charles V, soon after his accession in 1364. His councillor Pierre d'Orgement (who became chancellor in 1373) produced the new recension as 'official' propaganda, with an especial bias against the English. I owe this information partly to an unpublished Ph.D. thesis for Edinburgh University (1969) by Sarah Murphy Farley, French Historiography in the Later Middle A.ges with Special Reference to the 'Grandes Chroniques de France', (see especially pp. 276-84, 361-8, 380-98). See also D. Hay, 'History and historians in France and England during the fifteenth Century', Bulletin of the Institute of Historical Research, xxxv (1962), 111-5, and, with reference to a later period, P. S. Lewis, 'War propaganda and historiography in fifteenth-century France and England', Transactions of the Koya/ Historical Society, xv (1965), i-2i.
The Alleged Rape by Edward HI of the Countess of Salisbury
275
show a marked anti-English bias. Moreover, there is at least one continental example of a sophisticated propaganda piece. Les Voeux du Heron* a poem in French, was apparently written to denigrate the English army commanders. It laid the blame for the start of the Hundred Years War on Robert of Artois, recounting how, embittered against King Philip by exile and disgrace, he came to the court of Edward III to incite him to renew his claim to the French crown. It relates that Edward and his nobles, whom it satirizes en passant, swore at a feast, on a roasted heron, to fight in France. The same story appears in an even more bitterly anti-English form in the Chronographia, a product of the abbey of St. Denis.2 (It contains little historical truth, though some such ceremony may have been performed at Edward's court.) The story of the rape of the countess of Salisbury can be considered as a piece of political propaganda, a deliberate 'smear' on Edward III. It may be suggested that behind both Le Bel's account and the Chronographia lies some now lost tract, or even a poem, aimed at discrediting Edward III.3 If such a work existed, the discrepancies between the episodes in Le Bel and those in the Chronographia need explanation. It may be postulated that Le Bel copied his source closely, but because of his fair-mindedness and wish to praise Edward III,4 he suppressed some details. Thus he omitted the statement that the first earl of Salisbury had no male heir. This statement was particularly libellous because by inferring that the second earl was not the first earl's son, it could be regarded as an innuendo against the countess. Le Bel also suppressed the earl's betrayal of Clisson and Harcourt, which in the full story appears as the damaging result of Edward's crime. He could well himself have added that the earl of Salisbury died at Algeciras. Knowing that the earl was at the siege, and reading in his source that he was never seen again after leaving France, Le Bel may have concluded that he died in Spain. In fact, the most scurrilous version of the story without any romantic embellishments, is that in the Chronographia. This suggests the possibility that the hypothetical propaganda piece was composed, like the Chronographia itself, by a monk of St. Denis. i. The poem is printed and translated in Political Poems and Songs . . . .from the Accession of Edward III to that of Richard HI, ed. Thomas Wright (Rolls Series, 1859, 1861, 2 vols.), i- 1-25. It is fully discussed by B. J. Whiting, 'The Vows of the Heron', Speculum, xx (1945), 261-78, who suggests that Les Voeux was a piece of anti-war propaganda composed in Hainault. For Robert of Artois in England see H. S. Lucas, The 'Low Countries and the Hundred Years' War, 1326-1347 (University of Michigan, 1929), p. 124. 2. Chronographia, ii. 36-38. 3. To accept that Le Bel used a written source for the story makes it necessary not to interpret literally the statement with which he prefaces his account of the rape, that he 'heard it': 'Or vous vueil je conter le villain cas que fist le roy Edowart, dont on le pouoit blasmer, car il ne fut pas petit, ainsy que je Tay ouy dire*; Le Bel, ii. 30. 4. For passages eulogizing Edward III see Le Eel, ii. 65-67, passim, 168. (Cf. i. pp. xii-xiii).
276
Legends, Traditions and History in Medieval England
One of the literary affiliations of the rape-story seems obvious; it is hard to avoid the conclusion that it is indebted to Livy's account of the rape of Lucretia.1 The use of a classical model would not be surprising. The Scottish barons in 1320 quoted from Sallust's Cataline in the Declaration of Arbroath.2 Moreover, Livy, who had been outside the tradition of the medieval schools, received increasingly close attention in the fourteenth century, in the course of which a more complete text became available.3 He was much studied by humanists in France, Italy and England. Petrarch's annotated copy of Livy still survives,4 and the English Dominican Nicholas Trevet wrote the earliest known commentary on the first and third decades. He wrote at the request of Pope John XXII, whose court at Avignon was the centre of a group of humanists. There were copies of the commentary not only in France but also in Italy.5 The Benedictine Pierre Bersuire translated Livy at the order of King John II, the Good (i35o-i364).6 The Lucretia legend itself was studied both from Livy and from Ovid.7 (The moral problems it posed had fascinated writers from the earliest Christian times.)8 There is abundant evidence of its popularity in fourteenth-century England, which had very close cultural ties with France.9 The Dominican scholar Robert Holcot, lecturing at Cambridge sometime between 1334 and 1342, used Lucretiaas an example of a faithful (pagan) wife.10 Ranulph Higden discussed the Lucretia legend in the Polychronicon.™Both Chaucer and John Gower wrote poems on the theme.12 Chaucer,
1. Lay, i.57~592. See J. R. Philip, 'Sallust and the Declaration of Arbroath', Scottish Historical Review, xxvi (1947), 75~783. For the study of Livy by the early fourteenth-century humanists and the availability of the text see Beryl Smalley, English Friars and Antiquity in the Early Fourteenth Century (Oxford, 1960), pp. 60-61, 64, 86, 92-93. (For further references see the index under Livy.) 4. See R. J. Dean, 'The Earliest Commentary on Livy is by Nicholas Trevet', Medievalia and Humanistica, iii (1945), 86, and Smalley, op. cit. pp. 289-90, 295. Petrarch read Trevet's commentary on Livy; ibid. p. 64. 5. For Trevet's commentary see Dean, loc. cit. pp. 86-98, and Smalley, op. cit. pp. 60-65. 6. See Smalley, op. cit. p. 262. 7. Ovid, Fasti, ii. 721-852. The account of the rape of the countess of Salisbury follows Livy rather more closely than Ovid. 8. E.g. Augustine, Civ. Dei, i. 19. For further references see R. M. Ogilvie, A Commentary on Livy Books /-/ (Oxford, 1965), p. 220. 9. See Hay, loc. cit. pp. 11 i-i 5. 10. Smalley, op. cit. pp. 155-6, 322. 11. Polychronicon Ranulphi Higden, ed. Churchill Babington and J. R. Lumby (Rolls Series, 1865-86, 9 vols.), iii. 156 ff. Higden cites Livy for the Lucretia legend (ibid. iii. 158), but it is not certain that he used him at first hand; see John Taylor, The Universal Chronicle of Ranu/fHigden (Oxford, 1966), p. 76. 12. Chaucer, The Legend of Good Women, lines 1680-1885 (The Works of Geoffrey Chaucer, ed. F. N. Robinson, 2nd ed. Oxford, 1957, pp. 507-9); John Gower, Confessio Amantis, lines 4754-5130 (The Complete Works of John Gower, ed. G. C. Macaulay, Oxford, 18991902,4 vols., iii, 367-77).
The Alleged Rape by Edward HI of the Countess of Salisbury
277
who regarded Lucretia as a model of virtue and steadfastness, even mentions her as a saint and says that her day was hallowed.1 The parallels between the Lucretia legend in Livy and the story of the rape of the countess of Salisbury are striking. In both emphasis is laid on the virtue of the victim, her faithfulness to her husband and fear of dishonour. In both the rape is preceded by a friendly visit of the would-be-lover. Lucretia's husband Collatinus brought Sextus Tarquinius, son of Tarquinius Superbus king of Rome, with other young men to his house, where they were hospitably dined, and Sextus Tarquinius fell passionately in love with Lucretia's beauty and virtue. Edward III came to the earl of Salisbury's castle with ten or twelve knights and was hospitably received by the countess. He was astounded by her beauty, noble bearing and gracious manner, and fell so deeply in love that he could hardly eat the dinner she provided. However, she rejected his declaration of love. Both Sextus Tarquinius and Edward paid a second visit. Tarquinius rode, with one companion, from Ardea to Collatia near Rome, a distance of about twenty miles. Edward, apparently alone, undertook a longer journey, from London to Wark. Both were again kindly received, and each went to his bedroom and waited until the household was asleep (although Edward had to warn the countess's personal servants, who were still awake, not to interfere). Then each went to the lady's room. Sextus Tarquinius approached with a drawn sword, terrified Lucretia into silence and made her submit by fear and force. Edward's violence appeared in other ways; he stopped the countess's mouth before she could utter more than two or three cries, and 'having forced on her such grief and martyrdom', he left her bleeding, in a faint. Neither Livy nor Ovid mention that Lucretia fainted, but Chaucer says that she fainted 'for fer of sclaunder and drede of deth'.2 Sextus Tarquinius left immediately, and Edward returned to London the next morning. Subsequently each victim, overcome with grief, voluntarily confessed to her husband, whose sorrow was only equalled by his anger with the culprit. Although no blame in either case was imputed to the wife, neither couple continued to live together; Lucretia killed herself, and the earl of Salisbury left for ever. After the confession scene it is harder to find similarities between the two works. However, by pursuing the comparison it is possible tentatively to suggest the purpose of, and an approximate date for, the postulated work on the rape of the countess of Salisbury. The i. The Legend of Good Women, line 1871. This 'canonization' of Lucretia was suggested by the fact that Ovid gives her story under a particular day, analogous to a saint's day; cf. W. G. Dodd, Courtly Love in Chaucer and Gower (Gloucester, Mass., 1959), pp. 227-8. 2. The Legend of Good Women, lines 1814-15.
278
Legends, Traditions and History in Medieval England
Lucretia legend provided the cause of the fall of the Tarquins. Lucretia pledged Collatinus and her father Brutus to avenge her lost honour. As a result Brutus roused the people against Tarquinius Superbus, and closed the gates of Rome against him. Sextus Tarquinius himself was slain by 'revengers of old quarrels'. The author of our tract could have had this denouement in mind. Edward's vile act, like Sextus Tarquinius', had momentous political consequences. The earl of Salisbury betrayed Edward's new allies, Oliver de Clisson and Godfrey de Harcourt. Their trial and punishment, part of Philip VI's round-up of traitors in 1343, constituted a breach of the truce of Malestroit, concluded on 19 January 1343 between King Edward and King Philip to stop the Breton war, and marked the resumption of hostilities.1 It is noticeable that the rape-story is set throughout against a background of Anglo-Breton history: messengers from Brittany come to Edward immediately after the grand tournament of 1342; Edward rapes the countess after the earl had been sent to Brittany and before Edward's own departure there; and his crime resulted in the renewal of the Breton war. A monk of St. Denis, writing to please King Philip, would reflect the official French attitude to Edward's interference in Breton affairs. Brittany had become a battleground for Edward and Philip. On the death of Duke John III in April 1341, without direct male heir, Edward supported one claimant to the duchy, John de Montfort (consanguineous younger brother of John III), and Philip supported the other, Charles of Blois (his nephew and husband of the daughter of a younger full brother of John III). De Montfort's party, in return for military support, recognized Edward as king of France and suzerain of Brittany. It seems likely that the rape-story was written as a propaganda tract soon after the breach of the truce of Malestroit. Its object could well have been to emphasize Edward's ultimate responsibility for the renewal of fighting, and perhaps even to justify Philip's breach of the truce. Moreover, the tract could have had a prophetic purpose; to show, on the analogy of the Lucretia legend, that Edward was doomed to disaster, because of his crime. If, as suggested, the rape-story is an example of war propaganda, it was a more subtle and insidious composition than Les Voeux du Heron. It was better calculated to deceive because it was set in the context of well-known events and contained nothing which was likely to appear demonstrably false to a Frenchman. In it fact and fiction were so cleverly interwoven (it is impossible to be sure of their relative proportions), that the whole was credible.
i. For the view that Philip's purge of traitors constituted a breach of the truce of Malestroit see Avesbury, ed. Thompson, p. 35 5, and Borderie, op. cit. iii. 482-3.
12 A Fourteenth-Century Chronicle from the Grey Friars at Lynn ALL that remains today of the Grey Friars' house at Lynn is the fine late fourteenth-century tower; the only manuscript from their library is a book of sermons, preached early in the fifteenth century by Friar Nicholas Philipps in East Anglia and written down by him at Lynn (and elsewhere).1 To these may now be added two pages of a chronicle in the form of tabular annals for the years 1349, 1360-77, written apparently between those dates on folios 19-20 of a mid-fourteenth-century commonplace book, now British Museum Additional MS. 47214. The volume has forty-eight paper leaves 2 measuring about 11^ in. X y-J- in., arranged in three gatherings (at least one gathering is missing after the first). The back cover of the original binding of soft white leather on sheets of stiff vellum remains. Possibly the book was still in the hands of Grey Friars at the end of the fourteenth century; a hand of about 1400 has inscribed on folio 48, ' iste quaternus est fratris lohannis Medilton de dono fratris Nicholai Fakynham magistri'. A late sixteenth-century hand has added below ' vide Johannem Baleum de Nicholao Fackenhamo cent. 7* fol. 530 '. The reference is to the second edition of Bale's Catalogue (1557) which describes Nicholas Fakenham of Norfolk as professor and doctor of Oxford and provincial minister of the Franciscans in England 1395-?!401. He died in 1407. The sixteenth-century annotator's identification of the inscription may be right, but Fakenham is a common East Anglian name.3 The whole volume, except for the Grey Friar's annals, is written by one scribe in a mid-fourteenth-century charter hand. The contents are a miscellany of legal and historical pieces. It is impossible to say who the scribe was, or where he was writing, but there is evidence to suggest that it was somewhere in East Anglia. He begins, folios 2v-ij> with annals from the Incarnation to 1314. To 1295 these annals are brief extracts (usually one line long) from a chronicle compiled at Bury St. Edmunds,4 which was popular 1 Bodley MS. Lat. Th. D. i, described by A. G. Little, Franciscan Papers, Lists and Documents (1943), pp. 244-5. F°r ^e house of the Grey Friars at Lynn see A. R. Martin, Franciscan Architecture in England (1937), pp. 101-5. 2 The water mark of the first gathering is a T and of the second and third an eagle; cf. C. M. Briquet, Les Filigranes, no. 9089 (Palermo 1361) and no. 77 (Longwy, Udine and Pisa, mid-fourteenth century) respectively. 8 For Nicholas Fakenham see A. G. Little, op. cit. p. 198, and ibid. Grey Friars in Oxford (1892), pp. 252-3. There was a prosperous family of Fakenhams in Lynn; H. Ingleby, Red Register of King's Lynn (1919 &c.), i. 155-6. 4 For the Bury chronicle see above pp. 204 and n. 30, 205, 226—38 passim.
280
Legends, Traditions and History in Medieval England
in East Anglia but not elsewhere. The extract has a few unimportant additions to the Bury chronicle; one, under 672, is a synopsis of the history of the abbey of Ely. On folios zov-2iv are lists of the kings of the Heptarchy including notes on East Anglian kings and saints. Chief justice Thomas Weyland, a Norfolk man and great landowner in East Anglia, is mentioned in the annals under 1289, ' obsessio fratrum minorum apud Babbewelle pro Thoma Weiland', and he appears again in the last item of the volume (fos. 45~7V) among the French civilians and canonists who in 1292 advised Edward I on the law of succession to the throne of Scotland.1 The house of the Grey Friars at Bab well, Suffolk, is also mentioned under 1259, ( fratres minores post venerunt apud sanctum Edmundum et expulsi erant extra villam, propter privilegium ecclesie sancti Edmundi usque Babbewelle '.2 The annals quoted above for 1289 and 1259 are both lifted from the Bury chronicle, but the name itself, Babwell, is an addition. There is not enough evidence, however, to conclude that the commonplace book was compiled for the Grey Friars. The contents suggest rather that it was compiled for some East Anglian lawyer who took an interest in Scottish affairs. Its principal contents, excluding the items already noticed, are: a list of the kings of Scotland 844-1285 (fo. 22V), Boniface VIII's letter to Edward I, 1299, about the imprisonment of Scottish ecclesiastics,3 and notes for Edward's answer, 1304,* the version of the coronation service used at Edward IPs coronation 5 (fos. 24V, 25), Anglo-Saxon and Norman laws, incomplete at the beginning (fos. 26-44),tne reissue of King John's charter, 13 January 1215, granting freedom of election to the churches 6 and his letters patent, 1213, revoking ' ut legatio '7 (fos. 1 The lawyers' opinions are printed from a later text (mid-fifteenth century) in Johannis de Fordun, Scotichronicon cum supplementis ac continuatione Walteri Berveni, ed. W. Goodall (1759), ii. 139-45. Edward's consultation and the opinions have received little attention from historians. F. Michel, Les cjossais en France, Les Francais en £cosse (1862), i. 41-2, dates Edward's consultation with the lawyers 1286. But probably it took place during the adjournment of the proceedings at Norham in 1292; Foederay i. 777, records that at that time Edward ' per sollempnes nuncios de consilio suo, super facto illo, diversarum regionum consulet sapientes '. One of the lawyers who gave an opinion was master Giles Lambert, described as ' decanus' of Tours; he was dean 1290-1313; E. R. Vaucelle, 'La Collegiale de Saint-Martin de Tours, 397-1328, in Bulletin et Mfmoires de la Socittt Archeologique de Touraine: Memoires, no. 46 (1907), p. 441. See also below p. 288. 2 A late medieval hand has written in the margin by these entries ' nota pro inhabitacione fratrum Babwell', and ' obsessio pro Weylond '. 2
4
Foederay i. 907-8.
Ibid. i. 932, 933. Cf. Rishanger, Cbronica et Annales, ed H. T. Riley (Rolls Series), pp. 208-10. 6 The ordo printed by L. G. W. Legg, English Coronation Records (1901), pp. 83-112, reference O in the footnotes. The text in Add. MS. 47214 agrees closely with that in Harley MS. 2901, for which see H. G. Richardson in Bull. Inst. Hist. Research (1938-9), xvi. lo-ii. 6 Statutes of the Realm, i. 5. 7 T. D. Hardy, Rotuli 'Litterarum Patentium (1835), i. 100.
A Fourteenth-Century Chronicle from the Grey Friars of Lynn
281
V
44 , 45), and the form of homage made by Edward I to Philip IV in i294 1 (fo. 45). When the scribe had written the annals from the Incarnation to 1314 on folios 2v-i7, he tabulated the next three leaves (fos. i7v-zo) for tjie annals from 1316-77; he wrote the year date in Roman numerals and the Dominical letter in the left hand margin, leaving a space for the annal before writing the next year date. There are ten years on each page. The general appearance is like that of a modern diary. The scribe then copied in neatly annals for the years 1340, 1345-1348, 1356, 1357, and 1360. His entry for 1345, about the murder of the abbot of Combe, contains information not found elsewhere. The remaining spaces were used by the Grey Friar of Lynn. He wrote in annals for the years 1346, 1361-70, and 1373-7. The annals themselves provide the evidence of authorship. Four entries suggest that they were written by a Grey Friar (those for 1349, 1362, 1366, and 1375) as they contain references to the Franciscans. An entry under 1364 suggests that the friar belonged to the Grey Friars' house at Lynn; it notes that ( post complein dictum in choro quasi usque ad mediam noctem erat fulmen horribile quasi continuum et grando Lenn'. Events at Lynn are also mentioned under 1363, 1374, and 1377. It is well known that most medieval religious houses kept chronicles as part of their archives. The grand chronicles, such as those of St. Albans, have survived, partly because more than one copy was made. But many less important works have disappeared. This Grey Friar's chronicle is interesting as one of the few survivors. It represents the humble attempt of a small house to write history and may be compared with the short chronicle of Butley priory in Suffolk 2 and that of the Grey Friars of London.3 The page format, the arrangement of spaces headed by the year date for the insertion of annals, also occurs in the mid-thirteenth-century annals of St. Augustine's Canterbury (the scribe tabulated annals from the Incarnation to 1300 and made entries to 1234),* and in the late thirteenth-century annals of Hickling in Norfolk (the scribe tabulated annals from the Incarnation to 1532, made entries to 1294 and other scribes added more entries later).5 The Winchester chronicler advocated composing a chronicle year by year on a schedule attached to the end of a volume and copying the annal in neatly at the end of the year.6 But here was another way of writing a chronicle. 1
Foedera, i. 807. Register or Chronicle ofBut/ey Priory, Suffolk, ijio-ij)}, ed. A. G. Dickens (1951). Monumenta Franciscana, ed. R. Hewlett (Rolls Series), ii. 143-260. 4 5 Cotton MS. Julius P u, fos. 3-21. Egerton MS. 3142, fos. 83v-io4v. 6 Cotton MS. Vespasian E iv, fo. 153, printed from a derivative text (see N. Denholm Young, EHR, xlix (1934), 85-93 in Annales Monastici, ed. H. R. Luard (Rolls Series), iv. 355. Cf. above p. 205 and n. 43. 2
3
282
Legends, Traditions and History in Medieval England
The Grey Friar of Lynn writes bad Latin in a cramped charter hand. His Latin shows the influence of the vernacular (for example, compkin for comphtorium under 1364 and princissa for principissa under 1365). Only once, under 1373, does he associate cause with effect. But his chronicle gains value from its very unpretentiousness. The chronicler was not tidily copying written sources. From 1361 it is fairly certain that he was writing more or less contemporarily with the events described. The shade of ink and width of pen strokes varies from annal to annal. The entry for 1364 got so long that it spilt over into the space for 1365, which annal is longer still and almost crowds out that for 1366. The author seems to have obtained much of his news as an eye-witness and listener: under 1365 he says c hec et alia multa retulit qui interfuit', and ends the annal for 1375 with ' ut dicebatur a multis '. He is more than a parochial antiquary. He mentions his own house only once and makes a bold attempt at a general history; he notes events of national importance and does not ignore the European scene. He must have profited from being at Lynn, a busy market place and port. He adds a number of facts to our knowledge, particularly of East Anglian history. He is apparently the only authority for the removal and reburial of Elizabeth Ufford's body in 1362, for the fires at the Austin friary , Orford, and in the tolbooth and Carmelite convent, Lynn, under 1363, for the celebrations on the occasion of Princess Joan's purification in 1365, for the flood and high price of water at Lynn in 1374, for the information about the Franciscans in 1375 and for the interdict on Lynn in 1377. BRITISH MUSEUM ADDITIONAL MS. 47214. v
fo. i8 . 1340 In x festo nativitatis bead lohannis Baptiste,2 statim post horam nonam, fuit bellum navale in Swyna Flandrie inter regem Anglic [Edwardum] tercium a conquestu et navigium regis Francie, ubi Deo opitulante rex Anglic triumphavit et cepit circiter vin naves, defensoribus et gubernatoribus earundem submersis et occisis. 1345 Duo albi monachi de monasterio de Koumbe 3 pro morte eorum abbatis capti, et per aliquod tempus in prisona de Warwyk' detenti, demum die sancti Laurencii 4 apud Rokeby per Justiciaries domini regis E[dwardi] « 1
Punctuation, the use of capitals, u, vy /, Jt are the editor's. The years are written in Arabic instead of Roman numerals and the dominical letters omitted. 2 3 24 June. Combe, co. Warwick, Cistercian abbey. 4 10 August, at Rugby. For this murder, which is not mentioned by the other chroniclers, see Cal. Pat. RoMs, 1343-134;, pp. 573, 580, and CaL of Inquisitions, Misc. u. 483.
A Fourteenth-Century Chronicle from the Grey Friars of Lynn
283
tercii, anno regni sui xix°, de morte et prodicione eorum patris arrainjiati et coram eis caucius convicti, privilegio clerical! non obstante, tanquam proditores eodem die tracti fuerunt et suspensi. 1346 Bellum apud Cressy in Picardia inter Edwardum regem Anglie tercium a conquestu et Philippum de Valoys regem Francie, ubi ex parte dicti regis Francie ceciderunt rex Boemie, dux Lothoringie, comites de Flandrie, de Alasonn, de Bloys, de Alba Maria, de Bello Monte, de Harecourt cum duobus filiis et plures alii tarn de Francia quam Alemannia; x et dominus rex Francie cum filio regis Boemie fugit.2 Et cito post rex Anglie incepit eodem anno obsidionem de Caleys. Eodem anno David rex Scocie captus est.
1347 Villa de Caleys fere per unum annum per Edwardum regem Anglie tercium post conquestum obsessa, et demum fame subacta, iiiito die Augusti eo anno per diem sabbati eidem regi cum castro reddita fuit capitaneo, et omnibus aliis in eisdem existentibus in gratia domini Regis de vita et membris, bonis et catallis se ponentibus. fo. 19. 1348 Isto anno apud Melcoumbe in comitatu Dorsate parum ante festum nat' sancti lohannis Baptiste,3 due scaphe, quarum una erat de Bristollia, applicuerunt, in quibus naute de Vasconia venientes quadam inaudita pestilencia epidemia nominata infecti, homines illius ville de Melcoumbe primo in Anglia inficiebant, ubi incole morbo illius pestilencie trium dierum ad plus infirmati in vigilia sancti lohannis Baptiste primo inde ibi mori ceperunt. 1349 Isto anno circa pascha vel modicum ante incepit pestilencia in custodia 4 Cantebrigiense et duravit per totam estatem. 1356
Bellum iuxta Pictaviam inter lohannem regem Francie et Edwardum primogenitum regis Anglie principem Wallie, ubi dictus rex succubuit et captus est cum filio Philippo et multis comitibus et dominis Francie 1
This list is inaccurate and incomplete. The flight of Philip with the son of the king of Bohemia, Charles IV of Luxembourg, is mentioned by the continuator of Murimuth; Chronica Adae Murimuth et Roberti de Avesbury, ed. E. M. Thompson (Rolls Series), p. 247. Froissart, Chroniques, ed. S. Luce (Soc. de PHist. de France), iii. 179, says of Charles, c il s'en parti: je ne sgai pas quel chemin il prist', and states that Philip fled with Jean de Hainault and others to Labroye and on to Paris (fb. in. Ivii, and 184). JL&r Vrayes Chroniques de Messire ]ehan le Be/t ed. M. L. Polain, ii. 89, also mentions his flight with Jean de Hainault. 8 The authorities give various dates. Malvern, Polychronicon, ed. R. Lumby (Rolls Series), viii. 355, gives about the feast of St. John the Baptist. 4 Custodia ; an administrative division of the Franciscan province; that of Cambridge covered East Anglia. Easter was on 12 April; A. Jessop states that the plague entered East Anglia in March; ' Black Death in East Anglia' in The Coming of the Friars (1890), pp. 208-9. 2
284
Legends, Traditions and History in Medieval England
xix° die Septembris, . . primogenito filio suo Dolphino cum acie sua a prelio fugiente. *357 Edwardus princeps Wallie, regis Edwardi tercii a conquestu primogenitus, de Vasconia in Angliam rediit et xi die mensis Mail per diem iovis * apud Plummouth prospere applicuit, lohannem regem Francie, Philippum filium suum et quosdam alios in bello iuxta Pictaviam captos de Burdegale secum ducens. Eodem anno circa festum nat' sancti iohannis duo cardinales in Angliam venerunt pro pace inter regna Francie et Anglic reformanda.2 Eodem anno mense Septembris David de Brus rex Scotie a custodia regis Anglic liberatus fait, certis obsidibus pro redemptio[n]e sua in Angliam missis.3 fo. i9v. 1360 Isto anno incepit pestilencia apud London' iam circa festum sancti Michaelis4 ubi primo infantes in magno numero delevit; et post pascham5 proximam sequens (sic) homines et mulieres in maxima multitudine decesserunt. 1361 Isto anno fuit pestilencia in partibus australibus Anglic maxime et mors puerorum et iuvenum et divitum; hec tamen pestilencia fuit multo minor quam precedens (sic) anno 13^0.6 Eodem anno secundum cursum ecclesie Anglicane 18 Kal. Februarii7 et sequente nocte fuit ventus vehemens qui campanilia et ecclesias et domos destruxit et discooperuit et arbores evertit. 1362 Isto anno in die sanctorum Gervasii et Prothasii 8 fuit corpus domine Elizabet, uxoris domini Thome UfTord heredis9 et filii comitis SufFolchie,10 exhumatum apud Campese u et deportatum ad fratres minores Gipp'12 1 This date is wrong; Edward arrived on 5 May; Chronicon Angliae (Rolls Series), ed. E. M. Thompson, p. 37. He left Bordeaux on u April; Grandes Chroniques, ed. R. Delachenal (Soc. de FHist. de France), i. no. 2 Nicholas Cappochi, cardinal-priest of St. Vitalis; Talleyrand of Perigord, cardinalbishop of Albano. 3 David Bruce was released on the conclusion of the treaty of Berwick, 3 October 1357; Foedera, iii. 372-3. 4 This approximate date, about 29 September, is roughly six months earlier than that given by the other authorities. 6 8 5 April 1361. Thirteen years previously, i.e. 1348. 7 8 15 January 1362. 19 June. 9 The description of Thomas as * heir ' may mean that this annal was written after 1366, as Thomas, the second son (d.? 1368), had an elder brother Robert who is supposed to have been alive in 1366; there is, however, so little evidence about the date of Robert's death that it is impossible to say when Thomas became heir: G. E. C. Complete Peerage, xii. 432. 10 Robert, ist earl of Suffolk, 1337-69. 11 Campsey Ash, Suffolk; the Uffords were patrons of the house of Austin canonesses there; R. E. Chester Waters, Genealogical Memoirs of the Extinct Family of Chester of Chicheley (1878), i. 325. 12 Ipswich. That this was Elizabeth's burial place is confirmed by J. Weever, Funeral Monuments (1767), p. 487; the story of her reburial does not seem to survive elsewhere.
A Fourteenth-Century Chronicle from the Grey Friars of Lynn
285
ac sepultum coram magno altari in choro; quod iacuerat apud Campese plus quam per 24 ebdomadas in terra; discussa enim ventate quomodo corpus suum legaverat sepeliendum in loco fratrum predicto, comes, qui erat homo bone consciencie, permisit corpus levari per fratres. 1363 Isto anno in vigilia translationis sancti Thome Cantuariensis archiepiscopi,1 in nocte quando fratres fuerunt in matutinis, fuit fulmen tarn horribile quod combussit ecclesiam et chorum Carmelitarum Lenn' et tolbothe in eadem villa eodem tempore; et anno proximato (sic) precedente combussit fulmen locum fratrum Augustinensium apud Orforde et consimili tempore anni. Isto etiam anno fuit magnum gelu incipiens in medio Decembris et durans quasi usque ad medium 4ome proximato (sic) sequentis. 1364 Isto anno obiit lohannes rex Francie London' cito post pascha, qui ante captus fuit in bello iuxta Pictaviam cum filio Philippo et multis aliis nobilibus; iste vero lohannes predictus post captivitatem rediit ad Franciam et postea reversus ad Anglicam, London' mortuus est. Item eodem anno iv Kalendas Augusti,2 post complein 3 dictum in choro, quasi usque ad mediam noctem, erat fulmen horribile quasi continuum et grando Lenn' et in villis quibusdam 4 vicinis man, tarn horribile quod lapides descenderunt ad quantitatem pomi, et alibi multo maiores, ita quod impetu sui decensus terram penetraverunt ad magnam profunditatem et quasi in terra se occultaverunt, et blada destruxerunt, et feras et cuniculos, aves et lepores (destru) 5 interfecerunt. 1365 Isto anno fecit Edwardus, primogenitus regis Anglic et princeps Aquitanie et Wallie, in quindena pasche, in Aquitania civitate Angilem,6 maximam sole[m]pnitatem in purificatione princi[pi]sse7 post partum filii sui primogeniti nomine Edwardi. Ubi fuerunt viic et vi milites et 154 domine, et duravit sole[m]pnitas per dies 10, et equi8 in expensis principis fuerunt xvm per dictam sole[m]pnitatem; ubi et princi[pi]ssa habuit de camera sua 24 milites et tot dominas in 9 apparatu solempni pro hastiludiis, tripudiis et aliis solaciis transitoriis; pro qua sole[m]pnitate cera empta excessit in precio iiiic libras argenti. Hec et alia multa retulit qui interfuit. 1366 Isto anno in parliamento suo celebrato apud Westmonasterium post pascham dominus rex cassavit et adnullavit10 statutum novum editum 1
2 3 Thursday, 6 July. 29 July. gplem in manuscript. 5 quibusdam in margin with tie-mark in manuscript. Deleted in manuscript. 6 Sunday, 27 April, at Angoul£me. Cf. Avesbury's account of similar festivities on the queen's purification after the birth of Thomas of Woodstock in 1355; Chronica Adae Murimutb et Roberti de Avesbury, ed. E. M. Thompson (Rolls Series), p. 422. 7 pi interim, in a later hand in manuscript. 8 eq in manuscript. An ecu. 9 A word has been erased and crossed out in manuscript. 10 statu added and expunged in manuscript. 4
286
Legends, Traditions and History in Medieval England
per universitates Oxoniensem et Cantebrigiensem contra fratres ; * de hac materia plenius supra anno bt.2 Item isto anno in mense Septembris frater Marcus, fratrum minorum generalis minister, factus est cardinalis et alius de ordine predicatorum.3 1367 Bellum in Hispania iuxta aquam Nazers,4 ubi Edwardus primogenitus regis Anglic vicit bastardum Hispanic intrusorem 5 et legitimum 6 restituit in regnum, captis 2 milibus Valencium et plusquam quinque milibus occisis. Hoc factum est 3° die Aprilis. Eodem anno statim post pascham Urbanus Vtus recessit de Avinione cum cardinalibus transferens sedem Viterbium. fo. 20. 1368 Isto anno dominus archiepiscopus Cantuariensis factus est cardinalis.
1369 Hoc anno renovata est guerra inter Angliam et Franciam, per falsitatem quorundam de Francia, qui contra concordiam firmatam invaserunt terras regis Anglic in partibus transmarinis. Item pestilencia magna fuit magnatum et puerorum. 1370 Isto anno fuit magna caristia bladi per totam Angliam et maxime in estate, ita quod quarterium frumenti valuit xx solidos et in quibusdam partibus 2 marcas.7 1373
Isto anno transfretavit magnus excercitus contra Francos, videlicet dux Lancastrie, dux Britannic, comes Sutfolch, comes War wye,' et alii comites et nobiles; modicum tamen profecerunt, quia fame magna pars excercitus interiit, propter defectum capitaneorum,8 ut dicebatur a multis.9 1 Oxford University passed a statute, 1365, forbidding the friars to receive into their orders those who had not yet completed their eighteenth year; Statuta Antiqua Universitatis Oxom'ensis, ed. S. Gibson (1931), pp. 164-5. Such a statute for Cambridge University has not survived. The Parliament at Westminster, May 1366, quashed the Universities' statute; R0/. Par/, ii 290. For details of the controversy see J. R. H. Moorman, The Grey Friars in Cambridge (1952), pp. 105 seq. 2 The meaning of this passage is obscure; the manuscript reads de he m pleni' a — i s an ibi. 3 Mark of Viterbo, minister general of the Friars Minor, 1359-66, cardinal-priest of St. Praxidius, 1366. William Sudre, cardinal-priest of St. John and St. Paul, 1366. 4 .Aquam Naders ; the battle of Najera, 3 April 1367, was fought on the banks of the river Najerilla. 6 6 Henry of Trastamara. Peter, king of Castille. 7 Wheat was 3* a bushel (248. a quarter), as a result of the plague and floods of the preceding year, according to Cfaonicon Anglia ed. E. M. Thompson (Rolls Series), p. 65 ; the Annals of Bermondsey record, c vendebatur busshellus frumenti Calisiae pro iii8. iv d . (26s. 8d. a quarter), luna currente per iii'; Annales Monastici, ed. H. R. Luard (Rolls Series), iii. 478. 8 capit' in manuscript. 9 For contemporary comments on John of Gaunt 's inefficiency see Etflogium Historiarum, ed. F. S. Haydon (Rolls Series), iii. 336; Walsingham, Historia Anglicana, ed. H. T. Riley, (Rolls Series), i. 316.
A Fourteenth-Century Chronicle from the Grey Friars of Lynn
287
1374 Isto anno 3° die mensis Decembris circio x flante formidine fuit fluxus maris terribilis Lenn', cooperiens pro maiori parte magnum forum ibidem, terras etiam villarum adiacencium et domos intrans et submergens, naves etiam in mari ad terram per magnam distantiam elevans et proiciens, et pecora in pascuis submergens. Et cito post sequebatur magnum gelu durans usque ad vigiliam natalis Domini. Quo tempore fuit magna penuria aque recentis in Lenn' ita quod pipa aque recentis quandoque 2 vendebatur pro izd. 1375 Isto anno dominica 4a quadragesime,3 que fuit 4 in octavis annunciationis Virginis gloriose, tanta 5 aquarum inundacia Oxonie in conventu predicatorum 6 quod populus venit ad locum fratrum minorum 7 et ibidem frater predicatorum predicavit sermonem illorum ad populum in domo fratrum minorum. Item in eadem quadragesima fuit frater Leonardus 8 generalis minister fratrum minorum in Anglia et tenuit concilium suum cum magistris et custodibus Anglic Bedefordie, et fuit ibi tamen per 3 dies et statim disponit se ad recessum de Anglia; dies autem convocationis magistrorum et custodum fuit dominica in passione Domini.9 1377 Hoc anno in vigilia sancti Albani10 obiit rex Anglic Edwardus 3US post conquestum, anno regni sui 51, sepultus ad quindenam post eodem die, scilicet die dominica ;11-et statim post (in die sanctorum Quirici et lulitte12 coronatus est rex Ricardus filius filii supradicti Edwardi anno etatis sue i imo)i3 jy Kalendas Augusti14 rex Ricardus coronatus, filius filii supradicti Edwardi, etatis sue anno nmo. Item eodem anno fuit villa Lenn' supposita interdicto a 5*0 Idibus lunii15 usque ad vigiliam sancti Laurencii16 eiusdem anni, propter violenciam factam per quosdam fatuos eiusdem ville in personam domini Henrici de Dispensariis episcopi Nor wye'. 1
Cirtius ; a violent wind (in Roman times W.N.W.). ' va' added and expunged in manuscript. 3 i April. 4 * illo anno' added and deleted in manuscript. 5 * fuit ' added and deleted in manuscript. 6 The buildings (1245) of the Oxford Dominicans lay close by the Isis outside the south gate in St. Ebbes. 7 The Church of the Grey Friars, Oxford, stood ' from the south end of Paradise Place, where the wall juts out southwards for a few yards, to a point about the north end of King's Terrace ', A. G. Little, Grey briars in Oxford, p. 23. 8 Leonardus Rossi de Giffono, minister general of the Friars Minor 1373-8. Gregory XI in 1374 ordered him to visit Umbria and other provinces to reform observance of the rule; Analecta Franciscana, vii. 349-50. His visit to England does not seem to be mentioned elsewhere. 9 8 April. A. G. Little, Franciscan Papers, Lists and Documents, p. 215, notes that a provincial chapter was held in England in 1375, date and place unknown. For the problems connected with the composition, &c., of the provincial chapters and other less formal assemblies of the Friars Minor see ibid. pp. 156-78 passim. 10 u 12 21 June. 5 July. 16 June. 13 14 15 Deleted in manuscript. 16 July. 9 June. le 9 August. For a full account of Despenser's quarrel with Lynn see Chronicon A.ngliae, pp. 139-40; Walsingham does not mention the interdict. 2
288
Legends, Traditions and History in Medieval England
ADDITIONAL NOTE Page 280 and n. i. The lawyers' opinions copied into BL MS Additional 47214, ff. 45-47^ are most recently discussed and printed in E. L. G. Stones and G. G. Simpson, Edward I and the Throne of Scotland: An Edition of the Record Sources for the Great Cause (Oxford, 1978, 2 vols), ii. 358-64. See also G. J. Hand, 'The opinions of the Paris lawyers upon the Scottish Succession Case', The Irish Jurist, v (1970), 141—54.
13 The Date and Authorship of John of Glastonbury? s Cronica sive Antiquitates Glastoniensis Ecclesie John of Glastonbury's Cronica is a history of Glastonbury abbey from the legendary foundation of the first church by St Joseph of Arimathea until the succession to the abbacy of Walter de Monington (1342-75). Although the Cronica is a 'house-history', it includes occasional entries about the kings of England; these become rather more detailed in the fourteenth century. As a local history, it is based on William of Malmesbury's De Antiquitate Glastoniensis Ecclesie, and then to 1290 on Adam of Domerham's chronicle. The information from them is embellished with much legendary material and eloquent verbiage. From 1290 the text derives either directly or indirectly (perhaps in part from some intermediate Glastonbury annals) from documentary and also probably from oral evidence. The final version of the Cronica, together with the brief continuation to 1497 composed by William Wyche at the order of Abbot Richard de Beere, was printed by Thomas Hearne (Oxford, 1726). He based his text on a later fifteenth century manuscript now in Princeton University Library (Robert Garrett MS 153, ff.2-147 v ), and took into consideration two other manuscripts, also of the late fifteenth century. It is an excellent edition by the standards of its time, but unfortunately Hearne overlooked a manuscript of the Cronica which is much earlier than those he used, now Trinity College, Cambridge, MS R.5.16, pp. 27-214. Scholars therefore welcomed the modern edition (excluding Wyche's continuation) by James P. Carley, first printed in paperback in 1978 (Oxford, British Archaeological Reports, xlvii, i and ii), and reprinted in hardback, with some revision and an English translation by David Townsend, in 1985 (Boydell and Brewer, Woodbridge). Except for the translation, the 1985 edition, the one cited below, is substantially the same as the 1978 one. Carley based his text on the Trinity MS, giving variant readings from the other known manuscripts in footnotes. Because leaves at the beginning of 1 The present article is based on the two reviews I wrote of Carley's 1978 and 1985 editions respectively; English Historical Review, xcv (1980), 358-63, and Albion, xix (1987), 54-8. For another review of the 1985 edition see David Corner, EHR, ciii (1988), 173. Cf. below and next note.
290
Legends, Traditions and History in Medieval England
the text in the Trinity MS are missing (Carley, pp. 2-86), Carley reconstructed that part from the best readings in the other manuscripts. In addition, Carley provided subject notes at the end of the volume; of particular value are those on the sources oftheCronica. Carley's Introduction contains the fruits of his researches on the chronicle itself and on its historiographical and historical background at Glastonbury. He pays particular attention to the question of the chronicle's date of composition and its authorship. His conclusions, the substance of which had already appeared in his article in Mediaeval Studies, xl (1978), 478-83, deserve careful attention because his dating of the chronicle differs from the previously accepted one, and he is the first scholar to claim to have identified the author, 'John'. They have been accepted without question by some scholars,2 although the evidence presented by Carley does not warrant certainty. This is recognised by John Taylor in his definitive study of fourteenth-century English historical literature: 'a final judgement,' he says, 'cannot yet be given'.3 Since Taylor wrote a new piece of evidence has come to light which may be relevant to the date problem. But before discussing that evidence, it is necessary to examine the evidence available to Carley and to test the soundness of his conclusion about the date of composition to which it led him. His identification of the author depends on his dating and will therefore be considered last. Since the chronicler states in the prologue (p. 6) that he has written the history of the abbey to about 1400, Hearne and subsequent scholars concluded that Cronica was composed at that time. Carley, however, rejects this approximate date, postulating instead a date between 1340 and 1342 or more generally to the 1340s (pp. xxv-xxviii). This would explain the fact that the Cronica ends in 1342 and not c. 1400. Moreover, in Carley's opinion the Trinity MS, a composite volume, provides evidence of an earlier date than the accepted one. He bases his conclusion on the argument that the copy of the Cronica in the Trinity MS cannot have been made as late as c. 1400. The Trinity text is a copy and not the author's draft and was likely, Carley speculates (p. xxvi), to have been made quite a while after the original composition since 'from what we know about the Glastonbury library it seems unlikely that a text such as the Cronica would have been immediately duplicated'. (However, if the early draft was untidy, it seems probable that a fair copy would soon have been made). 2 Carley's redating is accepted by J.R. Maddicott in his review of Carley's 1985 edition; Journal of Ecclesiastical History, xxxvii (1986), 490-1. Both the redating and the identification of the author are accepted by the editors of the Annals of St. Neots; The Anglo-Saxon Chronicle, a collaborative edition, ed. D.N. Dumville and Simon Keynes, xvii, The Annals of St. Neots with Vita Prima SanctiNeoti, ed. Dumville and Michael Lapidge (Cambridge, 1985), p. cxv. 3 John Taylor, English Historical Literature in the Fourteenth Century (Oxford, 1987), p. 45.
John ofGlastonbury, Cronica sive Antiquitates Ecclesie
291
The Trinity text is in a good bookhand and is followed without a break or a change of handwriting by a list of sixty-three monks who professed under Monington. Since this list includes John Chinnock, Monington's successor to the abbacy (1375-1420) and refers to him as postea abbas, it must have been copied after Chinnock's succession; Carley believes that this would not have been done much after 1375. This argument is purely speculative, since obviously a copy can be made at any time. More crucial to Carley's conclusion is the date of the handwriting of the Cronica in the Trinity MS. Carley cites the opinion of an eminent palaeographer, Father Leonard Boyle, that the copy was made in the last half of the fourteenth century 'and can probably be dated to the last quarter of the century' (p. xxv). Carley concludes: 'A palaeographical analysis, then, shows that the text cannot have been written as late as 1400'. However, it is surely rash to argue that the copy could have been made just before 1400, but not c. 1400. It is a truism that an old man might write a hand current in his youth, and that a whole scriptorium might be old fashioned, especially one in an out-of-the-way place - like Glastonbury. Carley's redating of the Cronica to 1340-2 (or the 1340s) raises two problems. First, the Cronica refers to one of Chinnock's acts as abbot (p. 45). Carley warns the reader of this difficulty in a footnote (p. xxvi n. 5 and cf. p. xxviii), observing that 'By my calculations this must be an interpolation'. The second problem is the statement in the prologue that the author had continued the Cronica down until c. 1400. Carley argues that this was the result of a scribal error. The scribe miscopied the date because he was misled homoeoteleuton and so wrote ad annum domini millesimum circiter quadringentesimum instead of ad annum domini circiter quadragesimum tercentesimum. Thus Carley concludes that the author started writing in 1340. There are objections to these two ad hoc hypotheses. The manuscript evidence does not support them. The Trinity MS, owing to the loss of leaves, throws no light on the matter, but the other manuscripts all have the passage about Chinnock in textu and the date 'c. 1400' in the prologue. This makes it necessary for Carley to claim (pp. xxii-xxiii) that all the surviving manuscripts descend from a common exemplar, since it is most unlikely that the interpolation would have been made and the year date miscopied in two independent manuscripts. Carley identifies this exemplar with the Trinity MS, which, he argues, in view of the relatively high quality of its text, is a copy of the original. (He does not consider the possibility that it had benefited from scribal correction.) Nevertheless, on Carley's own showing (p. xxiii), one of the other copies, the Princeton MS, does not share the few errors in the Trinity MS: this fact (which Carley attributes to scribal correction) suggests that perhaps the Trinity MS is not its exemplar. Carley's evidence, therefore, for redating the Cronica is not, as he suggests (p. xxx) 'incontestable'. At most it suggests a possibility. And
292
Legends, Traditions and History in Medieval England
meanwhile a good case can be made for the traditional date. The textual evidence is on its side. Furthermore, Carley himself pointed out in his 1978 edition4 (i.xxxvii n. 1) that it is hard to find a chronicle composed in the fourteenth century comparable with the Cronica. But it would fit well into the historiography of the early fifteenth century, which includes a number of similar works: Abbot William Frocester's history of St Peter's, Gloucester;5 Thomas Elmham's history of St Augustine's, Canterbury,6 Thomas Burton's history of Meaux;7 and possibly the 'pseudo-Ingulf of Crowland abbey.8 Later in the century other monastic histories were written: Prior John Wessington's history of St Cuthbert's, Durham; 9 Thomas Rudborne's history of St Swithun's, Winchester;10 Prior John Flete's history of Westminster abbey;11 the history of Bermondsey abbey;12 and the history of Hyde abbey, Winchester.13 When John Chinnock was abbot the situation at Glastonbury would have been particularly conducive to the composition of a monastic history. The principal purpose of a work of this kind was to increase prestige and improve the esprit de corps of a house by demonstrating its long and creditable past. External hostility and internal trouble both tended to result in such productions. Glastonbury abbey, having enjoyed harmony and prosperity earlier in the fourteenth century, suffered discord and economic reverse under Chinnock. Chinnock's election was disputed by another monk, Thomas Coffyn, who had a substantial following, and whose opposition was only finally defeated in 1408.14 (It is noteworthy that Thomas Burton started writing in similar circumstances. He was elected abbot of Meaux in 1396, but there was a rival candidate, and his
4
Carley (1978), i. xxxvii n. 1. Historia et Cartularium Monasterii Sancti Petri Gloucestriae, ed. W.H. Hart (Rolls Series, 1863-7), i. 3-58. 6 Historia Monasterii S. Augustini Cantuariensis, ed. Charles Hardwick (R.S., 1858). 7 Chronica Monasterii de Melsa, ed. E.A. Bond (R.S., 1866-8, 3 vols). 8 Rerum Anglicarum Scriptorum Veterum Tom. /, ed. William Fulman (Oxford, 1684), pp. 1133. H.T. Riley, 'The history and charters of Ingulfus considered', Archaeological Journal, xix (1862), 114-28, dates the pseudo-Ingulf to c. 1414, but W.G. Searle, Ingulf and the Historia Croylandensis (Cambridge Antiquarian Society, octavo series, xxvii, 1894), p. 207, dates it to the mid-fifteenth century, and F. Liebermann, 'Ueber Ostenglische Geschichtsquellen des 12., 13., 14. Jahrhunderts, besonders den falschen Ingulf, Neues Archiv der Gesellschaft fur dltere deutsche Geschichtskunde, xviii (1892), 262-3, to the midfourteenth century. 9 Unpublished. See H.S. Offler, Medieval Historians of Durham (Durham, 1958), p. 17, and R.B. Dobson, Durham Priory 1400-1450 (Cambridge, 1973), pp. 379-81. 10 Anglia Sacra, ed. Henry Wharton (London, 1691), i. 177-286. 11 History of Westminster Abbey, ed. J.A. Robinson (Cambridge, 1909). 12 Annales Monastici, ed. H.R. Luard (R.S., 1864-9, 5 vols), iii. 423-87. 13 Liber Monasterii de Hyda, ed. E. Edwards (R.S., 1866). 14 For an account of Chinnock's rule see Ian Keil, 'Profiles of some Abbots of Glastonbury', Downside Review, Ixxxi (1963), 356-62. 5
John of Glastonbury, Cronica sive Antiquitates Ecclesie
293
election entailed so much discord that he resigned in 1398. He wrote his chronicle in retirement in the abbey.) The conflict at Glastonbury not only lowered morale, but also resulted in expensive litigation. The abbey's finances deteriorated further owing to a rise in prices and a decline in profits and rents. It is, therefore, likely that the monks would have turned to their legends for help, in the same way as they had done after the fire which devastated the abbey in 1184. Chinnock was the first abbot actively to promote the cult of St Joseph of Arimathea at Glastonbury, which cult had lain more or less dormant since its inception there in the thirteenth century.15 In 1382 he dedicated a chapel which he had restored in the cemetery to St Michael and St Joseph. He may also have commissioned the Magna Tabula and had it put in the abbey church for the benefit of pilgrims. The tabula^ comprising folding wooden boards with parchment pasted on them, is almost exclusively devoted to the legends of the abbey's foundation and early history, all derived from the Cronica.16 Chinnock's responsibility for it is suggested by the fact that its one entry not copied from the Cronica records his dedication of the chapel of St Michael and St Joseph.17 Circumstances at Glastonbury would have provided Chinnock with ample reason for commissioning the Cronica at any time during his abbatiate. However, he may have had an immediate incentive, at a specific date. Perhaps this incentive took the form of encouragement by the king, Henry IV, himself. Chinnock's contact with the Lancastrians is proved by the fact that in 1399 he was one of the delegation sent to Richard II to announce his deposition.18 From the twelfth to the fourteenth centuries there had been sporadic royal interest in the Glastonbury legends.19 Henry II probably had some part in Abbot Henry de Soilli's decision to turn the Arthurian legends to the abbey's advantage, and both Edward I 20 arid Edward III 2 1 visited King Arthur's tomb there. Royal fostering of the legends was not always the result only of concern for the abbey's welfare or of chivalric enthusiasm for the cult of King Arthur. At
15 See V.M. Lagorio, 'The evolving legend of St. Joseph of Glastonbury', Speculum, xlvi (1971), 213-18. 16 The tabula, now MS Lat. Hist. A. 2 in the Bodleian Library, is described and discussed by J.A. Bennett, 'A Glastonbury Relic', Proc. Somersetshire Arch, and Nat. Hist. Soc., xxxiv(1886), 117-22. 17 Ibid., p. 120. 18 Rot. ParL, iii. 422. 19 Seejourn. ofEccles. Hist., xxvii (1976), 354-5. 20 For Edward I's visit in 1278 see R.S. Loomis, 'Edward I, Arthurian enthusiast', Speculum, xxviii (1953), 115-16. 21 For Edward Ill's visit in 1331 see The Great Chartulary of Glastonbury, ed. Aelred Watkin, i (Somerset Record Soc., lix, 1947), 194-5.
294
Legends, Traditions and History in Medieval England
least Henry II and Edward I had in addition a political motive;22 demonstrable proof that Arthur was dead and would not return, as some believed, to lead the Britons to victory against the English, strengthened their hands against the Welsh. Henry IV may well have seen the legend of Joseph of Arimathea at Glastonbury as a weapon against the French at the Council of Pisa in 1409. The precedence of the 'nations' at the general councils was determined by the date of the country's conversion. A corollary of the legend of St Joseph at Glastonbury was the apostolic conversion of the English. This put Christianity in England on an equal footing with that in France; although the French claimed actual conversion by St Denis, they asserted that the first mission was by Mary Magdalen, Martha and Lazarus to Provence. In fact the legend of St Joseph was cited by the English party to back its claim to precedence at a series of general councils - at Pisa, Constance (1417), Siena (1424) and Basle (1434)23 Chinnock, therefore, may have commissioned the Cronica preparatory to the Council of Pisa, which he himself attended.24 Perhaps he was also at the Council of Constance,25 and his successor Nicholas Frome was at both the Council of Siena and that of Basle.26 If the composition of the Cronica was at least in part the result of the conciliar movement, it might have a historiographical parallel. It is likely that Henry V commissioned the Gesta Henrici Quinti as propaganda partly to fortify his position at the Council of Constance.27 However, no doubt Chinnock would not have commissioned the Cronica solely to help the English party at Pisa; he would also have intended that it would provide 22 For Henry IPs political motive for fostering the cult of King Arthur at Glastonbury see R.F. Treharne, The Glastonbury Legends (London, 1967), pp. 105-6, and for Edward I's see F.M. Powicke, King Henry HI and the Lord Edward (Oxford, 1947), ii. 724. The hope of Arthur's return may also have encouraged the Celts to resist Henry II in Normandy; see J.S.P. Tatlock, 'Geoffrey and King Arthur in Normannicus Draco', Modern Philology, xxxi (1933-4), 122-3. See also above pp. 170-1. 23 See CJ. Hefele, Histoire des conciles, ed. H. Leclercq (Paris, 1907-38, 20 vols), vii, pt. i, pp. 31-2 n. 2, and J. Ussher, Britannicarum Ecclesiarum Antiquitates (London, 1687), pp. 1314. 24 G.D. Mansi, Sacrorum conciliorum . . . collectio (Florence etc., 1759-1962, 31 vols), xxvii, col. 349. 25 The list of delegates to the Council of Constance (ibid, xxviii. coll. 626-54) does not include Chinnock, but a recent scholar presumes he was there; Logario, loc. tit., p. 223 n. 64. 26 For Frome's presence at Siena see Hefele, op. tit., vii, pt. i, p. 614, n. 3, and at Basle see Thomas Rymer, Foedera (London, 1704-35, 20 vols), x. 577, and Official Correspondence of Thomas Bekynton, ed. George Williams (R.S., 1872, 2 vols), ii. 259. 27 See J.S. Roskell and F. Taylor, 'The authorship and purpose of the Gesta Henrici Quinti: Part F, Bull. John Rylands Lib., liv (1971-2), 227-40. It is of course possible that the Cronica was likewise written preparatory to the Council of Constance. However, it seems more likely that it was written preparatory to the Council of Pisa; there is no certain evidence that Chinnock was at Constance, and the date of that council is further removed from the chronicler's target ofc. 1400 than that of Pisa.
John of Glastonbury, Cronica sive Antiquitates Ecclesie
295
proof, to be rehearsed with full publicity before an international audience, of his abbey's priority of foundation. We must, I believe, accept that the weight of the evidence favours the traditional date, c. 1400, for the composition of the Cronica. Nevertheless, the problem remains how to explain the fact that the Cronica ends in 1342 while the author claims in the prologue to have written to c. 1400. He may have composed the prologue before, or in the course of, writing the Cronica, and then failed to complete his task. Indeed, the Trinity MS contains , besides the John of Glastonbury's Cronica, numerous documents relating to the abbacy of Walter de Monington , which could be collectanea made in preparation for the composition of an account of his rule.) A parallel case would be Thomas Elmham (d.? 1426) who makes plain his intention to write the history of St. Augustine's, Canterbury, up to his own day but only reached 806, leaving only brief particulars to 1089 and then a collection of documents to 1191.28 Similarly, neither John Flete nor John Wessington ever completed their histories of Westminster abbey and the see of St. Cuthbert respectively.29 However, other possible reasons for the discrepancy between the actual terminal date of the Cronica and that given in the prologue is suggested by a piece of evidence which has recently come to light. Professor Donald Watt has called my attention to a passage in Walter Bower's Scotichronicon. Bower, writing in the 1440s, was proRichard II and in his annal for 1398 explains how Henry Bolingbroke was not the nearest heir to the throne. He continues: Iste Henricus Lancastrie dux Herfordie et comes de Darby quando sibi regni diadema de facto assumpsit misit abbati de Glasynbiri pro actis parliamenti et cronica que posuerunt filias Rogeri de Mortuomari debere succedere, et quia excusando negavit, recognovit temporalitatem donee cronicam optinuit et incendit ac novas pro se facientes fabricare jussit.30 There seems no reason to disbelieve this story, especially as it accords with the Lancastrians' propagandist use of chronicles in their attempt to prove that Bolingbroke had a hereditary right to the throne.31 According 28
See Historia Monasterii Sancti Augustini Cantuariensis, ed. Hardwick, p. 344 n. 1. Wessington, prior of Durham 1416-46, completed his history only to 1362 (his history is unprinted; see A. Gransden, Historical Writing in England, ii, c. 1307 to the Early Sixteenth Century (London, 1982), p. 393 n. 13), and Flete, prior of Westminster 1457-65, completed his history only to 1386 (The History of Westminster Abbey, ed. Robinson, pp. 1, 33). 30 Scotichronicon by Water Bower, ed. with an English translation, general editor D.E.R. Watt (Aberdeen, 1987-, 9 vols, in progress), viii, Books XV and XVI, ed. D.E.R. Watt (Aberdeen, 1987), pp. 21-2. 31 Adam of Usk was one of the commission appointed to examine the legal justification for Bolingbroke's succession to the throne; Chronicon Adae de Usk A.D. 1377-1421, ed., with an English translation, E. Maunde Thompson (2nd edn., Oxford U.P., London, 1904), pp. 29-31. 29
296
Legends, Traditions and History in Medieval England
to John Hardyng (writing in the 1460s), John of Gaunt had a chronicle forged and distributed to various monasteries, in support of the claim of his son, Bolingbroke, which alleged that Edmund Crouchback, the Lancastrians' ancestor, and not Edward I, from whom the Mortimers derived their claim, was Henry Ill's eldest son.32 The claim to the throne of Roger Mortimer, fourth earl of March, was through his mother, Philippa, daughter of Lionel of Clarence, Edward Ill's third son, and wife of Edmund, third earl of March. The Eulogium Historiarum asserts that in the October parliament of 1385 Richard II proclaimed that Roger was next heir to the throne.33 Although historians have tended to dismiss this statement,34 the March claim was undoubtedly good, and Roger's successor, Edmund, fifth earl of March, became the focus of revolt and sedition early in Henry IV's reign.35 Bower's reference to the right of succession to the throne of Roger's daughters is puzzling. Roger himself was killed on 20 July 1498. He had four children: Anne, born 27 December 1388; Edmund, born 6 November 1391; Roger, born 23 April 1393; and Eleanor, the date of whose birth is not apparently known.36 It is tempting to place Eleanor's birth between Anne's and Edmund's. In that case, if we accept Bower's evidence, the entry in the Glastonbury chronicle to which he refers was probably written in, or shortly before or after, 1390. Possibly the chronicle was continued after that date to the end of the century. If so, John's chronicle as it survives today may be a copy of a truncated text; perhaps originally it extended to the late fourteenth century, but Henry destroyed the section from 1342, which would have contained the offending passage or passages. An objection to this theory is that the chronicle is a local history of Glastonbury abbey, not a national history. However, it does have passages concerning political events, and these might have increased in proportion as the author reached Richard IPs exciting reign, which culminated in the deposition. An alternative interpretation is that Henry destroyed the whole of some other Glastonbury chronicle of a general nature, and the extant Cronica, which would have positively pleased him, 32
The Chronicle of John Hardyng, ed. Henry Ellis (London, 1812, repr. New York, 1974), pp. 353-4. 33 Eulogium Historiarum . . ., ed. F.S. Haydon, (R.S. 1858-63, 3 vols), iii, 361. 34 See T.F. Tout, Chapters in the Administrative History of Medieval England (Manchester, 1920-33, 6 vols), iii. 396 n. 1; M.V. Clarke, Fourteenth Century Studies, ed. L.S. Sutherland and M. McKisack (Oxford, 1937), p. 107 n. 5; Anthony Tuck, Richard II and the English Nobility (London, 1973), p. 205. However, Anthony Steel, RichardII (Cambridge, 1962), p. 214, gives the story more credence. 35 Usk, op. cit., p. 82. Cf. G.L. Harriss, Henry V, The Practice of Kingship (Oxford U.P., 1985), p. 32. 36 G.E. C[okayne], Complete Peerage of England, Scotland, Ireland, Great Britain and United Kingdom . . ., new edn by Vicary Gibbs et al. (London, 1910-59, 12 vols, vol. xii in two parts), viii. 449-450 and n. a.
John of Glastonbury, Cronica sive Antiquitates Ecclesie
297
was composed to replace it. Finally, there is the question of the identity of'John', the author of the Cronica. He declares himself in the prologue as 'Johannes, abiectissimus et humilimus eiusdem venerabilis loci quamuis indignus confrater et monachus' (Carley, i.5). Carley's candidate is John Seen, a monk of Glastonbury who received a D.Th. from Oxford in 1360, and who Carley thinks died before 1377 (Carley, pp. xxix-xxx). Obviously if the Cronica was written, as I believe, c. 1400, Seen cannot have been the author. Even supposing that the Cronica was written between 1340 and 1342, the case for his authorship is slim. He composed a now lost history of the Trojan war, and in addition, is referred to in a letter of John Grandison, bishop of Exeter, to Walter de Monington: '. . . venerabilis et religiosus vir dompnus Johannes Sene domus vestre commonachus inter omnes huius regni peritos de gestis antiquis et cronicis magis novit . . .'. Carley suggests that the gestae antiquae was Seen's history of the Trojan War, and that cronicae denoted the Cronica. However, the reference is probably to Seen's historiographical learning, not to any activity as a chronicler. It should perhaps be observed that twenty-three of the sixty-three monks in the list of those who professed under Monington have the Christian name 'John'.
Note Pages 290-7. Most recently Professor Carley's redating of John of Glastonbury's Cronica has been accepted without reservation as correct by C.T. Wood, 'Fraud and its Consequences: Savaric of Bath and the Reform of Glastonbury' in The Archaeology and History of Glastonbury Abbey, ed. Lesley Abrams and J.P. Carley (Woodbridge, 1991), p. 282. (See also Wood's review of Carley's edition of the Cronica in Speculum, Ixi (1987), 4279.) Wood claims that the De Gestis Mirabilibus Edwardi Tertii by Robert of Avesbury (d. 1359) provides conclusive evidence for Carley's dating. He asserts that the De Gestis, which he dates c.1350, traces King Arthur's ancestry back to Joseph of Arimathea. Since John of Glastonbury also gives this genealogy (ed. Carley, p. 56), which was otherwise unknown in the fourteenth century, Avesbury's information, Professor Wood argues, must have come from John's Cronica. Wood's reasoning here is not sound because Avesbury could have derived the genealogy from some other, now lost, source, perhaps the same one as John used. But more important, Wood's premise is itself false. Avesbury's in his De Gestis (written in 1356 or shortly afterwards) gives no genealogy of King Arthur. The genealogy in question occurs in some notes appended to the earliest extant manuscript of the De Gestis, BL MS. Harley 200, a copy made in the early fifteenth century. The notes are in the same hand as, or in a hand contemporaneous with, that of the scribe of the De Gestis. (See Adae Murimuth Continuatio Chronicorum. Robertus de Avesbury De Gestis Mirabilibus Regis Edwardi Tertii, ed. E. Maunde Thompson (Rolls Series, 1889), pp. xxiii-xxiv.) Thomas Hearne printed the notes as 'Minutae' in his edition of the De Gestis (Historia de Mirabilibus Gestis Edwardi HI (Oxford, 1720), pp. 256-66; Arthur's genealogy is on p. 259). Except for one note (a genealogy of Edward III) Hearne did not believe that the notes were by Avesbury (ibid., pp. x-xi). Carley in the introduction to his edition of John's Cronica (p. xxviii) describes the notes (which he dates £.1350) as anonymous. He cites R.H. Fletcher, The Arthurian Material in the Chronicles (Harvard Studies and Notes in Philology and
298
Legends, Traditions and History in Medieval England
Literature, x, 1906), pp. 176, 189; Fletcher, who cites Hearne p. 259, describes the notes as of uncertain authorship. Therefore, since the De Gestis does not contain Arthur's genealogy and the note which does was almost certainly not by Avesbury, Wood cannot use the evidence of the genealogy to prove that John of Glastonbury's Cronica was Avesbury's source. For the sake of accuracy it should also be mentioned that Wood's belief that the notes occur in only one manuscript of the De Gestis is inexact. The other two surviving copies have some of the notes appended, though slightly differently arranged. Both these copies of the De Gestis were made in the first half of the fifteenth century and derived from Harley 200 (Maunde Thompson, op. cit., pp. xxv-xxvi).
14 Antiquarian Studies in Fifteenth-Century England Antiquarian studies in medieval England have had a bad press. To Professor McKisack such studies begin with John Leland:l As has long been recognized, Leland's distinction as an antiquary derives, first and foremost, from the fact that he was a pioneer in the method of direct enquiry and first-hand observation, the forerunner of Camden. The originality of his approach [is] revealed in the Itinerary . . . And yet over twenty years before Sir Thomas Kendrick had done ample justice to the fifteenth-century antiquaries, John Rous and William Worcester.2 However, he treats them as heralds of a new era. While admitting that previously attempts had been made to study the past 'through the intelligent use of archives and visible monuments' (he cites a few examples from the twelfth and thirteenth centuries),3 he characterizes the antiquarianism of medieval England as fanciful and prone to fiction.4 The chronicler, Sir Thomas asserts, had no training in criticism, and his view of the past was dominated by the British History.5 This corpus of legend owed its origin to Geoffrey of Monmouth, but, in the words of Sir Thomas, it was 'not a static system of antiquarian belief. . . but a thriving garden of spurious history in which any transitory nonsense about the remote past might take root and flourish'.6 In fact Sir Thomas saw the 1
May McKisack, Medieval History in the Tudor Age (Oxford, 1971), p. 11. T.D. Kendrick, British Antiquity (London, 1950), pp. 18-33. 3 Ibid., p. 18 and nn. 3-4. 4 For example, Sir Thomas states that Leland, like many antiquaries of his day, remained 'medieval in mind'; he was 'a man two-faced, in one direction looking hopefully forward into a new era of empirical research and practical survey, and at the same time looking with affection backward to the writing-desk of the medieval scholastic chroniclerantiquary where a traditional fable might be repeated without unrestful inquiry or impertinent sixteenth century doubt'; ibid., p. 63. Similarly, Sir Thomas comments that in the sixteenth century 'antiquarian thought tended to remain medieval in kind - medieval in credulity and in recklessness of conjecture'; ibid., p. 65. 2
5 6
Ibid., pp. 2 et seq. Ibid., p. 15.
300
Legends, Traditions and History in Medieval England
history of antiquarianism from the sixteenth century onwards as a battle between the British History and the new criticism of the Renaissance. In my opinion this view is misleading: it exaggerates the impact of the Renaissance on historiography in England and underemphasizes the continuity from medieval to modern times. Indeed Sir Thomas neglected a vast area of medieval antiquarian study — local history. Sir Richard Southern has provided a partial antidote. He has paid tribute to the monastic historians in the Anglo-Norman period.7 Fired by the desire to prove the continuous history of their houses from AngloSaxon times, and so counter the threat to tradition posed by the Norman Conquest, they undertook minute research on monastic archives and antiquities. Unfortunately Sir Richard ends his account on a false note:8 In historical research they had no successors. The monastic historians of the later Middle Ages abandoned historical research for contemporary journalism, and relied on their predecessors for their record of the past. Thus he overlooks the important research done by the monks on local history in the reign of Henry II, not to mention Gervase of Canterbury's architectural history of Canterbury Cathedral and Matthew Paris's work on the antiquities of St Albans. He also passes over the fifteenth-century antiquaries, not only Rous and Worcester, but many others besides. Together these men make the fifteenth century (with which we are here concerned) remarkable for the quantity and quality of its antiquarian research. The fifteenth-century antiquaries fall into two groups: the monks and the seculars (we include in the term 'secular' both secular clergy and laity). The monks were active in the first half of the century, and the seculars in the second half. We will consider the monks first. A number of incentives urged monks to study the past, most emanating from their esprit de corps. Again, as in the Anglo-Norman period, the monks felt threatened, though this time their enemies were different. Since the twelfth century the regular and secular canons were their rivals and now, in the fifteenth, the Lollards and other 'dispossessioners' were their enemies. There were also litigious neighbours, interfering diocesans and rapacious tax collectors. And meanwhile some houses suffered from internal dissension. The monks, therefore, did everything they could to defend themselves and improve monastic morale. One method was by propagandist writing. Towards the end of the fourteenth century a monk of Bury St Edmunds
7 R.W. Southern, 'Aspects of the European tradition of historical writing: the sense of the past', T.R.H.S. 5th ser. xxxiii (1973), 246-56. 8 Ibid., 263.
Antiquarian Studies in Fifteenth-Century England
301
had already produced a tract proving the antiquity of monasticism.9 He traced its history back to its earliest origins, far beyond the date when, according to tradition, St Augustine instituted the regular canons. In the fifteenth century various versions of this tract were current in the monasteries and elsewhere. Individually each house might try to prove that it had a long and glorious past — the longer its tradition, the greater its reputation. The monastic historian would trace his house's history back to its origins. He would include information on its treasures and notable inmates — these subjects could also be dealt with in shorts tracts. Thus he strengthened the monastery's fortifications in a general way. But he might also fight its battles on specific issues. He might record evidence of its right in a particular legal case. Such information could be put into a narrative history or into a tract. The monastic historian had other motives. He wrote to edify his fellow monks by describing the past achievements of the community, and to inform them so that they would be knowledgeable guides for visitors. He also wrote as an act of piety. But another motive is often apparent: he wrote to record the results of painstaking research undertaken to satisfy his own curiosity — simply because he was interested. In his search for the past, the historian did not rely exclusively on verifiable fact. Vital evidence of early foundation might be lost, or the circumstances of the foundation be forgotten owing to the passage of time. So a historian might copy forged documents to substantiate what he himself believed, or he might adopt some legend to convince the reader of what he wanted him to believe. Legends, if derived from a standard authority such as Geoffrey of Monmouth, were used indiscriminately with verifiable fact. Generally speaking they only came under fire if they conflicted with an author's tendentious purpose. And then, having demolished one inconvenient legend, the author might well adopt another, equally unlikely, because it suited him. Neither forgery nor legend was totally inimical to the development of antiquarian studies. To achieve verisimilitude, the forger had to study the form and content of ancient documents and books — he might even plunder genuine material and copy archaic scripts. And a legend stimulated interest in the past, attracting immediate attention to any place or object connected with it. It is now necessary to discuss individually some of the works produced by the monasteries. First, the histories may be considered, and, secondly, a few examples of the tracts. The principal monastic histories, arranged as far as possible in chronological order of composition, are: John of
9 For this tract, its versions and their dissemination, see W.A. Pantin, 'Some medieval English treaties on the origins of monasticism', Medieval Studies presented to Rose Graham, eds. Veronica Ruffer and AJ. Taylor (Oxford, 1950), pp. 189-215.
302
Legends, Traditions and History in Medieval England
Glastonbury's history or Glastonbury Abbey (A.D. 31-1342), c. 1400;10 Walter Frocester's history of St Peter's, Gloucester (A.D. 681-£. 1412), early fifteenth century; 11 Thomas Elmham's history of St Augustine's, Canterbury (A.D. 597-806, with notes and documents to 1191), early fifteenth century; 12 Thomas Burton's history of the abbey of Meaux (1150-1396, with a continuation to 1417), first third of the fifteenth century; 13 the annals of the (Cluniac) abbey of Bermondsey (1042-1432), c. 1432; 14 John Wessington's history of St Cuthbert's, Durham (A.D. 6351195), first half of the fifteenth century; 15 Thomas Rudborne's history of St Swithun's, Winchester (A.D. 164-1141), mid-fifteenth century;16 John Flete's history of Westminster Abbey (A.D. 184-1386), mid-fifteenth century; 17 the Crowland chronicle (eighth century-1148), ? mid-fifteenth century; 18 and the history of Hyde Abbey, Winchester (the so-called Book of Hyde, A.D. 455-1023), which, I suggest, was written in the mid-
10
Printed The Chronicle of Glastonbury Abbey . . . John of Glastonbury's 'Cronica sive Antiquitates Glastoniensis Ecclesie3, ed. J.P. Carley, with an English translation by D. Townsend (Woodbridge, 1985). Professor Carley in the introduction to his edition argues against the traditionally accepted date, c. 1400, for the composition of the chronicle. However, I do not find his arguments convincing; see pp. 291-8 above. 1J Printed Historia et Cartularium Monasterii Sancti Petri Gloucestriae, ed. W.H. Hart (Rolls Series, 1863-6, 3 vols.), i, 3-58. For the composition and value of this chronicle see C.N.L. Brooke, 'St Peter of Gloucester and St Cadoc of Llancarfan', Celt and Saxon. Studies in the Early British Border, ed. N.K. Chadwick (Cambridge, 1963), pp. 260-70, 277-9. 12 Printed Historia Monasterii S. Augustini Cantuariensis, ed. Charles Hardwick (Rolls Series, 1858). Its medieval title was Speculum Augustinianum. 13 Printed Chronica Monasterii de Melsa, ed. E.A. Bond (Rolls Series, 1866-8, 3 vols.). 14 Printed in AnnalesMonastici, ed. H.R. Luard (Rolls Series, 1864-9, 5 vols.), Hi, 423-87. 15 John Wessington's Libellus de exordio et statu ecclesie . . . Dunelmensis, ac de gestis pontificum eiusdem [635-1195] is unpublished. For the three manuscripts of the Libellus see H.H.E. Craster, The Red Book of Durham', E.H.R., xl (1925), 504-14 passim. See also H.S. Offler, Medieval Historians of Durham (Durham, 1958), p. 17, and R.B. Dobson, Durham Priory 14001450 (Cambridge, 1973), pp. 379-81. 16 The Historia Major . . . Ecclesie Wintoniensis of Thomas Rudborne is printed in Anglia Sacra, ed. Henry Wharton (London, 1691, 2 vols.), i, 177-286. 17 Printed History of Westminster Abbey by John Flete, ed. J. Armitage Robinson (Cambridge, 1909). 18 Printed in Rerum Anglicarum Scriptorum Veterum Tom. /, ed. William Fulman (Oxford, 1684), pp. 1-132. Various dates have been suggested for the composition of the Crowland chronicle, but the most recent scholar to study the question argues in favour of the midfifteenth century; W.G. Searle, Ingulf and the Historia Croylandensis (Cambridge Antiquarian Soc., 8vo ser. xxvii, 1894), p. 207. Cf. H.T. Riley, 'The history and charters of Ingulfus considered', Arch. J. xix (1862), 114-28, and F. Liebermann, 'Ueber ostenglische Geschichtsquellen des 12, 13, 14 Jahrhunderts, besonders den falschen Ingulf, Neues Archiv der Gesellschaft fur dltere deutsche Geschichtskunde, xviii (Hanover and Leipzig, 1892), 262-3.
Antiquarian Studies in Fifteenth-Century England
303
fifteenth century, perhaps by Thomas Rudborne himself.19 This list illustrates that the author's primary interest tended to be in the foundation and early history of his house. Indeed only three of the ten works, Thomas Burton's history of the abbey of Meaux, the Bermondsey annals, and Walter Frocester's history of St Peter's, Gloucester, reach the author's own time. The rest stop short of it — some, notably Thomas Elmham, well short. A number were written by eminent men, which shows the importance attached to such histories. Two were well-known writers. Thomas Rudborne, author of the history of his monastery of St Swithun's, Winchester, and possibly of the Book of Hyde, also wrote other histories,20 and his reputation is witnessed by John Rous.21 Thomas Elmham, after he left St Augustine's in 1414, wrote a successful verse Life of Henry V. He had studied at Oxford, and left St Augustine's to become prior of the Cluniac priory of Lenton in Nottinghamshire, and in 1415 he became vicar-general of the Cluniacs in England and Scotland.22 Four of the authors were heads of their houses. Walter Frocester was abbot of St Peter's, Gloucester, from 1382 to 1412;23 John Wessington was prior of Durham from 1416 to 1446; John Flete was prior of Westminster from 1457 to 1465. And Thomas Burton was briefly abbot of Meaux from 1396 until 1399 - he was forced to resign because of discord in the convent, and wrote his chronicle in retirement. 24 It is relevant to note that John Rous tells us that the Book of Hyde was in the abbot's keeping.25 The esteem in which such works were held is underlined by the fact that all (except the
19 Printed Liber Monasterii de Hyda, ed. Edward Edwards (Rolls Series, 1866). The possibility that Rudborne was the author of the Liber is suggested by the similarity of its arrangement, under kings, with that of the Historia Major, and also by the fact that it uses two of the unidentified sources cited in the Historia Major — Girardus Cornubiensis's De Gestis Regum West-Saxonum (see Hist. Major, pp. 189, 201, 204 etc., and, for Girardus, T.D. Hardy, Descriptive Catalogue for Materials relating to the History of Great Britain and Ireland to 1327 (Rolls Series, 1862-71, 3 vols. in 4 pts., reprinted New York, 1963, iii, 50-1)), and 'Vigilantius' (for which see p. 304 and nn. 31, 32 below); see Edwards, op. cit., p. xxi. The evidence of the only extant manuscript is compatible with this suggestion; see p. 305 and n. 40 below. Perhaps further research would result in a firm conclusion. 20 For works attributed to Rudborne, see Wharton, op. cit., i, pp. xxvi-xxviii. However, Wharton's list needs revision. For Rudborne's possible authorship of the Book of Hyde see above and previous note. 21 Joannis Rossi Antiquarii Warwicensis Historia Regum Angliae, ed. Thomas Hearne (Oxford, 1716), p. 73. For references to the Historia Major see ibid., pp. 78, 82, 96, 98. 22 For Elmham's career see J.S. Roskell and Frank Taylor, 'The authorship and purpose of the Gesta Henrici Quinti: I', Bulletin of the John Ry lands Library, liii (1970-1), 455-61, and C.L. Kingsford, English Historical Literature in the Fifteenth Century (Oxford, 1913, reprinted New York, 1962), pp. 45-6. 23 For Frocester's authorship of the Gloucester history see Brooke, op. cit., p. 260 and n. 3. 24 See Chron, Mon. de Melsa, i, Ixii-lxx; iii, 239-40, 258-71, 274-5. 25 See Rous, Historia Regum, p. 96. Cf. p. 305 and n. 39 below.
304
Legends, Traditions and History in Medieval England
Bermondsey annals) are on a massive scale, and some survive in handsome manuscripts. Most magnificent is the manuscript of Elmham's history, now in the library of Trinity Hall, Cambridge.26 We must now consider the contents of these histories. The one which contains the most forgery is the Crowland chronicle — it is spurious from beginning to end. It shows the lengths to which a forger would go to make his work appear authentic. The author foisted the section to 1089 on Ingulf, abbot of Crowland from 1085/6 to 1109, and the section from 1089 to 1148 on the twelfth-century writer, Peter of Blois.27 'Ingulf starts in traditional style by declaring his intention to commemorate the abbey's founders and benefactors, and to record its history and acquisition of property.28 Besides citing numerous (forged) charters in full, he claims to have consulted Domesday Book - which he thought was in roll form.29 'Peter of Blois' opens by quoting a (spurious) letter of the abbot, Henry de Longchamp; in it Abbot Henry asks him to continue the chronicle and undertakes to bring documents to London from the abbey muniments. In his reply Peter praises Ingulfs work, promises to correct it, and says he will use the documents provided.30 Many of the other writers copied forged charters, but Thomas Rudborne probably used a literary forgery as well. He cites Vigilantius, De Basilica Sancti Petri, for the pre-Conquest history of the Old Minster.31 'Vigilantius' is also cited in some of the other works which may be by Rudborne.32 Professor Robert Willis used Rudborne's citations for his architectural history of the Old Minister, and concluded that 'Vigilantius' wrote shortly before the Conquest.33 More recently the late Roger Quirk has accepted 'Vigilantius's' evidence.34 But there are apparently no references to the work, which does not survive, prior to those just mentioned. Professor D.J. Sheerin of the University of North Carolina has
26 Trinity Hall, Cambridge, MS.l. See M.R. James, A Descriptive Catalogue of the Manuscripts in the Library of Trinity Hall, Cambridge (Cambridge, 1907), pp. 1-3. 27 The Crowland chronicle is discussed: Riley, op, cit., pp. 32-49, 114-33; Liebermann, op. cit., pp. 249-67; Searle, op. cit., passim. 28 Fulman, op. cit., p. 1. 29 Ibid., pp. 80, 83. Cf. Searle, op. cit., pp. 7-12. 30 Fulman, op. cit., 108-10. 31 e.g. Rudborne, Hist. Major, pp. 181, 186, 199, 223. 32 The chronicle from Lucius to the beginning of Henry VI's reign, Corpus Christi College, Cambridge, MS. 110, pp. 213 et seq. (for which see M.R. James, A Descriptive Catalogue of Manuscripts in the Library of Corpus Christi College, Cambridge (Cambridge, 190912, 2 vols.), ii, 235, and R.N. Quirk, 'Winchester Cathedral in the tenth century', Arch.J. cxiv (1957), 57, n. 1); the Epitomes Historiae Majoris Ecclesiae Wintoniensis (for which see Wharton, Anglia Sacra, i, xxvi); the Book of Hyde (see p. 303 and n. 19 above). 33 R. Willis, 'Architectural history of Winchester Cathedral', Proc. of the Royal Architectural Institute, Winchester (1845), p. 3, n.a. 34 Quirk, op. cit., p. 29, n. 2.
Antiquarian Studies in Fifteenth-Century England
305
pointed out to me35 that the account of the foundation of the Old Minster attributed to Vigilantius cannot be earlier than Geoffrey of Monmouth. This is proved by the fact that it states that the minster was founded by King Lucius and consecrated by the missionaries Phagan and Deruvian36 — all three first appear in Geoffrey's pages.37 Probably 'Vigilantius' was written later still. When the monks of St Swithun's wrote in defence of their privileges in the late thirteenth century and claimed a British origin for their house, they made no mention of King Lucius, or of Fagan and Deruvian.38 Indeed Rudborne's is the earliest known reference to them in this context. Perhaps 'Vigilantius' never existed as an independent work and Rudborne's appeal to it to substantiate his history was in imitation of Geoffrey of Monmouth's appeal to an ancient book as his authority. It is worth noting that Rous describes the Book of Hyde, which the abbot showed him, as a fine copy in a recent hand made from a very old manuscript in case the latter perished.39 A recent scholar has dated the extant text to the fourteenth or fifteenth century. It may be suggested that the manuscript is mid-fifteenth-century and that the work was composed then.40 Rudborne wrote partly to establish the Old Minster's priority of foundation. He may have had in mind the claims of the Glastonbury monks, who alleged that their abbey was founded by Joseph of Arimathea. Perhaps he chose Phagan and Deruvian as the consecrators of the minster because Glastonbury named them as the refounders of their abbey.41 He
35
In a letter of 7th December 1976. For a fuller discussion see A. Gransden, Historical Writing in England, ii, c.1307 to the Early Sixteenth Century (London, 1982), pp. 493-4 (Appendix D). 36 Hist. Major, pp. 181-2. 37 The Historia Regum Britanniae of Geoffrey of Monmouth, ed. Acton Griscom (London, 1929), pp. 328-30 (Bk IV, cap. xix). 38 The monks were attempting to prove that the king, not the bishop of Winchester, was their patron; see Registrum Johannis de Pontissara Episcopi Wyntoniensis, ed. Cecil Deedes (Canterbury and York Soc., xix, xxx, 1915, 1924, 2 vols.), ii, 609-15. For the conflict between the bishop and his chapter see ibid., i, xx-xxi; 676-94. The monks even earlier had falsified the history of the Old Minster; see H.P.R. Finberg, The Early Charters ofWessex (Leicester, 1964), pp. 226-44passim. 39 For the foundation of Cambridge University, Rous cites a chronicle of Hyde which must surely be the extant Book of Hyde (for the account in it of the foundation of Cambridge University see Liber Monasterii de Hyda, ed. Edwards, p. 11). He writes that he found the passage 'in dicta abbathia [de Hyde] in quadam bene indictata nobili Chronica, quae de vetusta et antiqua manu iterum, ne periret, nova manu et placida scripta est, ut egomet vidi benevola licentia domini abbatis ejusdem loci'; Historia Regum Angliae, p. 96. 40 G.R.C. Davis, Medieval Cartularies of Great Britain (London, 1958), p. 121, no. 1051. The manuscript is owned by the earl of Macclesfield and is in the library of Shirburn Castle. A page is reproduced in facsimile as a frontispiece to Edwards, op. cit. 41 Chronicle of Glastonbury Abbey, ed. Carley, p. 56.
306
Legends, Traditions and History in Medieval England
states categorically that Glastonbury Abbey was founded by King Ine.42 Moreover, he writes slightingly of King Arthur, who was supposedly buried there. For example, he proves from literary evidence that Arthur could not have defeated an emperor called Lucius, because no emperor of that name lived in his (supposed) time.43 Legends are a common ingredient of the monastic histories, but none contains as many as John of Glastonbury's. John wrote initially to counter Higden's assertion that the St Patrick buried at Glastonbury was not the apostle of Ireland, but a ninth-century abbot. And he wrote to show that previously the abbey was more privileged and prosperous than it was in his day, but had suffered from the neglect of 'certain prelates', the depredations of Danish and Norman invaders, and from the oppression of the bishops of Bath.44 Using the earlier histories of Glastonbury and perhaps his own imagination, John produced a definitive history of the abbey to his own day, replete with legends of Joseph of Arimathea and King Arthur. 45 It is not, however, for the use of forgeries and legends that most of the monastic historians are remarkable, but for the sound research they did. They cite charters and other domestic documents, and even public records — presumably from extracts in monastic registers. Dr Craster imagined John Wessington 'sitting down, with a pile of manuscripts from the conventual library at his elbow, to compile a history of his monastery'.46 Some also used visual evidence. Even Rudborne, who leant heavily on written sources, described the tombs in Winchester Cathedral, and copied the inscriptions.47 Flete's notices of the tombs of the abbots in Westminster Abbey are crucial evidence for the abbatial succession.48 But most outstanding in their use of documentary and visual evidence are Thomas Elmham and Thomas Burton, who will be discussed in more detail. Elmham's commemorative intention, his need to express his piety and satisfy his curiosity, are all evident in his work. But he wrote primarily to defend the privileges of St Augustine's. The abbey's perennial enemy was
42
Hist. Major, p. 194. Ibid., pp. 187-8. Chronicle of Glastonbury Abbey, ed. Carley, p. 8. 45 For the legends in John of Glastonbury see J. Armitage Robinson, Two Glastonbury Legends: King Arthur and St Joseph of Arimathea (Cambridge, 1926), passim. 46 Craster, op. dt.9 p. 514. 47 For his record of the original burial place of St Swithun (Hist. Major, p. 203) see Quirk, op. cit., 65 and n. 6, and Martin Biddle and R.N. Quirk, 'Excavations near Winchester Cathedral, 1961', Arch.J. cxix (1962), 174 and n. 6. 48 See Flete, pp. 22-4. Flete also mentions the tapestries in the church, and the sumptuous mosaic pavement before the high altar; ibid. pp. 105 (cf. pp. 24-9), 113. 43
44
Antiquarian Studies in Fifteenth-Century England
307
the archbishop of Canterbury:49 the archbishops had periodically challenged its right to exemption from episcopal control.50 By Elmham's day the matter had been settled in the monks' favour, but he mentions certain trouble-makers who scoffed at their claims.51 Elmham explicitly states his respect for documentary evidence. When, he writes, the guiding light of Bede failed, by good fortune he could turn to the archives of his house, to papal and royal documents, to bulls and charters, sealed and indented; few historians before him had enjoyed such a benefit.52 He studied both the content of documents and their physical appearance - he had a developed visual sense which was supported by talent as a graphic artist. He paid close attention to the abbey's four 'earliest' documents, all in fact late eleventh-century forgeries,53 the privilege of St Augustine and the three charters of King Ethelbert. It was no doubt to reaffirm their authenticity, which had been called in question by the archbishops of Canterbury in the twelfth and thirteenth centuries,54 that Elham reproduced them in facsimile as well as in the handwriting of his own time. He copied the privilege of St Augustine (pi. 30) and Ethelbert's third charter in uncials, and Ethelbert's first and second charter in Anglo-Saxon script;55 he has reproduced, he claims, referring to Ethelbert's third charter, the size and script of the original, for the benefit of posterity. During the abbey's struggle with Archbishop Richard in Henry IPs reign, one of the faults found with the privilege of St Augustine was its leaden bull. The archbishop's party objected that only popes used such bulls.56 No doubt to demonstrate its genuineness Elmham drew its 49
For a synopsis of the causes of contention between St Augustine's and the archbishops of Canterbury, see Wilhelm Levison, England and the Continent in the Eighth Century (Oxford, 1946), pp. 182-3. 50 See ibid., and in particular M.D. Knowles, 'Essays in monastic history, iv. The growth of exemption', Downside Review 1 (1932), 401-15. 51 Elmham, op. cit., pp. 87-8. Elmham also had to counter the attacks of the canons of St Gregory's, Canterbury, who claimed to have the relics of St Mildred; ibid., pp. 218-19, 2256. See M.L. Colker, 'A hagiographic polemic', Mediaeval Studies, xxix (1977), 61-2. 52 Ibid., pp. 309-10. 53 Levison, op. cit., pp. 205-6; Elmham, op. cit., pp. xxvii-xxxiv. 54 Knowles, op. cit., p. 414 and nn.; Elham, op. cit., pp. xxvii-xxxiv; William Thome's Chronicle of Saint Augustine's Abbey Canterbury, translated by A.H. Davis, with a preface by A. Hamilton Thompson (Oxford, 1934), pp. liv-lvi, 116 et seq. 55 Elmham, op. cit., pp. 109 and n. 4, 110, 111 and n. 1, 112-13, 114 and n. 1,115-16,119 and n. 3, 120-1. The facsimiles are described and some reproduced in Michael Hunter, 'The facsimiles in Thomas Elmham's History of St Augustine's, Canterbury', The Library, 5th ser. xxviii (1973), 215-20. 56 Elmham, op. cit., pp. 122-4. Elmham erroneously dates this dispute to Henry Ill's reign, apparently because he identified Archbishop Richard as Richard Grant, archbishop 1229-31. However, the reference (see below) to Philip, count of Flanders (1168-91) makes it clear that the archbishop concerned was Richard of Dover (1173-84). This conclusion is
308
Legends, Traditions and History in Medieval England
obverse and reverse, with an explanatory note and a copy of the inscription.57 And he pointed out that St Augustine, as a Roman and a papal legate, might have used a bull. Indeed, he asserts, in later times popes had occasionally granted a bishop the right to use one. He records that, to counter Archbishop Richard's objection, Philip, count of Flanders, had sent the abbot a bull to use in evidence: Philip said that it had been given to him by a bishop, who alleged that he and his predecessors had used it.58 The abbey's enemies had objected to the three charters of Ethelbert because they had no seals. Elmham rejoins, more or less correctly, that the practice of sealing documents was introduced after the Conquest: previously only King Canute, a foreigner, had used a seal; the Anglo-Saxon kings had authenticated documents with the sign of the cross.59 Since the eleventh-century forgeries are lost, we do not know if they were in facsimile, or whether the idea of producing fascimiles was Elmham's own. Examples of the imitation of archaic scripts occur from the tenth century onwards.60 (At St Augustine's plenty of palaeographical models were at hand among its early books and charters.) 61 Nor was it unprecedented to copy authenticating devices: twelfth-century examples survive from St Augustine's itself.62 However, reproduction of the total format of a document does seem to be a new concept. Elmham's interest in documents extended beyond the four 'earliest' charters of St Augustine's. Although, as is to be expected, he copied the eleventh-century forgeries without question, his treatment of documents was in general well-informed and judicious. He could read Anglo-Saxon63 confirmed by the account of the dispute between St Augustine's and Archbishop Richard of Dover given by Gervase of Canterbury, which mentions that the leaden bulla was then called in question; The Historical Works of Gervase of Canterbury, ed. William Stubbs (Rolls Series, 1879-80, 2 vols.), i, 296-7. 57 Reproduced Hunter, op. cit., pi. III. 58 Elmham, op. tit., p. 123. 59 Ibid., p. 118. Rous made a rather similar statement; see p. 320 and n. 123 below. 60 See N.R. Ker in Margaret Deanesly, 'The court of King ^Ethelberht of Kent', Cambridge Historical Journal, vii (1942), 107, n. 11. 61 Elmham's library list (see p. 310 below) includes at least one ancient manuscript in uncials. For the school of uncial writing at Canterbury, c. 700, see E. Maunde Thompson, An Introduction to Greek and Latin Palaeography (Oxford, 1912), pp. 384-5. Professor Deanesly argues that Elmham's facsimiles of Ethelbert's charters are in Merovingian script and are based on genuine documents; Deanesly, op. cit., pp. 103-10 passim, and the same author's 'Early English and Gallic minsters', T.R.H.S. 4th ser. xxiii (1941), 53-66 passim, and 'Canterbury and Paris in the reign of ^Ethelberht', History, xxvi (1941), 101-4. However, Levison, op. cit., pp. 174 et seq., has demonstrated that this view is not sound. 62 BL MS. Cotton Vitellius A II, folios 14, 19, 19V; see Hunter, op. cit., 218 and pi. VI (a)-(c). See also the picture of the rota of William II of Sicily in 'Benedict of Peterborough': BL MS. Cotton Vitellius E XVII, fo. 28; see p. 184 and pi. 4 above. 63 Elmham copies a passage in Anglo-Saxon from a charter and supplies a Latin translation; Elmham, op. cit., p. 332.
30 Thomas Elmham's facsimile of the (forged) privilege of St Augustine to St Augustine's, Canterbury. (Cambridge, Trinity Hall, MS I Jo. 2)
31 Thomas Elmham's picture of the seal of an abbey dedicated to St Stephen. (Cambridge, Trinity Hall, MS I Jo. 24)
32 Thomas Elmham's plan of the high altar and sanctuary at St Augustine's, Canterbury. (Cambridge, Trinity Hall, MS I Jo. 77)
33 Thomas Elmham's map of Thanet. (Cambridge, Trinity Hall, MS I,fo. 42v)
34 Arthgallus, a legendary British earl of Warwick, from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976)
35 Ufa, from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976)
36 The seal, 1353-54, of Thomas Beauchamp, earl of Warwick (1329-62). (London, British Library, seal no. xliii. 18)
37 Thurkill, from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976)
38 William de Newburgh, earl of Warwick (d. 1184), from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976)
39 Waleran de Newburgh (d. P1203), from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976)
40 Thomas Beauchamp, earl of Warwick (1370-1401), from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976)
41 Richard Beauchamp, earl of Warwick (1403-39), from John Rous's 'Yorkist' Warwick roll. (London, British Library, MS Add 48976)
42 Two pages from William Worcester's Itinerary. The page on the right shows a cross section of the mouldings of the south porch of St Stephen's church, Bristol. (Cambridge, Corpus Christi College, MS 210Jo. 129)
Antiquarian Studies in Fifteenth-Century England
309
and appreciated that documents are usually more authoritative than literary sources; he twice corrects the accepted length of a reign on charter evidence.64 He also states that an original document has more authority than a copy: he collated two originals with copies in registers, noting discrepancies, which he attributed 'either to the carelessness of the scribes or to the inability of the compiler to read Anglo-Saxon script correctly'.65 Nor was Elmham's interest in authenticating devices limited to the problems raised by the abbey's four 'earliest' charters. For example he drew the leaden bull attached to the privilege of Boniface IV,66 and the bene valete and rota of Lucius II,67 at the end of his transcriptions of their grants to St Augustine's. Sometimes his interest was objective - his observations could not have helped the abbey defend its rights. He comments that the leaden bull given by Philip, count of Flanders, was not, as the count asserted, a bishop's: the inscription and picture on the obverse showed that it had belonged to some monastery dedicated to St Stephen; he drew the obverse (pi. 31), and remarked that it was almost impossible to discern the picture on the reverse, on account of its age.68 He also describes two curious examples of seals. One, a seal of William, first earl of Warenne, attached to a charter in the Cluniac priory of St Pancras at Lewes, contained some of the earl's hair. The other, the seal of an earl of Lincoln attached to a charter at the priory of Castle Acre, a cell of Lewes, bore the imprint of the earl's teeth; it had the subscription: 'In evidence of this I have impressed the seal with my teeth, as Muriel my wife witnesses.'69 64
Ibid., pp. 137,324. Ibid., pp. 233, 237-8. 66 Trinity Hall MS. 1, fo. 26V. 67 Ibid., fo. 90V. See Hunter, op. cit., p. 217, and his pi. I (b). 68 Elmham, op. cit., p. 123. 69 Ibid., pp. 118-19. The foundation charter of Lewes priory, granted by William de Warenne, first earl of Surrey, makes no reference to the presence of the earl's hair in the seal; Early Yorkshire Charters, ed. C.T. Clay (Yorkshire Archaeological Soc. record ser., extra ser. 1935-65, 10 vols.), vi, 54-5, no. 2. However, the confirmation by William de Warenne, third earl of Surrey (1138-48), notes that he gave seizin of a tenth penny of his rents 'by hair of his own head and that of Ralph de Warenne his brother, cut with a knife by Henry, bishop of Winchester, before the altar'; printed ibid., vi, 84-5, no. 32. Cf. V.H. Galbraith, 'Monastic foundation charters of the eleventh and twelfth centuries', Cambridge Historical Journal, iv (1934), 211. The entry concerning the seal at Castle Acre is even more problematical. Elmham states that the earl of Lincoln in question 'pluribus possessionibus eandem ecclesiam dotavit'. But the founder of Castle Acre priory was William de Warenne, first earl of Surrey (1088) and its benefactors were his successors as earls of Surrey; see V.C.H., Norfolk, ii, p. 356, and G.E. C[ockayne], The Complete Peerage, ed. Vicary Gibbs et al. (London 1910-59, 13 vols.), xii, pt. i, 494-7 passim. None of the known names of their wives is Muriel; however, the name of the first earl's second wife is unknown (G.E. C[ockayne], op. cit., xii, pt. i, 494); possibly she was called Muriel. None of the countesses of Lincoln was called Muriel, as far as is known. 65
310
Legends, Traditions and History in Medieval England
Elmham was interested in other remains of the abbey's past besides its early documents. Again his attention was first attracted to antiquities associated with St Augustine himself. He intended to record the translation of his relics in 1091 by Abbot Scotland, but never reached that date. However, he included a note on St Augustine's previous burial places, and a pictorial plan of the high altar and sanctuary with their shrines (pi. 32). 70 He also gives an account of the eight books which, according to tradition, St Augustine gave to the monks. He lists their contents and describes their appearance,71 noting, for example, that the Bible in two volumes had pages tinted purple or pink, and that the psalter on the high altar bore on its bindings silver images of Christ and the four evangelists. The accuracy of his descriptions is testified by the fact that Humphrey Wanley was able to suggest an identification for one of the books, an identification accepted by later scholars.72 Elmham used Bede to discover the plan of the earliest church. He concluded that it was narrower than the church of his own day, that the altar of St Gregory was more or less in the middle, and that the porticus of St Martin was on the south side.73 On the rare occasion when Elmham cites a legend, he adds sound antiquarian observations. He tells the story of Domne Eafe's hind. (He was interested in the minster on the Isle of Thanet, of which Domne Eafe's was the founder and first abbess, because after its destruction by the Danes King Canute gave its lands to St Augustine's.) According to legend King Edgar promised to give the minster the area delineated by the course of the hind running across the island. But a wicked thegn, Thunor, tried to frustrate his generous intention by stopping the hind. Luckily the earth opened and swallowed Thunor up. 74 To illustrate this tale Elmham supplied a map of Thanet, marking the hind's course and the spot where Thunor disappeared. But he also showed, with pictures, the churches assigned to the sacristy of St Augustine's. Although the map itself is inaccurate, the pictures and relative positions of the churches seem to be correct. There was apparently a linch along the boundary of St Augustine's lands which roughly followed the hind's course as plotted on the map. There was also probably a pit on the spot marked on the map
70
Trinity Hall MS, I, fo. 63; cf. Elmham, op. cit., pp. 286, n. 1, 346, n. 1. The plan is discussed by W. Urry and reproduced in Local Maps and Plans from Medieval England, ed. R.A. Skelton and P.D.A. Harvey (Oxford, 1986), pp. 107-17. 71 Elmham, op. cit., pp. 96-9. 72 See The Vespasian Psalter ed. D.M. Wright (Copenhagen, 1967), p. 37-43. 73 Elmham, op. cit., pp. 132-3 (cf. Bede, H.E., II, 3). This passage is not cited as evidence for the church built by King Ethelbert in H.M. Taylor and Joan Taylor, AngloSaxon Architecture (Cambridge, 1965, 2 vols.), i, pp. 135-7. 74 Elmham, op. cit., pp. 207-9passim. See D.W. Rollason, The Mildrith Legend: A Study of Early Medieval Hagiography in England (Leicester, 1982), chs 3-4.
Antiquarian Studies in Fifteenth-Century England
311
where Thunor was engulfed (pi. 33).75 We turn now to the other outstanding monastic antiquary, Thomas Burton, whose history of Meaux is the most comprehensive and scholarly of the local histories written in medieval England. Burton's purpose was partly commemorative. He wanted to commemorate the past abbots. (Perhaps he hoped that he himself would be similarly commemorated and his troubled rule vindicated: in the event, the monk who continued his chronicle did render him this service.76) He was grieved, he states, because he saw 'the memory of those illustrious men, the abbots of Meaux, almost lost and their light obscured owing to the sloth of the negligent'.77 He also wanted to commemorate the achievements of outstanding monks, and the generosity of benefactors. His desire to commemorate the abbey's benefactors helps explain the presence in his history of numerous pedigrees of local families. But there is also another explanation. Burton wanted to record the abbey's acquisition of property. The benefactors' pedigrees were inextricably connected with the descent of the properties which they had given. Burton's meticulous record of the descent of such properties, if possible from the date when the abbey acquired them until his own time, is as remarkable as that of the pedigrees themselves.78 Two of the pedigrees were included not from the commemorative motive, but as part of Burton's record of a legal dispute. They are the two versions of the pedigree of a bondman, Richard de Aldwyne, who rebelled against Abbot Robert de Beverley (1356-67).79 Aldwyne argued that he belonged to the king, not the abbot, and produced a pedigree to prove it. But the abbot produced a pedigree showing that Richard descended from one of his bondmen. The desire to substantiate the abbey's case in various disputes was one reason why Burton wrote. Before becoming abbot Burton had been the bursar. This no doubt accounts for his particular interest in any dispute 75
Trinity Hall MS, I, fo. xxviiiv. Discussed by F. Hull and reproduced in Skelton and Harvey, op. tit., pp. 119-26. 76 Chron. de Melsa, iii, 237-76. Cf. ibid., i, xlviii-xlix. 77 Ibid., i, 71. 78 e.g. the pedigrees of the families of Sayers (ibid., i, 96), Scures (ibid., i, 97-8), Fossard (ibid, i, 104), and Sculcottes (ibid., i, 169-70). The pedigree of the Etton family of Gilling, which is combined with the history of its estates (ibid., i, 316-18, and other references, for which see the index in ibid, iii), provided much of the information used in John Bilson, 'Gilling Castle', Yorks. Arch.J. xix (1907), 105-22 passim. Burton also gives the pedigree of the Forz family (the counts of Aumale), the founders of the abbey, and the descent of the honour and of the lordship of Holderness; Chron. de Melsa, i, 89-93. 79 Chron. de Melsa, iii, 130, 134. For this case see R.H. Hilton, 'Peasant movements in England before 1381', Essays in Economic History, ed. E.M. Carus-Wilson (London, 195462, 2 vols.), ii, 89 (the article is reprinted in the same author's The Decline of Serfdom in Medieval England (London-New York, 1969)).
312
Legends, Traditions and History in Medieval England
involving money. The abbey was in financial straits — Burton gives many details about its debts. The burden of taxation must have fostered interest in the abbey's estates. This conclusion is suggested by the fact that it was usual for most of the clergy to be taxed with the laity on property acquired under licence of mortmain after 1291, that is for spiritualities and temporalities acquired after the valuation in that year of Pope Nicholas IV.80 Burton explains clearly about the impact on the monks of Meaux of the ninth granted by parliament in 1339. Of the clergy only the bishops and 'parliamentary' abbots (who did not, of course, include the abbot of Meaux) were liable for the tax, since they had agreed to the grant in parliament. Nevertheless, in 1341 it was specially ordered in parliament that lands acquired by the clergy after 1291 should be taxed for the ninth. Since the assessment was severe, the clergy raised objections. Burton gives details, apparently copied from the public records, of the negotiations of the abbot of Meaux with the royal officials, to obtain exemption from the ninth for certain estates, which had been assessed although the abbey had acquired them before 1291.81 To prove the pre-1291 date of an estate involved, of course, research into its origins. Taxation may also have encouraged interest in the topography of some of the abbey's estates. Its properties in the levels of Holderness had been diminished by erosion by the sea and by the river Hull.82 In Burton's day the monks were trying to obtain a reduction of the tax assessment on these properties.83 Burton himself wrote a tract putting the case to the royal officials. Research on the extent of the erosion must have involved examining it on the spot. This must help explain why Burton gave such detailed and vivid descriptions of the floods and of the damage they did, a topic which will be discussed below. Finally, the abbey's financial 80 For the practice with regard to the taxation of clerical properties see W. Stubbs, Constitutional History of England, ii (3rd edn., Oxford, 1887), pp. 416 n. 1, 443, n. 4, 580; J.F. Willard, 'The taxes upon movables of the reign of Edward II s , E.H.R. xxix (1914), 318; W.E. Lunt, The Valuation of Norwich (Oxford, 1926), p. 72, n. 6; J.F. Willard, Parliamentary Taxes on Personal Property 1290 to 1334 (Cambridge, Mass., 1934), pp. Wetseqq. For the ninth granted in 1339 in particular see Natalie Fryde, 'Edward Ill's removal of his ministers and judges, 1340-1', Bulletin of the Institute of Historical Research, xlviii (1975) ,152. For help on the subject of clerical taxation and for the above references I am indebted to Dr Gerald Harriss and Professor E.B. Fryde. 81 Chron. de Melsa, iii, 24-9. Cf. Rot. Parl ii, 129, 130. Collation of the relevant entry on the Memoranda Rolls, in the Recorda for Trinity term, 18 Edward III (P.R.O., E 159/120 membrane 245) with the text in the chronicle shows that the latter is slightly shorter, has the king in the third person instead of the first, and uses the past instead of the present tense. 82 Ibid., in, 123,247etseqq. 83 The tract is among the material appended to one of the two manuscripts of the chronicle, BL MS. Egerton 1141, the revised version; see Chron. de Melsa, i, liii-liv. Cf. the account of the struggle between the abbey and the royal assessors in the continuation to Burton's chronicle; ibid., iii, 279 et seqq.
Antiquarian Studies in Fifteenth-Century England
313
difficulties drew Burton's attention to other aspects of local history. There was apparently resentment at a royal imposition on the East Riding, for which Burton is the only authority. It was a due of four measures (thraves) of corn from every carucate, dating back to the reign of Athelstan. It was called 'hestcorn' — but, Burton relates, now certain wicked people call it 'bestcorn'.84 The wish to commemorate the abbey's inmates and benefactors and to provide information about its disputes were certainly among Burton's motives for writing the history. But objective curiosity was a powerful incentive. The work was the occupation of his retirement, and his researches went far beyond the needs of his house. As Meaux was of comparatively recent foundation, Burton had no need to resort to forgery or legend for its early history. He himself mentions his sources in the preface: 'No one who finds anything in what follows, which he did not know before, should think I invented it: let him rest assured that I have only included what I have found written in other works, or in a variety of documents, or have heard from reliable witnesses, or have myself seen.'85 As bursar Burton must have been well trained in the use of documents. Besides the chronicle he has left collectanea — a register and the documentary material which he appended to his chronicle. These comprise series of documents, digests of charters, and tracts; all relate to various aspects of the abbey's history.86 For the chronicle, Burton writes, T collected together many ancient documents and long forgotten parchments: I found some which had been exposed to the rain, and others put aside for the fire.'87 Unlike most chroniclers Burton did not copy documents in full. As he explains C I have abridged their great length and illuminated their obscurities . . . and finally I have combined the total results into this one volume with the greatest care.'88 Like Elmham he understood that an original document has more authority than a copy. He claims to have 'read through the registers and added from the original documents whatever they omit'.89 Burton's history, therefore, rests on a substratum of documentary evidence, and where comparison can be made between the text of the chronicle and the underlying document, Burton's accuracy is apparent. For example, his account of the abbey's foundation is corroborated by the 84
Ibid., ii, 236. See Early Yorkshire Charters, ed. William Farrer (Edinburgh, 1914-16, 3 vols.),i, 95-6. 85 Ibid.,i,7-Z 86 The register is now BL MS. Cotton Vitellius C VI; see Chron. de Melsa, i, Ivi-lviii. For the collectanea appended to one of the two manuscripts of the chronicle see p. 312 n. 83 above. Q1 Chron. deMelsa, i, 71. 88 Ibid.,i,7. 89 Ibid., i, 71.
314
Legends, Traditions and History in Medieval England
count of Aumale's foundation charter.90 Most of Burton's information on the antiquities of the abbey buildings he derived from written sources, but he also used visual evidence. He has practically nothing on the abbey church itself except the usual record of burial places.91 However, his notice of one of the epitaphs, that of Abbot Robert de Beverley, has a critical comment: the dates were wrong; the sculptor invented them because he lost the sheet given him to copy.92 Two of Burton's observations concern other buildings in the precincts. He notes the beauty of the crucifixion in the lay brothers' church, which he attributed to the sculptor's piety (he would only carve on Fridays when fasting on bread and water), and to his use of a nude model.93 And he records that the foundations of the chantry chapel which Abbot Adam de Skyrne (131039) had begun to build above the great gate, but which Abbot Hugh de Leven (1339-49) had pulled down, were still visible in his day: Abbot Hugh used the stone to improve the brewery; he made a handsome tank next to the malt kiln for the fermentation of the malt and barley.94 More remarkable are Burton's observations on the topography of the abbey's estates. He had obviously tried to trace the original boundary of the manor of Meaux. Using charter evidence he records that it was marked by a mound in the form of a kiln, a large stone under a bridge, a buried cow, and holes in the ground three feet wide and a stone's throw apart. 'But,' he comments, 'as the marks are now wholly unknown, we must examine the boundary as it exists today.'95 He also has an excellent account of the topographical changes resulting from silting and erosion. He explains how the vill of Wick, which was washed on two sides by the Humber and the Old Hull, had once been in Holderness but was now part of Harthill. The New Hull to the east of Wick had grown, while the Old Hull had become a mere trickle ('hardly deserving to be called a drain'). And so the New Hull 'now divides Holderness and Harthill', leaving Wick on the Harthill side. In Burton's time the grange there, 'now called Grangewick', had been abandoned, though 'its ruins can easily be seen'.96 90 Ibid., i, 81. For the charter see Early Yorkshire Charters, ed. Farrer, iii, 89-91, 93. Compare also Burton's detailed account of the foundation of a chantry in the abbey with the inspeximus of 1238 on the Charter Rolls; Chron. de Melsa, ii, 59-62; Cal. Charter Rolls, i, 233-4. 91 For the burial places of abbots see e.g. Chron. de Melsa, i, 107, 234, 380; ii, 119; iii, 167, 234. Burton states that the burial place of Abbot Robert de Skyrne (1270-80) is unknown; ibid., ii, 157. For the burial places of laymen see ibid., i, 260; ii, 106; cf. i, 212.
92
93
Ibid., iii, 152.
Ibid., iii, 35-6. Ibid., iii, 36. 95 Ibid., i, 78-81. Cf. ibid., iii, 1 and n. 2. 96 Ibid., i, 168-9. For the vill of Wick, later the site of Kingston-upon-Hull, see ibid., ii, 186, 192, and V.C.H., Yorks., East Riding, i, p. 16. For Grangewick see also Chron. de Melsa, ii, 192. 94
Antiquarian Studies in Fifteenth-Century England
315
Burton gives vivid descriptions of the flooding, twice in the midfourteenth century, of Ravenser Odd, a small port at the extreme end of Holderness.97 The first flood devastated the abbey's church there, washing the bodies and bones from their graves in the cemetery — 'a horrible sight'. The second time the village was surrounded by ca towering wall of water'. The inhabitants fled, never to return. And so Ravenser Odd disappeared into the Humber, although its ruins remained, a danger to sailors. Burton explains that it must not be confused with Old Ravenser, a manor set back from the sea and the Humber, which still existed in his day. He describes the remains of the road which once led from Old Ravenser to Ravenser Odd.98 It is sandy and scattered with round, yellow pebbles, and is raised very little above water-level, and is hardly a bowshot in width, but it has resisted the flooding of the sea on the east and the constant battering of the Humber on the west in a truly wonderful fashion. Before leaving the subject of the monks as antiquaries, something must be said about their historical tracts. Each tract tends to highlight one of the principal motives which led monks to antiquarian studies - the commemorative, the instructive and the legalistic. And many betray antiquarian curiosity. Two examples of non-legalistic tracts from St Albans, written early in the century, may be considered first.99 One, a commemorative tract, 97 Ibid., iii, 79, 120-1. The latter passage is noticed by N. Denholm-Young in his account of Ravenser Odd in 'The Yorkshire estates of Isabella de Fortibus', Yorks. Arch. J. xxxi (1934), 404, n. 2. For Ravenser Odd and Old Ravenser, with references to Burton's chronicle, see George Poulson, The History and Antiquities of the Seigniory of Holderness (Hull, 1840-1, 2 vols.), ii, pp. 529-40, Barbara English, The Lords of Holderness 1086-1260 (Oxford, 1979), pp. 211-13. 98 The passage occurs twice in the chronicle; ibid., ii, 30; iii, 121-2. 99 A number of other examples can be cited of tracts written with a commemorative intention but which show antiquarian interests. Some of John Wessington's tracts (see p. 316 and n. 108 below) are primarily commemorative. One, for example, gives the scriptural references to the inscriptions beneath the images (or pictures) of the hundred and forty-eight monks at the altar of SS. Jerome and Benedict in Durham Cathedral, and names some of the hitherto unidentified figures: Durham Cathedral Library MS. B III 30, folios 6-25v; partly printed in Rites of Durham, ed. J. T. Fowler (Surtees Soc., cvii, 1903), 124-36; cf. Historiae Dunelmensis Scriptores Tres, Gaufridus de Coldingham, Robertus de Graystanes, et Willielmus de Chambre, ed. James Raine (Surtees Soc., ix, 1839), cclxix, Pantin, 'Some medieval treatises on the origins of monasticism', 200, Dobson, Durham Priory 1400-1540, p. 382. Wessington worked on the books in the cloister and cathedral library; Raine, op. cit., cclxx. The desire to instruct and antiquarian interest were probably the main reasons why at Christ Church, Canterbury, William Glastonbury described the scenes in the twelve windows of the choir, partly from his own observation. His description is noticed and printed The Chronicle of William Glastynbury, Monk of the Priory of Christ Church, Canterbury, 1418-48, ed. C. E. Woodruff in Arch. Cant, xxxvii (1925), 123-5, 138, 139-51.
316
Legends, Traditions and History in Medieval England
describes the altars, monuments and graves in the abbey church.100 The author records the names of those buried, whom he identified by reference to 'the Book of Benefactors on the high altar',101 documents in the keeping of the almoner, William Wintershull,102 and Thomas Walsingham's Gesta Abbatum.103 He was interested in past as well as in present arrangements. He points out, relying on written evidence and pictures, that the cult of the Virgin Mary was observed in the abbey before the chapel where her altar now stood had been built. Masses for her were celebrated in the chapel of St Blase which used to be by the abbey gate; no trace of it remained in the author's day.104 The other tract describes the statues and pictures in the abbey church and explains their iconography.105 It was probably written partly for the benefit of visitors,106 but again it shows antiquarian interest. For example, it relates that formerly the statues of St Laurence and St Grimbald had been in the almonry chapel. But when the latter was pulled down, to make room for the new gatehouse and infirmary, they had been moved, 'in accordance with the tenor of the constitution then drawn up', to their present place, so that they could be venerated as before.107 As examples of legalistic or quasi-legalistic tracts, those by John Wessington may be cited. He produced at least forty.108 A contemporary wrote of him:109 He compiled [the tracts] not without labour and study, for the perpetual preservation and defence of the rights, liberties and possessions of the church . . . against the malice and machinations of would-be molesters. The scribe of the tract defending the prior's archidiaconal jurisdiction over his churches, remarked that it 'would be of value against the 100 Printed in Annales Monasterii S. Albani ajohanne Amundesham, ed. H. T. Riley (Rolls Series, 1870-1, 2 vols.), i, 431-49. Described ibid., ii, lix-lxii. 101 Ibid., i, 431-2. Cf. ibid., i, 434-41 passim. Perhaps this Book of Benefactors is to be identified with that by Thomas Walsingham, now BL MS. Cotton Nero D VII, for which see The St. Albans Chronicle 1406-1420, ed. V. H. Galbraith (Oxford, 1937), pp. xxxvi-xxxvii. 102 Amundesham, ed. Riley, i. 448. 103 Ibid., i, 433-4. 104 Ibid., i, 445-6. 105 Printed ibid., i, 418-30. Described ibid., ii, Iviii-lix. 106 SeeifoV.,i,418. 107 Ibid., i, 421. 108 VVessington's tracts, not all of which survive, were listed by a contemporary on a roll, three copies of which are preserved in Durham Cathedral Library; Dobson, op. cit., p. 379, and n. 2. One copy is printed in Raine, op. cit., cclxviii-cclxxi. See also Craster, op. cit., 515 and n. 2. Rather similar tracts were written at Bury St. Edmunds early in the fifteenth century, e.g. the Visitatio Thome deArundel, the Contentio cum Episcopo Eliensi, and the Pensio de Woolpet', printed Memorials of St. Edmund's Abbey, ed. Thomas Arnold (Rolls Series, 1890-6, 3 vols.), iii, 183-8, 188-211, 78-112, respectively. 109 Raine, op. cit., cclxviii-cclxix.
Antiquarian Studies in Fifteenth-Century England
317
archdeacon of Durham'.110 Not all of Wessington's tracts, however, have a litigious or even a tendentious purpose. For example, he wrote the Jurisdictio Spiritualis simply to try to determine whether during a vacancy the see of Durham's spiritual jurisdiction should be administered by the cathedral chapter or by the archbishop.111 For evidence of past vacancies he examined not only Robert Graystanes's history of Durham, but also the archives of the prior and chapter. He cites a certificate of the archdeacon of Durham, which had 'many seals5,112 and an indenture 'with the seals of Archbishop John, Antony bishop of Durham, Ralph bishop of Carlisle, and the chapter of York'.113 I have tried to show that some generalizations about the motives and methods of the monastic antiquaries are possible. The case with the seculars, John Hardyng, William Worcester and John Rous, is different. It is hard to decide in what ways they were unlike the monks but like each other: however, their interests were not so centred on one locality; loyalty was primarily to a patron, not a place (Rous is a partial exception). And they travelled around much more. But, like the monks, they accepted legends equally with verifiable fact. Of the three, Rous most closely resembled a monastic antiquary. Having studied at Oxford, he became a chaplain of the earl of Warwick in the chantry chapel at Guy's Cliff, near Warwick.114 He had a strong local attachment to both his homes, Oxford and Warwick, and as a chantry priest had a powerful incentive to commemorate his patrons in the first instance, but also others. His Historia Regum Angliae shows a close interest in Oxford and Warwick. He has a number of passages on the antiquities of Oxford. Although his account of the foundation of the university is legend, he follows it by accurately locating some of the halls in his day.115 (Elsewhere he has left a topographical list of all the halls, including six 110
Ibid., cclxix. Durham Dean and Chapter Archives, Register III, fo. 211-21 l v . Discussed and printed by Robert Brentano, 'The Jurisdictio Spiritualis: an example of fifteenth-century English historiography', Speculum, xxxii (1957), 326-32. Professor Brentano (op. cit., 327) only tentatively ascribes the work to Wessington (it is not in the contemporary list of his works), but his authorship is accepted by Professor Dobson (op. cit., p. 386). Another tract written to establish the truth about the past, without litigious or even tendentious purpose, was written at St Albans: it explains exactly how the abbot of St Albans had lost his right precedence in parliament in the reign of Richard II; printed Amundesham, i, 414-17. 112 Brentano, op. cit., 331. 113 Ibid., 331 and n.g. Gf. ibid., 327-8. 114 For Rous's career see William Courthope's introduction to This rol was laburd and finished by Master John Rows of Warrewyk (London 1845-59, repr. with an historical introduction by Charles Ross, Alan Sutton, 1980), and Kendrick, op. cit., pp. 19-20. 115 Ibid., pp. 77-8. For another example in the Historia showing Rous's antiquarian interest in Oxford, see his account of the processional cross and stone cross owned by the university; Historia, pp. 201-2. 111
318
Legends, Traditions and History in Medieval England
pulled down 'before my time', and six more pulled down c in my time in Cat Street for [All] Souls College'.)! 16 Similarly, although he recounts the legendary origins of Warwick, he carefully describes where a monastery and a nunnery, destroyed by the Danes, had been situated.117 The history of Warwick is one theme of Rous's two Warwick rolls, but their principal subject was the history of the earls of Warwick. In their production Rous was strongly influenced by the heraldic tradition. Heraldry, with its dependence on genealogy, made an important contribution to antiquarian studies. Rous's Warwick rolls are not the only example of armorial genealogies,118 but they are the best. One of them was executed in about 1477 but revised in Richard Ill's reign, when the other was executed.119 The purpose of both was to commemorate the earls and countesses, and others closely connected with the family and town, who are depicted with biographical captions and their coats of arms. Rous begins with the legendary British earls, but as soon as he reaches historic times his account is sound. The antiquarian interest of the rolls has long been recognized. Rous appreciated the value of pictures to the historian. Indeed he claims to have anticipated Bernard of Breydenbach who took an artist with him to the Holy Land in 1483, to provide him with a pictorial record: apparently Rous had earlier advised John Tiptoft, earl of Worcester, to do just that when he visited the Holy Land.120 The pictures on the Warwick rolls not holograph of Rous's list has not survived, but there are a number of copies, the best of which is printed in 'Survey of the Antiquities of the City of Oxford' composed in 1661-6 by Anthony Wood', i. The City and Suburbs, ed. Andrew Clark (Oxford Historical Soc. xv, 1889), pp. 638-41. See T. H. Aston, 'Oxford's medieval alumni', Past and Present, no. 74 (1977), 36-8 and nn. It should be noted that the remarkably accurate bird's eye view (executed c. 1463) of New College by the warden Thomas Chandler, belongs to this period; see A. H. Smith, New College, Oxford, and its Buildings (Oxford, 1952), p. 179 (see also pp. 43, 49, 109) and frontispiece. 117 Historia, ed. Hearne (see p. 303 n. 21 above), pp. 45-6, 58, 60, 104. i is por other examples of genealogies, some with armorial and/or historical details, see the note at the end of this article (pp. 326-7 below). 119 The latter, the so-called Yorkist roll, the text of which is in English, now BL MS. Additional 48976, is printed with line reproductions of the pictures, by Courthope, op. cit. See also C. E. Wright, 'The Rous Roll: the English version', British Museum Quarterly, xxx (1955-6), 79. The revised roll, the so-called Lancastrian roll, preserved in the College of Arms, has not been published but its text, which is in Latin, is printed in footnotes to the descriptions of the plates in Courthope, op. cit., and there is a good account of it in A. R. Wagner, A Catalogue of English Mediaeval Rolls of Arms (Harleian Soc., c (1948) and Oxford, 1950), 116-18. Pictures from it are reproduced in A. G. B. Russell, 'The Rous Roll', Burlington Magazine, xxx (1917), 31 and pi. I; J. G. Mann, 'Instances of antiquarian feeling in medieval and renaissance art', Arch.J. Ixxxix (1932), pis. II (2), III, IV, opposite pp. 259, 260, 261, respectively, and in Kendrick, op. cit., pis. II (b)-IV. 120 Historia, p. 5. For the life of Bernard de Breydenbach, dean of Mainz, and the artist, Erhard Reuwich of Utrecht, who accompanied him to the Holy Land, see H. W. Davies, Bernhard von Breydenbach and his Journey to the Holy Land 1483-4 (London, 1911, reprinted
Antiquarian Studies in Fifteenth-Century England
319
only record the coats of arms of the individuals depicted, but also show each earl in the armour which Rous considered appropriate to his time. Thus Rous tried to show the evolution of medieval armour. In a general way he got it right: he traces the development from chain mail, to mail and plate, to full plate; he was also correct on some of the details.121 He obviously had no idea what the (legendary) British earls of Warwick might have worn, and depicts them in long chain mail with coifs in one piece, a style characteristic of the late twelfth century (pi. 34). He shows the early Anglo-Saxon earls in short mail hauberks, and hemispherical ribbed helms, styles in use from the seventh to the eleventh centuries. He represents the helms, some of which the earls carry on staves or lances, as chained to the pommel of the sword (pi. 35). Here he made a mistake, no doubt led astray by the practice, which was fairly widespread in the fourteenth century, of chaining loose pieces of equipment to the hauberk. Rous shows the late eleventh-century and twelfth-century earls in mail with knee and elbow caps, a type of armour introduced in the midthirteenth century (pi. 37). The thirteenth- and early fourteenth-century earls are in the mixed mail and plate worn in the first half of the fourteenth century. Thereafter the chronology of the progression from mixed mail and plate to full plate in the last half of the fourteenth century and in the fifteenth century is correct (pis 39-41). Rous dates correctly the introduction of two pieces of outfit: the war hat, a protective head-piece introduced in the late twelfth century, and the armorial surcoat, introduced in the early thirteenth century; William de Newburgh (d. 1184) is the first earl depicted in a war hat, and Waleran de Newburgh (d. 1203 or 1204) the first wearing a surcoat (pis 38, 39). It may well be that Rous obtained information about armour from suits preserved in Warwick castle. He may also have examined the effigies in St Mary's church at Warwick. The suit shown in his portrait of Thomas Beauchamp, earl of Warwick 1329-69, is identical with that represented on the earl's effigy. (Rous depicts Thomas's son and heir, Thomas Beauchamp, earl 1370-1401, in the same suit; pi. 40.) Similarly, the suit Utrecht, 1968), pp. i-iii, xxi. See alsoj. R. Mitchell,>/m Tiptoft (1427-70) (London, 1938), p. 28, 121 Rous's pictorial history of medieval armour is discussed by Kendrick, op. cit., pp. 289, and by Mann, art. cit., p. 262. For full mail with coifs in one piece, resembling that worn by the British earls of Warwick in Rous's pictures, see G. F. Laking, A Record of European Armour and Arms through Seven Centuries (London 1920-2, 5 vols.), i, 66-70. For a ribbed helm like those worn by the early Anglo-Saxon earls in Rous's pictures, see ibid., i, 8, fig. 11. For a reconstruction of an Anglo-Saxon warrior not unlike those depicted by Rous, see ibid., i, 31, fig. 39. For the introduction of mixed mail and plate in the mid-thirteenth century see ibid., 121 et seqq., and for its development in the fourteenth century see ibid., i, 145.Forthe introduction of war-hats and surcoat see respectively ibid, ii, 57, 66; i, 124. For the suit on Richard Beauchamp's effigy, itself an important landmark in the history of armoury, see Laking, op. cit., i, 163-70. See also the next note.
320
Legends, Traditions and History in Medieval England
in which Rous depicts Richard Beauchamp, earl 1403-39, is the same as that represented on his effigy (pi. 41). In addition, Rous may have used the evidence of seals. Possibly his mistake over the chaining of the early Saxon earls' helms was the result of misinterpreting the picture on one of the Beauchamp seals. A surviving example shows the sword chained to the shoulder of the hauberk in such a way that it could give the impression of being chained to the helm (pi. 36). 122 Rous's interest in seals appears in the Historia Regum Angliae. He notes more or less correctly that Henry I started the practice of sealing documents with wax seals, instead of authenticating them with a cross and sign manual, as had been the previous practice.123 And he remarks, rather less correctly, that after the mid-fourteenth century ('the capture of King John of France' in 1356), the English nobility replaced equestrian figures on their seals with coats of arms.124 Rous was also interested in the history of costume; although he did not try to establish the sequence of fashions, he did depict or describe individual styles in particular periods. He shows the countesses and other ladies wearing old-fashioned clothes, but of the fourteenth century, not those appropriate to the era in which they lived. Thus King Alfred's daughter ^Ethelflaed wears clothes similar to those of the effigy of Catherine, wife of Thomas Beauchamp, earl of Warwick 1329-69.125 The Historia has a number of (brief) descriptions of dress: in Anglo-Saxon times, under William II, Edward III and Richard II, and in his own day, besides that of the Jews and that of the canons of the Holy Sepulchre.126 However, not all these passages are antiquarian in tone. Those
122 For an English example of the helm chained to the hauberk see the brass (c. 1300) of Sir Roger de Trumpington; reproduced Daniel Lysons and Samuel Lysons, Magna Britannia (London, 1806-22, 6 vols.), ii, opposite p. 65. For a German example see the effigy of Heinrich von Seinsheim (d. 1360) in Wurtzburg Cathedral; reproduced Paul Martin, Armour and Weapons, trans. Rene North (London, 1968), p. 70 and pi. 63. For an example of the sword and dagger chained to the hauberk see the brass (1370) of Ralph de Knevyngton in Aveley church, Essex (reproduced Laking, op. cit., iii, 5); for similar fourteenth-century examples from Germany see Martin, op. cit., pp. 52, 57, 70, and pis. 51, 52, 55, 58, 62, 65.1 am indebted to Mr A. R. Dufty for supplying some of the references for this note. 123 Historia, p. 138. In fact Edward the Confessor was the first king to use a great seal, which is of wax; Facsimiles of English Royal Writs to A.D. 1100 presented to Vivian Hunter Galbraith, ed. T. A. M. Bishop and Pierre Chaplais (Oxford, 1957), pp. xix, xxii, and Pierre Chaplais, English Royal Documents, King John to Henry VI, 1199-1461 (Oxford, 1971), p. 2. However, surviving examples of Henry I's seals are more numerous than of those of previous kings; see W. de G. Birch, Catalogue of Seals in the Department of Manuscripts in the British Museum (London, 1887-1900, 6 vols.), i, pp. 2-8. 124 Historia, p. 204. In fact this change took place early in the thirteenth century; C. H. Hunter Blair, 'Armorials upon English seals from the twelfth to the sixteenth centuries', Archaeologia, Ixxxix (1943), -26passim. 125 Kendrick, op. cit., p. 28. 126 Historia, pp. 106, 110, 204, 205, 131, 202, 139-40, respectively.
Antiquarian Studies in Fifteenth-Century England
321
concerning the fashionable clothes of William IPs time and of Rous's own day have a homiletic ring, and were no doubt partly inspired by the preaching tradition. (It should be noted that a number of fourteenthcentury chroniclers described the extravagent fashions worn by their contemporaries, while inveighing against the worldliness they revealed.) 127 The Historia proves that Rous's horizon was not limited by his own loyalty to Oxford, or to the town and earls of Warwick. Indeed, he travelled quite widely. He visited London (where he worked in the libraries of St Paul's and the Guildhall) and Windsor, and the abbeys of St Albans, Osney, and of St Swithun's and Hyde, Winchester, besides going further afield, to North Wales and Anglesey.128 He cared for England as a whole, and wrote the Historia partly to commemorate the great people in English history. He started the Historia after a visit to 'the newly built college at Windsor'. There he saw niches left for statues of the famous. John Seymour, whom Rous calls master of the works, asked him 'to write a little book on the kings, the princes of the church and the founders of cities, so that statues of them could be honourably placed in the niches for the perpetual remembrances of their names'.129 The founders of towns and universities included by Rous are nearly all legendary and belong to the remote past,130 but his founders and benefactors of churches, colleges and the like in the post-Conquest period are historical.131 His interest in founders was combined with an interest in origins in general, whether of the shire system, tournaments or the side-saddle.132 However, some of Rous's best research was elicited neither by his commemorative intention nor by objective curiosity, but by the desire to persuade. The monks had used historical evidence to defend their houses; Rous used it to defend the common weal.133 One reason why he wrote the
127 Chronicajohannis deReading . . ., ed. James Tait (Manchester, 1914), p. 167; Chronicon Henrici Knighton, ed. J. R. Lumby (Rolls Series, 1889-95, 2 vols.), ii, 229; Eulogium Historiarum, ed. F. S. Haydon (Rolls Series, 1858-63, 3 vols.), iii, 230-1; Historia Vitae et Regni Ricardi Secundi, ed. G. B. Stow (University of Pennsylvania, 1977), p. 168. 128 Historia, pp. 69 and 200 (London), 120 (Windsor), 60 (St Albans), 203 (Osney), 73 (St Swithun's), 96 (Hyde), 54 (North Wales and Anglesey). 129 Ibid., p. 120. There is no evidence confirming the statement that John Seymour, a canon of St George's chapel, Windsor, was master of works at the chapel at the end of the reign of Edward IV. However, he was in charge of repairs to the college early in Henry VIFs reign. See The History of the King's Works, ed. H. M. Colvin (London, 1963-76, vols. iiii, v, vi, and a volume of plans published to date), iii, pt. i, 305-6. 130 e.g. Historia, pp. 22-7 passim, 96, 119. 131 e.g. ibid., pp. 140-1, 203, 204, 210-11, 215-16. 132 Ibid., pp. 66, 194-5, 205. 133 Rous denies that he undertook his research on enclosures for its own sake; he states that 'pro certo nunquam aliud intendebam quam honorem dei, et regis ac totius rei publicae proficuum, ut deus novit'; ibid., p. 86.
322
Legends, Traditions and History in Medieval England
Historia was because he and others objected to the enclosure movement. He had, he tells us, petitioned the parliament held at Covenry in 1459 to remedy the evil by legislation.134 The petition itself has not survived but it is tempting to think that the long tirades against enclosures in the Historia are extracts from it.135 His arguments are various - philosopical, theological, legal, humanitarian and expediential. To prove that enclosures were not in the country's best interests, Rous appealed to history. He listed fifty-eight of the villages near Warwick, and, to show the extent of depopulation, studied twelve in more detail. He compared their present condition with that in Edward I's reign as recorded in the Hundred Rolls.136 Using his own observation, he expatiates in general on the ruin of villages and the deterioration of roads,137 and sometimes gives specific details. For example,138 At Fulbrook, where there was formerly a rectory, the church is destroyed, the villeins fled, and only the manor remains. The rest was enclosed for a park by John duke of Bedford, brother of King Henry V, who built a noble square tower for the castle, but now almost nothing is there. Also Joan, Lady Bergavenny, built a splendid gate-house inside the park fence, suitable to welcome her noble lord — to please him on arrival; now this too is destroyed. In his research on enclosures Rous was a pioneer; no one had previously tried to assess their extent and results, and the government undertook no comparable inquiry until the early sixteenth century.139 The desire to persuade was one reason why John Hardyng turned to antiquarian research. He had started life as an esquire in the household of
134 Ibid., pp. 120-1. For his petition to the Coventry parliament of 1459 see M. W. Beresford, The Lost Villages of England (London, 1954), pp. 102, 148, idem, 'The deserted villages of Warwickshire', Transactions . . . of the Birmingham Archaeological Soc. for . . . 1945 and 1946, Ixvi (1950), 53-4, C. C. Dyer, 'Deserted Medieval Villages in the West Midlands', Economic History Review, xxxv (1982), 25.
135 136
Historia, pp. 39-43, 87-96, 112-37.
Ibid., pp. 122-4. Rous was a pioneer in his use of the Hundred Rolls; Beresford, op. cit., p. 282. Rous also cites Domesday Book; Historia, p. 107. However, recent research shows that, although Rous seems to be fairly reliable on the desertion of villages in his own locality in his own day, his information about the movement in the past should be treated with caution. See Dyer, op. cit., 25-6, C. J. Bond, 'Deserted villages in Warwickshire and Worcestershire', Field and Forest, An Historical Geography of Warwickshire and Worcestershire, ed. T. R. Slater and J. P. Jarvis (Norwich, 1982), pp. 150-2.1 am indebted to Dr Christine Carpenter for the references to recent work on Warwickshire enclosures. 137 Ibid., pp. 125-6. 138 Ibid., pp. 123-4; cf. V.C.H., Warwick., iii, 91-2. For Joan, Lady Bergavenny (d. 1435), wife of William Beauchamp, Lord Bergavenny (1392-1411) see G. E. C [ockayne], Complete Peerage, ed. Vicary Gibbs et al. (London, 1910-59, 13 vols.), i, p. 26. 139 See Beresford, op. cit., pp. 81-2, 117, 148-9, and Deserted Medieval Villages: Studies, ed. M. W. Beresford andj. G. Hurst (London, 1971), p. 11.
Antiquarian Studies in Fifteenth-Century England
323
Sir Henry Percy ('Hotspur'). 140 Hardyng then entered the service of Sir Robert de Umfraville, and came into contact with Henry V. In 1418 the king, who was contemplating enforcing the claim of the English crown to overlordship of Scotland, sent Hardyng there 'to spy out with all kinds diligence' how best to invade the country, and to collect evidence proving his claim.141 He stayed three years and subsequently presented documents to Henry V, Henry VI and Edward IV successively: modern research has shown that seventeen of the nineteen extant documents are forgeries. It is believed that Hardyng himself forged them,142 and, in the opinion of Sir Francis Palgrave, they show that he was 'a diligent antiquary'. 143 Hardyng began his Chronicle in about 1440.144 One of his reasons for writing was to persuade the king (he presented copies to both Henry VI and Edward IV)145 to enforce his right by invading Scotland. He recites every instance he knew of a Scottish king doing homage to an English one.146 Although he attributes a legendary origin to Scotland, he makes sound antiquarian observations. He notes, for example, that until Henry V's reign the king's head on Scottish coins faced sideways, as a sign of submission c to his sovereign lord of England5, while thereafter it looked straight ahead, as a sign of equality. Numismatically Hardyng's observation is more or less correct.147
140 por J0hn Hardyng's career see C. L. Kingsford, 'The first version of Hardyng's chronicle', E.H.R., xxvii (1912), 462-9. 141 Ibid., 463-7 passim, 741-3, 751. 142 Palgrave prints the eight documents preserved in the Public Record Office; Documents and Records illustrating the History of Scotland, ed. Francis Palgrave (London, 1837, one vol. only printed), i, pp. 367-76. Cf. Kingsford, op. cit., p. 468, and, with particular reference to the documents relating to the Great Cause, Edward I and the Throne of Scotland, 1290-1296: an Edition of the Record Sources for the Great Cause, ed. E. L. G. Stones and G. G. Simpson (published for the University of Glasgow, Oxford, 1978, 2 vols.), ii, pp. 385-7. 143 Palgrave, op. cit., p. ccxxiii. Palgrave was referring to the content of the forgeries, not to their handwriting which is that of Hardyng's own time. 144 -r;ne original version is discussed and part printed in Kingsford, op. cit., 469-82, 74053. The revised version is printed The Chronicle of John Hardyng, ed. Henry Ellis (London, 1812, reprinted New York 1974). For the date when Hardyng started the chronicle see Kingsford, op. cit., 465. 145 See iW., 465-7. 146 Chronicle, ed. Ellis, pp. 42-3, 159, 166, 210, 212, 214, 223, 228, 235-6, 240, 243, 247, 253-4, 256, 262, 269, 270, 276, 283, 294, 296, 299, 323. 147 Ibid., pp. 87-8. The bust of the king on Scottish coins was in profile probably until the reign of Robert III (1390-1406); thereafter the head was full-face: see H. A. Grueber, Handbook of the Coins of Great Britain and Ireland in the British Museum (revised ed., London, 1970), pp. 170 et seqq., and P. F. Purvey, Coins and Tokens of Scotland, Seaby's Standard Catalogue of British Coins, iv (Seaby's Numismatic Publications Ltd., London, 1972), pp. 13-34. However, the full-face portrait may have been introduced at the end of the reign of Robert II (1371-90); Edward Burns, The Coinage of Scotland (Edinburgh, 1887, 3 vols.), i. 364. Hardyng also suggests etymologies for the name of the Scots, and gives the (legendary) history of the Stone of Scone; Chronicle, ed. Ellis, pp. 86, 87, respectively.
324
Legends, Traditions and History in Medieval England
To help the royal campaign, Hardyng supplied an itinerary from Berwick via Edinburgh to Ross,148 the route taken by Henry IV in 1400, with distances between the towns.149 He also provided a map, the earliest known of Scotland: it marks the towns, with conventional pictures of fortifications, the principal rivers and seas. 15° Neither itinerary nor map is accurate, but they show remarkable interest in, and knowledge of, Scottish geography. As Hardyng's first editor, Richard Grafton, wrote, with more truth than elegance, in his verse preface:151 Neither is there any that ever wrote, Which in matters of Scotland could better skill. . ., Or better knew water, wood, town, vale and hill. The desire to persuade the king to foreign conquest was also one factor leading William Worcester to historical study, but in this case the foreign country was France. Worcester was born in Bristol in 1415, and spent his working life as secretary to Sir John Fastolf, a veteran of the French wars, who had served John, duke of Bedford, as major-domo when the duke was the king's lieutenant in Normandy (1422-35).152 Worcester wrote the Boke of Noblesse shortly after 1451 partly to please Fastolf.153 But his principal object was to persuade Henry VI to enforce his claim to the throne of France. He uses stock chivalric arguments in favour of war, but he supports his thesis with considerable research on the chronicles from which he drew precedents for such a conquest in ancient and medieval history. In addition, he collected relevant documents, most illustrating the duke of Bedford's success in France, which he appended as a 'codicil' to his Boke. Later he adapted the preface of the 'codicil', substituting Edward IV for Henry VI in the address, so that he could present a copy of both Boke and 'codicil' to Edward IV in 1475 on the eve of his French
148 There are two versions of the itinerary, one in verse, in the original chronicle, and one in prose, in the revised version. Both are printed by Ellis; Chronicle, pp. 422-9, 414-20, n. 12, respectively. 149 Ibid., p. 414, n. 12. 150 Three copies are known. One is in the original chronicle and is reproduced by D. G. Moir, The Early Maps of Scotland to 1850 (Royal Scottish Geographical Soc., Edinburgh, 1973), facing p. 5. The two others are in the revised version; one is reproduced in Rfichard] G[ough], British Topography (London, 1780, 2 vols.), ii, p. 579, and in Facsimiles of National Manuscripts of Scotland, photozincographed by Colonel Sir Henry James (Record Publications, Edinburgh, 1867-72, 3 pts.), pt. ii. 151 Chronicle, p. 11. For the value of Hardyng's maps see Moir, op. cit., pp. 6, 163. 152 por Worcester's life see K. B. McFarlane, 'William Worcester: a preliminary survey', Studies presented to Sir Hilary Jenkinson, ed. J. Conway Davies (Oxford, 1957), pp. 196-221. 153 Printed The Boke of Noblesse, ed. J. G. Nichols (Roxburghe Club, London, 1860, reprinted New York, 1972). It is discussed by McFarlane, op. cit., pp. 210-15.
Antiquarian Studies in Fifteenth-Century England
325
campaign.154 Fastolf worked Worcester hard, and even after his death in 1459, his business affairs, which he had left in confusion, occupied Worcester for nearly twenty years more. Worcester was only able to retire in 1478, and only then, less than five years before his own death, was he free to concentrate on antiquarian studies. At once he set out on his travels: in 1478 he rode from Norfolk to London and on again to St Michael's Mount; in 1479 he travelled in Norfolk; and in 1480 he rode from London to Glastonbury.155 It is true that on his journeys he exercised his piety,156 transacted business157 and discovered what he could about Sir John Fastolf s family158 - and his own.159 But his overwhelming motive was curiosity. Everywhere he went he took sheets of paper160 and jotted down what he saw and heard (pi. 42). His interests were multifarious 161 and history was one of them. He was an enthusiastic student of the British History, enquiring at Glastonbury Abbey for chronicles on King Arthur.162 But he also copied inscriptions and any notice of historical interest in churches, and examined and extracted from numerous martyrologies, chronicles and the like.163 However, his principal concern was topography. He described the appearance of places, and recorded (not always accurately) the distances between them and the measurements of buildings: no man previously had paced out buildings with such assiduous regularity.164 And his survey of Bristol was the first methodical topographical survey of an English town, foreshadowing John Stow's survey of London.165 The fifteenth century was, therefore, a period of importance in the development of antiquarian studies. Alongside traditional belief in legends, and even the willingness to accept forgeries, sound methods of documentary research and first-hand observation were growing. Many
154 The 'codicil' is printed in Letters and Papers illustrating the Wars of the English in France, ed. Joseph Stevenson (Rolls Series, 1861-4, 2 vols. in 3 pts.), ii, pt. ii, 521-742. It is discussed in McFarlane, op. cit., pp. 210-13. 155 For a map of Worcester's itineraries see the end of the printed edition; William Worcestre, Itineraries, ed., with an English translation, J.H. Harvey (Oxford Medieval Texts, 1969). 156 Ibid., pp. xiii-xiv. 157 See e.g. ibid., pp. 18,76,260. 158 See e.g. ibid., pp. 180, 184, 220. 159 See ibid., W6-12passim. IGO por ^ manuscript of the Itinerary see ibid., pp. xviii-xxi. 161 For his various interests see ibid., pp. x-xi. 162 Ibid., pp. 260, 292, 293, n. 2. 163 See e.g. ibid., pp. 100, 101, n. 1, 112, 122, 148, 154, 164,224,236,312. 164 See ibid., pp. xi-xii. 165 Worcester's survey of Bristol is not included in Harvey's edition, but is printed by James Dallaway, Antiquities ofBristowe (Bristol, 1834).
326
Legends, Traditions and History in Medieval England
scholars contributed to these studies, and ensured that the medieval antiquarian tradition survived until the eve of the sixteenth century, when it was to be taken over and extended by scholars in the Tudor period. The monastic antiquaries, though their views were distorted and their outlook limited by local loyalty, were well educated and intelligent. The seculars were less restricted in vision and scope, but were perhaps less thorough. And both groups show that objective curiosity was often a motive for studying the past.'This was history for its own sake, or, in the words of V.H. Galbraith, 'history for the sake of the historian - the indulgence of boundless curiosity, the thrill of discovery and the satisfaction of detecting error'.166 Note A now lost chronicle composed in the 1360s and 1370s in the interest of the Mortimers, earls of March, included a genealogy back to the legendary kings of Britain; see John Taylor, 'A Wigmore Chronicle, 1355-77', Proceedings of the Leeds Philosophical and Literary Society (Literary and Historical Section), xi (1964), pt. v, 81-94. Another Mortimer genealogy was compiled soon after 1385 (to prove the claim of Roger Mortimer to the throne) probably under the direction ofjohn Othelake, March herald; see M. E. Giffin, 'A Wigmore manuscript at the University of Chicago', National Library oj Wales Journal, vii (1951-2), 321-4. A genealogy and armorial roll of the earls and dukes of Gloucester was compiled c. 1435, almost certainly for Isabella Beauchamp, widow of Thomas Despenser, earl of Gloucester. It is now Bodleian Library MS Lat.misc.6.2 (R). An early sixteenth-century armorial genealogy of the Clares is in the benefactors' book of Tewkesbury abbey. The best surviving copy is now Bodeleian Library MS Top.Glouc.d.2. It is described by C. H. Bickerton Hudson in Transactions of the Bristol and Gloucestershire Archaeological Society, xxxiii (1910), 60-6. A version is printed in W. Dugdale, Monasticon Anglicanum, ed. John Caley, Henry Ellis and Bulkeley Bandinel (London, 1817-30, 6 vols. in 8 pts.), ii. 59-65. A page from each of these Clare armorial genealogies is reproduced in Colin Platt, The Abbeys and Priories of Medieval England (London, 1984), pp. 108-9. I am grateful to Dr Martin Kauffmann, Assistant Librarian at the Bodleian, for information about these manuscripts. Besides armorial genealogies made for members of the nobility a number of genealogical chronicles of the kings of England were produced in the fifteenth century. Some had a propagandist purpose similar to that of the Mortimer genealogy; they demonstrated the legitimacy of the king's claim to the English throne. In the case of Henry VI, they showed that he had a right to both the throne of England and that of France; see J. W. McKenna, 'Henry VI of England and the dual monarchy: aspects of royal political propaganda, 1422-1432', Journal of the Warburg and Courtauld Institutes, xxviii (1965), 145-62. For royal genealogical chronicles in the vernacular see A Manual of the Writings in Middle English 1050-1500, ed. A. E. Hartung, viii, E. D. Kennedy, XII Chronicles and Other Historical Writings (New Haven, Connecticut, 1989), pp. 2674-9, 2888-91. A rather similar roll to the Rous rolls was executed in the mid-fifteenth century for the Sudeleys and Botelers of Sudeley castle. The roll is now in the New York Public Library, Spenser Collection MS 193. It comprises a history of England, the royal pedigree with portraits of the kings in roundels, and the pedigree of the Sudeleys and Botelers, with heraldic shields; see Lord Sudeley, 'Medieval Sudeley. Part I. The Sudeleys and Botelers 166
V.H. Galbraith, Historical Research in Medieval (London, 1951), pp. 42-3.
Antiquarian Studies in Fifteenth-Century England
327
of Sudeley Castle', Family History, the Journal of Heraldic and Genealogical Studies, x (1977), 920, and D. Winkless, 'Medieval Sudeley. Part II. The fifteenth century roll chronicle of the kings of England, with Sudeley and Boteler pedigree. The Latin text and roundels', ibid., 21-39. For another roll chronicle of the Sudeleys and Botelers see Heralds3 Commemorative Exhibition 1484-1934 . . . Catalogue (London, 1936), p. 38, no. 68.
This page intentionally left blank
Appendix
Additional Notes to Chapter 6 Page 155 and n. 2. Edmund Bishop's attribution of the pruning of the calendar of Christ Church, Canterbury, to Lanfranc (F.A. Gasquet and E. Bishop, The Bosworth Psalter (London, 1908), pp. 27-39), is now known to be wrong. The revision took place in the early eleventh century. See Nicholas Brooks, The Early History of the Church of Canterbury (Leicester, 1984), pp. 252, 265 and refs. Page 156 n. 2. Since the publication of this article (in 1976) a new edition of John's chronicle, but not of the continuation, has appeared: The Chronicle of Glastonbury Abbey . . . John of Glastonbury's Cronica sive Antiquitates Glastoniensis Ecclesie, ed. J.P. Carley, with an English translation by D. Townsend (Woodbridge, 1985). Cf. above, pp. 289-97. Page 157 n. 4. See now the new edition: The Early History of Glastonbury . . . William of Malmesbury's De Antiquitate Glastonie Ecclesie, ed., with an English translation, John Scott (Woodbridge and Totowa, 1981), 41, 42. Page 158 n. 4. For 'Adam de Domerham', i. 71' read 'De Ant., ed. Scott, 144'. — n. 5 line 5. Insert after 1-25 'and in De Ant., ed. Scott, 27-33 and end notes passim'. Pages 159 and n. 2. For 'Adam de Domerham, i. 3; cf. ibid., 24, 113' read 'DeAnt., ed. Scott, 40'. — penultimate line. Insert after 'ii. 199', 'and in De Ant., ed. Scott, 34-9'. Page 160 n. 4. For 'Adam de Domerham . . . 53-4' read 'De Ant., ed. Scott, 52-4, 66-8, 94'. — n. 6. For 'Adam de Domerham, i. 54' read 'De Ant., ed. Scott, 94'. — n. 8. For 'Adam de Domerham, i. 44-5' read 'De Ant., ed. Scott, 84'. Page 161 n. 3 top line. Delete 'William on'. For 'Adam de Domerham., i. 18-20 passim' read 'De Ant., ed. Scott, 64-78 passim 193-5'. — n. 3 third line from end. For 'Adam de Domerham, i. 29' read 'De Ant. ed. Scott, 68'. Page 163 n. 1. For 'Adam de Domerham, i. 18' read 'De Ant., ed. Scott, 54'. — n. 3. For 'Adam de Domerham, i. 19-22' read 'De Ant., ed. Scott, 54-8'. — n. 5. For 'Adam de Domerham, i. 35-8; cf. ibid.' read 'De Ant., ed. Scott, 72-8; cf. Adam de Domerham'. Page 165 n. 4. For 'Johannes Glastoniensis, i. 7-8' read 'John of Glastonbury, ed. Carley, 6, 8'. Page 166 n. 3. Add 'Cf. now also R. Barber, 'Was Mordred buried at Glastonbury? Arthurian tradition at Glastonbury in the Middle Ages', in Arthurian Literature, iv, ed.
330
Legends, Traditions and History in Medieval England
R. Barber (Woodbridge and Totowa, 1985), 44-50'. Pages 168-9 n. 7. Add 'The words on the leaden cross probably derive from Geoffrey of Monmouth's Historia Regum Britanniae; see S.C. Morland, 'King Arthur's Leaden Cross', Somerset and Dorset Notes and Queries, xxxi (1984), 215, reprinted in idem, Glastonbury, Domesday and Related Studies (Glastonbury Antiquarian Society, 1991)', pp. 59-60. Page 174 n. 5. For 'Adam de Domerham, i. 20' read 'De Ant., ed. Scott, 44, 56'. — n. 7. For 'Johannes Glastoniensis, i. 56-7' read 'John of Glastonbury, ed. Carley, 54, 280-1 n. 83'.
Additional Note to Chapter 8
Pages 211, 235. Professor Joan Greatrex has called my attention to, and kindly sent me transcripts of, entries in the fourteenth- and fifteenthcentury precentors' accounts of Worcester cathedral, which concern the writing of a chronicle. Possibly the payments which these entries record resulted from the use of the method of up-dating a chronicle as described in the prologue of the thirteenth-century Winchester/Worcester chronicle. (A lost Winchester chronicle to 1281, which was apparently a contemporary account of events from about 1264, was the source of the thirteenth-century Worcester chronicle.)1 The prologue specifies that a sheaf of loose leaves should be attached to the end of the chronicle, upon which important current events should be noted; at the end of the year an annal was composed, using the notes, and copied neatly on to the end of the chronicle; the sheaf of rough notes was removed and a new sheaf substituted. 2 The entries in the Worcester precentors' accounts, which may relate to this practice, number nearly twenty, range in date from 1346/7 and 1425/6, and usually occur under the heading 'Minute Expense' but occasionally under 'Forinsece Expense'. The accounts themselves are in Worcester Cathedral Muniments, reference numbers C.351 (1346/7), C.352 (1348/9), C.353 (1349/50), C.354 (1350/1), C.355 (1354/5), C.356 (1358/9), C.357 (1360/1), C.358 (1361/2), C.361 (1374/ 5). C.361a (1376/7), C.363 (1383/4), C.364 (1384/5), C.367 (1390/1), C.370 (1400/1), C.375 (1419/20), C.376 (1422/3) and C.377 (1425/6). The entries in question record various payments of 2d, 3d, 4d, 6d, 8d, 12d or 2s to a scribe for writing, 'altering' or 'emending' the chronicle (e.g., 'In cronicfis] scribend[is]', 'scriptori scribentfi] cronicas', Tn cronicfis] mutand[is]', 'In emendacfione] cronic[arum]', 'pro 1
See N. Denholm-Young, 'The Winchester-Hyde chronicle', English Historical Review, xlix (1934, repr. in idem, Collected Papers on Medieval Subjects (Oxford, 1946), pp. 86-9. 2 See above pp. 211 and n. 64, 235 and n. 184.
Additional Notes
331
cronicfis] emend [andis']. The exact meanings here of the verbs 'mutare' and 'emendare' are unclear but the words seem to be used synonymously with 'scribere'. The scribe worked, it seems, on leaves added to the chronicle; in two cases the cost of parchment for him is mentioned (4d in 1354/5 and 8d in 1358/9). Moreover, the work was done annually; the scribe often includes 'hoc anno' in his entry. Thus far the precentors' accounts concur with the evidence in the prologue of the Winchester/Worcester chronicle. They have in addition two pieces of information for which the prologue provides no parallel. Two mid-fourteenth century (1346/7, 1348/9) entries state that the chronicle was 'written' in preparation for Easter (Tn cronicfis] scribendfis] ad pasch[am] 6d', Tn cronicfis] scribendfis] contra Pascham 6d'). The second entry, also mid-fourteenth century (1349/50), unparallelled in the Winchester/Worcester prologue, records that 12d was paid to the scribe of the 'great chronicle hanging in the church' (solutfos] cuidam scribent[i] magnas cronicas pendent[es] in ecclesia 12d). This suggests the possibility that neither this entry nor those already mentioned concern the up-dating of a chronicle of general history akin to the twelfth and thirteenth-century Worcester chronicles. Perhaps 'the great chronicle hanging in the church' served the same purpose as, or even was, a tabula of the kind which hung in a number of great churches from the fourteenth to the sixteenth centuries. A tabula was a board, or boards hinged to form pages, as it were, of a 'book'. On it were pasted sheets of parchment bearing the history of the church to which it belonged and/or, if the church were a conventual one, of the appropriate religious order. The purpose of the tabula was to inform visitors either at first hand or by supplying their guides with necessary information. A good example, of c.1400, survives from Glastonbury. It comprises a folding wooden frame 3 ft 8 ins high and 3 ft 6 ins broad when open. It encloses two smaller folding boards. All six sides of the boards are faced with parchment and bear the legendary history of Glastonbury, nearly all derived from John of Glastonbury's chronicle.3 However, it does not seem very likely that the references in the 3 For the Glastonbury tabula see above p. 293 and n. 16. For other examples of tabulae see: H.E. Savage, The Lichfield Chronicles (Lichfield, 1915), pp. 8-10; G.H. Gerould, ' 'Tables' in medieval churches', Speculum, i (1926), pp. 439-40; N. Denholm-Young, 'The birth of a chronicle', Bodleian Quarterly Record, vii (1933), pp. 236 and n. 5, 237-8; W.A. Pantin, 'Some medieval English treatises on the origin of monasticism' in Medieval Studies presented to Rose Graham, ed. Veronica Ruffer and AJ. Taylor (Oxford, 1950), pp. 200-1, 207-8. J.S. Purvis, 'The Tables of the York Vicars Choral', Yorkshire Archaeological Journal, xli (1966), 741-8. (I owe this last reference to Professor Greatrex; the example, Additional MS 533 in York Minister Library, is described in Medieval Manuscripts in British Libraries iv: Paisley-York, ed. N.R. Ker and AJ. Piper (Oxford, 1992), pp. 824-5). Cf. A. Gransden, Historical Writing in England, ii, c.1300 to the Early Sixteenth Century (London, 1982), p. 495 (Appendix E).
332
Legends, Traditions and History in Medieval England
Worcester precentors' accounts are to a tabula. The entries recording the purchase of parchment for the chronicle indicate a book of the normal kind. Nor does it seem probable that a chronicle of local history or an historical monograph, such as are found on tabulae, would have required the annual writing and revision revealed by the precentors' accounts. These procedures seem more appropriate to a chronicle of general history. The reference to a chronicle hanging in the church recalls the words of Gaimar (writing c.1140) who states that at King Alfred's command a copy of the Anglo-Saxon Chronicle was kept chained in Winchester cathedral.4 The truth of Gaimar's assertion has been doubted5 (it has been suggested that he confused the Chronicle with Alfred's translation of the Pastoral Care), and no surviving copy has marks indicating that it was ever chained.6 However, considered in the light of the other evidence, there seems just a faint possibility that Gaimar was right. If, as is possible, the Alfredian chronicle was originally composed at Winchester,7 this fact lends weight to Gaimar's mention of a chained copy in the cathedral. It is not unlikely that Worcester was among the first churches to receive a copy of the Chronicle, and a church which had a copy would continue it. Therefore, the instructions on the writing of a chronicle in the prologue of the thirteenth-century Winchester/Worcester chronicle, and the entries in the Worcester precentors' account showing that a chronicle in book form hung in the church and was 'written in' and 'emended' every year, suggests the merest chance that a method of chronicle writing introduced under King Alfred survived at Winchester at least until the thirteenth century, and at Worcester at least until the mid-fifteenth century.8
Additional Note to Chapter 10 On pages 259-61 above, I suggest that the 'Merton' Flores was written after the death of Edward I (7 July, 1307) at the command of Edward II, perhaps for presentation at his coronation. As I explained, the subject matter supports this dating. However, Adelaide Bennett, while accepting my contention that the 'Merton' Flores is an 'official' history, argues, for
4
Gaimar, L'Estoire des Engleis, ed. A. Bell (Anglo-Norman Text Society, xiv-xv, 1960), 11.2327-36. 5 See The Anglo-Saxon Chronicle, ed. Dorothy Whitelock with D.C. Douglas and S.I. Tucker (London, 1961), pp. xix-xx. 6 I owe this information to Dr Mildred Budny. 7 The Anglo-Saxon Chronicle, a Collaborative Edition, ed. David Dumville and Simon Keynes, iii, MS A, ed. J.M. Bately (Cambridge and Woodbridge, 1986), pp. xiv and n. 6, xxxii-xxxiii. 8 See above p. 202 and n. 15.
Additional Notes
333
'codicological, artistic and iconographic reasons', in favour of a late thirteenth-century date. (A. Bennet, 'A late thirteenth-century PsalterHours from London', in England in the Thirteenth Century, Proceedings of the 1984 Harlaxton Symposium, ed. W.M. Ormrod (Harlaxton College, 1985), 21 n. 20). I had decided that since the 'Merton' Flores is a revised version of the 'Westminster' Flores it would have been composed after the latter work was completed; the last entry in the 'Westminster' Flores is for early February 1307 and so presumably the chronicler stopped writing shortly after that date. The changes of hand in the Eton MS of the 'Merton' Flores which, as Dr Bennett points out, occur from the annal for 1295 onwards, and the other codicological irregularities, could have been the result of the manuscript being handed around the scriptorium to various scribes in the process of copying. (Cf. my Historical Writing in England, [i], C.550-C.1307 (London 1974), 458 n. 157). It could also be argued that artistic evidence seldom leads to exact dating. Nevertheless, in certain circumstances Dr Bennett's dating of the Eton MS could be right. Possibly from the late thirteenth-century the annals of the 'Westminster' Flores and those of the 'Merton' Flores were composed concurrently at Westminster, fairly close in time to the events they recorded. The text, therefore, of the 'Westminster' Flores would have been in a fluid state and wide differences would have been possible between it and its derivative, the 'Merton' Flores', the latter could easily have been adapted to suit the needs of the king. The relationship of the 'Westminster' Flores to the 'Merton' Flores would be analogous to that of the chronicle of Bury St Edmunds to the chronicle of St Benet of Hulme in the 1290s (above pp. 235-8). If, indeed, the 'Merton' Flores was started in 1295 or shortly afterwards, this would lead to the interesting conclusion that the Westminster monks started producing an 'official history for Edward I during his lifetime and just when royal propaganda was busy trying to stimulate enthusiasm for the Scottish war. (See M. Prestwich, War, Politics and Finance under Edward I (London, 1972), 240-2, D. Burton, 'Requests for prayers and royal propaganda under Edward I', in Thirteenth Century England, III, ed. P.R. Coss and S.D. Lloyd (Woodbridge, 1991), 25-35, and A. Gransden, 'John de Northwold, Abbot of Bury St Edmunds (1279-1301) and his defence of its liberties', in ibid., 111-12.) This conclusion would not affect my overall argument that: the 'Merton' Flores was an official history; a copy might have been made for presentation to Edward II on the occasion of his coronation; Murimuth preserves otherwise lost material from it.
This page intentionally left blank
Index
Aachen 36, 37 Abbo of Fleury, abbot of Fleury (988-1004) 99 — , Passio Sancti Eadmundi 47-51, passim,
54, 75, 77, 83-9 passim, 100, 104 Abingdon, Berks., Benedictine abbey 36, 56-65 passim, 70, 81, 118 — , version of Anglo-Saxon Chronicle at 202
— , copy of John of Worcester at 203 — , chronicle (12th c.) 59,65,66,70 — , abbots of, see: yEthelwine; Ordric; Siward; Spearhafoc Abraham, as type 261 n. 95 Acca 18, 179 n. 21 ^Elfgar, father-in-law of Edmund I 46 ^Elfheah, St, Bishop of Winchester (984-1006), archbishop of Canterbury (1006-12), 32, 43 and n. 48, 49, 60, 84 n. 7,90, 164 ^Elfhere, ealdorman of Mercia 41 and n. 42, 42 ^Iflaed 46 vElfric, bishop of East Anglia 46 and n. 64 ^Elfric, the grammarian 5, 6, 35, 39, 43, 44, 54, 55, 72, 75-9 passim, 83, 86, 87 n. 3, 94, 114, 145 — , Vita S. Ethelwoldi 39 and n. 33
yElfric, 'prior of Ely', 'bishop of Beadericesworth' 48, 93, 95 ^Elfthryth, queen of King Edgar 36, 58, 121
^Elfweard, abbot of Evesham, then bishop of London (1035-44) 56, 60, 64-6, 68, 99 ^Elfwine, bishop of Elmham (?x 1019-1023 x 1038) 48, 77, 93-6 passim, 98, 101, 103 ^Ifwine, abbot of Ramsey (1043-79/80) 91, 100, 102 Aeneas Sylvius Piccolomini (later Pius II) 29 n. 166 vEthelflaed, daughter of King Alfred, queen ofMercia320
yEthelflaed, queen of Edmund I 46 ^thelflaed, queen of King Edgar 121, 122 /Ethelric (or yEtheric) , bishop of Dorchester (1016-34) 47,66-7,69,70, 183 ^Ethelstan, see Athelstan ^Ethelthryth, St, see Etheldreda yEthelweard, ealdorman of Wessex, Chronicle 73 n. 226, 83, 1 18, 137, 202 ^Ethelwig, abbot of Evesham (1058-77) 60, 63,72, 109 n. 10, 110 — , Life of 65, 71 ^Ethelwine, ealdorman of East Anglia 41 and n. 42, 53, 64 n. 170, 66, 68, 69, 78, 81,99, 100 ^Ethelwine, abbot of Abingdon (1018-30) 56
^Ethelwine, sacristan of St Edmund's 94, 101
^Ethelwold, see Ethelwold yEtheric, thegn 47. See also vEthelric Agatho, pope 678-81,62 Aidan, St 15, 161 n. 3 Ailred, St, abbot of Rievaulx (1 147-67) 17 and n. 93, 156, 171, 179 n. 20, 187 Alban, St 5, 154 Alcuin 186, n. 63 Aldhelm, St 142, 145, 147, 177 n. 6, 181 and n. 31 Aldwin, prior of Winchcombe 7,8, 73 Aldwyne, Richard de, rebellious bondman 311 and n. 79 Alexander the Great, as type 22 Alexander II, pope 1061-73 14 n. 83, 16 Alexander I, king of Scotland (1 107-24) 15, 122 Alexander, bishop of Lincoln( 11 23-48) 10, 133
Alfonso XI, king of Castile (1312-50) 273 Alfred, king of West Saxons (871-99) 3, andn. 13, 4 and nn. 14, 16,32,33,34,41, 51, 63, 78 n. 253, 109, 110, 116, 118, 148,
336
Legends, Traditions and History in Medieval England
180,212 Alfred (or Alured), son ofWestou 6, 17, 179 n. 20 Algeciras, Andalusia, Spain, seige (1344) 272,273,275 'Alice', countess of Salisbury — , identity of 269-70 and nn. 1-3 — , alleged rape of 268-78 passim Alice, suojure countess of Lincoln and Salisbury, wife of Thomas of Lancaster (d. 1348), 270 and n. 2 Alice, daughter of Thomas of Brotherton 270 n. 3 Alice, wife of Edward Montagu 270 and n. 3 Al-Kharismi 119 Alured, see Alfred, son ofWestou Amalric I, Latin king (1163-74) 127 Ambrosius Aurelianus 19 anachronism, sense of 319-20 Angers, 191-2, 193 and n. 1 15, 194 Andrews, St, bishopric of 15, 122 Anglesey 321 and n. 128 Anglo-Saxon armour, see armour — , art 57,63, 72 n. 219 Anglo-Saxon Chronicle — , 'Alfredian' 4 and nn. 14, 15, 202 and n. 14, 212 and n. 65, 217, 262 n. 101 — , Edwardian 212 and n. 66, 262 n. 101 — , continuations, 10th- 12th cc., 63-4, 72 and n. 218, 202 and nn. 15-17, 212 and n. 67, 217-18 — , debt of Anglo-Norman chroniclers to 73, 118,203,207 — , Latin translations of, 9th- 12th cc. 73 and n. 226, 118,202-3,207 Anglo-Saxon dress 320 Anglo-Saxon kings, of Heptarchy, lists 280 Anglo-Saxon language 184, 212, 308-9 Anglo-Saxon laws, see laws, Anglo-Saxon Anglo-Saxon martyrology, see martyrology, Old English Anglo-Saxon script, facsimiles of, 15th c. 307 and n. 55 Annales Londonienses, use ofFlores Historiarum
in 204 and n. 29. See also Horn, Andrew Anselm, St, archbishop of Canterbury (1093-1109) 32,37,73,76,113,114, 146, 182, 186 anthropology, social, interest in, 12th c. 193, 194 and n. 120, 195, pis 5, 6, 8 Antioch 191 antiquarianism and archaeology, medieval examples of 168-9 n. 7, 175, 177 and
nn. 6, 1 1, 183, 186, 299-327 passim Aquitaine 193, 194 Abroath, Declaration of 276 architectural descriptions 68, 177 and nn. 6-8, 182 and n. 44, 183 and n. 45, 185-6 Arfast (or Herfast), bishop of Elmham (1070-2) bishop of Thetford (1072-85) 88, 89, 94 armour, John Rous's depiction of evolution of 319 and n. 121, 320 and n. 122 ARNOLD, THOMAS 239 and n. 2, 240-3 passim Arthur, legendary king 18-25 passim, 153, 245, 264, 293, 325 — , alleged sepulchral cross of 166, 167 and n. 5, 168 and n. 7, 175 note — , see also Glastonbury . . . abbey, invention and cult of King Arthur at Artois, count of, see Robert Arundel, Thomas, archbishop of Canterbury (1396-7, 1399-1414) 216 Ashingdon (Assandun), Essex, battle of (1016) 53, 100 Asser 34, 37, 41 and n. 41, 73 n. 226, 78 n. 253, 113, 118, 202 and n. 18 Athelney, Somerset, 180 and n. 30 Athelstan, king of England (925-39) 32, 33,45,50,69,83, 159, 187 n. 63 Athelstan, the 'Half King' 69 Athelstan, abbot of Ramsey (1020-43) 66, 67
Athelstan, one of King Alfred's helpers 1 1 0 Augustine, St, of Canterbury 14, 15, 16, 23, 26, 27, 34, 37, 48, 64, 93, 118, 154, 184, 195,307,311 Augustine, St, of Hippo, De Civitate Dei, 134, 147-8 and n. 155, 209 n. 59 Augustinian canons 301 Aumale, counts of see: Forz family; William le Gros authenticating devices 309. See also: bulls, papal; rotae; seals 'authority', literary, medieval respect for 2-3, 13-14,26-7, 142. See also under Bede Avalon, Isle of 21, 22, 166, 169 and n. 7, 170, 1 72 and n. 2, 173 and n. 1 Aveley, Essex, church 320 n. 122 Avesbury, Robert of 206 and n. 48, 214 — , De Gestis Mirabilibus Regis Edwardi Tertii
214,217,249,263 Avignon, papacy at 276, 286 Aylesham, John of, abbot of St Benet of Hulme (1325-46) 243 and n. 2
337
Index 'B' biographer of St Dunstan 159, 162 Babwell, Suffolk, Franciscan friary, chronicle 205, 280 and n. 2 Bache, Alexander, bishop of St Asaph (1390-4) 232 Baker, Geoffrey le (fl. 1350) 216, 257, 274 Baldwin, archbishop of Canterbury (1184-90) 172, 189 n. 83, 194 Baldwin, abbot of Bury St Edmunds (1065-97/8) 65, 72, 85, 88, 92, 95, 101 n. 2, 104, 145, 240, 242 Balsham, Hugh, bishop of Ely (1257-86) 235
Bale, John, bishop of Ossory, topo-bibliographer (d. 1563) 279 Bamburgh, Northumb. 1 1 Basle, Council of ( 1 434) 294 Bassingbourn, Sir Warin de 215 n. 80 Bateman, William, bishop of Norwich (1344-55) 244 Bath — , description of 177 n. 6, 192 and n. 102a castle 188 bishops of 155, 156, 306. See also: Savaric; Jocelin Bath and Wells, bishop of, see March, William Battle, Sussex, Benedictine abbey, chronicles 183, 184, 199,205 Beadericesworth, see Beodricesworth bear 196 Beauchamp, Catherine, countess of Warwick 320 Beauchamp, Isabella, widow of Thomas Despenser, earl of Gloucester 326 Beauchamp, Richard, earl of Warwick (1403-39) 320, pi. 41 Beauchamp, Thomas, earl of Warwick (1329-69) 319,320 Beauchamp, Thomas, earl of Warwick (1370-1401) 286, 319, pi. 40 Beauchamp, William, Lord Bergavenny (1392-1411) 322 n. 138 Beaumont, John (d. 1342), son of Henry Beaumont earl of Buchan 270 and n. 6 Bee, Eure, Benedictine abbey 73 Beccles, Suffolk 46 Becket, Thomas, St, archbishop of Canterbury (1162-70), 154, 185, 190 n. 83, 192, 193 — , Life of, see Tewkesbury, Alan of Bede 2-3, 36, 73-4, 175, 177, 186 n. 63, 310 — , cult, 6, 27 and n. 158, 28, 29 and n. 166 — , praise of, and eulogies on 2, 6 n. 25,
9-10, 11-13, 23, 27-8, 75, 109, 137, 140 — , WORKS — , — , commentary on Song of Songs 13 n. 74 — , — , De Natura Rerum 6, 55 n. 113 — , — , De Temporibus 6, 55 n. 1 13, 56 and n. 114, 119 — , — , — , chronicle in, 2, 200 and n. 6, 20 1-2 and n. 13,209 — , — , De Temporum Ratione 6, 55 n. 113, 56
and n. 1 16 — , — , — , chronicle in, 2, 200 and n. 6, 201-2 andn. 13,209
— , — , Ecclesiastical History 18, 35, 47, 52,
53, 73-4, 78, 108, 111, 135, 141, 142, 143, 150
— , — , — , manuscripts 1-2, 29 — , — , — , epitome 4 and n. 15 — , — , — , popularity 1-29 passim — , — , — , influence of, on, and use of, by, chroniclers etc. 1-3, 4, 5, 10-13, 19-20, 22-3, 24-5, 26-9, 39-41, 75-6 and n. 240, 109, 139, 148, 158, 160 and n. 4, 180 — , — , — , influence of, on actions and events 5, 7-8 and n. 36, 17-18, 35, 37, 43, 51
— , — , — , use of, in controversy 13-16, 23, 26-7, 109 — , — , Letter to Ecgbert 34 and n. 14, 35 andn. 15,37-8,40 — , — , Lives of the Abbots ofWearmouth 2, 5,
25 n. 142, 55, 64, 109 — , — , Lives of St Guthbert 2, 4, 5, 13, 27, 28,35,45,50, 177 BEERJEANETTE 127
Bedford 287 — , dukes of, see John of Lancaster Beere, Richard de, abbot of Glastonbury (1493-1524) 289 Bek, Anthony, bishop of Durham (1284-1311) 317 Bel, Jean le, chronicler 267-78 passim Benedict, St, of Nursia 28, 48, 51, 98, 104, 315 n. 99 — , Rule 16,31,38,81 'Benedict of Peterborough' 308 n. 62 Benedict Biscop, abbot ofWearmouth/ Jarrow (674-90) 7, 16, 26 and n. 149, 62, 161 n. 3 Benignus, St, 159 BENNETT, ADELAIDE 265
Beodricesworth (or Beadericesworth, later — , Bury St Edmunds, q.v.), Suffolk clerical community 10th c., 45-50
338
Legends, Traditions and History in Medieval England
passim, 54, 59, 77, 81, 83, 85, 88, 90-8 passim, 100, 101, 104 — , 'bishop of, see yElfric Bergavenny — , Lady, see Joan — , Lord, see William Berkeley, Roger de,jn. (d. 1131) 119,238 Bermondsey, Surrey, Cluniac abbey, annals, 292 and n. 12, 302, 303-4 Bernham, William de, abbot of Bury St Edmunds (1135-61) 244 Bersuire, Pierre, Benedictine scholar 276 Berwick 324 bestiaries 197 and nn. 140, 141, pis 8-1 1 Bever, John (? alias John of London), monk of Westminster 253 and nn. 40, 41, 259-60. See also London, John of . . . author Beverley, Alfred (or Alured) of 19-20 — , Annales . . . de Gestis Regum Britanniae
19,24, 130, 133, 142, 143, 149 Beverley, Robert de, abbot of Meaux (1356-67) 311,314 Bewcastle, Cumberland 161 Bible, allusions to, etc. 200, 249 and n. 30, 250-2, 255-6 bibles, de luxe copies 1 78, 310. See also Gospel-books Billfrith, hermit 1 79 Black Death, see under plague BLAIRJOHN 33 Blois, Henry of, bishop of Winchester (1129-71) 19, 157 n. 4, 181, 189 n. 83, 309 n. 69 Blois, Peter of (fl. 11 90) — , description by, of Henry II 252 — , section of Crowland chronicle falsely ascribed to 304 Boethius 137 Bohun family, earls of Hereford 216 Bohun, Humphrey de, earl of Hereford (1299-1322) 216 Bohun, William de, earl of Northampton (1337-60) 271 Boisil, abbot of Melrose 13 Bolingbroke, Henry, see Henry IV Boniface IV, pope 608-15 16 and n. 90, 40 — , forged bull of 16 n. 91 Boniface VIII, pope (1294-1303) 264 n. 108, 280 'Book of Benefactors', see under: St Albans; Tewkesbury 'Book of Hyde' see under Winchester, Hyde . . .
Book of Kells, see under Gospel-books 'Borgoyne', Westminster 259 Boroughbridge, battle of ( 1 322) 250 Bosworth Psalter 159 n. 3, 162 n. 1, 175 Note Boteler family, see Sudeley and Boteler families Bower, Walter, Scotichronicon, 226 and n. 133, 230, 231 and n. 159, 232, 295, 296 BOYLE, LEONARD 291
Brabant, duke of, see John II Bradfield, Suffolk 46 and n. 64 Bradford-upon-Avon, Wilts., Anglo-Saxon church 180n. 30 Brakelond, Jocelin of, Cronica (i.e. Life of Abbot Samson) 85, 178, 199, 203, 241 and n. 3 bramble 197 and n. 140, pi. 12 BRENTANO, ROBERT 31 7 n. I l l
Bretons, dissident and rebellious, 12th c. 21-2 Breydenbach, Bernard of (d. 1502), traveller, 3 18 and n. 120 Bridget, St 161 n. 3 Bridlington, Yorks., Augustinian priory, chronicle 236 n. 186 Brihtnoth, thegn, 46 Bristol 192, 324, 325 and n. 165 — , castle 188 Britons, ancient 18-23 passim, 294 Brittany, county of — , count of, see Geoffrey — , countess of, see Constance — , see also Bretons Britanny, duchy of 271, 273, 278 — , duchess of, see Montfort, Jeanne de — , dukes of, see: John (de Montfort) III; John (de Montfort) IV Brockford, Suffolk, 46 and n. 64 BROOKS, NICHOLAS 45
Brotherton, Thomas of, earl of Norfolk (131 2-38) , daughter of, see Alice Bruce, David, see David II, king of Scots Brut chronicles 206 and n. 50, 207, 2 16 and n. 84, 223 — , probably written in London 206 Bruton, Somerset, early church 181 and n. 34 Brutus, father of Lucretia 278 Brutus, legendary founder of Britain 157, 173
buffaloes, Richard of Cornwall's 178 Bugga, St, see Eadburga bulls, papal 307-8, 309 and n. 66
Index burglary, of royal treasury (1303) 251 and n. 35, 252, 254 Burgundy, duke of, see Philip Burton, Thomas, abbot of Meaux (1396-9) 292-3, 303 — , history by, of Meaux 292, 302, 311-15 and nn. Burton upon Trent, Staffs., Benedictine abbey 59 — , chronicle 205 Bury chronicle, see Bury St Edmunds, Benedictine abbey, chronicles Bury St Edmunds (previously Beodricesworth, q.v.), Suffolk, Benedictine abbey 61, 62, 72, 77, 81-104, 119, 154,280 — , foundation of 47-9, 57, 59, 77, 81-2, 88- 104 passim, 240, 243 — , chapels — , — , Lady 238 —,—, St Edmund's 237-8,242-3 — , connections of, with St Benet of Hulme, 95-8 passim, 101-4 passim 240 and n. 6, 243-4 — , relations of, with town, see Bury St Edmunds, town — , copy of 'Florence' and John of Worcester at 118, 203 and n. 2 1 , 204 and n. 24, 205 and n. 34, 240, 241 -, CHRONICLES — , — , Annales Sancti Edmundi 241 and n. 4 — , — , chronicle ofjohn de Taxter (to 1 265) 204, 205, 226-9 passim — , — , — , continuation of (to 1 296) 204-5, 226-38 passim, 241-2 and nn. 3, 4, 243, 279-80 — , — , — , — , continuation of (to 1 30 1 ) ascribed to John de Everisden 204, 226, 241 and n. 7, 244 Note — , — , Cronica Buriensis (1020-1346) 238 n. 142, 239-44 — , — , — , to 1327 written at St Benet of Hulme 244 — , — , Depraedatio Abbatiae 240 — , — , Electio Hugonis 241 and n. 5, 244 Note — , — , see also: Brakelond, Jocelin of; Hermann, archdeacon — , abbots of, medieval lists of 46-7 and n. 64, 240. See also: Baldwin; Bernham, William de; Draughton, Richard de; Leofstan; Northwold, Hugh de; Samson; Ufi — , prior of, see Stowe, William de
339
Bury St Edmunds, town, relations of, with abbey, 240 and n. 2, 243-4 Butley, Suffolk, Augustinian priory, chronicle of 281 and n. 2 Byrhtferth, monk of Ramsey 6 and n. 25, 39, 54 and n. 103, 55-6 and n. 1 14, 64, 99, 100 Byrhtnoth, ealdorman of Essex 41 Caedwalla, king of West Saxons (685-8) 27 Caen, John of, clerk of Edward I 264 Caerleon, Monmouths. 196 Cair Segeint (Segontium), Caernarvon 177 n. 6 Calais, siege of (1347) 283 Calixtus II, pope (1119-24) 14 Cambridge, university of 276, 285-6 and n. 1 Campsey Ash, Suffolk, priory of Augustinian canonesses 284 and n. 1 1, 285 Canan, John, monk of Glastonbury 164 Canterbury 192 Canterbury, metropolitan see, primacy of 14, 15, 16 n. 91, 23, 76, 109, 122, 236 n. 186 — , archbishops 154. See also: yElfheah; Arundel, Thomas; Becket, Thomas; Blois, Henry of; Dover, Richard of; Dunstan; Escures, Ralph d'; Lanfranc; Langton, Stephen; Oda; Reynolds, Walter Canterbury, cathedral 27, 163, 164, 300 — , architecture of, description of, 12th c. 185 Canterbury, Christ Church, pre-monastic community of 45, 47, 81 — , — , cathedral priory 16 and n. 89, 57, 62, 63, 70, 71 n. 214, 78-9, 91 n. 4, 108, 121, 145, 154, 155, 161 n. 3, 172 n. 2, 182, 183, 185-6 — , — , hagiography and historiography at 16 and n. 89, 72, 73, 182, 185-6, 202-3 and n. 20. See also: Eadmer; Canterbury, Gervase of; Osbern. — , — , prior of, see Conrad — , — , subprior, see Ethelred Canterbury, St Augustine's, Benedictine abbey 16 n. 91, 26 and n. 147, 27, 202, 306- 10 and nn. — , historiography at 26 n. 147, 281. — , — , see also: Elmham; Thomas, Sprott, Thomas; Thorne, William — , library of, book lists and books from
340
Legends, Traditions and History in Medieval England
204, 253 and n. 41, 308 n. 61, 310 — , abbots of, see: Hadrian; Scotland (Scollandus) Canterbury, St Gregory's, Augustinian priory 307 n. 51 Canterbury, Gervase of 130, 185 and n. 58 — , Chronicle 172 n. 2, 184-6 passim, 300, 308 n. 56 — , — , prologue 132-4 passim, 136-8 passim — , Mappa Mundi 185
Canute, see Cnut Capgrave, John, Augustinian friar (d. 1464) 223
— , Liber de Illustribus Henricis 2 1 3 and n. 70
Cappochi, Nicholas, cardinal-priest of St Vitalis (1350-61), cardinal bishop of Frascati (1361-8) 284 and n. 2 Caradoc of Llancarfan, Life ofSt Gildas 156, 162, 163, 168, 169 CARLEYJ.P. 289-92, 297, 302 n. 10 Carlisle, bishop of, see Ireton, Ralph Carmarthan 196
Chelsworth, Suffolk 46 and n. 61 Chertsey, Surrey, Benedictine abbey 42, 76, 145 Chester, description of 12th c., 191, 192, 193
Chester-le-Street, Durham, St Cuthbert's 45,50,81 Chichester, Sussex 192 Chinnock, John, abbot of Glastonbury (1375-1420) 29 1-2 and n. 14, 295 — , political role of 293-4 and nn. 25, 27 chivalry, code of, heroes of, etc. 214, 263 n. 101, 274, 293. See also romance Chronicle of the Rebellion in Lincolnshire 2 1 3
and n. 71 chronicles, royal use of 2 19-20 and nn., 264 Chronique de Flandres 267 n. 4, 273 nn. 1, 7 Chronique normande, 267 n. 4, 273 n. 1 Chronographia Regum Francorum, see under
Paris, St Denis, Benedictine abbey, chronicles compiled in chronology, study of, 8th- 12th cc. 6, 55-6, 142, 143. See also Dionysian era CARPENTER, CHRISTINE 322 n. 136 Cicero 200 n. 1 Cass mac Glaiss, legendary swineherd 1 73 Cistercian Order, monasticization of north Cassiodorus 139 of England by 17-18 Castile, kings of, see: Alphonso XI; Peter II Clare, earls of Gloucester Castle Acre, Norfolk, Cluniac priory 309 — , patrons of Tewkesbury abbey 2 1 6 n. 83 — , armorial genealogy of 326 and n. 69 Clare, Osbert de, monk of Westminster 68 Cato, Disticha de Moribus 136 classics, influence of, on writers, 12th c. Catulla 87 125, 128, 133-9 passim, 146, 175-6, Caxton, William, printer 25, 223 186-97 passim Celle, Peter de, abbot of St Remigius, clerical communities, 10th, 1 1th cc. 43 and Rheims 132 n. 48, 44 and nn., 45-51 passim, 54, 76, Ceolfrid, abbot of Wearmouth/Jarrow 77 and nn., 78, 84, 90, 94-5 (681-716) 62, 161 n. 3 — , invective against 39-42 passim, 78-9 Ceolwulf, king of Northumbria (729-37) clerks, secular, as chroniclers and 143 historians 206, 215 and n. 80, 216, 262 Chandler, Thomas, warden of New College, andn. 101,263 Oxford (1450-75) 3 18 n. 116 Clisson, Amaury de 271 n. 5, 273 n. 8 CHAPLAIS, PIERRE 68 Clisson, Oliver de 273 and n. 8, 274, 275, Charlemagne 87 278 Charles (of Luxembourg) IV, emperor, king of Bohemia (1346-78) 283 and n. 2 Cluniac Order, in England 303 Cnut, king of England (1016-35) 48, 56-60 Charles IV, king of France (1322-8) 257 passim, 62, 66, 67, 77, 90-6 passim, 98, Charles V, king of France (1364-80) 274 104, 114, 183, 190 n. 84,308,310 n. 6 Cockfield, Suffolk 46 — , as dauphin 284 Coffyn, Thomas, monk of Glastonbury Charles of Blois (d. 1364) 271, 273, 278 (fl. late 14th c.) 292 Chartres, Ivo of 131 Coggeshall, Ralph de, Chronicle 167, 172, Chatteris, Cambs., Benedictine nunnery 191 54,99, 100, 101 Chaucer, Geoffrey The Legend of Good Women Colchester, Essex, false etymology 1 73 Colchester, St John's, Benedictine abbey, 276-7
341
Index chronicle 205 and n. 36 Coldingham, Reginald of, monk of Durham, 101 n. 2 Cole, eponymous king 1 73 Coleman, St Wulfstan's biographer 6, 71 and n. 216, 72, 1 14, 115, 120, 158. See also under Malmesbury, William of, works Collatinus 277,278 Combe, Warwicks., Cistercian abbey, abbot of, see Geoffrey commemoration, of individuals, as motive for some historical writing 1 75, 1 78, 311, 313, 315 and n. 99, 316, 317, 318, 321 Commendatio Lamentabilis in Transitu Magni
Regis Edwardi [I] 252 and n. 36, 253 and nn. 40,41,260-1 and n. 95 common-place book 279, 280 confraternities 97, 102, 103, 240 and n. 6, 244
Conrad, prior of Christ Church, Canterbury (1108/9-26) 186 Constance, Council of (141 7) 294 Constance, countess of Brittany (d. 1201) 21
Constantinople 191 Cornwall, Richard of, Holy Roman Emperor (1257-72) 178 coronation — , ordo, see under Edward II — , pictures of — , — , King Arthur and Edward the Confessor 245 and n. 3 — , — , William I-Edward I 245 and n. 3, 247-8 —,—, Henry III pi. 28 —,—, Edwardi pi. 29 — , regalia 213, 247-8 and n. 20, 259, 260 costume, see dress Cotton, Bartholomew, monk of Norwich, Historia Anglicana 226 n. 129
Coventry, Warwicks., Benedictine abbey 59, 203 Coventry and Lichfield, bishop of, see Langton, Walter CRASTER, H.H.E. 306
Crecy , Somme, battle of ( 1 346) 283 and n. 1 Crediton, Devon, see of 59 Crendon, Richard of, abbot of Nutley (fl. 1339) 219 Cricklade, Wilts., castle 188 Cronica Buriensis, see under Bury St
Edmunds, . . . abbey, chronicles
Cronica Maiorum et Vicecomitum Londoniarum
206 n. 49. See also Fitz Thedmar, Arnold Crowland, Lines., Benedictine abbey 53, 60,61 n. 147 — , chronicle of, see under Ingulf Crusade, Third, chronicles of, see: Itinerarium Peregrinorum L'estoire de la guerre sainte
curiosity, to satisfy, a motive for historical writing 209-10 and n. 62, 325, 326 Cuthbert, St 18, 29 and n. 166, 50, 51, 55, 56,81,84,87, 101 n. 2, 114, 154 — , Lives of, see: Bede, works, Lives of St Cuthbert; Lindisfarne, monk of — , see also St Cuthbert's, see Cyprian 138 DANBURY, ELIZABETH 253 n. 39a, 272 n. 1
Danegeld 119 Danelaw 50, 83 Daniel and the lions 86, 261 Danish invasions, see Vikings DARLINGTON, R.R. 78
David, as type 261 n. 95 David I, king of Scots (1085-1 153) 17 David II, king of Scots (1329-71) 231,269, 283, 284 and n. 3 DAVIS, R.H.C. 212 n. 65 DEANESLY, MARGARET 308 n. 61
dedications, of historical works 125, 127, 129,214-15 Dee, river 193 DENHOLM-YOUNG, N. 225 and n. 122
Denis, St, apostle of France 86-8 passim, 92, 294 Derby, earl of, see Henry of Grosmont Deruvian (or Duvian), legendary missionary 24, 174,305 descriptions of people 175, 176, 186 and n. 63, 187 and nn. 65, 66, 68, 188, 189 and n. 83, 190 Despenser, Henry, bishop of Norwich (1370-1406) 287 and n. 16 Despenser, Thomas, earl of Gloucester (1397-1400), widow of, see Beauchamp, Isabella Devizes, Richard of, Cronicon de Tempore Regis Ricardi Primi 1 32, 1 35, 1 5 1 , 1 88, 1 92 Diceto, Ralph, dean of St Paul's (1 180/1 1201) 24, 130, 188-9 and n. 80, 208 n. 56 — , Ymagines Historiarum 191-2, 193, 194,
237
— , — , prologue 138, 139, 141-3 passim, 149-51 passim
342
Legends, Traditions and History in Medieval England
Dionysianera 112, 143 Dionysius Exigus 112 diplomatic, medieval interest in 175, 184 divine providence, concept of, in historical writing 134, 147, 148. See also edification
Durham, Symeon of 7, 8 n. 36, 12, 13, 14, 27,28, 75, 101 n. 2, 118, 131, 151, 179, 181
Durham, Reginald of 18 Duvian, see Deruvian
DOBSON, R.B. 317 n. I l l
documents, use of, by chroniclers etc. 184, 210-1 1, 236 and n. 186, 289, 304, 307, 309, 3 1 3- 1 4 and n. 90, 3 1 7, 324-5 and n. 154 Domerham, Adam of, continuation of William of Malmesbury's De Antiquitate Glastoniensis Ecclesie by 156, 158, 164, 166, 167, 170, 171,289 Domesday Book 304, 322 and n. 136 Dominic, prior of Evesham 109 Domne Eafe, abbess of Minster, Thanet 310
Dorchester, bishops of, see: ^Ethelric; Eadnoth I; Eadnoth II Doulting, Somerset 181 Dover, Richard of, archbishop of Canterbury (1 174-84) 307 and n. 56, 308
Draughton, Richard de, abbot of Bury St Edmunds (1312-35) 243 dress, John Rous's interest in historic 320- 1 Drythelm 5, 8 Dublin, archbishops of, see: Hotham, William de; Sanford, John de DUFTY,A.R. 320 n. 122
Dunfermline, Fifes., Benedictine abbey 247 Dunstable, Beds., Augustinian priory, chronicle 220 n. 104, 226, 227, 237 Dunstan, St 4, 5, 36-8 passim, 42, 43, 45, 49 n. 76, 50, 52-4 passim, 59, 76-8 passim, 83, 113, 122, 154, 156 n. 4, 157-160 passim, 161 and n. 3, 163-6 passim, 168, 169 n. 5, 181, 182 Dunwich, Suffolk 47 Durham, city 192 Durham, cathedral 8, 12, 29 and n. 166 Durham, St Cuthbert's — , see 317. Bishops of, see: Bek, Anthony; Flambard, Ranulf; Langley, Thomas; St Carilef, William of; Walcher. See also St Cuthbert, see — , cathedral priory 6, 7, 17, 27, 45, 74, 118, 154, 155, 179, 315 n. 99, 316-17 — , — , chronicler, see Durham, Symeon of — , — , priors of, see: Turgot; Wessington, John — , — , monks of, see: Durham Reginald of; Durham Symeon of; Graystones, Robert
Eadburga (or Bugga), abbess of Minster, Thanet 177n. 6 Eadberht, bishop of Lindisfarne (688-98) 160 n. 4 Eadfrith, bishop of Lindisfarne (698-721) 179
Eadmer, monk of Christ Church, Canterbury 15, 16 n. 89, 32, 37, 71 nn. 213, 214, 74, 76, 78, 79, 108, 109, 1 12, 113, 120-2 passim — , as hagiographer 73, 108, 113, 146, 155, 156 and n. 4, 163, 182 — , as architectural historian 185 — , WORKS — , — , Historia Novorum 73, 118, 137, 146, 148
—,—,—, prologue 130,132,136,141, 145, 151 —,—,LifeofStAnselm 73, 132, 146 Eadnoth, abbot of Ramsey (993-1006), bishop of Dorchester (1007x1009-1016) 53,54,60,66,99, 100, 104 Eadnoth II, bishop of Dorchester ( 1 034-49) 66 and n. 190 Eadsige, clerk, then monk, of Old Minster, Winchester 43, 78 Eadulf, vision of 8 Easter Tables, annals on 201 and n. 10 Ecgbert, (arch)bishop of York (P732-66), Bede's letter to, see under Bede, works Ecgwin, St 11,64,99, 109 Eddius Stephanus, Life of Wilfrid 11, 137, 1 77 and n. 8 Edgar, king of England (959-75) 4, 36, 37, 39, 41-5 passim, 46 n. 61, 51, 52, 53, 56, 57, 62, 68, 75 and n. 233, 76, 77 n. 242, 92, 95, 100, 121, 122, 145, 217-18, 310 — , queen of, see ^Elfthryth edification, as motive for historical writing 135, 138,209,301 Edinburgh 324 Edith, queen of Edward the Confessor 58, 59,70 Edmund, St, king of the East Angles (855-69) 82, 154 — , passion of 84, 86-7 and nn. 3, 6 — , burial place, first 88-9 — , — , second (final), see: Beodricesworth;
Index Bury St Edmunds — , body (incorrupt) 49, 53, 84, 85-6, 145 — , — , translations of 85, 241 n. 3 — , — , viewings of 49, 85, 94, 145, 241 n. 3 — , cult of 45-6, 49, 53, 54, 70 n. 212, 77, 81 — , feast of 243 n. 2 — , chapel of, see under Bury St Edmunds Edmund I, king of England (939-46) 45-6,
50,82,83,92, 164
Edward the Elder, king of the West Saxons (899-939) 262 n. 101 Edward the Martyr, king of England (975-9) 76, 121 Edward the Confessor, king of England (1042-66) 65, 72, 88, 98, 186 n. 63, 218, 320 n. 123 — , coronation of, picture of 245 and nn. 3, 4 — , as patron/benefactor of monasteries 58, 59, 60, 62, 67, 70 — , appearance of 186 n. 63 — , cult of 110, 154, 156 — , queen of, see Edith — , Lives of 114, 156, 177, 186 n. 63, 264 Edward I, king of England (1272-1307) 233, 234 — , as Lord Edward 258, 261 — , relations of, with —,—, France 281 — , — , Scotland 280 — , — , — , succession case ( 1 29 1 -2) 21 9-20, 232, 236, 243, 247 and n. 13, 264 and nn. 107-8. See also Processus Scotiae _? __, __ 9 war 250-1, 254, 255, 258 and n. 82, 260. See also Stirling, siege of —,—, Wales 294 — , — , Westminster abbey 251 and n. 35, 254, 259 and n. 86. See also Westminster . . . abbey, royal connections of — , interest of, in Arthurian legends 171, 264, 293-4 — , commissioned account of siege of Stirling castle 259-60 — , use by, of history 264 — , eulogies on 250-1, 252, 259-61. See also Commendatio Lamentabilis in Transitu Magni Regis Edwardi — , appearance of 1 78 — , body and tomb of 248 and n. 21 — , queens of, see: Eleanor of Castile; Margaret of France — , daughters of, see: Elizabeth; Mary Edward II, king of England (1307-27) — , coronation of 259-60 — , — , ordo for 260 and n. 90, 280 and n. 5
343
— , — , picture of 213-14, 247-8, 260, pi. 29 — , loss of Gascony by 256,261 — , relations of, with — , — , Scotland 256, 261, 262 — , — , Westminster abbey 259 and n. 88. See also Westminster . . . abbey, royal connections of deposition of 103, 214, 247 n. 13, 258, 261-2 opinions about, of chroniclers etc. 214, 255-62 passim queen of, see Isabella de Valois Edward III, king of England (1327-77) — , as heir-presumptive 257 — , accession 262 — , relations of, with — , — , Brittany 271 and n. 7, 272, 273, 278 —,—, France 214,271,274,278,282-6 passim — , — , Scotland 214, 272 n. 1 — , — , Spain 273 —, itinerary of (1342) 271 and n. 7 , 2 7 2 and n. 1 — , alleged rape by 267-9, 271-8 passim — , interest of, in Arthurian legends 293 — , appeal by, to chronicle evidence 219 — , ?Flores Historiarum (1307-26) written for 261-2 — , Gesta of, see Avesbury, Robert of Edward IV, king of England (1461-83) 323, 324. See also Historic of the Arrivall of.. Edward of Woodstock, prince of Wales (1343-76) ('Black Prince') 283, 284 and n. 1, 285-7 passim Edwy, king of England (955-9) 46 effigies, etc. 316, 320 and n. 122 Eggestein, Heinrich, printer 25 Eilaf, hereditary priest of Hexham 17 Einhard 113, 146 Eleanor of Castile, queen of England I 171, 189 n. 83, 232, 234 Eleanor of Provence, queen of Henry III, a work dedicated to, by Matthew Paris 222, 264 elephant, pictures of 178, 197 Eleutherius, pope 24, 160, 174 Elfwy, monk of Evesham 7, 8 Elizabeth (alias Walkiniana), daughter of Edward I 234 Ellington, Hunts. 67 Elmham, Norfolk, cathedral 10th c. 49 n. 74 Elmham, see of 47, 49
344
Legends, Traditions and History in Medieval England
— , bishop of, see: ^Elfwine; Arfast Elmham, Thomas, monk of St Augustine's, Canterbury, then prior of Lenton (1414P1426) 26 and n. 146 — , Speculum Augustinianum 26 and n. 149,
27, 236, 292 and n. 6, 295 and n. 28, 302 and n. 12, 303 and n. 22, 304, 306-11 and nn., pis 30-33 Eltham, nr London 270 Ely, Cambs., bishops of, see: Balsham, Hugh; Eustace; Northwold, Hugh de Ely, clerical community 10th c. 54 Ely, Benedictine abbey, then cathedral priory (1 109) 5, 41 and n. 42, 43, 51, 53, 60, 61, 62, 69, 81, 90, 91 n. 4, 93, 96, 97, 98, 240, 280. See also Liber Eliensis — , abbot of, see Leofsige Emma, queen of Cnut 58, 59 and n. 126, 91,92, 104, 183 enclosure movement, 15th c. 322 and n. 136 English, Middle, language, chronicles in 207
eponyms 172-3, 191 Ernald, abbot of Rievaulx ( 1 189-99) 20-22 passim, 132, 133, 136, 137 Escures, Ralph d', archbishop of Canterbury (1114-22) 14, 15 Espec, Walter, lord of Helmsley (d. 1 153) 187
Essarts, Guerin des, abbot of St Evroul (1123-37) 132 Ethelbert, king of Kent (560-616) 11,16, 22, 37, 307, 308 and n. 61 Ethelbert, St, see Ethelred and Ethelbert, SS Ethelbert, St, of Hereford 47, 49, 64 n. 170, 89,99, 100, 101 Ethelreda (or ^thelthryth), St 5, 12, 51, 54,81, 183 Ethelred (the Unready), king of England (978-1016) 52,57,76, 148,217 Ethelred and Ethelbert, SS 64 n. 170, 99 Ethelred, monk of Worcester, formerly subprior of Christ Church 121 Ethelwold (or ^Ethelwold), St, bishop of Winchester (963-84) 4, 5, 35-44 passim, 51-5 passim, 59, 69, 72, 75, 76, 77 and n. 242, 94, 114, 145, 179 — , treatise by, on establishment of the monasteries 75 and n. 233 Ethelwulf, king of the West Saxons (839-55) 147 Etienne of Rouen, see Stephen of Rouen etymologies 177 n. 6, 323 n. 147
eulogies 6 n. 25, 9-10, 1 1-13, 23, 214, 215,
216 Eulogium Historiarum 296
Eusebius, Chronicle, translation of, by St Jerome 25, 200 and n. 5 Eustace, bishop of Ely (1 198-1215) 241 n. 5 Evesham, Worcs., battle of ( 1 265) 255, 260 Evesham, Benedictine abbey 1 1 , 42 n. 46, 57, 60-1, 63, 64, 65, 71, 73, 81, 99, 109, 202 and n. 15 — , chronicle 56, 66, 70, 155 — , abbots of, see: ^Elfweard; ^Ethelwig; Mannig; Walter — , prior of, see Dominic — , monks, see: Elfwy; Reinfrid Everisden, John de, monk of Bury St Edmunds, chronicle ascribed to, see under Bury St Edmunds, . . . abbey, chronicles exemption, monastic 26 and n. 147, 61 and n. 150,62,88,92 Exeter 192 — , castle 188 — , see of, bishop of, see Leofric Eynesbury, Hunts. 53 and n. 99 Eynsham, Adam of 188, 189 and n. 73, 190, 196 Fabyan, Robert (d. 1513), New Chronicles of England and France 207 and n. 53, 209
Fagan (or Phagan), legendary missionary 24, 174, 305 Fakenham, Nicholas, provincial minister of the Franciscans in England ( 1 395-? 1 40 1 ) 279 and n. 3 Fantosme, Jordan, Chronique 199 Faringdon, Berks., castle 188 Faritius, abbot of Abingdon ( 1 1 00- 1 7) 1 42 Fastolf, Sir John 98 n. 3, 324-5 Felix, St 48,66,93, 177 Finan, bishop of Lindisfarne (65 1 -6 1 ) 1 60 n. 4 Fitz Stephen, William 188,192 — , Life ofSt Thomas 190 n. 83, 196 n. 134 — , — , description of London in 189 and n. 76, 191, 192-3, 194, 196 n. 134 Fitz Thedmar, Arnold, Cronica Maiorum et Vicecomitum Londoniarum attributed to
206 n. 49 Flambard, Ranulf, bishop of Durham (1099-1128) 101 n. 2 Flanders, count of, see Philip Flanders, chronicles of, see: Chronique de Flandres Istore et cronique de Flandres
Index FLEMING, ROBIN 50 Flete, John, prior of Westminster (c. 145765), history by, of Westminster abbey 292 and n. 1 1, 295 and n. 29, 302, 306 and n. 48 Fleury, Loiret, Benedictine abbey 32, 37, 43,51,99 — , abbot of, see Abbo of Fleury floods 282, 287 and nn. 6, 7 Flores Historiarum, see under: Paris, Matthew, works; St Albans . . . chronicles written at; Westminster . . . chronicles written at Foldbriht, abbot of Pershore (c. 970-88) 5-6 and n. 24 Fontevrault, Maine-et-Loire, abbey of Benedictine nuns 190 food 194 and n. 117 food-rents 46 n. 64, 61 Fordun,John, Chronicon Gentis Scotorum 226 n. 133,230 foreigners, dislike of, 10th, 1 1th, 13th cc. 217-18 and n. 97 forgery — Archaeological 22, 163-4, 165-6 — , of charters 15-16 and n. 91, 26, 46 and n. 60, 68, 71, 82, 88 n. 6, 90-1, 98, 162, 163, 184 and n. 55, 301, 304, 307, 308, 323 and n. 143 — , literary 82, 296, 304. See also: 'Girardus Cornubiensis'; 'Vigilantius'; and under Ingulf Fortescue, Sir John 223 Fortunatus 86 Forz family, including counts of Aumale 311 n. 78, 315 n. 97 foundation, monasteries anxious to prove their ancient 9, 11, 174, 175, 181-2,289, 293-5 passim, 301, 303, 305-6 founders, of towns, churches, etc., interest in, 12th, 15th cc. 157, 321. See also origins France — , conversion of, to Christianity 294 — , claim to, of English crown 274, 275, 324 — , kings of, see: Charles V; Philip IV; Philip VI Franciscan Order — , in England 281, 282, 283 and n. 4, 285-6 and n. 1, 287 and n. 9 — , — , provincial minister, see Fakenham, Nicholas — , — , friars, see: Medilton, John; Philipps, Nicholas
345
— , — , houses, see: Babwell; Lynn; London, Newgate — , ministers general, see: Mark of Viterbo; Rossi, Leonardus Feculfus, bishop of Lisieux (823 x 824-853) 160 Frocester, Walter, abbot of St Peter's, Gloucester (1381-1412) 303 history by, of St Peter's, Gloucester 292 and n. 5, 302 and n. 1 1, 303 and n. 23 Froissart, Jean 201 n. 12, 268-9 and n. 3, 270 and n. 5,274 Fulbrook, Warwicks., manor and castle 322 Fulco, archbishop of Rheims (883-900), letter from, to King Alfred 41, 78 n. 253 Fulda, Germany, abbey 1 12 Furness, Lanes., Cistercian abbey, chronicle 221 GALBRAITH, V.H. 48, 93, 176, 211 n. 64, 224, 225, 241 n. 7, 326 Gascony 256,261,283,284 Gembloux, Sigebert of 138, 139, 140 genealogical chronicles 326-7 genealogies 311 and n. 78, 318, 326-7 Geoffrey, count of Brittany, son of Henry II 21 Geoffrey, abbot of Combe (1332-45) 282 geographical descriptions 10-11, 12-13, 159 Germanus, prior of Ramsey (969-c. 970) 99 Germany, merchants from, 12th c. 193 Gesta Henrici Quinti 213, 263, 294 Gesta Stephani 262 n. 101 Ghent, St Peter's, Benedictine abbey 37 Gildas 18, 20-4 passim, 148, 161 n. 3, 162, 163 and n. 1, 165, 169, 177 and n. 6, 191 n. 91 'Girardus Cornubiensis' 303 n. 19 Giso, bishop of Wells (1060-88) 59 Glaber, Radulfus 149n. 165 Glasteing, eponymous swineherd 172, 173 Glastonbury, Somerset, false etymology of name 172-3 Glastonbury, Benedictine abbey 43 n. 18, 62,81, 182,290,291,292-3,325 — , cellofStDunstan 182 — , fire (1184) 155,160,164,165,174 — , 'pyramids' in cemetery 161 and n. 3, 181 — , invention and cult of King Arthur at 22, 153, 163, 165-74 and nn. passim,
346
Legends, Traditions and History in Medieval England
293-4, 306, 325. See also Guinevere — , — , royal interest in 170 and nn. 2-4, 171 andnn. 2, 5, 273-4 and n. 22 — , invention of St Dunstan at 163-4, 165, 181
— , other cults and relics at 153, 161 and n. 3, 163, 164-5 and n. 1 — ,— , Gildas 162, 165 — , — , Joseph of Arimathea 153, 154, 174, 293, 294 — , — , Patrick 159, 161 n. 3, 162, 163 and n. 3, 164, 165 — , legendary history of 82, 153, 154, 289, 293, 294-5, 305 — , — , royal interest in 293-4 — , Plost annals of, 13th- 14th cc. 289 — , chroniclers etc. of 83. See also: Domerham, Adam of; Glastonbury, John of; Malmesbury, William of, works, De Antiquitate Glastoniensis Ecclesie;
Wyche, William. See also Tabula Magna — , abbots of: Bere Richard de; Chinnock, John; Dunstan, St; Monington, Walter de; Robert; Soilli, Henry de; Taunton, John of — , monks of, see: Canan; Coffyn, Thomas; Seen, John; Toe, Ralph; Watelege, John de; Wyche, William Glastonbury, John of, chronicler, identity of 290, 297
— , Cronica . . . Glastoniensis Ecclesie by 156,
165, 173-5 passim, 289-97 passim, 301-2 andn. 10, 306 and n. 45 — , — , date 290-6 Glastonbury (Glastynbury), William 315 n. 99 Gloucester, town 191 — , earls of, see: Clare; Despenser, Thomas; Robert — , earls and dukes of, genealogies and armorial roll of 326 Gloucester, St Peter's, Benedictine abbey — , copy of Worcester chronicle at 118, 119,203,238 — , history of, see Frocester, Walter — , abbot of, see Frocester, Walter Gloucester, Robert of — , chronicle 215 n. 80 Godgifu, wife of Leofric 59 Godwin, earl of Wessex (d. 1053) 217-18 Gorron, Geoffrey de, abbot of St Albans (1119-46) 58 n. 131 Goscelin, hagiographer 58, 71 n. 215, 156, 191 n. 91
Gospel-books 177 n. 8 — , Book of Kells 196 and n. 133 — , Lindisfarne Gospels 1 79 — , Stonyhurst Gospels 179 n. 24 — , See also bibles, de luxe Gower, John, Confessio Amantis 276 Grafton, Richard, chronicler and printer 324 Grandes chroniques, see under Paris, St Denis . . . abbey, chronicles compiled in GRANSDEN, K.W. 268 n. 1 Graystanes, Robert, Libellus Dunelmensis Ecclesie 317
GREATREXJOAN 33, 330, 331 n. 3 Gregory the Great, pope (590-604) 14, 15, 34 andnn. 13, 14,37,38,51, 110, 183, 192 — , Whitby Life of 55 Gregory XI, pope (1371-8) 231, 287 n. 8 Grimbald, St 117,316 Groton, Suffolk 229 and n. 151 GUENEE, BERNARD 127-31 passim, 139, 150, 151,201
Guerno, forger 16 n. 91 Guinevere, legendary queen 22, 153, 163, 165, 167 n. 5, 168, 169, 171, 172 and n. 3 Gundulf, bishop of Rochester (1077-1 108) 114 Guthlac, St 177n. 11 Guy's Cliff, nr Warwick, chantry chapel at 317 HADCOCK, R.N. 81
Hadrian, abbot of St Augustine's (669-708) 27
'Haeglesdun', place of St Edmund's 'martyrdom' 84, 88 hagiographers, professional 156 and n. 4, 162 Hainault, Jean de, brother of Queen Philippa 282 n. 2 Hainault, Holland and Zealand, count of, see William II Hall, Edward (d. 1547) 207 Harcourt, Godfrey de, Norman lord 273-5 passim, 278 hare 196 HARDY, T.D. 253 n. 40 Hardyng, John, soldier and chronicler (d. 1465) 223, 317, 322-3 and n. 140 — , Chronicle 215 n. 80, 296, 323-4 and nn. HARMER, F.E. 91, 93 Harold, king of England (1066) 48, 93, 148 Harthill, Yorks. 314
Index Hastings, Sussex, battle of 70 HEARNE, THOMAS 289, 290
hedgehogs 197 and n. 140, pi. 1 1 Hemming, monk of Worcester, cartulary and Life of Wulfstan by 71,110
Henry I, king of England (1100-35) 9, 117, 119, 320 and n. 123 — , of, 12 the. 190n. 84 Henry II, king of England (1154-89) — , relations of, with — , — , Brittany 21 — , — , Normandy 294 n. 22 — , — , Welsh 294 and n. 22 — , interest of, in Arthurian legends etc. at Glastonbury 22, 170-2, 293-4 and n. 22 — , pen-portraits of, 12th c. 189 and n. 83, 190, 252 — , historiography in reign of 183, 300 Henry III, king of England (1216-72) — , baronial opposition to 210, 218 and n. 94, 227, 258, 259, 260 — , relations of, with Westminster abbey 258-9. See also Westminster . . . abbey royal connections of — , pen-portrait of, 14th c. 178 — , picture of, pi. 28 — , alleged posthumous miracles of 254 — , chroniclers' attitudes to 222, 229, 254-5, 260, 263 and n. 104, 264 — , queen of, see Eleanor of Provence Henry IV (Henry Bolingbroke), king of England (1399-1413) 219-20, 231, 296 — , claim of, to throne 295 and n. 31, 296 — , censors Glastonbury chronicle 295 — , ?interest of, in Glastonbury legends 295-6 Henry V, king of England (1413-22) 214, 294, 322, 323 — , Thomas Elmham's Life of 303 — , see also Gesta Henrici Quinti
Henry VI, king of England (1422-61, 1470-1) 323,324,326 Henry VII, king of England (1485-1509), as Henry Tudor 223 Henry (the Fowler), king of Germany (919-36) 32 Henry of Grosmont, earl of Derby (133761), duke of Lancaster (1351-61) 273 Henry of Trastamara 286 and n. 5 heraldry 178 n. 14, 318, 326-7 Note. See also: genealogical chronicles; genealogies herbals 197 and n. 140, pis 12, 13 Hereford
347
— , earl of, see Bohun, Humphrey de — , bishop of, see Losinga, Robert Hermann, bishop of Ramsbury (1945-55), bishop of Sherborne (1058-78) 59 Hermann, hagiographer 65, 77, 85, 89, 94, 103 — , De Miraculis Sancti Eadmundi 65, 77, 85,
89, 90-6 passim, 103,240-1 Herfast, see Arfast Hexham, Northumb., hereditary priest of, see Eilaf Hexham, St Wilfrid's monastery 177 Hexham, Augustinian priory 171 — , translation of saints (1 155) 18, 156 Hickling, Norfolk, Augustinian priory, chronicle of 281 and n. 4 Higden, Ranulf (or Ranulph) 23 n. 131, 24-5,219 — , Polychronicon 23, 24 and n. 133, 26, 165, 201 and n. 12, 209, 223, 224, 276, 305 Hilduin, abbot of St Denis (814-C.842) 86 hind, legend of a 310-11 Histoire du chateau et sires de Saint-Sauveur-leVicomte 268 n. 4 Historia Aurea 28 n. 158 Historieof the Arivall of Edward IV . . . 213 and n. 71 HOHLER, CHRISTOPHER 54 n. 99
Holcot, Robert, Dominican scholar 276 Holderness, Yorks. 312,314-15 — , descent of honour and lordship of 31 1 n. 78 Holinshed, Raphael (d. ? 1580) 207 Holme, St Benet of, see St Benet of Hulme Holy Land, the 318 Holy Sepulchre, canons of, their dress 320 homiletic, influence of, on historical writing 321-2 passim Horace 192, 193 Horn, Andrew, fishmonger of London, Annales Londonienses attributed to 206 n. 49 Horningsheath (Horringer), Suffolk 46 horses 196n. 134 Hotham, William de, archbishop of Dublin (1297-8) 233-4 Howden, Roger of 130 — , chronicle 151,263 Hoxne, Suffolk 47, 49, 88, 89 and n. 4, 100, 101, 103 Hugh, 'dean of York' 12 Hugh, St, bishop of Lincoln (1186-1200) 190, 196 HughofFleury 142
348
Legends, Traditions and History in Medieval England
Hull, Yorks., see Kingston-upon-Hull Hull (Old and New), river 314 Hulme, St Benet of, see St Benet of Hulme humanists, 14th- 15th cc. 276 Humber, river 314-5 Hundred Rolls, John Rous consults 322 n. 136 Hundred Years' War 274-8 passim, 282-6 Huntingdon, Henry of 11, 130, 209 — , Historia Anglorum 9, 23 n. 131, 90 n. 4, 151
— , — , prologues 10, 134-8 passim, 147, 148, 150 — , Letter to Warin 18 Hyde, Benedictine abbey, see under Winchester Horsa, derivation of name 177 n. 6 Ickworth, Suffolk 46 Indract, St 159, 164, 165 Ine, king of West Saxons (685-9) 160, 306 Ingulf, abbot of Crowland (1085/6-1 109), Crowland chronicle (the 'pseudo-Ingulf) falsely ascribed to 82, 292 and n. 8, 302 n. 18,304 Innocent III, pope (1 198-1216) 241 n. 5 Ipswich, Franciscan friary 284 and n. 12 Ireland 193. See also Irish, the Ireton, Ralph, bishop of Carlisle (1280-92) 317
Irish, the, description of, by Gerald of Wales 194-5 Isabella de Valois, queen of Edward II 214,257,261,262 Isidore of Seville 87 n. 3 Istore et croniques de Flandres 267 n. 4, 273 n. 1
itineraries 263 n. 101, 324, 325 and n. 155. See also: Leland, John; travels Ivarr, Viking leader 49, 84 Ivo, St 53, 99 James, St 1 74 Jarrow, Durham, Anglo-Saxon monastery
6, 12, 13,35,62,73, 109
Jarrow, Benedictine priory 7 Jerome, St 28, 187 n. 63, 315 n. 99 Jerusalem 191 Jews 192, 242 n. 4, 320 Joan, daughter of Henry II 184 Joan of Kent (d. 1385), wife of Edward of Woodstock 282, 285 Joan, Lady Bergavenny (d. 1435), wife of William Beauchamp, Lord Bergavenny 322 and n. 138
Jocelin, bishop of Bath (1206-42) 155 n. 4 John XV, pope (985-96) 62 John XXII, pope (1316-34) 276 John, king of England ( 1 1 99- 1 2 1 6) 1 90 — , as prince 194 — , coronation of, picture of 248 n. 20 — , reissue of his charter to the church, etc. 280
John II, king of France (1350-64) 274 n. 6, 276, 283-4, 285, 320 John II, duke of Brabant (1294-1312) 234 John (de Montfort) III, duke of Brittany (1312-41) 278 John (de Montfort) IV, duke of Brittany (1364-99) 271,274,278,286 John of Gaunt, duke of Lancaster ( 1 362-99) 22 1, 286 and n. 9 — , allegedly has chronicle forged 296 John of Lancaster, duke of Bedford (141435) 322,324 John, abbot of St Albans (1 195-1214) 97 n. 6 John, author, 14th c. 251-3 passim. See also London, John of JOHN, ERIC 32, 41,44 n. 54 JONES, M.C.E. 219 n. 101, 268 n. 1, 271 nn. 5, 7 Joseph, St, of Arimathea, legend of foundation by, of Glastonbury abbey 1 74, 289, 305. See also under Glastonbury Jumieges, Robert of, archbishop of Canterbury (1051-2) 91 n. 4 Justin, early Christian apologist 131 Juvenal 192 Katharine, wife of William Montagu I, earl of Salisbury 269, 273 — , alleged rape of, by Edward III 274 KENDRICK,T.D. 299 and n. 4 300
Kenilworth, Dictum of (1266) 229 KEYNES, SIMON 31 note, 33, 42 n. 47
Kildare, Ireland 195 Kingston-upon-Hull, Yorks. 314 n. 96 Kirby Cane, Norfolk 91 Knevyngton, Ralph de 320 n. 122 Knighton, Henry, Augustinian canon and chronicler (d. c. 1396) 216 KNOWLES, DAVID 31, 32,33, 38,44, 78,81, 92, 97, 244 LAISTNER, M.L.W. 1, 2 n. 9, 3. 29
Lambert, Giles, dean of Tours (1290-1313) 280 n. 1
Index laments, eloquent 250-3 Lancaster —, earl of, see Thomas of Lancaster —, earls and dukes of 216 —, dukes of, see: Henry of Grosmont: John of Gaunt Lanfranc, archbishop of Canterbury (1070-89) 9, 14 and n. 83, 16, 32, 37, 70 and n. 212, 71, 73, 74, 76, 88, 91, 107, 110, 113, 121, 155, 176andn. 5, 182, 186 —, Constitutiones 70 and n. 211 Langley, Norfolk, Premonstratensian abbey 244 —, abbot of, see Strumpeshagh, John de Langley, Thomas, bishop of Durham (1406-37) 28 Langtoft, Peter of, chronicle 201 n. 12 Langton, Stephen, archbishop of Canterbury (1207-28) 241 n. 5 Langton, Walter de, bishop of Coventry and Lichfield( 1296-1321) 232,233 Lantfred, monk of the Old Minster, Winchester, Translatio et miracula Sancti Swithuni by, 52, 54 LAPIDGE, MICHAEL 31 note, 86 n.4 Latin —, influence on, of vernacular, 14th c. 282 —, florid 68, 249 and n. 30, 250-7, passim. See also rhetoric Laurence, St 316 Laurence, archbishop of Canterbury (604x609-619) 15, 16 Laurence, Master, abbot of Westminster (c. 1158-1173) 156, 171 laws, Anglo-Saxon and Norman 280 lawyers, advise Edward I (1292) 280 and n. 1 laymen, as historians, 206-8 passim. See also: Hardyng, John; Worcester, William Lazarus 294 legends and legendary people 276-8, 289, 299, 301, 305-6, 310-11, 317, 318, 321, 323 and n. 147, 325. See also: Arthur, King; Glastonbury . . . legendary history of; Joseph, St of Arimathea Leicester 191 Leicester, St Mary's in the Meadows, Augustinian abbey 216 Leland, John, antiquary 299 and n. 4 —.Itinerary 299 Lenton, Notts., Cluniac priory, prior of, see Elmham, Thomas Leo, Byzantine emperor (457-74) 25 n. 141
349
Leofric, earl of Mercia 59 Leofric, bishop of Exeter (1050-72) 59, 114 —, Missal 161 n. 3 Leofric, abbot of Peterborough (1052-66) 62 Leofsige, abbot of Ely (1029-35) 61 Leofstan, abbot of Bury St Edmunds (1044-65) 85,92,94, 145 Leofstan, 'a proud thegn' 49 L'estoire de la Guerre Sainte 262 n. 101 Les Voeux du Heron 275 and n. 1 278 LETTENHOVE, r KERVYN DE 268-9 and n. 1 Leven, Hugh de, abbot of Meaux (1339-49) 314 Lewes, Sussex, battle of (1264) 220, 255 Lewes, St Pancras's, Cluniac priory 309 and n. 69 Lewknor, Oxon. 59, 70 L'Histoire de Guillaume le Marechal 215 n. 80 Liber Eliensis 11, 56, 69 n. 206, 95-6, 183 Limousin 231,234 Lincoln 189 and n. 74, 190, 192 —, bishop of, see Alexander Lincoln and Salisbury, countess of, see Alice Lindisfarne island, Northumb. 177 n.6 Lindisfarne, St Cuthbert's see, church and community 13, 27, 28, 81, 160 n. 4, 179 —, bishops of, see: Eadberht; Eadfrith; Finan —, monk of, Life ofSt Cuthbert by 55 Lindisfarne Gospels, see under Gospel-books Livy 276 and nn. 3-7, 11,277 local history 8-9, 11-12, 26-9, 64-70, 109, 115, 157, 174, 175, 199,210,211,223, 292, 300-18 and nn. passim, 326 Lollards, as 'dispossessioners' 300 London 14, 15, 60, 101, 173, 191, 268, 325 —, description of, 12th c. 189, 191-3 passim —, called 'New Troy' 191,193 —, bishop of, see yElfweard London, Guildhall 321 London, Newgate, Franciscan friary, chronicle and continuation 207 and n. 54,281 andn. 3 London, New Temple 259 London, St Paul's cathedral 321 —, copy ofFlores Historiarum at 204, 246 —, tradition of chronicle writing at 208 and n. 56. See also: Diceto, Ralph; Murimuth, Adam London, Tower of 193, 251
350
Legends, Traditions and History in Medieval England
—, see also Thames London Chronicles 206 and n. 49, 207 and n. 52, 208, 214 —, not 'official' 208 and n. 57, 223-4 —, see also Annales Londonienses Brut chronicles; Cronica Maiorum . . .; Fabyan, Robert; London, Newgate, Franciscan friary, chronicle . . .; Stow, John London, John of, monk of Westminster, 253 and n. 39a London, John of (? alias John, q.v., and/or John of London, monk of Westminster, q.v., and/or John Bever, monk of Westminster, q.v.), author 252-3 and nn. 40, 41 London, John of, rector of Newland, Forest of Dean (1264-1302) 253 n. 39a Longchamp, Henry de, abbot of Crowland (1190-1236) 304 Losinga, Herbert, bishop ofThetford (1090 x 1091-1094 x 1096), bishop of Norwich (1094 x 1096-1119), 48, 49 Losinga —, quarrel of, with Bury St Edmunds abbey 47,48,88,89,93, 101 —, —, use of historiography to support his cause in 47, 48, 93-6 passim, 101, 103 Losinga, Robert, bishop of Hereford (1079-95) 111, 112,203 Louis I (the Pious), Roman emperor and king of the Franks (778-840) 36 Lucan 192 Lucian, monk of St Werburgh's, Chester De Laude Cestrie 188, 191, 193 Lucina, St 87 Lucius, 'emperor', 25 n. 141 Lucius, legendary king of Britons 24, 305 Lucius II, pope (1144-5) 309 Lucretia, rape of 276-8 passim —, her 'canonization', 277 and n. 1 Lucy, Richard de, justiciar (1154-78/9) 184 Lud, eponymous hero 173 Lynn, Norfolk 281, 282, 285, 287 —, interdict on (1377) 287 and n. 16 —, tolbooth in 285 Lynn, Carmelite friary 282, 285 Lynn, Franciscan friary 279 and n. 1 —, chronicle 205, 279-87 McKISACK, MAY 299
Mainz 112 Malestroit, Morbihan, truce of (1343) 278
and n. 1 Malmesbury, Wilts., Benedictine abbey 77 and n. 244, 81, 108, 145, 147, 180, 181 andn. 31 —, prior of, see Warin Malmesbury, William of 27, 71 n. 214, 74, 76, 77 and n. 244, 111,120, 175, 176, 181 andn. 36, 185, 186, 191 —, his admiration for, and influence on him, of Bede 9, 11, 12, 23 n. 131, 26, 75, 78-9, 137, 138, 150 —, on Anglo-Saxon church 74-6,108, 145-6, 179, 180 —, on Glastonbury'pyramids' 160-1, 182 —, on Malmesbury abbey 76-7, 180, 181 —, WORKS —, —, De Antiquitate Glastoniensis Ecclesie 156-8 and nn. 4, 5, 159-62 passim, 165, 174, 180, 182,289 —, —, Gesta Pontificum 11, 75, 112, 120, 147, 180, 181, 191,221 —, —, —, prologues 130, 136-44 passim, 147 —, —, Gesta Regum 11, 26, 112, 118-19, 147,221 —, —, —, prologues 130-44 passim —, —, Historia Novella 146, 215 —, —, —, prologues 132, 135 —, —, Life ofSt Benignus (lost) 159 —,—,LifeofStDunstan 159 —, —, Life ofSt Dunstan (lost) 159 —, —, Life ofSt Indract (lost) 159 —, —, Life ofSt Patrick (lost) 159 —, —, Life of Wulfstan, translated from Life by Coleman, q.v., 71 n. 212, 72, 115, 118-19, 120 Malvern, Great, Worcs., Benedictine priory 111 Mannig, abbot of Evesham (1044-58) 63, 65,66 manuscripts, see Index of Manuscripts Map, Walter, archdeacon of Oxford (1196-?!208), satyrist 176,188,189 andn. 73, 190 maps, topographical plans, etc. 186, 310 andn. 70,311 andn. 75, 324 and nn. 150, 151 March, earls of, see: Mortimer family; Mortimer, Edmund; Mortimer, Roger; Mortimer, Roger March, William, bishop of Bath and Wells (1293-1302) 232 Margam, Glamorgans., Cistercian abbey, chronicle 167, 172 and n. 3, 206 and
Index n. 45 Margaret of France, queen of Edward I 252,264 Margaret of Provence, queen of Louis IX 178 Marianus Scotus 203 —, Universal Chronicle 203, 112, 116, 117 Mark of Viterbo, minister general of the Franciscans (1359-66) 286 and n. 3 Markyate, Christine of 58 n. 131 Marlborough, Thomas of, monk of Evesham 109 n. 10 Martha 294 martyrology, Old English 4 and n. 16 Mary, Virgin 48, 93, 98, 122, 259, 316 —, Seven Joys 209 Mary, St, of Egypt 87 Mary, daughter of Edward I 264 Mary Magdalen 294 Matilda, Empress (d. 1167), 118, 146 Maurice abbot of Rievaulx (1145-7), 17 Meaux, Yorks., Cistercian abbey 311-15 —, a history of, see under Burton, Thomas —, abbots of see: Beverley, Robert de; Burton, Thomas; Leven, Hugh de; Skyne, Adam de Medilton, John, Francisan friar 279 Melcombe (Regis), Dorset, Black Death starts at 283 Mellitus, bishop of London (604-17), 51 Melrose, Roxburghs., Anglo-Saxon monastery 7 —, abbot of, see Boisil Melrose, Cistercian abbey, chronicle 206, a n d n . 45, 226, 227, 230, pi. 18 Melvas, legendary king 169 Merlin 22 Merton, Surrey, Augustinian priory, copy ofFlores Historiarum made for, see under Westminster, . . . abbey, chronicles written at, Flores Historiarum . . . ('Merton' version) . . . Michael, St, archangel 293 Mildred, St 307 n. 51 Milton, Dorset, Benedictine abbey 42, 76, 145 monasteries and monasticism —, revival of Benedictine, 10th c. 4-6, 31-52, 60, 74, 75-9 passim, 81-2 —, predicament of, under Anglo-Normans, 8-16 passim, 70-9 passim, 88, 107 etseqq. 154-5, 176, 182, 183 —, hostility to, 14th-15th cc. 25, 223 —, historians of, 14th-15th cc. 301
351
Monington, Walter de, abbot of Glastonbury (1342-75) 289, 291, 295, 297 Monk Soham, Suffolk 46 and n. 64 Monmouth, Geoffrey of 18-19, 24, 27, 130 — Historia Regum Britannie 19-23 passim, 131-4 passim, 142, 157, 162, 169, 170, 173 and nn. 5, 6, 188, 191, 193, 195 and n. 130, 253 n. 41 —,—, critics of 19-23, 131, 143, 150 —, —, popularity and influence of 19, 24, 27, 157, 173, 174, 188, 193, 299, 301, 305 Montagu, Edward, brother of William Montagu earl of Salisbury (1337-44) 269 n. 5, 270 and n. 3 —, wife of, see Alice Montagu, William, earl of Salisbury (1337-44) 268, 269 a n d n . 6, 271,272, 273,274,277,278 Montagu, William, earl of Salisbury (1349-97) 269, 270 and n. 3, 273 and n. 3 Montfort, Jeanne de, duchess of Britanny (d. 1384) 271 and n. 3, 273 and n. 3 Montfort, John de, see under Brittany, duchy of Montfort, Simon de, earl of Leicester
(1239-65) 221, 229, 255, 258, 260, 261, pi. 15 Moray, earl of, see Randolph, John Mordred 172n. 3 More, Sir Thomas de la, patron of Geoffrey le Baker, q.v., 216 Mortimer, Edmund, earl of March (1413-25) 296 Mortimer, Roger, earl of March (1328-30) 214,251,257,261,262 Mortimer, Roger, earl of March (1381-98) —, claim of, to throne 295-6 —, issue 296 Mortimer family, earls of March 216 —, genealogies 326 —, herald of, see Othelake, John Moses 138 —, as type 261 n. 95 murder, a 281, 282 and n. 4 'Muriel', wife of'earl of Lincoln', 309 and n. 69 Murimuth, Adam, prebendary of St Paul's
206, 208 n. 56, 247 n. 13, 248, 249 and n. 27, 257 and n. 81, 258 and n. 82, 261, 265, 270, 271-2 n.7, 273 n. 4, 274 Najera, Spain, battle of (1367) 286 Nedging, Suffolk 46
352
Legends, Traditions and History in Medieval England
natural history, interest in, 12th c. 196-7 and nn. Nennius 169, 177 and n. 6 Neot, St 53 and n. 99, 64 Nero, as type 254 Neville, Robert, bishop of Durham
(1438-57) 29
Newburgh, Yorks., Augustinian priory 20 —, canon of, see Newburgh, William of Newburgh, Waleran de, earl of Warwick (1184-1203 or 1204) 319 Newburgh, William de, earl of Warwick (1153-84) 319 Newburgh, William of, canon of Newburgh 20 —, Historia Rerum Anglicarum 20-4 passim, 130-3 passim, 136, 137, 143, 149-51 passim, 188 Newland, Forest of Dean, rector of, see London, John of, rector . . . Newmarket, Suffolk 243 newsletters 222 and n. 114 Nicholas, prior of Worcester (c. 1113-24) 15 and n. 88, 109, 112, 120, 121, 122 Norfolk, travels in, 15th c. 325 Norfolk, earl of, see Brotherton, Thomas of Norman laws, see laws, Anglo-Saxon and Norman Normandy, chronicles of 214-15 —, see also Chronique normande Northampton, tournament at 270 Northampton, earl of, see Bohun, William de 'Northo' PSuffolk 47 Northumbrian annals 4 n. 15 Northumbrian renaissance 2, 5, 7, 9, 13, 16-18, 26, 31, 33-5 passim, 52, 64, 74, 148 Northwold, Hugh de, abbot of Bury St Edmunds (1215-29), bishop of Ely (1229-56) 241 nn. 5, 6 —, account of election of, see Bury St Edmunds . . . abbey, chronicles, Electio Hugonis Norwich, city 103, 192 Norwich, see 88 —, bishops of, see: Bateman, William; Despenser, Henry; Losinga, Herbert Norwich, cathedral priory 89, 102 —, copy at, of Bury chronicle 205 and n. 32 —, —, of Flares Historiarum 204, 246 —, chronicle 205 and n. 34, 226 and n. 129, 228 —, chronicler, see Cotton, Bartholomew Nowton, Suffolk 46
numismatics 323 and n. 147 nunnery, Anglo-Saxon, at Ramsey 57-8 Nutley, Bucks., Augustinian abbey, abbot of, see Crendon, Richard of Oda, archbishop of Canterbury (942-58) 43, 49, 50, 69 Odulf, St 60, 99, 109 Offa, king of Mercia (757-96) 89 Olaf, St 87 official and quasi-official historiography 208 and n. 57, 212-15 and nn., 257-62 and n. 101, 263-5 passim, 274 and n. 6 Omer, St 186 n. 63 oral information 7-8, 65, 67 and n. 196, 91,
100, 102, 141, 142, 208, 275 n. 3, 282, 285, 286, 289 Onalafball, Viking 50 Ordinances of barons (1311) 256,257 Ordric, abbot of Abingdon (1052-66) 65 Orford, Suffolk, Augustinian friary 282, 285 Orgement, Pierre d', chancellor of Charles V 274 n. 6 origins, interest in 310-11, 321, 323. See also: foundation; founders Orm, vision of 8, n. 36 Orosius 19, 134, 135 Osbern, monk of Christ Church, Canterbury 78, 108, 121, 122, 155, 158, 161, 181, 182 Osney, Oxon., Augustinian priory 321 andn. 128 —, chronicle 206 and n. 46, 218, 220-1 and n. 106 Ossory, bishop of, see Bale, John Ostrevantz, William, count of, later William VI of Holland (d. 1417) 232 Oswald, St, kingofNorthumbria (633-41) 11,83,121, 164 Oswald, St, bishop of Worcester (961-92), archbishop of York (972-92) 4, 5, 32, 36, 37, 39, 43, 44, 48-55 passim, 59, 66-9 passim, 77-83 passim, 94, 96, 99, 104, 110, 113, 122, 154, 179 n. 20, 183 Oswald, monk of Ramsey 67, 76 Oswaldslaw 62 n. 154 Oswen, St Edmund's carer 84, 87 Oswy, king of Bernicia (641-70), king of Northumbria (654-70) 5, 37 Othelake, John, March herald 326 Ovid 192, 193, 276 and n. 7, 277 n. 1 'Oxenedes, John de,' chronicle, see under St BenetofHulme
Index Oxford, town 189, 192, 318 —, castle 188 Oxford, archdeacon of, see Map, Walter Oxford, Dominican friary 287 Oxford, Franciscan friary 287 Oxford, schools and university 189 n. 72, 214 and n. 76, 285-6 and n. 1 —, colleges and halls 317, 318 and n. 116 —, chroniclers etc. who had studied at 214,297,303, 317 and n. 115 Oxford, Provisions of (1258) 250, 255 palaeography, medieval study of 175, 184, 307, 308 Palgrave, Suffolk 47 and n. 64 PALGRAVE, FRANCIS 323 and n. 143 Paris, St Denis, Benedictine abbey 213, 274, 275 —, chronicles —, —, Chronographia Regum Francorum 267 and nn. 3, 4, 268-75 passim —, —, Grandes chroniques 265, 268 n. 4, 274 and n. 6 Paris, Matthew —, as artist 178 and n. 14, 245, 247 —, politics of 218 n. 94, 221-2, 254 —, relations of, with Henry III 222, 263 and n. 104, 264 —, as chronicler 23 and n. 131, 24 n. 133, 150, 209, 224, 225, 237 n. 191, 300 —, WORKS _? _? Chronica Majora 201, 204, 209, 241 and n. 6, 263 —, —, —, continuation of (to 1420), by Thomas Walsingham 206 and n. 51 —, —, Flores Historiarum (Creation - 1249) 204, 245, 263 —, —, —, copy of, made for Westminster abbey 204, 238 n. 197, 245 and n. 4 —, —, —, see also St Albans, chronicles written at, Flores Historiarum —, —, Gesta Abbatum Monasterii Sancti Albani 201 n. 12 —, —, Historia Anglorum 263-4 —, —, Liber Additamentorum 236 and n. 186 —, —, Life of Edward the Confessor 264 parliaments 285-6 and n. 1, 296, 312, 317 n. I l l 'parliament' (1327) 261 and n. 97, 262 and n. 100 Patrick, St, apostle of Ireland 154, 159, 161 n. 3, 162-5 passim, 173, 174, 306 Patrick, St, 'the younger' 159 patriotism 216-18
353
patrons and patronage 36-7, 46-7, 56-9, 60-1, 215, 216 and nn., 284 n. 11, 317-21 passim, 322-5 passim Paul of Monte Cassino 138 Paulinus, archbishop of York (626-33) 15, 160, 186 n. 63 peacock, motif 181 and n. 32 Peasants' Revolt (1381) 210 Pelagius 187 n. 63 Percy, Henry ('Hotspur') 323 Pershore, Worcs., Benedictine abbey, abbot of, see Foldbriht Persius 192 Peter, St 122 —, as type 251,257 Peter II, king of Castile (1350-68) 286 and n. 6 Peter Alphonso, Spanish Jew 111 Peterborough, Northants., Benedictine abbey —, in 10th, l l t h c c . 5,60,61,81 —, in 13thc. 97 —, CHRONICLES —, —, version of Anglo-Saxon Chronicle 62,72,202,218 —, —, chronicle, 12th c. 183 —, —, continuation of chronicle of John of Worcester (to 1295) 203, 220 n. 104, 234, 237, 238 —, abbot of, see Leofric Petrarch 276 Phagan, see Fagan Philip, St 160, 174 Philip IV, king of France (1285-1315) 281 Philip (de Valois) VI, king of France (1328-50) 269 n. 6,272, 273, 274 and n. 6, 275, 278 and n. 1, 283 and n. 2 Philip, duke of Burgundy etc., son ofjohn II of France 283-5 passim Philip, count of Flanders (1168-91)307 n. 56, 308, 309 Philipps, Nicholas, Franciscan friar 279 Phocas, Byzantine emperor (602-10) 40 pigeons 197 and n. 40, pi. 10 pilgrim trade 154 and n. 1, 163, 293 Pipewell, Northants., Cistercian abbey, chronicle 261 Pisa, Council of (1409) 294 and n. 27 Pius II, pope (1458-64), see Aeneas Sylvius Piccolomini plague 148 —, (1348-9) Black Death 283 —, (1360-1) 284 —,(1369) 286
354
Legends, Traditions and History in Medieval England
Plegmund, archbishop of Canterbury (890-923) 109 Plymouth, Devon, port 284 Poitiers, Vienne, battle of (1355) 283, 284 Poitiers, William of 127, 262 n. 101 Polychronicon, see Higden, Ranulph Pompeius, Trogus 139 prices —, water (1374) 287 —, wheat (1370) 286 and n. 7 printing, first, of chronicles 223-4 and n. 118 Processus Scotiae (1291 -2), 264 prologues 10, 20-3, 125-51, 211 and n. 64, 235 and n. 184 'propaganda' —, Anglo-Saxon 39 and n. 31, 41-2, 79 —, Anglo-Norman 9, 39 n. 31, 71 n. 213, 74, 78-9 —, Angevin 21-3 —, Plantagenet 264, 265 —, during Hundred Years' War 274-8 passim —,Lancastrian 214, 217, 219, 220 and nn. 104, 105,221-2,295-6 —, monastic 89,94, 155-7 passim, 172, 174 Provence 231,234 QUIRK, ROGER 304 and n. 32 Ramsbury, Wilts., see of 59 Ramsey, Hunts., Benedictine abbey 41, 42, 47-50 passim, 53, 57, 60-9 passim, 77, 81, 90, 98-101 passim, 104 —, monks of, commission Abbo of Fleury 50, 54, 83, 84 —, chronicle, 12th c. 42, 44, 56, 58, 66, 69, 70, 77, 183, 184 —, abbots of: yElfwine; Athelstan; Eadnoth; Losinga, Herbert; Walter; Wulfsige; Wythman Randolph, John, earl of Moray (1332-46) 269 n. 6 Ravenser Odd (or Ravenspur), formerly on coast of Holderness, Yorks. 315 and n. 97 Reading, Berks., Benedictine abbey 180 Reading, John of, monk of Westminster, chronicle, 14th c. 249 n. 27 Reading, Robert of, monk of Westminster, continuation of Flores Historiarum ascribed to 213 and n. 72, 214, 248 and n. 24, 249-50, 253-8 passim, 261-2
Regularis Concordia(c. 972) 4-5, 35 n. 15, 36-41 passim, 58, 60, 63 Reinfrid, monk of Evesham 7, 8 Reading, Berks., Benedictine abbey 180 relics, cult of, see saints, cults of Rheims, archbishop of, see Fulco rhetoric 129-31, 150, 249-57 passim —, see also Latin, florid Richard I, king of England (1189-99) 21-2 passim, 190, 191, 262 n. 101 Richard II, king of England (1377-99) 216-20 passim, 232, 287, 295, 296 Rievaulx, Yorks., Cistercian abbey 17 —, abbots of, see: Ailred; Ernald; Maurice Ripon, Yorks., Anglo-Saxon monastery 5, 43, 51, 177 and n. 8, 179 n. 20, 202 Robert II, king of Scotland (1371-90) 323 n. 147 Robert III, king of Scotland (1390-1406) 323 n. 147 Robert III, count of Artois (d. 1343) 271, 275 and n. 1 Robert, earl of Gloucester (1122-47) 132, 133,146, 170 Robert, abbot of Glastonbury (1173-80) 170 ROBINSON,J.ARMITAGE 33,44 Rochester, Kent, bishop of, see Gundulf Rochester, cathedral 180,192 Rochester, cathedral priory 69 n. 206, 180, 192, 204 and n. 26, 220 n. 104, 246 Roger, abbot of St Evroult (1091 -1123) 132 romance, some historiography influenced by 157, 165-74 passim, 199, 215 and n. 80 —, see also chivalry Rome 88, 173, 191, 193 Romeyn, John le, archbishop of York (1286-96) 317 Romulus and Remus 173 Ross, co. 324 Rossi, Leonardus, de Giffono, minister general of the Franciscans (1373-8) 287 and n. 9 rotae 184 and n. 53, 308 n. 62, 309, pi. 4 Rous, John, antiquary (d. 1491) 24, 223, 299,303,305,317-22 —, Historia Regum Angliae 320-2 —, 'Warwick Rolls' 318 and n. 119,319-20 Rudborne, Thomas, monk of St Swithun's, Winchester 24, 25, 303 and nn. 19, 20 —, history of St Swithun's by 302, 304-5, 306 Rugby, Warwicks. 282
Index Runcton, Norfolk 46 and n. 64 Ruthwell, Dumfreiss., Anglo-Saxon cross at 161 St Albans, Herts., Benedictine abbey 89, 97, 103 n. 2, 154, 178, 263 n. 104, 300, 315-16, 317 n. I l l , 321 and n. 128 —,'Book of Benefactors' 316 —, chronicles written at 242 n. 2, 254, 281 —, —, Flores Historiarum, continuation (1250-65) of Matthew Paris's 204, 218, 245 —, —, —, copy of, at Westminster 245 and n. 4. See also Westminster, Benedictine abbey, chronicles written at —, —, see also: Paris, Matthew; Walsingham, Thomas; Wendover, Roger of —, abbots of, see: Gorron, Geoffrey de; John —, almoner of, see Wintershull, William St Asaph, bishop of, see Bache, Alexander St Benet of Hulme (or St Benet of Holme), Norfolk, Benedictine abbey 57, 95-8 passim, 101-4 passim, 240 and n. 6, 243-4 —, chronicles written at —, —, 'John de Oxenedes' 96, 226, 228 and n. 142, 234-7 passim, 240, 242 andnn. 1-4, 243, pi. 26 —, see also under Bury St Edmunds . . . abbey, chronicles, Cronica Buriensis —, copy of Flores Historiarum at 204 —, abbot of, see Aylesham, John of St Carilef, William of, bishop of Durham (1081-96) 7, 8 and n. 38, 13 n. 74, 74 StCuthbert's,see 12, 13,28,37,74,81,317 —, clerical community attached to 7, 45, 47,49,50,81 —, see also: Chester-le-Street; Durham; Lindisfarne St David's, Pembrokes., bishopric 23 —, ?Worcester chronicle at 204, n. 24 St Ives, Hunts, (formerly Slepe) 53, 54, 99, 100, 101 St Neots, Hunts. 53, 64, 100 St Omer, Pas-de-Calais, St Bertin's, Benedictine abbey 156 Saint-Sauveur-le-Vicomte, Manche, chateaux and lords of, history of see Histoire du chateau et sires de St Stephen's, unidentified monastery 309 St Victor, Hugh of 130, 131, 139, 141, 150 saints, cults of 45-6, 49-50, 51-4 and nn., 60,81etseqq., 108-9, 113, 154-5, 161 and
355
n. 3, 162-3, 165, 179 and n. 20, 185 —, see also Edmund, St Salisbury —, countess of, see 'Alice'. See also Lincoln and Salisbury, countess of —, earls of, see: Montagu, William; Montagu, William Salisbury, John of, Historia Pontificalis 132, 135, 136, 139, 141, 150 Sallust 128, 176, 192 —, Bellum Catalinae 128, 133-43 passim Samson, abbot of Bury St Edmunds (11821211) 85, 178,241 n. 4 —, Life of, see Brakelond, Jocelin of Sanctilogium, see Tynemouth, John of Savaric, bishop of Bath (1192-1205) 155 Sanford, John de, archbishop of Dublin (1286-94) 233-4 Scarborough, Yorks., castle 188 Scollandus, see Scotland (Scollandus) Scotichronicon., see Bower, Walter Scotland 267,323 —, relations of, with Edward I, see Edward I, relations of, with —, succession question (1291-2), 219-20 and nn., 236, 243, 264 and nn. 107-8, 280 —, map of 324 —, kings —, —, list 280 —, —, regalia 259 —, —, see also: Alexander I; David II Scotland (Scollandus), abbot of St Augustine's (1070-87) 310 scripts, see: Anglo-Saxon; uncial seals, 184, 308, 309 and n. 69, 317, 320 and nn. 123, 124, pi. 36 Sebastian, St 86, 87 Seen, John, D.Th., monk of Glastonbury
297
Seinsheim, Heinrich von 320 n. 122 Semer, Suffolk 229 and n. 151 Segontium, see Cair Segeint Seneca 193 Sens, William of, master builder 185 Sergius II, pope (844-7) 62, 181 Seven Ages, the, periodization of 209 and n. 59 Sextus Tarquinius 277 Seymour, John, canon of St George's, Windsor 321 and n. 129 SHEERIN, DJ. 304-5 and n. 35 Sherborne, Dorset, bishops of, see: Hermann; Wulfsige Sherborne, cathedral priory 57, 59, 145
356
Legends, Traditions and History in Medieval England
Shillington, Beds. 67 Sicily, king of, see William II Siena, Council of (1424), 294 and nn. 25, 27 SIMON, GERTRUD 127 Siward, abbot of Abingdon (1030-44) 65 Skyrne, Adam de, abbot of Meaux (131039) 314 Slepe, see St Ives SMALLEY, BERYL 253 n. 40 Smithfield, London, fair 193 Soham, Cambs. 66 —, see also Monk Soham Soilli, Henry de, abbot of Glastonbury (1189-93) 165, 168,170 Solomon, as type 261 n. 95 SOUTHERN, R.W. 32, 78, 300 Southwark, Surrey, St Mary's, Augustinian priory, owned copy ofFlores Historiarum 204 Southwold, Suffolk 46 and n. 64 sows, legendary 172-3 Spain 193,272,273,286 —, see also Castille Spalding, Lines., Benedictine priory, chronicle 205 and n. 35 Spearhafoc, abbot of Abingdon (1047-51) 65 Spott, Wulfric, thegn 59 Sprott, Thomas, monk of St Augustine's, Canterbury, chronicle 26 n. 147 Stapleton, Thomas, Catholic polemicist 25 Stephen, king of England (1135-54) 119, 133, 183, 184, 262 n. 101 —, see also Gesta Stephani Stephen of Rouen, monk of Bee (d. c. 1170) 22 Stirling, castle, siege of (1304) 250-1, 254, 255,259-60 'Stone of Scone' 259, 323 n. 147 Stonehenge, Wilts. 195 and n. 130 STONES, E.L.G. 268 n. 1 Stonyhurst Gospels, see under Gospelbooks Stow, John (d. 1605) —, Chronicles of England 207 —, Survey of London 325 Stowe, William de, prior of Bury St Edmunds (d. 1333/4) 243 and n. 2 Strumpeshagh, John de, abbot of Langley (1335-61) 244 Sudeley and Boteler families, of Sudeley castle, Glos., armorial genealogy, and genealogical chronicle of, 15th c. 326-7
Sudre, William, cardinal-priest of St John and St Paul (1366-7), cardinal bishop of Ostia( 1367-73) 286 and n. 3 Suetonius 19, 113, 146, 175, 186, 187, 190 Suffolk, earls of, see: Ufford, Robert; Ufford, William Surrey, earl of, see Warenne, William de swan 196, 197 and n. 140, pi. 9 Swithun, St, bishop of Winchester (852 x 853 - 862 x 865) 306 n. 47 SYMONS, THOMAS 32, 35 Tabula Magna, of Glastonbury 293 tabulae 331 and n. 3, 332 Talleyrand, Elias, of Perigord, cardinal bishop of Albano (1348-64) 284 and n. 2 Tarquinius Superbus, king of Rome 277, 278 Taunton, John of, abbot of Glastonbury (1274-91) 171 Taunton, Richard of, monk of Glastonbury 164 taxation, clerical 259 n. 86, 312 and nn. 80, 81,83 Taxter,John 228 —, chronicle, see under Bury St Edmunds . . . abbey, chronicles TAYLOR, JOHN 201 n. 12, 216 n. 84, 290 Tewkesbury, Glos., Benedictine abbey 180 —, 'Book of Benefactors' of 326 note —, chronicles 205, 216 n. 83 —, abbot of, see Tewkesbury, Alan of Tewkesbury, Alan of, abbot of Tewkesbury (1186-1201/2) Life ofSt Thomas [Becket] 236 n. 186 Thames, river 193 Thanet, Isle of —, map 310-11 a n d n . 75 —, Anglo-Saxon minster on 177 n. 6, 310 —, —, abbess of, see Eadburga (or Bugga) Theodore, archbishop of Canterbury (668-90) 15,26,27 Theodred, bishop of London (909 x 926 951 x953) 46 andn. 63 Therfield, Herts. 67 Thetford, Norfolk, see of 88 thistle, milk 197 and n. 140, pi. 13 Thomas of Bayeux, archbishop of York (1070-1100) 113, 115, 121 187 n. 65 Thomas of Lancaster, earl of Lancaster, Leicester and Derby (1298-1322) 216 and n. 84, 256-7 —, widow of, see Alice
357
Index THOMPSON, SALLY 58 n. 131
Thorne, William, monk of St Augustine's, Canterbury, chronicle 26 n. 147 Thorney, Cambs., Benedictine abbey 51,
60,61 n. 147
'Thorpe', Suffolk or Norfolk 46, 47 n. 64 THORPE, BENJAMIN 241, 242 n. 2 Thunor, legendary wicked thegn 310-11 Thurstan, archbishop of York (1119-40) 13, 14 Thurstan, abbot of Glastonbury (c. 1077/8-1096+) 154 Tintern, Monmouths., Cistercian abbey, copy of Flores Historiarum at 204 and n. 28, 246 Tiptoft, John, earl of Worcester (1449-70) 318 Toe, Ralph, monk of Glastonbury 164 topography, observations of 175, 177 nn. 6-8, 180, 188-9, 190-4 passim, 314-8 passim, 325 topoi 35, 44, 76-8, 86-7, 95, 125-38 passim, 141 Torigni, Robert of, chronicler 138 tournaments 267,270-1,273,278 TOUT, T.F. 246 and n. 9, 248 and n. 24, 249, 250 traditionalism, Anglo-Saxon 33-52 passim, 54-7 travels 318 and n. 120,325 —, see also itineraries Trevet, Nicholas 178 and n. 19 —.Annales 247 n. 13 —, chronicle in Anglo-Norman 264 and n. 110 Trogus Pompeius 19 'Troy, New', see under London Trumpington, Sir Roger, his brass 320 n. 122 Tudor hisoriography 224 and n. 119, 326 Turgot, prior of Durham (1087-1109) 13 Tynemouth, Northumb. 103 n. 2 Tynemouth, John of, Sanctilogium 28 n. 158 Tyre, William of, Gesta Amalrici 127 Ufi, abbot of Bury St Edmunds (1020-44) 96, 102 Ufford, Elizabeth (d. 1362), wife of Thomas Ufford,q.v., 282, 284 and n. 12, 285 Ufford, Robert, earl of Suffolk (1337-69) 284 n. 9, 285 Ufford, Thomas (d. P1368), son of Robert first earl of Suffolk 284 n. 9
—, wife of, see Ufford, Elizabeth Ufford, William, earl of Suffolk (1369-82) 286 Uhtred, precentor of Worcester (d. 1132) 117, 119,238 Umfraville, Sir Robert (d. 1421) 323 uncial script, 15th c. 307 and n. 55, 308 n. 61 Urban V, pope (1362-70) 286 Usk, Adam of, chronicler 216 and n. 87, 295 n. 31 Usuard 86,87 Vannes, Morbihan 274 VAUGHN, SALLY 37, 76 n. 240
Vergil, Polydore 24 VESSEY, D.W.T. 126, 127 VIARDJULES 269 'Vigilantius', supposititious historian of Old Minster, Winchester 304 and n. 32, 305 Vikings 46, 49, 50, 74, 76, 78 and n. 253,
84/100, 101, 104, 110, 147, 148,210,212
n. 65, 306, 310, 318 Villehardouin 127 Virgil 173, 191-3 passim visual evidence, use of, by chroniclers etc. 128, 159-61, 167-8, 175-97 passim, 306-10 passim, 314, 318, 325 Vitalis, Orderic 9, 11, 75, 117, 120, 130, 137, 138, 141 Viterbo, papacy at 286
Walcher, bishop of Durham (1071 -80) 7, 8 a n d n . 37, 73, 109, 110, 119n. 78 Waleran, count of Meulan and earl of Worcester (d. 1166) 133 Wales 23,321 —, see also Welsh, the Wales, Gerald of 23, 24 n. 133, 156, 166-8 passim, 169n. 5, 170-2 passim, 176, 188-9 —, Descriptio Kambriae 194 and n. 120, 195, 196 —, Topographia Hibernica 194, 195 and n. 130, 196 and n. 135, 197 and n. 138, pis 5, 6, 8 Walkelin, bishop of Winchester (1070-98) 16 n. 89 Walkiniana, see Elizabeth Wallingford, John of, chronicle 241 and n. 2 Walsingham, Thomas, monk of St Alban's 214,221-2,225 —, WORKS
358
Legends, Traditions and History in Medieval England
—, —, continuation of Matthew Paris's Chronica Majora 206 and n. 51, 221, 224, 225 —, —, Gesta Abbatum Monasterii Sancti Albani 317 —,—, Ypodigma Neustriae 214-5 Walter, abbot of Evesham (1077-1104) 108, 109 Walter, abbot of Ramsey (1135-60) 68 Walter, archdeacon of Oxford (?1104-51), allegedly gave Geoffrey of Monmouth 'very old book' 173 n. 6 Walter, Hubert, archbishop of Canterbury (1193-1205) 24 WANLEY, HUMPHREY 310 WARDJ.O. 127 Warenne, Ralph de, brother of William III de, earl of Surrey 309 n. 69 Warenne, William III de, earl of Surrey 309 and n. 69 Warin, prior of Malmesbury (c. 1124c. 1143) 115 Wark(-on-Tweed), Northumb., castle 269 and nn. 4, 5, 272, 277 Wars of the Roses 223 Warwick, castle 319 Warwick, St Mary's church 319 Warwick, earls of —, legendary 319, pis 34, 35 —, patrons of John Rous 317-18 —, see also: Beauchamp, Richard; Beauchamp, Thomas; Beauchamp, Thomas; Newburgh, Waleran de; Newburgh, William de Warwickshire, enclosure movement in 322 and n. 136 Watelege, John de, monk of Glastonbury 164 WATT, D.E.R. 225, 295 Waverley, Surrey, Cistercian abbey, chronicle 206 and n. 45, 220 n. 104, 225 and n. 122 Wearmouth, Durham, Anglo-Saxon monastery 7, 13, 35, 62, 73, 109 weather —, storms in Norfolk (1362, 1364) 281, 284, 285 —, great frosts (1363, 1374) 285, 287 Wells, Somerset, see of 59 Welsh, the 21-2, 194-5,294 Wendover, Roger of, monk of St Alban's, chronicle 89 Waerferth, bishop of Worcester (869 x 872
-907x915) 109, 110
Werwulf, one of King Alfred's helpers 110 Wessington, John, prior of Durham (1416-46) 27 and n. 158, 28 and n. 160, 292 and n. 9, 295 and n. 29, 303, 315 n. 99, 316-17 and n. I l l —, history of St Cuthbert's by 302 and n. 15,306 West Saxons, kings of, see: Alfred; Edmund I; Edward the Elder Westbury-on-Trym, Glos., Benedictine priory 32, 113 Westminster, alleged council at (c. 1071) 110 Westminster, palace 259 Westminster, Benedictine abbey 61, 82, 97, 154, 177, 257 and n. 81, 258 n. 82, 262, 306 and n. 48 —, royal connections of 58, 92, 213, 247-8, 251 and n. 35, 252-65 passim —, royal treasury in 251 and n. 35, 252, 254, 258 —, see also 'Borgoyne' —, CHRONICLES; 82 —, —, continuation of the St Albans' Flores Historiarum (1265-1307) —, —, —, ('Westminster' version) 204 and nn. 26, 28, 246 and nn. 12, 13, 247-8, 249 n. 27, 250, 253, 254, 259, 262, 265 —, —, —, —, dissemination of 204 and nn. 26, 28, 246 —, —, —, 'Merton' version 204, 213 and n. 72, 220 n. 104, 246 and nn. 12, 13, 247-8 and nn. 14, 16, 17, 250-65 passim —, —, —, —, an 'official' history 259-65 passim —, —, (1307-27), see Reading, Robert of —, —, continuation of Polychronicon (to 1394) 216,225,226,227,232-3 —, —, —, also Flete, John —, abbot of, see Laurence, Master —, monks of, see: Bever, John; Clare, Osbert de; London, John of; Reading, John of; Reading, Robert of —, prior of, see Flete, John Weyland, Thomas, Chief Justice (1278-89) 280 and n. 2 Whepstead, Suffolk 46 Whethamsted, John, abbot of St Albans (1420-40, 1452-65) 24 Whitby, Yorks., Anglo-Saxon monastery 7 —, Life ofSt Gregory written at 55 Whitby, Synod of (664) 5, 37 Whithorn, Galloway 179 Whitland, Carmarthens., Cistercian
Index abbey, chronicle 205 and n. 38 Wick, Holderness, Yorks. 314 and n. 96 Wigstan (or Wistan), St 99 Wilfrid, St, archbishop of York (664-78) 11, 27,51, 142, 160 n. 4, 177 William the Conqueror, king of England (1066-87) 9, 37, 70, 71, 72, 74, 76,88,91, 110, 117, 143, 187 n. 63, 221 William Rufus, king of England (10871100) 8n. 38,9, 13, 110, 117, 187,221 William II, king of Sicily (1165-89) 184, 189 n. 83, 308 n. 61 William le Gros, count of Aumale (1138-79) 314 and n. 90 William II, count of Hainault, Holland and Zealand (1337-45) 270 Willibrord, missionary 186 n. 63 WILLIS, ROBERT 304 and n. 33 Wilton, Wilts., Benedictine nunnery 58 Winchcombe, Glos., Benedictine abbey 99, 104, 109, 117 —, Annals 226, 227, 233 —, prior of, see Aldwin Winchester, city 192 Winchester, Council of (c. 972) 4, 32, 45, 60 Winchester, Council of (1072) 14 Winchester, royal palace 36 Winchester, bishops of, see: ^Elfheah; Blois, Henry of; Ethelwold; Swithun; Walkelin Winchester, New Minster (later Hyde, q.v.) —, community of secular clerks, 39-40 —, Benedictine abbey 39, 42, 44, 57, 76,91 n. 4, 145 Winchester, Hyde, Benedictine abbey 321 andn. 128 —, 'Book of Hyde' 292 and n. 13, 302-3 andn. 19,305 Winchester, Old Minster (later St Swithun's, q.v.), cathedral priory 39, 42, 43, 52, 54, 57, 62, 76, 78, 81, 94, 145 —, monks of, see: Eadsige; Lantfred; Wulfsige; Wulfstan Winchester, St Swithun's, cathedral priory 82, 306 and n. 47, 321 and n. 128, 305 —, chronicle, 13th c. 205 and nn. 31, 43, 208 andn. 55, 211 and n. 64 —, histories of, see: Rudborne, Thomas; 'Vigilantius' Windsor, Berks., royal palace 271, 273 Windsor, St George's collegiate chapel 321 andn. 129 Wintershull, William, almoner of
359
StAlbans 316 Wistan, St, see Wigstan, St wolves, in legends 84, 87, 173 Wood, C.T. 297-8 Worcester, city 192 Worcester, earl of, see Waleran Worcester, bishops of, see: Oswald; Waerferth; Wulfstan I; Wulfstan II Worcester, cathedral priory 154, 158 —, in Anglo-Saxon period 42, 44 and n. 54, 48, 62, 78, 81, 94, 100, 104, 109-10, 154 —, in Anglo-Norman period, 108-9, 110-12, 113, 120-3 —, biographical tradition at 113-15 —, as centre of historiography 72, 115-20, 202 andn. 15,203-4 —, CHRONICLES —, —, Chronicon ex Chronicis, to 1118 ascribed to 'Florence of Worcester' (q.v.), and continued to 1140 x 1143 by John of Worcester (q.v.), 9, 10, 11, 42, 44, 72, 73, 76, 77, 91, 94, 96, 99, 104, 116-21 passim, 203 and n. 21, 226, 227, 238 —, —, —, dissemination of, 118-19, 120, 203-4 and n. 24, 205 and n. 34, 240, 241 —, —, chronicle, 13th c. 205 and n. 44, 208 and n. 55, 211 and n. 64, 220 n. 104, 225 and n. 122, 235 and n. 184 —, priors of, see: Oswald, St; Wynsige —, precentor of, see Uhtred —, monk of, see Ethelred 'Worcester, Florence of 116, 203 and n. 21 See also under Worcester, cathedral priory, chronicles, Chronicon ex Chronicis Worcester, John of, 117-18, 119 n. 78 —, see also under Worcester, cathedral priory, chronicles, Chronicon ex Chronicis Worcester (or Worcestre), William, 'gentleman bureaucrat' and antiquary (d. 1482), 98 n. 3, 299, 317, 324 and n. 152,325-6 —, Boke of Noblesse 324 and n. 153, 325 Worcestre, William, see Worcester, William WORMALD, FRANCIS 32 WORMALD, PATRICK 31 note, 35 n. 15, 61 n. 150 Wulfsige, bishop of Sherborne (P993-1001/2) 77, 145 Wulfsige, abbot of Ramsey (1006-16) 53 Wulfsige, clerk, then monk of Old Minster, Winchester 43 Wulfstan I, bishop of Worcester (1002-16), archbishop of York (1002-23) 57 n. 120 Wulfstan II, St, bishop of Worcester
360
Legends, Traditions and History in Medieval England
(1062-95), 6, 70, 72, 108-17 passim, 120, 121, 123, 154, 180, 187 Wulfstan, monk of Old Minster, Winchester, biographer of St Ethelwold 35, 39 and n. 33, 43, 44, 54, 75, 76, 145 Wulfstan, thegn 47 Wymondham, Norfolk, Benedictine priory 89 Wynsige, prior of Worcester (?969-c.985) 44 and n. 53, 77 and n. 242, 99 Wytham, abbot of Ramsey (1016-20) 66, 67 Wurtzburg, cathedral 320 n. 122 Wyche, William, monk of Glastonbury, continuation by, of John of Glastonbury's Cronica 289 YAPP, BRUNSDON 197 n. 141
Yarmouth, Norfolk 233 York, city 192 —, royal Exchequer and King's Bench moved to (1319) 259 York, earliest church 160 n.4 York, Minster 202 and n. 15 —, chapter of 317 York, metropolitan see 38 —, dispute with Canterbury over primacy 13, 14, 15, 122, 236 n. 186 —, primacy over Scotland 109,122 —, and bishopric of Worcester 110, 115 —, archbishops of 12, 13. See also: Oswald, St, bishop . . .; Paulinus; Romeyn, John le; Thomas of Bayeux; Thurstan York, Augustinian friary, copy of John Taxter's chronicle at 205 and n. 39
Index of Manuscripts Bury St Edmunds Moyses Hall Museum MS
228-32 passim, 234, 241 n. 7, pi. 17
Cambridge Corpus Christi College 178 2 304 n. 32 110 230, 231 andn. 159, 171 232 183 45 n 57 241 iJQ9 j£ 32, pi. 22 197A 325, pi. 42 210 3QJ 15,pl.1 *jji Trinity College R.5.16 R.5.33 R.17.1 Trinity Hall 1
University Library Ff.2.33 Ii.4.26
289-91,295 158 n. 5, 166 n.4 pi. 7 304 andn. 26, 307, 309 and nn. 66, 67, 310andn. 70,311 and n. 75, pis 30-3 46n.62 197, pi. 11
Durham Cathedral Library BII35 Bill 30 Eton College 123
London British Library Additional 47214 48976 Cotton Claudius D IV Cleopatra A XVI DomitianA VIII Faustina B I GalbaEII Julius A I DII
D VII NeroD I D II Vespasian E IV
7n.35 315n.99 246 andn. 12,247 andn. 14, 248 and n. 20, pis 28, 29
279 andn. 2, 280-7 318andn. 119, pis 34, 35, 37-41
28 n. 158 249 andn. 27 202 and n. 20 233, pi. 24 96 andn. 2, 240 and n 6j 11. 227, 228-9, 233, pis 15,23 281 and n.4 241 n. 2 178 n. 12 98 n. 3, pi. 26 281 andn. 6
361
Index VitelliusAII CVI EX VII Egerton 1141 3142 Hdrley 200 641 743 1005
2901 Royal 13 B VI I I Stowe 944 College of Arms Arundel 6 30
Muniment Room 18/19 Rons Roll Lambeth Palace 12
308 n. 62 313n.86 pi. 4
312n.83 281 and n. 5 297 note 253 and n. 41 240 97 nn. 3,4, 102 and n.3, 177 n. 16,240 and n. 6 280 n. 5 pis 6, 8 57 n. 126
227,229-30,233 227-34 passim, 237, 238, 241 and n. 7, 242 and nn. 3-5, 243 and n. l,pls 14,16,20,21,25 327 318n.ll9 28 n.158
Manchester Chetham Library £772
245 and nn. 3,4, 247, 249
New Haven, U.S.A. Yale University Library 426 ('Tenison MS', 246 n. 13 formerly Phillipps 15732) New York, U.S.A. New York Public Library Spenser 193 326 Oxford Bodleian Library Ashmolel511 197, pi. 9 BodleylSO 197, pis 12, 13 240 89 n. 2, 95-6 passim 297 240 Lat Th. D. I 279 and n. 1 Laud Misc. 247 197,pi. 10 Top Glouc. d.2 326 Corpus Christi College 157 120, pis 2, 3 Magdalen College 172 180 and n. 27 Princeton Princeton University Library Garrettl53 289

T H E R O M A N FA M I LY I N T H E EMPIRE
This page intentionally left blank
The Roman Family in the Empire Rome, Italy, and Beyond
Edited by MICHELE GEORGE
1
1
Great Clarendon Street, Oxford   Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan South Korea Poland Portugal Singapore Switzerland Thailand Turkey Ukraine Vietnam Published in the United States by Oxford University Press Inc., New York © Oxford University Press  The moral rights of the author have been asserted Database right Oxford University Press (maker) First published  All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose this same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available ISBN –––X           Typeset by Hope Services (Abingdon) Ltd. Printed in Great Britain on acid-free paper by Biddles Ltd., King’s Lynn, Norfolk
This volume is dedicated to Beryl Rawson
This page intentionally left blank
P R E FAC E A N D ACKNOWLED GEMENTS
M papers in this volume were given at the Fourth E. T. Salmon conference in Roman Studies held at McMaster University in Hamilton, Canada, in September . Exceptionally, Antti Arjava’s paper, ‘The Roman family in the Greek east’, which he delivered at the conference, will appear in a forthcoming volume, and Mary T. Boatwright’s paper, included here, was a welcome later addition. The conference was made possible by generous financial support from the E. T. Salmon Fund for Roman Studies of McMaster University and the Social Sciences and Humanities Research Council of Canada, to which I am greatly indebted. Many hands went into the organization of the conference and the making of the volume, but special thanks must go to Carmen Camilleri and to Marina Salmon, whose unfailing interest and warm encouragement were constant positive forces. Alexa Holbrook valiantly assumed responsibility for the preparation of the final text, with editorial assistance from Oxford University Press. From their inception, the shape and scope of conference and book benefited from the judicious guidance of Keith Bradley, to whom I extend my profound gratitude. Finally, and above all, the conference owes its genesis to Beryl Rawson, whose contribution to the study of the Roman family cannot be underestimated; this volume is dedicated to her. M.G.
This page intentionally left blank
CONTENTS
List of Contributors List of Figures Map List of Tables Abbreviations Introduction M G . Putting the Family Across: Cicero on Natural Affection S T
xi xiii xv xvi xvii  
. Family Imagery and Family Values in Roman Italy M G

. The Roman Child in Sickness and in Health K B

. Parent–Child Conflict in the Roman Family: The Evidence of the Code of Justinian J E G . Searching for the Romano-Egyptian Family R A
 
. The Jewish Family in Judaea from Pompey to Hadrian—the Limits of Romanization M W

. Family Relations in Roman Lusitania: Social Change in a Roman Province? J E

. Family History in the Roman North-West G W


Contents
. Family and Kinship in Roman Africa M C

. Children and Parents on the Tombstones of Pannonia M T. B

References Index
 
LIST OF CONTRIBU TORS
R A is Professor of Roman History in the Department of Classics, Royal Holloway, University of London. M T. B is the Professor of Ancient History in the Department of Classical Studies at Duke University. K B is the Eli J. Shaheen Professor of Classics, Concurrent Professor of History, and Chair of the Department of Classics at the University of Notre Dame. M C is Director of Research at CNRS and Director of L’Année épigraphique. J E is Associate Professor of History and Classical Studies, York University, Canada. M G is Associate Professor in the Department of Classics, McMaster University. J E G is Professor of Classics at Washington University. S T is the Anne T. & Robert M. Bass Professor Emeritus at Stanford University, and Member of the SubFaculty of Ancient History, University of Oxford. M W is Associate Lecturer at the Open University, and an Honorary Fellow in the Department of New Testament Language, Literature and Theology at the University of Edinburgh. G W is Professor of Ancient History at the University of St Andrews.
This page intentionally left blank
LIST OF FIGURES
.. Funerary relief of the Gratidii (Vatican, Museum Pio Clementino, Sala dei Busti , Inv. ) (DAI .) .. Funerary relief of the Maelii (Raleigh, NC, North Carolina Museum of Art, Inv. ..) .. Funerary relief of the Vettii, Rome, Via Pio a (DAI .) .. Funerary relief of the Furii (Vatican, Museo Gregoriano Profano, Inv.  ) .. Funerary relief (Rome, Palazzo dei Conservatori, Museo Nuovo, Sala VI, , Inv. ) (DAI .) .. Funerary relief of the Alennii (Bologna, Museo Civico Archeologico, Lapidario ) (DAI .) .. Funerary relief of the Firmii (Ravenna, Museo Nazionale ) (DAI .) .. Funerary relief of the Montani (Ravenna, Museo Nazionale ) (DAI .) .. The family of Thaubarion I .. The family of Thaubarion II .. Family of Thaubarion and Didyme .. The birth family of Tryphon of Dionysios .. The family of Tryphon in the s and s  .. The active family of Kronion .. The probable family of Soter and Kleopatra alias Kandake .. The family of Makarios: P. Kell. Copt.  .. Funerary monument with portraits of Asellia Hygia and M. Publicius Felix, from Emerita (Museo Nacional de Arte Romano, Mérida). Photo: M. de la Barrera. Reproduced by kind permission of the Museo Nacional de Arte Romano .. Funerary monument of Lubaecus Antae lib., Binarea Triti f., Boutius Lubaeci f., and Cilia Caenonis f., set up by Claudia Tangina, from the civitas capital of the Igaeditani (Museu de Idanha-a-Velha). Photo: D. Ferreira. Reproduced by kind permission of the Instituto de Arqueologia, University of Coimbra

List of Figures
.. Types of funerary commemoration within the nuclear family in Lusitania. .. Funerary stela of Cloutia Ambini f., from Salmantica (Museo de Salamanca). Photo: J. Edmondson. Reproduced by kind permission of the Museo de Salamanca .. Stela with married couple, Mactar (Tunisia). Photo Y. Thébert. .. Stela with banquet scene, region of Bou Arada (Bardo Museum, Tunis) .. Cippus with couple and daughter, provenance unknown (Bardo Museum, Tunis) .. Stela of C. Sulpicius Primus and Laetoria Rufina, Sidi Mohammed Lazrag (Museum, Chemtou) .. Stela of Sulpicius Faustus and Sempronia Urbica, Sidi Mohammed Lazrag (Museum, Chemtou) .. Stela of Sulpicius Primus and [Sulpicia] Faustina, Sidi Mohammed Lazrag (Museum, Chemtou) .. Stela of Aelius Munatius, Aurelia Cansauna, and family, Intercisa (Hungarian National Museum, Budapest, RIU ) .. Stela of Comminia Valagenta and family, Savaria (Hungarian National Museum, Budapest, RIU ) .. Stela of Demiuncus and family, Intercisa (Hungarian National Museum, Budapest, RIU ) .. Stela of Otiouna, Intercisa (Hungarian National Museum, Budapest, RIU ) .. Stela of the Flavii, Ulcisia Castra (Aquincum Museum, Budapest, RIU ) .. Stela of Suriacus Secuindinus and family, Brigetio (Hungarian National Museum, Budapest, RIU ) .. Stela of M. Attius Rufus and family, Ulcisia Castra (Balassa Bálint Museum, Esztergom, RIU ) .. Stela of Germanius Valens and family, Intercisa (Ferenczy Museum, Szentendre, RIU ) .. Stela of Claudia Julia, Scarbantia (City Museum, Wiener Neustadt, CIL . )
MAP
.. Map of the Roman Province of Lusitania (drawn by Donna Gillespie)
L I S T O F TA B L E S
.. Families and households in cities and villages .. Funerary commemoration in various regions of Lusitania .. Comparison of commemorative practices in Lusitania and other parts of the western empire .. Gender balance between commemorators .. Gender balance between those commemorated .. Children commemorated by parents according to age cohorts .. Gender balance in the commemoration of children at various stages of the life cycle at Emerita, in the conventus Pacensis, and at Olisipo .. Balance of commemorator gender and age spread in conjugal commemoration .. Age at death for men and women: modern province of Salamanca .. Funerary commemorations from Emerita and the Civitas Igaeditanorum according to the method of D. B. Martin .. Joint burials in various regions of Lusitania by type .. Joint acts of commemoration in various regions of Lusitania
A B B R E V I AT I O N S
AE BGU
L’Année épigraphique Aegyptische Urkunden aus den Königlichen (later Staatlichen) Museen zu Berlin, Griechische Urkunden ChLA Chartae Latinae Antiquiores CIJ Corpus Inscriptionum Iudaicarum CIL Corpus Inscriptionum Latinarum CLE Carmina Latina Epigraphica CMBad. Catálogo monumental de España: Provincia de Badajoz, J. R. Mélida (Madrid, –) Cod. Just. Codex Justinianus, ed. P. Krueger (; repr. Frankfurt, ) Cod. Theod. Codex Theodosianus, ed. T. Mommsen and P. M. Meyer (, repr. Berlin, ) CPILC Corpus provincial de inscripciones latinas: Cáceres, R. Hurtado de San Antonio (Cáceres, ) CPJ Corpus Papyrorum Judaicarum, V. A. Tcherikover, A. Fuks, and M. Stern,  vols. (Cambridge, Mass., –) CPR Corpus Papyrorum Raineri CRINT The Jewish People in the First Century, Compendia rerum Iudaicarum ad Novum Testamentum, ed. S. Safrai and M. Stern ( and ),  vols. (Assen) CSIR Österreich . Die Reliefs der Stadtgebiete von Scarbantia und Savaria, M. L. Krüger (Vienna, ) D. Digesta Iustiniani, ed. T. Mommsen (), in The Digest of Justinian, ed. A. Watson,  vols. (Philadelphia, ) DJD Discoveries in the Judaean Desert EE Ephemeris Epigraphica
xviii ERAE
FE FIRA Frag. Vat. HAE HEp. ILAfr. ILAlg. ILER ILS ILTun. IRCP IRT LICS LTUR O. Florida P. Amh. P. Berl. Bork. P. Brem. P. Cair. Isid.
P. Col.
Abbreviations Epigrafía romana de Augusta Emerita, L. García Iglesias, unpublished doctoral dissertation, Universidad Complutense (Madrid, –) Ficheiro Epigráfico (supplement to the journal Conimbriga) Fontes Iuris Romani Antejustiniani Fragmenta quae dicuntur Vaticana (FIRA), ed. J. Baviera, vol. : Auctores (nd edn., Florence, ) Hispania Antiqua Epigraphica Hispania Epigraphica Inscriptions Latines d’Afrique Inscriptions Latines de l’Algérie, ed. S. Gsell (repr. Paris, ) Inscripciones latinas de la España romana, J. Vives (Barcelona, ) Inscriptiones Latinae Selectae, H. Dessau (Berlin, –) Inscriptions Latines de la Tunisie, ed. A. Merlin (Paris, ) Inscrições romanas do Conventus Pacensis, J. d’Encarnação (Coimbra, ) Inscriptions of Roman Tripolitania Latin Inscriptions from Central Spain, R. C. Knapp (Berkeley, Los Angeles, and Oxford, ) Lexicon topographicum urbis romae The Florida Ostraka: Documents from the Roman Army in Upper Egypt, ed. R. S. Bagnall (Durham, NC, ) The Amherst Papyri Une description topographique des immeubles à Panopolis, ed. Z. Borkowski (Warsaw, ) Die Bremer Papyri The Archive of Aurelius Isidorus in the Egyptian Museum, Cairo, and the University of Michigan, ed. A. E. R. Boak and H. C. Youtie (Ann Arbor, ) Columbia Papyri
Abbreviations P. Coll. Youtie P. Fam. Tebt. P. Fay. P. Graux P. Iand. P. Köln P. Kell. Copt. P. Lips. P. Lond. P. Mert. P. Meyer
P. Mich. P. Michael.
P. Mil. Vogl. P. Oslo. P. Oxy. P. Princ. P. Ryl. P. Sakaon PSI P. Stras.
xix
Collectanea Papyrologica: Texts Published in Honor of H. C. Youtie A Family Archive from Tebtunis, ed. B. A. van Groningen (Leiden, ) Fayum Towns and their Papyri, ed. B. P. Grenfell, A. S. Hunt and D. G. Hogarth (London, ) Papyrus Graux, ed. H. Henne in BIFAO  (),  ff. Papyri Iandanae Kölner Papyri Coptic Documentary Texts from Kellis, ed. I. Gardner, A. Alcock, and W. P. Funk (Dakhleh Oasis Project vol. ) (Oxford, ) Griechische Urkunden der Papyrussammlung zu Leipzig Greek Papyri in the British Museum A Descriptive Catalogue of the Greek Papyri in the Collection of Wilfred Merton Griechische Texte aus Aegypten, : Papyri des Neutestamentlichen Seminars der Universität Berlin; : Ostraka der Sammlung Deissmann, ed. P. M. Meyer (Berlin, ) Michigan Papyri Papyri Michaelidae, being a Catalogue of Greek and Latin Papyri, Tablets and Ostraca in the Library of Mr G. A. Michailidis of Cairo, ed. D. S. Crawford (Aberdeen, ) Papiri della R. Università di Milano, ed. A. Vogliano (Milan, ) Papyri Osloenses The Oxyrhynchus Papyri Papyri in the Princeton University Collections Catalogue of the Greek and Latin Papyri in the John Rylands Library, Manchester The Archive of Aurelius Sakaon, ed. G. M. Parássoglou (Bonn, ) Papiri greci e latini Griechische Papyrus der Kaiserlichen Universitäts- und Landes-bibliothek zu Strassburg
xx P. Tebt. P. Vind. Bosw. P. Wash. Univ. P. Wisc. P. Würzb. RIU SB Sent. Pauli UPZ ZPE
Abbreviations The Tebtunis Papyri Einige Wiener Papyri, ed. E. Boswinkel (Leiden, ) Washington University Papyri The Wisconsin Papyri Mitteilungen aus der Würzburger Papyrussammlung, ed. U. Wilcken (Berlin, ) Die römischen Inschriften Ungarns Sammelbuch griechischer Urkunden aus Aegypten Sententiarum Receptarum libri quinque qui vulgo Iulio Paulo adhuc Tribuuntur, text in FIRA , at pp. – Urkunden der Ptolemäerzeit (ältere Funde), ed. U. Wilcken Zeitschrift für Papyrologie und Epigraphik
Introduction Michele George
T volume follows from the three Australian conferences begun by Beryl Rawson in , a time when, quite coincidentally, a number of scholars in different parts of the mainly AngloAmerican world were beginning to focus on the topic of the Roman family as a distinct theme in ancient social history. Research on family studies in the ancient world was in part the result of a growth of interest in the s and s in the history of the family among historians at large (e.g. Laslett, MacFarlane, Stone), and, for some, the growth of a very ‘scientific’ approach to demographic studies, represented especially by the Cambridge Group for the History of Population and Social Structure.1 From these new historical methodologies emerged Keith Hopkins’s pioneering application of demography to the ancient context, which opened up a new form of analysis that continues to thrive while remaining highly controversial.2 Rawson, who herself had had interests in the Roman family since the time of her graduate studies, had the acumen to recognize this nascent movement in Roman history and to gather these scholars together to produce the conferences in Australia.3 The subject has matured since the s and each of the three volumes shows a growing refinement of approach and a better set of results. One of the problems that was faced by these scholars but which has never been satisfactorily resolved is that of how the 1 Family history: e.g. Laslett , , ; Stone , ; Laslett and Wall ; MacFarlane, Harrison, and Jardine ; Cambridge Group for the History of Population and Social Structure: Wachter, Hammel, and Laslett ; Wall, Robin, and Laslett . 2 Hopkins  and ; demography: e.g. Shaw ; Scheidel , a, b; Sallares , . 3 Rawson . The first three conferences on the Roman Family, held at the Australian National University in Canberra, resulted in three volumes on the subject: Rawson a, a, Rawson and Weaver .

Michele George
Roman family is to be defined. The need for definition seems essential but the problem is more complicated than appears at first sight. In modern western societies ‘family’ is an ambiguous term, but it connotes something that most people most of the time can understand. In , using funerary inscriptions from the western empire, Saller and Shaw argued that the Roman family was essentially nuclear, not extended, a claim that has dominated the discussion ever since. There are, however, problems with this view. First, the Romans themselves had no term for ‘family’ in any modern sense, and there are many contexts in which they show interest in non-nuclear or partial-nuclear family members in ways that are distinct from the modern cell-like concept of the nuclear family. Secondly, the demography of Roman Egypt shows that many households in that province had non-nuclear family members and offers a far different model from the paradigm traced by Saller and Shaw of what was normative in Roman society at large. For example, although it is only one region of the Roman empire, the existence of sibling marriage in the Fayum raises questions about household composition and domestic organization at large, and about how to identify and account for discrepancies from the Italian norm. The insistence of some scholars on regional demographic variability and scepticism about applying model life tables to ancient populations underlines the difficulties of the demographic approach. Valid as they may be, however, such debates do not affect the legitimate questions about household composition that are related to the issue of family definition. Thirdly, the term familia to the Romans meant ‘household’ rather than ‘family’ in any biological sense, and it is in terms of household organization and structure that family relations are therefore best understood.4 A significant complicating factor that modern historians of the family (in Europe at least) did not have to face was the presence of slavery. When, as in Rome, slaves were responsible for most childcare, at least in elite circles, to think in terms of nuclearity without any nuance or allowance for cultural specificity is an oversimplification. Because Romans had no concept of the 4 Saller and Shaw ; Roman Egypt: Bagnall and Frier ; criticisms: Scheidel b; Sallares , . Variability in household organization in North Africa: Bradley b.
Introduction

family as a collective unit, referring instead to their wives and children specifically, or just to ‘their own’, it is questionable whether a definition of the Roman family is possible without injecting anachronistic and therefore culturally irrelevant elements. A challenge to Saller and Shaw was raised by an examination of funerary commemorations in Asia Minor which identified an emphasis on the ‘extensive’, rather than the nuclear, family, proving that epigraphic evidence can be used in different ways to paint a variety of pictures and highlighting the problems inherent in seeking a specific family type that dominated the Roman Mediterranean. In turn, however, the conclusions of this study have been criticized for not fully acknowledging the regional diversity that emerged in its results.5 Definition, however, is not the only issue which has engaged historians of the family. The last generation of scholarship has focused on the component elements of family life that are regarded now and were regarded then as important: paternal power, family law, marriage patterns (including divorce and remarriage), the history of children and childbearing, the life course, old age, relations between family members (spouses, parents, and children), as well as between kin and non-kin members of the household.6 Instead of trying to contain the Roman family within a particular set of parameters, such studies have adumbrated the diverse elements of domestic life and considered their interplay. Rather than resolving the question of structure, they have complicated it by enhancing our understanding of the many dimensions of experience which fall within the category of ‘family life’, but for which the issue of structure has only minor relevance. This fourth volume builds logically on its predecessors and on the scholarship on the family that has appeared in the intervening years. It has a twofold approach. A number of the studies complement the emphasis of the earlier conference volumes on Study of Asia Minor: Martin ; corrective to Martin: Rawson b. To cite a few key contributions: Wiedemann , Bradley , Kertzer and Saller , Treggiari , Dixon , Saller , Evans Grubbs , Arjava , Parkin , Rawson . For ongoing research into the family, though without an ancient component, see the series edited by Kertzer and Barbagli . 5 6

Michele George
Rome while also looking to a wider orbit. Using evidence from Italy, and most often Rome, the first three articles (Treggiari, George, Bradley) explore notions surrounding the family in the abstract and in reality. Treggiari examines the way the idea of ‘family’ was used in the forensic works of Cicero as a touchstone for elite morality, especially for men, and how the social family norms of pietas and affection informed the identity of the Roman nobility. George’s discussion of family portrait groups on Republican and early imperial funerary commemoration takes up the same set of attitudes toward family life and shows how the emerging urban middle class of Italy, former slaves in Rome and citizens of mixed origins in Cisalpine Gaul, used family imagery to position themselves in the mainstream culture. Bradley, by contrast, investigates the harder side of ancient family life in his survey of diseases and treatments of illnesses, thus retrieving a sobering dimension of ancient experience which is radically different from the modern. The remaining chapters begin the examination of family life in the Roman world outside Italy in a systematic way focusing on specific regions. These studies revisit the issue of family structure, both directly and obliquely, as they tackle the enormous problem of how the Roman family, or forms of the Roman family, may have revealed themselves in Rome’s provinces in the imperial age. Using a lesser known source, rescripts mostly from the east which are preserved in the Justinianic Codex, Evans Grubbs examines the domestic tensions that arose between parents and adult children, and shows how certain social values such as marital happiness were prized above patria potestas, the central doctrine of Roman family law. In raising the question of precisely how ‘Roman’ these third-century families from the Greek east were, this study introduces the problem of Roman identity, the issue at the heart of the remaining chapters. Williams and Alston examine family life in the eastern Roman empire and the different points of access, in the form of various kinds of evidence, which might allow us to discern its shape and track its development. Williams looks at the impact of Roman political rule on Jewish family life in the early imperial period, charting the relatively minor changes in circumcision, onomastics, and burial practices and drawing attention to the crucial role of Hellenization. Adopting methodologies from anthropological theory, Alston
Introduction

considers family structure in Roman Egypt, the role of local practices such as sibling marriage, and variations in patterns between urban and rural families. Pursuing similar questions, but in quite different cultural contexts, the four final studies consider regions within the western empire, ranging from areas such as Spain and Gaul, which might justifiably be expected to have adopted more Roman practices and institutions than the Greek east, to the farther corners of North Africa and Pannonia, where multicultural layering poses special challenges. In his search for family structure, Edmondson’s statistical analysis of funerary commemorations in the province of Lusitania provides an illuminating pendant piece to Alston’s. Examining factors such as gender, age, and onomastic conventions, Edmondson argues for a direct relationship between the Roman political presence and the shape of social relations in the province, while at the same time tracing the presence of indigenous cultural attitudes, such as the particularly high valuation of women, which diverged from Roman norms. In Lusitania, as in Egypt, it seems that families might have assumed a Roman form to some degree while at the same time maintaining attitudes and behaviours rooted in the pre-Roman context. This might well have been true in Roman Gaul as well, but, as Woolf explains, the three Gauls taken as a whole offer far less evidence on which to base an analysis. In the absence of sufficient conventional sources, Woolf contemplates the potential value of certain aspects of culture, such as a rhetorical education and Roman law, as agents of change and as inducements for family members to adopt Roman family values and structure. ‘Going Roman’ had the greatest appeal for men, Woolf suggests, but in many aspects Gallo-Romano families, like those in Egypt and Lusitania, probably exercised choice in the ways in which they conformed to Roman models. The delineation of cultural identity presents similar problems in North Africa for Corbier, not for lack of evidence but for its complexity. In extricating possible Roman influence on family structure and practice from among the varied cultural influences (indigenous, Hellenized Greek, Punic, Roman, and eventually Christian) which shaped family life, Corbier finds conventional Roman approaches to funerary commemoration (albeit with the occasional local twist),

Michele George
but speculates that distinctive North African customs might be visible in, for example, regional variations in marriage customs. In distant Pannonia, Boatwright’s funerary stelae portray affectionate family groups attired in both Roman and local traditional dress, making a statement about family identity in a way that is not exclusively Roman, yet is congruent with Roman attitudes. Behind these studies lies the thorny question of ‘Romanization’, the process of acculturation that has conventionally been viewed as a kind of local imitation of Roman attitudes and institutions. In recent years, however, this assumption has come under careful scrutiny, and increasingly it is argued that, far from being a straightforward mimicking of the dominant power, the evidence from the provinces illustrates the formation of new regional cultures and identities through the blending of Roman and local forms. New debates about the meaning of ‘Romanization’ and the extent to which local cultures adopted and absorbed Roman norms, practices, and ideologies add another complicating element in the search for the family in the Roman provinces.7 The collection does not pretend to cover the whole Roman empire, nor do the authors presume to offer definitive coverage of the family in their respective areas. On the contrary, a shared characteristic of the provincial chapters is an awareness of the difficulties inherent in exploring issues such as family structure and more intractable matters such as affective relationships in social contexts which cannot be assumed to have been entirely ‘Roman’. The authors in this volume exploit different kinds of evidence in diverse ways, reflecting the reality of often vastly heterogeneous material which exists in differing degrees of quantity and quality. Some (Alston, Edmondson) rely extensively on papyrological and epigraphic material, while others (George, Boatwright) focus on visual imagery. Treggiari and Bradley draw from conventional texts, while Evans Grubbs concentrates on an undervalued juridical source. Still others (Williams, Woolf, Corbier) adopt an eclectic approach, taking advantage of the evidence available in their respective regions. Yet, even in the face of limited sources, it seems clear enough 7
Romanization: Woolf ; MacMullen ; Fentress and Alcock .
Introduction

that in different parts of the empire variations in family practice existed within a set of accepted social values, whose precise shape admittedly we cannot always see. This is not surprising, considering the flexibility the populace at Rome itself enjoyed in matters of marriage, adoption, and inheritance, while operating within the bounds of Roman law. Although preliminary in many respects, this volume takes a definite step outwards from Rome and Italy to try to understand family life in the Roman imperial period on a wide geographical basis in a way that has not been previously attempted. The exploratory approaches selected here lay the groundwork for further provincial studies and, ultimately, for a better understanding of family life across Rome’s empire as a whole. In opening up new lines of inquiry into the Roman family, this volume raises questions about how the terms ‘Roman’ and ‘family’ are defined, and suggests numerous avenues which might be followed in future research. The anthropological and theoretical methodologies in Alston’s and Woolf ’s chapters, for example, might fruitfully be applied to other provinces, as might Williams’s and Corbier’s selection of particular kinds of epigraphic and archaeological material for family life and social values. Another potential topic for discussion is the possible variation in attitudes and practices over time, both within particular regions and in comparison with one another. For example, did Apuleius or Plutarch mean the same thing as Cicero when they wrote about family, and to what extent is modern scholarship able to elucidate the distinctions that must have existed among them? What are the differences between the west and the east, where the Greek traditions of family life automatically come into play? Scholarly interest in the Greek family has followed from the growth of Roman family studies, but the extent to which family behaviour as evidenced in classical Greece was the same as or different from family life among the Greeks of the Roman imperial age remains to be examined. What, if any, were the ramifications of the spread of Christianity on family life and forms of family interaction? While family studies within early Christianity have been undertaken by New Testament scholars, only rarely has there been any significant engagement with the conventional Graeco-Roman texts or secondary literature, nor in general have classical scholars ventured into the unfamiliar territory of biblical sources. Finally,

Michele George
family life in late antiquity, for which there are numerous good sources, has received less attention than it deserves.8 The subject is far from exhausted. No one has as yet produced the definitive book on the Roman family that combines all these approaches, uses a comprehensive theoretical framework, or considers how the Roman family changed over time. Despite the problems of definition and of available evidence, it is clear that there is still much to be learned about Roman family life, and that great advances in scholarship have been made in the last twenty plus years. This collection, like all conference volumes, is partial and lacks the consistency demanded of a single-authored monograph. It does, however, offer answers to a number of specific questions while setting new challenges for another generation of scholars in a field still ripe for investigation. 8 Christianity: Moxnes ; late antiquity: Shaw b; Evans Grubbs ; Arjava ; Nathan .
 Putting the Family Across: Cicero on Natural Affection Susan Treggiari
That he (Cicero) lied,—is as I have said a matter of course because he was a Roman; and equally a matter of course that he lied successfully, because he was gifted with the use of words. (Trollope, Letters, i. –, to G. H. Lewes,  June )
Introduction O of the things most Roman citizens had in common was experience of a family. Even ex-slaves, deprived of legal parents and perhaps of actual children, aspired to family life. In an age when politicians tremble to think that if they indicate approval of marriage they will make some of the electorate feel excluded, it is interesting to see how Roman orators, such as Cicero, assume norms of family affection and proper behaviour and deploy them in advocacy, invective, and political speeches, with audiences of all types.1 I am grateful to Michele George for organizing an outstanding conference and for editing this volume, to the participants and to the two readers. This chapter is part of a larger study on morality. I acknowledge the generous help of the J. S. Guggenheim Foundation and of All Souls College. 1 It was pointed out by Professor Woolf that Hillary Clinton, in the immediate aftermath of  Sept.  (in which the conference was held) linked her feelings for her own family with sympathy for the families of victims of the attack. I have the impression that similar rhetoric was used by other politicians. This may be, partly, because a major human crisis concentrates the mind on essentials. Woolf’s formulation was different: ‘How does talking about the family let you avoid talking about something else?’ Cherie Booth in  provided an example when, in a broadcast defending the circumstances in which she had

Susan Treggiari
The focus in this chapter is on Cicero’s speeches, though I shall appeal on occasion to the letters and rhetorical or philosophical treatises. I shall begin with some general considerations, and go on to two speeches which highlight the relationship between parent and child, then focus on moralizing about the family and end up with Cicero’s presentation of himself as a good family man. We still debate the relevance of what we regard as private life to our assessment of the qualifications of a person to lead us in public life. The Romans took that relevance for granted. In theory, any human society had its origins in the mating of a couple (according to ‘natural law’): from this derived the relationship with children, other kin, fellow-citizens, and the whole human race.2 Romans talked about Rome, their polity (civitas, res publica), their native land (patria, which is linked with pater, a father), but the state was constitutionally the Roman people, the collectivity of all citizens, represented on public occasions by social ranks and by men, women, and children.3 The state depended on the family.4 Cicero might divide various experiences up into public, common, or individual. For instance, public occasions involved the whole citizen body: games, festivals, or war. Shared experiences might include harvests or cold or hot weather. Individual (singulare) events were experienced privately, privatim: weddings, sacrifices, funerals, a party, or going to sleep (Inv. . ). A number of these involve the family, friends, and staff, not just one person, but there was no general admission for members of the public. There is a concept of the individual operating in a personal and private sphere. Nevertheless, how a person behaves with his bought two flats with money held in a blind trust, wept when explaining that her eldest son, an undergraduate, was living away from home for the first time. Works of Cicero are cited by title alone. In citing the letters (A. 0001 ad Atticum, F. 0001 ad familiares, QF 0001 ad Quintum fratrem), I give Shackleton Bailey’s number before the slash and then the traditional number. The fourth number is that of the paragraph and does not appear in all translations. Unspecified dates are , except in this footnote. Translations are mine. 2 Fin. . , Off. . ; cf. Fin. . , Tusc. . , Off. . ; D. . . . , Ulp.; Treggiari : –. 3 Discussed further in Treggiari, forthcoming a. 4 e.g. Rep. . , Treggiari, forthcoming b.
Cicero on Natural Affection

nearest and dearest spills over into the public sphere, where he is observed by outsiders. When Aeneas took his father on his shoulders and his son by the hand to get them out of burning Troy, he was behaving like a proper leader. The family conduct or irregular sex-lives of emperors mattered to the overall assessment. The Greeks and Romans, Plutarch tells us, took an interest in their leaders’ private lives. Men in public life are not responsible merely for their public words and actions. Their dinners, their bed, their marriage, their amusements and interests are all objects of curiosity. (Plut. Rules for politicians, tr. Russell )
Roman politicians, who had repeatedly to submit their characters and records to the judgement of the electorate, needed to appear to be ‘good men’. Lucilius in his description of the indefatigable and wily public man includes using charm as a weapon and pretending to be bonus vir (‘a good man’; Warmington () Lucil. –). One element of that capacious word bonus was attested by conduct and feeling to kin and wife. Plutarch says that Caesar’s funeral eulogy for his dead wife, Cinna’s daughter Cornelia, and his observed grief won him the favour of ordinary people, who ‘loved him as a gentle man’ (Plut. Caes. . ). The defendant’s ‘private’ behaviour, as far as it could be perceived, indicated his character and proved him likely or unlikely to have committed the crime of which he was accused. Cicero’s courtroom speeches, mostly for the defence, allow us to see how he and the prosecution attempted to represent character.5 Cicero’s mature treatise on the perfect orator ( ) hammers home the point that the orator must speak so as to be understood by ordinary people.6 He must understand the emotions which nature has given to all mankind: the whole point is to quieten or excite their minds.7 He must appeal to their shared sentiments, sensus hominum communis.8 The adroit orator Antonius says he tries to scent the audience’s perceptions, opinions, expectations, and wishes, and fit his speech to them in order to get his hearers 6 de Orat. . ; cf. .  (Antonius speaking). Cf. Fantham . de Orat. . ; cf. , ,  (Crassus speaking), . – (Antonius), .  (Crassus). 8 de Orat. . , cf. .  (Antonius). 5 7

Susan Treggiari
moving in the direction he wants (de Orat. . ). One of the three things he thought about in deciding on his strategy in making a speech was what would fit in with the hearers’ ideas.9 The task is to present oneself in a way that fits in with the preconceptions of the hearers. In the run-up to the British election of  a leaked Tory memo was alleged to have included this statement: The more conservatives talk like (and as a party look like) the rest of Britain—in both language and content—the more credible our messages will be and sound.
Janet Daley commented: ‘Will be and sound?’ Are they the same thing? And what (or who) is the ‘rest of Britain’? All the bits outside the Tory party? Does the rest of Britain have one homogeneous style of speech and appearance that the Tories have somehow missed out on over the years in office and must now catch up with? (Janet Daley, Daily Telegraph,  April )
Obviously, Roman audiences were not homogeneous either (except that those present in the Forum on formal occasions would be mostly adult citizen men). Antonius is thinking especially of forensic oratory and of judges or arbitrators of the upper classes. Mutatis mutandis, the orator also had to think about his hearers when addressing Senate or People. There was plenty of room for different opinions on politics or guilt or innocence even when the people who had to be convinced came from the wealthier strata: the judges (iudices) of Cicero’s time were deliberately drawn from three groups. There was also the corona, the bystanders who listened to forensic speeches, from a wider crosssection of society (Fin. . ).10 Their reactions might influence the verdict and their own assessment of the orator’s standing. I 9 de Orat. . –: ‘accomodatum ad eorum animos . . . ad id, quod volumus, commovendos.’ In sum: ‘. . . () that we prove our case to be true, () that we win over our hearers to us, () that we call their minds to whatever feeling our cause demands’. () depends on evidence and argument; () is commendation of ourselves and our clients. It was inept and tactless to misjudge the audience (de Orat. .  (Crassus speaking), . – (Antonius speaking) ). Cf. Inv. . . See Fantham : – and May : – for the Aristotelian background. 10 See Frier : –, chapter vi ‘The corona’.
Cicero on Natural Affection

shall suggest that the ideas about family which the orators deploy had broad appeal to all sections of society. Antonius in Cicero’s treatise explains how the orator must put both himself and his client across by getting the audience to approve their characters and way of life: ‘hype’ is easier than sheer invention (de Orat. . , cf. . –). The speaker’s reasonable tone will portray him as moral, of good conduct and bonus (de Orat. . , cf. . , . , and Part. or. , ). Good behaviour to kin was one of the indices of goodness. The Defence of Plancius ( ) has a particularly significant passage, arguing that pietas towards parents and other relations was the best possible indication of probity in all social dealings, including politics. I omit those things which are less in the limelight but are certainly praised when they are publicized, how he lives with his people, first of all with his parent—for in my judgement pietas is the foundation of all virtues—whom he venerates like a god—and indeed a parent is not very different from a god to his children—and loves like a companion, a brother, a contemporary.11 What shall I say about his relationship with his father’s brother, his kin by marriage, his relatives, with Cn. Saturninus here, that distinguished man? . . . What shall I say about me, when I feel at his trial that I am the accused? What shall I say about the great number of good men whom you see here in mourning clothes? These are the solid and clear proofs, judges, these are the signs of a probity which is not painted with the cosmetics of public display but branded by the private marks of truth. When we canvass and court the people, it’s an easy thing; it can be looked at but not handled; it shows up well at a distance, it isn’t shaken out and scrutinized, it can’t be picked up in the hand. (Planc. )12
Constant play is made with family affection in appeals to the jurors in speeches for the defence. The roll-call of family members who can be exploited in this way depended chiefly on the kin a man had at the time, though ancestors could be invoked at need. So the presence of M. Marcellus’ weeping cousin Gaius could evoke all the dead ancestors (Marc. ). Calling up ghosts was, Cicero makes Antonius say, a trick of the great Crassus (cos. 11 ‘amat . . . ut sodalem, ut fratrem, ut aequalem.’ Cf. QF /. . : ‘When I miss you, am I missing just a brother? For me you are brother for pleasantness, (?) almost a contemporary, a son for obligingness, a parent for counsel’. (‘. . . suavitate fratrem, aequalem, obsequio filium, consilio parentem.’) 12 Cf. A. /. .  for the contrast between show and real feelings.

Susan Treggiari
), who would conjure up a dead father before the eyes of the centumviri and make him embrace his son and commend him tearfully to the court (de Orat. . ).13 Constraints and Opportunities Created by the Defendant’s Circumstances For demographic realities meant that the surviving kin of a Roman defendant were often few. They are invoked especially in the peroration, but may be referred to elsewhere in a speech. Young sons, in mourning garb, might supplicate the People or the judges.14 Both the spoken word and the sight of those who stood in various family relationships suggest that the defendant is a man like the judges. For instance, the son of L. Valerius Flaccus was too young to understand much of what was going on and may be the one whom Cicero mentions elsewhere as a prop he used to great effect, by picking him up and filling the Forum with tears and lamentations. He is treated as a little suppliant. If the judges leave him his father, they will leave him a model of a good citizen; if not, they will show him that goodness is not rewarded.15 M. Caelius Rufus too was short of close relatives, so that Cicero makes much of the loneliness faced by his father, whose only son he is and who depends utterly on him. The appeal, as so often, is to men who will understand the strength of paternal affection because they remember their own fathers or have children of their own. Caelius will repay his debt not only to them but to their children (Cael. , –). Cf. further Treggiari : –. Berry (: –) usefully assembles sources and discussions. Cf. Rhet. Her. . , . ; Brutus ; for the success of fathers pleading for defendants see Clu. . 15 Flac. , probably referred to in Orat. . For appeals in the name of the defendant’s offspring (liberi) see e.g. Quinct. , for allusions to the orator’s children Mil. , . For the judges thinking of their own liberi in making their decision see Cael.  (final paragraph), Planc.  (penultimate paragraph). Cicero commemorates Flaccus’ services to the judges and their wives and children (Flac. ). Flaccus’ son supplicates the judges and their children (Flac. ). The tax-contractors who voted for Plancius were thinking of their own children (Planc. ). In political speeches about his own status Cicero can make promises of service to the descendants of his hearers (Red. Pop. ) or invokes his liberi at the very end (Dom. ). 13 14
Cicero on Natural Affection

The clever orator deployed kin who existed and kin who did not. In both his speeches on his return, Cicero compares his own small contingent with those of other illustrious exiles. In his speech of thanks to the Senate after his recall, he lists those who had pleaded for men who had earlier been threatened with banishment: For me there was no adolescent son to plead as young sons and a multitude of kin had begged the Roman People for mercy on behalf of that noblest of men, P. Popilius. Q. Metellus, that eminent and famous man, had a son whose youth had already been observed; he had L. Metellus and C. Metellus, both of consular rank; their children; Metellus Nepos, who was then a candidate for the consulship; the Luculli, Servilii and Scipiones, sons of the Metellae, all of whom weeping and in mourning (flentes ac sordidati) supplicated the Roman People. I had just one brother, who for pietas proved himself a son to me, for counsel a parent, for love the brother which in fact he was.16 By his grieving appearance (squalore) and tears and daily prayers he compelled the renewal of longing for my name and the cherishing of the memory of my deeds. (Red. Sen. )
Cicero avoids the appearance of self-pity and keeps a stiff upper lip by invoking his family, a good way to move his audience to compassion. He must have thought the theme worked, for he reused it to the People, stressing the popular hero Marius17 and, while continuing to give his brother Quintus the major credit, mentioning also his son-in-law, whose grief was also in the public eye, and his wife, daughter, and little son (Red. Pop. –). It is interesting that the evocation of the extended family of Metellus Numidicus, which requires extensive footnoting for us, with the picture of dignified senators and probably around ten little future politicians, was expected to be so powerful with ordinary citizens and not just with the upper classes.18 16 ‘. . . sed unus frater, qui in me pietate filius, consiliis parens, amore, ut erat, frater inventus est.’ Cf. n.  and Red. Pop. : ‘This one man was found to be a son to me in dutifulness, in conferring benefit a parent, in love, as he always was, a brother.’ (‘Unus hic . . . mihi pietate filius inventus est, beneficio parens, amore idem qui semper fuit frater’). 17 He was another consular exile, but came back by violence, not intercession. He appears in Red. Sen. only as a footnote (). 18 P. Popilius C. f., consul , was exiled . Q. Caecilius Metellus Numidicus (exiled , recalled late ) had a young son, Q. Metellus Pius (the future praetor of /, consul ); his cousins the ex-consuls L. Metellus Diadematus (consul , censor , with his son Q. Celer) and C. Metellus

Susan Treggiari
The orator in lament or complaint, conquestio, to evoke pity, was advised (among other things) to detail the pleasure he had taken in a dead son, to deplore separation from loved ones, to commend his children or parents to the audience, to claim that he grieved for his loved ones’ suffering, not his own (Inv. . , ). He might ask the audience, when looking at him, to remember their own children or parents or others who ought to be dear to them (Inv. , cf. ). The insistence on the impact of judicial or political decisions on the family of defendant or politician not only creates pathos but underlines the man’s domestic virtues and innocent and deserving kin. These may be a theme, not only in the peroration, but throughout a speech. Exploiting Relationships Affection between family-members is held up as the social norm.19 Taken as a given, it is often evoked to great emotional effect, as in a highly wrought passage of On his house (below). Love of one’s own family, sui, says Cicero, is demanded by common humanity: we naturally hold them dear, cari, and find them agreeable, iucundi (Dom. –).20 In this passage, Cicero uses the blanket-word sui. In general, the relationship most often exploited in the speeches is that between parent and child. So that is the area I shall focus on here, first giving a general sketch.21 Caprarius (consul , censor ) and his sons Gaius (senator in ), Q. Creticus (consul ), Lucius (consul ), Marcus (praetor ); his cousin Balearicus’ son, Q. Metellus Nepos (consul ); children of Metellae, his sister and three cousins. See especially Wiseman b: –, –. 19 Natural law: Rhet. Her. . ; cf. Amic. . Some of the texts are discussed in Lambert (: –) . 20 The adjectives carus and iucundus occur frequently together to describe family members, e.g. Font. , Planc. . 21 The Defence of Ligarius and Cicero’s own exile give prominence to brothers, but I touch only lightly on that relationship. Wives are not often highlighted in speeches (despite the priority of the sexual bond to that between parent and child in philosophical theory: Off. . , as a reader pointed out), in part because it was not normally proper to drag their names in: nevertheless Cicero might choose to mention his own wife in public and he will attack Antony for betraying Antonia (Phil. . ) or for being hand-in-glove with Fulvia (e.g. Phil. . ; cf. Treggiari : –). Proper behaviour in marriage is one theme of the Defence of Cluentius.
Cicero on Natural Affection

Children The common sentiment of children about parents or parents about children is constantly invoked (Cael. –). Philosophers held that it was natural to desire to procreate and to love one’s children (A. /. . ; Fin. . ; . , –; . , . , –, –, ; Off. . ). Children give parents delight, through their iucunditas (e.g. Dom. , Cael. ). An only son is the prop of his father’s old age; hope in his future is his father’s comfort, fear for him his major worry.22 We love our children out of duty and natural predisposition to favour them. They are a gift from heaven (Red. Pop. ). P. Sulla is said to love his young son more than life (Sulla ). This is a conventional expression, but we need not doubt that the emotion existed (Red. Pop. , F. /. . ). Indulgentia is the natural feeling of any parent towards a child (Verr. . . ; de Orat. . ). Cicero portrays it in action in the Defence of Ligarius ( ), when he asks Caesar to exercise forgiveness like a father to Q. Ligarius who had opposed him in the civil war. I have pleaded many cases with you, Caesar, while your political career kept you in the forum, but never one like this: ‘Forgive him, judges; he made a mistake, he slipped up, he did not realize; if he ever does anything like it again . . .’ This is how one speaks to a parent. To judges one says, ‘He did not do it, he never even thought of doing it. The witnesses are liars, the charge is invented.’ Say, Caesar, that you are a judge about what Ligarius did; ask in what forces he was engaged: I hold my peace, I do not even collect points like these, which might perhaps have some weight even with a judge: ‘he went out as a lieutenant before war broke out, he was abandoned in peace, overwhelmed by war, in war itself he was not irreconcilable, he is now all yours in heart and commitment.’ That is how I might speak to a judge, but I am speaking before a parent: ‘He made a mistake, he acted rashly, he is sorry; I take refuge with your mercy, I ask pardon for his offence, I pray for forgiveness.’ (Lig. )
Caesar’s reaction is attested. Reckoning that Cicero could not make him change his mind, he chose to listen to his oratory and was so much moved, especially by the treatment of Pharsalus (Lig. –, which immediately precedes the passage comparing Caesar with a father) that he forgave Ligarius (Plut. Cic. . –). 22
Cael. –; cf. Rhet. Her. .; Treggiari : .

Susan Treggiari
Cicero also plays on the paternal instinct in political speeches. The old appeals to the need to defend the status quo against political opponents who are defined as people who would act like foreign foes, are couched in terms of defence of hearth, home, fortunes, household gods, wives and children. Children alone can evoke all the rest. They are called liberi, the term which highlights the idea of legal descent, and which could be used even if a man had only one child.23 (The English ‘issue’ has nothing like the same emotional charge.) Cicero took the position in the Catilinarians that he was defending the children of his hearers against threats of fire (especially the alleged plot to start a conflagration in the City) and the sword (Catil. .; Flac. , , , ). Clodius was the next major threat. Building on the Bona Dea scandal and the alleged incest with sisters, Cicero can make Clodius out to be a direct threat to family life. If he had become consul, he would scarcely have kept his hands off the wives and children of the citizens (Mil. ). The watching multitude at Milo’s trial thought the struggle was on for themselves, their children, country, and fortunes. Cicero hoped for a happy future for them, which was impossible if Clodius had lived (Mil. , ). The Antonii, in arms against the Senate in , might inspire a more rational fear, especially after L. Antonius had allegedly massacred women and children at Parma (Phil. . –; cf. Phil. . ).24 Parents Filial affection is also invoked. We owe to parents our life, liberty, citizenship, patrimony, and innumerable blessings. Cicero equates the Senate to his parents in a passage of high emotion in his elaborate thanks for his recall from exile in : If we ought to hold our parents very dear, because it is from them that our life, patrimony, freedom and citizenship are handed down; if we owe the same love to the immortal gods, by whose grace we have possessed these things and have been given other blessings; if we ought to love the Roman People, by whose election to offices we have been 23 Ter. An. ; S. Rosc.  (cf. ); Verr. . . , Phil. . ; D. . . , Gaius. 24 The sack of cities, involving rape and murder, is a commonplace from Homer onwards, e.g. Rhet. Her. . ; Paul .
Cicero on Natural Affection

placed in the most distinguished council and in the highest rank of dignity and in this citadel of the world, and if we should love this very order [sc. the Senate], by whose splendid decrees we have been honoured, immense and infinite is what we owe you who by your remarkable support and unanimity have restored to us all in one moment the benefits of our parents, the gifts of the immortal gods, the offices bestowed by the Roman People, your many testimonials to me. Now, though we owe much to you, great things to the Roman People, innumerable gifts to our parents and everything to the immortal gods, we have recovered all these things together through you. (Red. Sen. )
Here parents are set in parallel with the gods, the Roman electorate, and the senatorial order. Elsewhere, they may be compared with the gods and the country, patria, as benefactors to whom gratitude is due (Rhet. Her. . ; Fin. . ).25 Violence to parents was a sin, parallel to sacrilege or treason (S. Rosc. ; cf. Fin. . , . ; Rhet. Her. . , , ). Fathers Sometimes the theme of family was central to a case and we see how advocates on opposing sides deployed stereotypes about family life and interpreted the behaviour of individuals in opposite ways.26 Contemplating in the Defence of Sextus Roscius of Ameria ( ), the worst conceivable murder, parricide (Parad. –), Cicero remarks that a mere look could mark a failure of pietas. Human and divine law might compel a child to die to save his father (S. Rosc. ).27 It was unbelievable that a son would attempt to kill his father without very serious reasons. Similarly, it was impossible to believe that a father would hate a son without grave cause. In defending Roscius against a charge of having instigated his father’s murder, Cicero makes the most of the unnaturalness of such an act. 25 Benefactors are habitually compared with parents (e.g. Pis. ). The consul of , P. Lentulus Spinther, architect of Cicero’s recall, receives extravagant praise, as parent, god, and saviour of Cicero’s life, fortune, memory, and name (Red. Sen. , Red. Pop. ; cf. Red. Sen.  on Plancius, Mil. , on Milo as father to Cicero’s children). 26 Cf. the (later) surviving rhetorical declamations and contrasting views of character in historians. 27 On dying for one’s country cf. Rhet. Her. . ; Phil. . .

Susan Treggiari
Unless there are many manifest proofs, such a criminal, atrocious, and wicked deed cannot be believed. For the strength of humanity is great; shared blood has a strong influence; nature herself cries out against such suspicions. An aberration of nature, a monster clad in human form— that is what a man must certainly be who can so surpass beasts in ferocity that he can foully deprive of the light of day those parents thanks to whom he gazes on this lovely light. Even wild animals are brought together by birth and rearing and nature. (S. Rosc. –)28
The parricide must be dehumanized, either a wild youth given over to vice and corrupted by his companions or a man hardened to violence and driven by extravagance and debt or selfish desires (S. Rosc. –, ). Cicero paints the younger Roscius as an austere and mature countryman (S. Rosc. , ). Moreover, only strong cause could motivate parricide. So the prosecution had alleged that Sex. Roscius’ father disliked him, had given his affection to another son who died young, saw very little of him, and intended to disinherit him. To show the context of dislike, the prosecution stressed the fact that the son lived near Ameria, managing some family property, while the father was often in Rome. They apparently claimed that the father had in effect banished the son, an argument which Cicero rebuts by saying that he could have achieved that by sending his son to live at one farm with just an allowance for food and necessities, whereas (Cicero hints) it was generally known that he was in charge of several farms and drew an income direct from some of them, in his father’s lifetime (S. Rosc. –, cf. ). This was kindness, not hatred, an honour, not a punishment. The prosecutor, Erucius, had made the tactical mistake of arguing that country life was less desirable than town life and sophisticated society, from which the younger Roscius was excluded by coarse manners and antisocial character (S. Rosc. –). Erucius perhaps did not have the luck to know who his father was and so find out for himself what a father’s feelings were towards his children. But his own talents and ambition had given him an education, so he could find out from comedy, a reflection of everyday life, that keeping a son in the country was
28 Cf. Inv. . : to produce indignation, the orator may show that a crime is rare and would not even be committed by animals, e.g. cruelty against parents, children, wives, etc. or against the helpless (women, children, the old).
Cicero on Natural Affection

no indication of disapproval (S. Rosc. –).29 In short, Erucius had failed to bring evidence of any faults in the younger Sextus Roscius which could have angered his father. Erucius’ conjecture (clearly intended to account for the timing of the murder) that the elder Roscius was thinking of disinheriting his only surviving son was without foundation (S. Rosc. –, ). Cicero argues that the instinct of a father to love his son is so strong that only serious faults would cause him ‘to manage to conquer nature herself, to cast out from his heart that deeply rooted love, to forget that he is a father’ (S. Rosc. ). Only a madman could hate a son without cause (S. Rosc. –, –). Yet the prosecutor had been unable to prove such cause. To disinherit a sole surviving son was incredibly harsh (S. Rosc. –, ; Clu. ). The judges should not believe that the elder Roscius hated his son, for that would be unnatural. Nor had the prosecution proved that the son was unnatural and capable of killing his father. Mothers In a later defence ( ), Cicero, driven to take a different line, produces the ultimate family soap-opera. The speech on behalf of the eques A. Cluentius Habitus, in its classic portrayal of an unnatural mother, gives a partial idea of what a good mother should be like and how she should be treated.30 In the Defence of Plancius ( ), Cicero had to argue that Plancius’ father was not an electoral liability (as the prosecution said) and certainly not turpis (shameful, disgraceful) or sordidus by reason of his position as an eques and tax-contractor. Even if he had been, ‘yet he would have influence on merciful and compassionate judges, he would, I say, have influence because of the common feeling of all mankind (communi sensu omnium) and the sweet commendation 29 Ordinary observation of his fellow-tribesmen and neighbours had convinced Cicero that they valued sons who were engaged in farming. A farming life was not only considered honourable by the patresfamilias of Umbria and the country towns but had long been the best preparation for service of the Roman state (S. Rosc. –). On this speech see recently Riggsby : –, Alexander : no. . 30 On this speech see recently Riggsby : –, Classen : –, esp. –, –, –, or : –, esp. –, –, –, and bibliography in Alexander : no. . On stereotypes of good mothers see Dixon : –, –, passim. Individual mothers occur rarely in the speeches.

Susan Treggiari
of nature’ (Planc. – at ). A different twist to the turpitudo of a parent is given in the Defence of Cluentius. Here defending counsel wants to argue that his client’s mother is disgraceful (morally, not socially) and, quite unlike the reputable Plancius who came to his son’s defence, was engineering the prosecution and was, in fact, a long-standing enemy. The ramifications of the families in Larinum to which Cluentius was attached are too complicated to be explored here. Suffice it to say that Cluentius was prosecuted on a charge of having procured the death of his stepfather, Statius Abbius Oppianicus, several years earlier. The prosecution was brought by Oppianicus’ son (of the same name) who was married to Cluentius’ half-sister, and Cicero alleges that the person who stood behind the prosecution was Cluentius’ own mother, Sassia, widow of the dead man. Cicero’s intention is to portray Sassia and the elder Oppianicus as the chief villains of the piece, the younger Oppianicus as a tool (Clu. , ), and in particular to deflect suspicion of murder away from his client. In this character-assassination, Cicero had first to break the rules of etiquette and chivalry which forebade attacks on women and to win over the judges who would normally have considered that the fair name of the defendant’s mother had to be protected, especially by the defendant and his advocate. I am perfectly well aware that, whatever a mother is like, it is scarcely fitting to speak of the shamefulness (turpitudo) of a parent at a son’s trial. I would not be competent for any case, judges, if, when I am called in to defend people against danger, I did not see this, which is firmly implanted in the common feelings of mankind and in nature herself (in communibus hominum sensibus atque in ipsa natura). I thoroughly understand that people ought not only to keep quiet about injuries done them by their parents but also to endure them with equanimity. But I think that they should endure those which can be endured and be silent about those which can be passed over in silence. (Clu. )
All Cluentius’ sufferings had been caused by his mother and he would have remained silent about them, but he could not do so when she had him charged with poisoning and used wealth and influence to bring witnesses against him. Cicero therefore claims to be entitled to bring her name into the defence: he is not lightly breaking the taboo against naming women in public or attacking kin (Clu. –; cf. ).
Cicero on Natural Affection

When he deprecates the judges’ reaction, Cicero has already alleged that he must answer a two-pronged attack by the prosecution (Clu. ). Not only must he tackle the accusation of poisoning, which is what the case is really about, but he must deal with allegations intended to produce prejudice (invidia)— Cicero’s coloured representation of the prosecution’s routine argument about the character of the accused—that Cluentius, when he had prosecuted his stepfather on a charge of having tried to poison him in , had secured a condemnation by bribing the judges (Clu. –, –, , –).31 He must therefore say much about the elder Oppianicus. He then launches into the narrative of previous events and rapidly comes to the first mention of Sassia. It is intended to shock the audience, much as the abrupt introduction of Sulla’s freedman Chrysogonus had been, in the speech on behalf of Roscius at the start of Cicero’s career (S. Rosc. ). Like Chrysogonus or Clodia in the speech for Caelius,32 Sassia is the sinister figure who stage-manages the prosecution. The passage begins calmly, in what sounds like an ordinary sketch of the client’s background. His father was an eminent man in Larinum. He died in , when Cluentius was  and his sister marriageable. The sister made a good match with her cousin A. Aurius Melinus. Then comes a change of tone: This marriage was eminently respectable and harmonious. Then a reckless woman’s wicked lust was suddenly fired and brought with it crime as well as disgrace.
Who is the woman? Has the bride suddenly fallen in love with someone else? Is there some alien seductress? Much worse: For Sassia, the mother of my client Habitus—for mother is what I will call her throughout the case, although she is a cruel enemy to my client, mother, I say, I will call her, nor, even when I mention her wickedness 31 This was embarrassing, since Cicero here wants to argue that Cluentius was innocent of judicial corruption, while Oppianicus had tried bribery, but in the Verrines ( ) he had accepted that both parties had bribed judges (e.g. Verr. . ) and in / he attacked the senator C. Fidiculanius Falcula for having been bribed to condemn (Caec. –). Cluentius’ prosecutor T. Attius taunted Cicero with his previous statements (Clu. , with Classen : – or : –). 32 A general allusion in Cael.  points forward to her appearance as Palatine Medea in , but she is not named until .

Susan Treggiari
and monstrosity shall she lose the name nature gave her. For the more loving and kindly the name of mother is, so much the more shall you think that you should hate the wickedness of this mother, who for many years, and more than ever now, has desired that her child might be killed. (Clu. )
The fractured syntax leads in to the mother’s love-affair with her son-in-law. Since Sassia was overtly supporting the prosecution, siding with her dead husband Oppianicus and his son (her son-in-law), against her own son, Cicero had to deflect the judges’ natural assumption that there was something badly wrong with Cluentius.33 He therefore portrays Oppianicus as an arch-villain, a profiteer, a serial husband34 and the murderer of his brother-inlaw (Clu. –), the man who engineered the proscription and/or death of several other Larinates, including A. Aurius, husband of Sassia and previously her son-in-law (Clu. , –); the man who killed two of his own children to persuade Sassia to marry him (Clu. ), the poisoner of his mother-in-law, Dinaea (Clu. –), and of one of his own wives and of his brother’s wife, unborn child, and the brother himself (Clu. –) and, in collusion with the widow of his wife Magia’s brother, of her unborn child (Clu. –). It is credible, then, that he also tried to poison his stepson A. Cluentius Habitus.35 Sassia is portrayed as a mother so unnatural that she emulates the conventional stepmother.36 She desires death and ruin for her 33 Alexander : nos –. See especially Clu. – for the alleged attempt to poison Cluentius and the subsequent trials. Cicero claims that Cluentius was driven to initiate the prosecution (Clu. , , ). Much space is devoted to rebutting the suspicion of corruption. For Cicero’s boast that he had succeeded in obscuring the truth see Quint. Inst. . . . 34 He was married to Sassia; Magia, by whom he had a son (Clu. , ); Novia, by whom he had a son, who died an infant before he married Sassia (Clu. ); Papia, by whom he had a son, who was being brought up by his (divorced) mother at Teanum and who died on a visit to his father before the marriage to Sassia (Clu. ); Cluentia, the aunt of Cluentius (Clu. ); an unnamed woman, previously married to Cn. Magius (Clu. –), unless she is identical with one of those just named. The chronological order cannot be completely established. Cicero claims that he was a polished performer in the murder of wives (Clu. ). 35 Clu. –—and once Sassia had inherited from her son, he intended to get rid of her too (Clu. )! There is a neat summary of the murders at Clu. . 36 See Noy ; Watson : especially – on Roman stepmothers in literature and life and – on Sassia.
Cicero on Natural Affection

son; she breaks up her daughter’s marriage. She brings about the death of two young children who would have been her stepchildren. Her hand is detected behind the prosecution (e.g. Clu. , –). Accusations are inevitable as long as she lives (Clu. ). The list of her crimes is shorter than that of Oppianicus, but she is smirched by the association with him. Like him (e.g. Clu. , , , , , ), she is characterized by reckless effrontery (audacia) (Clu. , , ). As the narrative of Oppianicus’ career and of Sassia’s marital history dominate the narratio at the beginning of the speech, so her machinations against her own son dominate the rebuttal of the actual charge against Cluentius at the end. After Oppianicus was condemned, Cicero claims that she was overfamiliar with a lusty farmer while her husband was lying sick. On his way to Rome (to get away from the farmer), Oppianicus injured himself by a fall from his horse and subsequently took a fever and died (Clu. , cf. ). Suspicion ought to have attached to Sassia rather than to Cluentius, but she at once used the death as a pretext for accusing him and savagely tortured slaves in order to obtain evidence. The slaves admitted nothing and one of the witnesses suggested that she was not trying to extract truth but falsehood. The widow therefore turned for home, grieving—that her son was now safe from her hidden plots (Clu. –). In summing up, Cicero lists monstrosities which allow him to portray her as an unnatural mother: What a portent is this, immortal gods! What monster like her could we say has ever been born anywhere? what dangerous and terrible crime has there been like this? Where did she spring from? Now at last, judges, you see that it was not without great and necessary reasons that I spoke about his mother at the beginning of my defence. There is no evil, no crime which she did not from the beginning will for her son, long for, think out, bring into effect. I pass over that first wrong motivated by lust, I pass over the wicked marriage with her son-in-law, I pass over the daughter driven out of her marriage by the desires of her mother. These things were relevant to the shared disgrace of the family, but not to the risk of my client’s life. I make no complaint about the second marriage with Oppianicus, when, after taking his sons from him, dead, as hostages, she married into the family’s mourning and the obsequies of her stepchildren. I leave out the fact that when she learned that A. Aurius, whose mother-in-law she had once been and whose wife she was only a short time before, had been proscribed through Oppianicus’ agency and slaughtered, she chose for her home and dwelling the house

Susan Treggiari
in which she might see every day the proofs of the death of her previous husband and the trophies of his fortunes. The first point about which I complain is about the crime which has only now been revealed, about the Fabrician poison [sc. the attempt on Cluentius’ life], which when it had just occurred was suspected by others but incredible to my client, but now seems open and manifest to everyone. His mother was not kept in the dark about that poison; Oppianicus thought nothing out without input from this woman; if she had been uninformed, she would of course not have left him as a bad man, but would have fled from him as from a cruel enemy and left for ever that house which flowed with all sorts of crime. Far from doing that, from that time onwards she let slip no opportunity to plot some trap and all night and day with all her mind the mother thought about how to destroy her son. (Clu. –)
Where a normal mother would pray and sacrifice for her son’s welfare, Sassia performs secret nocturnal sacrifices and makes vows for her son’s destruction (Clu. ).37 Where an advocate would normally beg judges to forgive a son’s transgressions in mercy to his parents, Cicero must ask them not to surrender Cluentius to his mother’s cruelty (Clu. ). This is an extreme instance of the thirteenth way of achieving pathos recommended in De inventione, when we lament that we are treated badly by the least appropriate people, such as relatives and dependants (Inv. . ; cf. Scaur. ). But what a mother! You see her carried away by cruelty and crime, whose lust has never been checked by any disgrace, who, by her faults of character, turns all human laws (iura) to the bad, so stupid that no one can call her a human being, so violent that no one can call her a woman, so cruel that no one can call her a mother. It is not just the name and laws of nature she has changed, but the names of relationships: wife of her son-in-law, stepmother of her son, mistress of her daughter’s husband. She has gone so far that only her outward shape makes her seem a member of the human race. So, judges, if you detest crime, stop the mother getting her son’s blood, give a parent incredible pain from the victorious acquittal of her child, allow a mother not to rejoice bereaved of her son, but rather to leave the court conquered by your fairness. (Clu. –)
All the usual pleas on behalf of sorrowing parents are turned inside out: in each imperative clause of the final sentence of this extract there is a sudden shock as Cicero asks for the opposite of what 37
For such clandestine activities cf. Catil. . , ; . ; Vat. .
Cicero on Natural Affection

would usually come next. The end of the peroration, in which he argues positively for acquittal that the judges should do their job and satisfy the good people of Larinum, is routine by comparison. Despite the denigration of Sassia, Cicero must make his client act dutifully towards her as long as possible. Even a mother as disgraceful as Sassia had a claim on the loyalty or discretion of her son. Even when she was conducting an adulterous affair with her son-in-law and Cluentius not only disapproved strongly but owed a duty to his sister, he merely stopped seeing his mother so that he would not appear to condone her behaviour (Clu. ). Later, when she was married to Oppianicus, Cicero claims that the reason Cluentius had never made a will was that he could not bear either to leave such a mother anything or to pass over a parent entirely (Clu. ). So there was animosity (simultas) between son and mother and reconciliation (in gratiam redire) continued to be impossible (Clu. ). Cicero could, he says, pass over the wickedness which did not directly affect his client. He imputes to her acting as accessory in Oppianicus’ attempted poisoning of her son (Clu. ), persuading the younger Oppianicus to prosecute Cluentius for the murder of his father (Clu. , ), suborning witnesses and masterminding the prosecution (Clu. –). Though he then turns away from Sassia to concentrate the judges’ attention on the innocence and suffering of Cluentius, almost at the end of the speech he describes his client as dear and agreeable, carus and iucundus, to many people, a delicate reminder that he was not loved by his mother as a son would be loved by any normal mother (Clu. ). Moralizing on Behaviour within the Family The topic of the family could not be avoided when a man was accused of the murder of father or stepfather. But it is striking how often the orator introduces family virtues or offences against them when their relevance is less apparent. Pietas, dutiful affection, was demanded between all family members.38 No offence must be done to a parent (Catil. . –), 38 On pietas see especially Saller : –; Bradley b: –, . For Cicero’s emphasis on it in rhetorical theorizing see e.g. Inv. . ; Part. –; for actual cases cf. e.g. Suet. Rhet. . Philosophers had views on duty towards e.g. parents and brothers, ‘how to live with them’ (Div. . , cf. Planc. ).

Susan Treggiari
a standard which Cluentius upheld. The good mother was the opposite of Sassia. Deductions about behaviour in more normal families were freely made by both prosecution and defence. Caelius’ opponents built on the known fact that he had left the family house to live independently in an apartment to criticize him for failure in pietas. The defence had to show the judges that a man’s own family and fellow-townsmen held him in esteem. A young man of this age could not be adequately recommended to you if he were disapproved of by such a distinguished and serious-minded town, let alone by his excellent parent. (Cael. )
Cicero uses the tears of Caelius’ mother and the mourning clothes of his father as visual evidence of their support for their son and their judgement of his character (Cael. ). By extension, it was deplorable for a third party to offend against the natural pietas of others. The Verrine orations use Verres’ destruction of other people’s families as a recurrent theme. The slaughter of the innocent and the wrecking of parents’ hopes for their children run through the whole of Verres’ record as young official, judge, and governor. In contrast, his father’s normal family affection for him is turned to corrupt ends and his son is brought up as a worthy successor to his father’s vices (Verr. . –; . . –; . . , , , ). Impiety towards families is one of the sins which drives Verres mad according to the highlywrought exordium to the published version of what Cicero would have said in the second part of the trial, if it had taken place. The di patrii 39 drag him to punishment because he brought himself to conduct to execution sons torn from the arms of their parents and demanded from parents a fee for their children’s burial. (Verr. . . )
This is later given full-scale treatment. Sicilian sea-captains were made scapegoats for a naval disaster at the hands of pirates, for which Cicero held Verres responsible. Their mouths had to be stopped, so only the death-penalty would do (Verr. . . –). The relatives of the young men (they are young, to increase the pathos) flock to Syracuse to plead for mercy. They include an aged father, a former host of Verres. 39 Gods who protected fathers ( patres) and children, home and country ( patria).
Cicero on Natural Affection

You were unmoved by the tears of the father for the peril of his innocent son. You had left your father behind at home; you had your son with you. Did not the son who was with you remind you of the dearness of our children, nor your father at home make you think of the kindness a father has for his child? (Verr. . . )
Cicero identifies several of the parents involved, with pathetic detail (Verr. . . , , ). All are pictured as lying at the threshold of the prison the night before the execution, paying the gaoler for the privilege of taking food and clothes to their sons, or for the promise of a quick death or proper burial (Verr. . . –, cf. ; for another instance . . , –). Verres is portrayed as having trampled on all the dearest family-feelings of those in his power. Cicero as a Family Man We have seen how family solidarity and dutiful conduct to kin can be used as a norm by which people are judged. Cicero points out that a man’s good repute spreads outward from his household, and boasts that this had helped build his own reputation.40 Cicero consistently portrays himself in public statements as a devoted family man. This is particularly apparent in the speeches which follow his return from exile, after more than a year’s separation from wife and children and when he had not seen his brother since Quintus went to his province in . In his thanks to the Senate in  he stresses the point that the favour they have conferred on him is also one to his brother and children. He insists on the reciprocity of their affection. You have restored to me the brother I longed to see, me to my loving brother, parents to our children, to us our children, dignity, rank, fortunes, the great commonwealth, country, the most delightful of all things, in sum us to ourselves. (Red. Sen. ; cf. , ; QF /. . )
The same ideas are recycled in his speech to the People two days later: For, Quirites, although nothing is more to be wished by a human being than a prosperous, equable and unceasing fortune with a favourable course of life without mishaps, yet, if everything had been tranquil and 40
Cael. . Cf. p.  above on the self-presentation of the orator.

Susan Treggiari
peaceful for me, I would have missed an incredible and almost divine pleasure of happiness which I now enjoy thanks to you. What sweeter thing is given to the human race by nature than his own children to each of us? To me both because of my kindness for them and because of their own excellent character my children are dearer than my life: yet I did not feel as much delight in acknowledging them as babies as I now feel in having them restored to me. Nothing was ever pleasanter to anyone than my brother to me: I never felt that so strongly when I enjoyed it as I did when I went without it and after you restored him to me and me to him. . . . Through my parents I was born, inevitably, as a little baby, by you procreated as a consular. They gave me a brother, but we could not know how he would turn out; you gave him back to me tested and known for his incredible devotion. . . . The immortal gods gave me children; you gave them back. We won many things besides which we desired from the immortal gods; had it not been your will, we would have been without all the gifts of the gods. Finally the honours you gave, which we had gradually achieved one by one, now we have all together from you, so that whatever we owed before to parents, gods and you yourselves, we now owe it all to the whole Roman People. (Red. Pop. –, )41
It is clear that he thinks mention of his family will strike a chord with his audience. In speaking to the priests about his house later in , he assumes that common humanity demands that we love our families (Dom. ). The theme is fully exploited in the following passage, where Cicero, while duly specifying the pain his exile inflicted on his family, does not refrain from piling on his own agony, in order to boast of the sacrifice he had allegedly made for his country: When I had judged the situation, I wept for the separation from my unhappy wife, the loneliness of my dearest children, the ruin of my loving and excellent brother (who was abroad), the sudden ruin of a wellestablished family; but I put all these things second to the lives of my fellow-citizens, and I preferred that the commonwealth should fall stricken at the departure of one man rather than perish by the destruction of all . . . Could I, when I was torn from so many different things, which I do not list because even now I cannot mention them without weeping, deny my humanity and repudiate the common feeling of our nature (communem naturae sensum)? I would not claim to have done a 41 Cf. Red. Sen.  (above, pp. ‒). For rebirth and family affection cf. Sest. . In a formal letter to Ap. Claudius Pulcher in , F. /. . , Cicero’s restoration to country, children, safety, dignitas, and himself could be attributed to Pompey, the father-in-law of Pulcher’s daughter.
Cicero on Natural Affection

praiseworthy deed nor to have conferred any benefit on the state, if I could have lost with equanimity the things I left for the sake of the state, and I would have thought such toughness of mind (like that of a body which does not feel a burn) an insensitivity, not courage. To take on such mental agony, for one man to suffer while the city still stands the things which happen to the conquered when a city is captured, and to see oneself torn from the embrace of one’s family, the house taken stone from stone, one’s fortunes plundered, to lose one’s very country for the sake of country, to be despoiled of the glorious privileges conferred by the Roman People, . . . to undergo all this and that when you are present and sorrowing, not such a philosopher as those who care for nothing, but loving your own people and things as common humanity (communis humanitas) demands: that is glorious and godlike fame. For a man who with equanimity for the sake of the commonwealth abandons things which he never thought dear and pleasant, demonstrates no remarkable benevolence towards the commonwealth; but a man who leaves for the sake of the commonwealth things from which he is agonizingly torn, truly loves his country, whose survival he puts before his love of home and family. (Dom. –)
Cicero’s chief aim must be to evoke the sympathy of the various audiences for suffering which they could easily imagine and to carry them away on a wave of shared exultation in his joy. Not only must he boast of his courageous self-sacrifice on behalf of his country, but he must persuade people that he deserves full restoration of his house and fortunes, so important to his wife and children. New popularity for Cicero with Senate and People would enhance his political authority, enabling him to pay off the debts incurred for his recall, and launch him, his brother, and their sons on new political careers. This is not to say that his own emotions, as described in the speeches, are false. They are consistent with those rehearsed in the letters to Atticus, Quintus, and Terentia, which similarly sound both rhetorical and deeply felt. His training and talent as an orator must inevitably shape what he chooses to say and how he says it. But conversely his personality and his range of emotions must always have fed his rhetorical abilities. If he had not known how to feel, as well as been able to weep, would he have reduced his audience to tears?42 He could 42 Cf. Treggiari : –; Hutchinson : –. For Cicero’s ability to cry and, sometimes, his claim that he has made his hearers cry see e.g. Planc.  (final paragraph), Mil. , , ,  (where Milo fails to weep). Cicero weeps, along with Rabirius’ friends (Rab. Post. ). Pathos was only sometimes

Susan Treggiari
risk the delicately balanced mixture of political self-justification and strong emotion even to the priests, his fellow-politicians. Even in forensic speeches, Cicero can allude to his family in order to evoke fellow-feeling. In the Verrines, he castigates Verres for having enforced the Voconian law against an only daughter to whom her father had wished to leave his estate. I am sure that this seems harsh and unworthy to each of you, as it does to me. I delight in my daughter; you are moved by a similar feeling (sensu) and kindness (indulgentia) towards your daughters. What is there that nature meant to be pleasanter or dearer to us? What more deserves all our care and kindness? Cruel man, why did you do such injury to the dead P. Annius? Why did you brand this pain on his bones and ashes? Why did you snatch away from his child the father’s property which was handed down by her father’s wish, by right and by the laws? (Verr. . . –)
A little later, thinking of the vulnerability of fatherless wards, he again identifies with fathers’ hopes for their posterity: ‘We have small children (liberos)’ (Verr. . . ).43 Manipulation of himself as a family man will be chiefly used in political speeches. Representation of his own family worked in the Catilinarians much as it was to work in –. Cicero could inspire the Senate to stand firm by pointing out how much he himself was risking. If Cicero’s story that the conspirators had tried to assassinate him at his morning reception is true, it was possible for him to claim that his wife and children might have been closely involved with violence (Sulla ). He is not too tough to be moved by the suffering of his family (Catil. . ) and he commends Marcus to the senators as the son of the saviour of the state (Catil. . ). He speaks repeatedly of the need to save the wives and children of his hearers (Catil. . , cf. ; . , , , , ). Later he was able to claim that he had achieved this (Mil. –). The rhetoric used by other senators matches Cicero’s. From late  to at least , every time Crassus saw his wife, house, and country, he remembered that he owed them all to Cicero. Or the keynote in perorations, but was used especially when the defendant faced exile (usually involving separation from family). 43 Tullia in  (aged about ) represented all his descendants, as did Annia for her father.
Cicero on Natural Affection

so he said, in a flattering speech in the Senate which Cicero relays with glee.44 Cicero will refer to his own family in speeches in particular circumstances and to heighten emotion. The Catilinarian crisis, allegedly threatening massacre and rapine to all classes, provides one type of context. Rejoicing after the end of the crisis of his own family makes up the pair. Had he spoken in public to oppose Clodius in the spring of , instead of creeping into exile, he would hardly have missed the chance to paint a picture of the suffering of his family as he had done for so many men accused in the lawcourts. Once he had recovered his house in , there was no need to bring his family into his speeches. But for years he continues to refer to the suffering inflicted on them by Clodius and the debt he owes to those who worked for his recall.45 He even claims in the Defence of Plancius in   to have earned the right, by his sacrifice for the sake of the commonwealth, to work for his own advantage and that of his family (Planc. –; cf. A. /. .  fin.). He can permit himself some personal remarks about his hopes for his son (Planc. ). In the Defence of Milo ( ), he is again ready to risk his and his children’s property for the sake of rescuing his old protector (Mil. ). By the time Cicero was again able to speak on great crises, his family was much reduced. The divorced husband of Terentia and Publilia could not introduce a wife, and Marcus represented all his liberi by the time of Cicero’s crusade against Antony. Marcus appears in a low-keyed statement that he has received the surrender of an Antonian legate (Phil. . ). Marcus’ senior and cousin, the younger Quintus, is staunchly defended against an attack by Antony which included the allegation that he had plotted to murder his father and uncle. Cicero declares the unshakeable harmony and affection of the family and his conviction of his nephew’s ability and good character in terms which contradict much of what he had said over the years in confidential discussion with Atticus (Phil. . –).46 A nephew must A. /. . ,  Feb. ; for the tropes cf. Pis. , Planc. . Allusion to mei and their sufferings: Dom. ; Sest. , ; Har. ; Cael. ; Balb. ; Planc. , –; Rab. Post. ; Mil. , . 46 For the relationships of Cicero, his brother Quintus, and Atticus, with Quintus’ son, the nephew of both Cicero and Atticus, as they varied over the period June  to July , see especially A. /. . , /.. –, 44 45

Susan Treggiari
be supported against Antony, for the discredit of an individual would discredit the family. How amazingly shameless, bold and rash to dare to write this against a young man whom my brother and I vie in loving because of his sweet and virtuous character and outstanding talent, and whom at all times we keep in our sight, hearing and embrace! (Phil. . )
Cicero gets a chance to idealize the strong family affection of all his surviving kin. Both individual virtue and a proper relationship of pietas and duty (officium) between members of the family justify the trust and respect of fellow-citizens. The orator could appeal to the common people as well as to senators or equites when he asked them to think of their own affections and sense of duty to their families.47 Conclusion It is obvious that many of Cicero’s digressions, which in a modern law-court would be ruled out of order, were a red herring to put the judges off the scent. The attack on Sassia, like those on Chrysogonus or Clodia, distracted attention from the actual charges. It was carefully judged to appeal to the judges’ prejudices.48 Here, then, talking about family allows Cicero to smother suspicions that his clients may well have been guilty (though not to avoid talking about their conduct as familymembers). The whitewashing of Roscius and Cluentius as family men is likely to be as false as the denigration of Sassia.49 Here I have used these two examples of Cicero’s discussion of parent–child relations to suggest what Romans’ gut-reactions were about reciprocal duties. Even where the circumstances in which a speech was delivered did not directly raise questions of /. . , /. . , /. . , /. . , /. . , /. . , /. . , /. . . 47 Dr George’s chapter documents the testimony of one section of the lower classes to their assertion of family affection and solidarity. Warmington : nos.  and  are vivid examples. 48 Especially fears about the conduct and influence on family members and property of wealthy women who married several times 49 But it is important to remember that Cicero was not usually the only advocate for the defence. His role was especially to arouse the judges’ emotions. It was his job to play on their feelings and beliefs.
Cicero on Natural Affection

‘how people should live with their family’, he often appeals in passing to ‘the common feeling of humanity’, especially about the innocent young and about obligations and love between relatives. It is these almost incidental passages which are in the end more significant. Certain purple passages of pathos, for instance, though highly rhetorical, follow lines of thought which seem natural to both speaker and audience. They allow a crescendo of emotion. They suggest that there were accepted beliefs and experience of life common to Cicero and his hearers, unquestioned and pervasive. The family was central to people’s concerns.
This page intentionally left blank
 Family Imagery and Family Values in Roman Italy Michele George
A the best extant visual imagery of families in Roman art is a form of funerary commemoration that emerges in the archaeological record around the beginning of the first century . Appearing first at Rome, and then throughout Italy, with its zenith falling in the years of the late Republic to early Augustan era, the formal and minimalist genre features relief portraits of individuals presented in groups and arranged frontally to face the viewer. The duo of husband and wife is the most common composition, but the trio of parents and child is well represented, and there are also examples which appear to fall outside the nuclear family model, in which an inscription, facial resemblance, or gesture indicates a degree of group identity. At Rome, the genre was apparently developed for the late Republican freedmen who emerged in this period as an increasingly wealthy and influential social group. Commemorative reliefs that feature family portrait groups are found throughout Italy in the first century , and, although less well-preserved and smaller in number than their counterparts at Rome, they were probably inspired by those in the capital. At Rome they were originally displayed on the exteriors of chamber tombs in the cemeteries that were strung out along the major roads leading into the city, where, with rare exceptions, they commemorated libertini.1 By contrast, outside Rome the genre was more often used on grave stelae and was adopted by the freeborn as well as former slaves. Both libertini at Rome and the rising local elites of the provincial north saw in the family motif an effective image through which they could display their 1 For a view of the solitary example still in situ at Rome, see Kockel : pl. a.

Michele George
social ascendancy and lay claim to a public profile, albeit one conditioned by practical and cultural limitations. The following discussion considers the meaning of the family collective in this funerary imagery and its common appeal for two diverse social groups. Relief Portraits of Freedmen from Rome Freedmen and the family To a contemporary audience, the relief portraits from Rome are a familiar sight due to their frequent use as illustrations of the ‘typical’ Roman family, generally without mention of their libertine ownership. It is at once ironic and yet completely logical that former slaves, individuals whose origins were not Roman but were very likely in this period to have been Greek, and whose path to manumission probably involved experiences which differed considerably from that of the ‘average’ Roman, were responsible for the production of images so often viewed in the modern era as quintessentially Roman. Scholarship on the reliefs has generally focused on stylistic aspects of the portraiture while giving some attention to their expression of status and legitimacy.2 Building on these studies, the discussion of these monuments can be reframed in order to examine more closely the implications of the slave experience to their form and contents, and in particular the role of the family motif to their commemorative message. The importance of this material as the earliest and most explicit evidence for libertine self-representation should not be underestimated. Unlike subsequent forms of commemoration, the epitaphs on these reliefs consistently include the epigraphic indicator of libertination (‘l’ for libertus), making them 2 On the reliefs from the city of Rome: Zanker , Kleiner , Volpi /, Kockel , Koortbojian . Two related groups of portraits, portrait busts in the round and the full-figured standing type, are sometimes included with the reliefs, but are fewer in number, e.g. in Kockel relief portraits represent % of the total, while portrait busts in the round make up % and the standing type only % (Kleiner includes the standing type but omits portrait busts). I have chosen to focus on the panel reliefs because they constitute the largest group, and, in contrast to the busts or standing type, they much more often have accompanying inscriptions which identify them as belonging to libertini.
Family Imagery and Values in Roman Italy

the only extant group of funerary monuments that can be identified so closely with freedmen. The reliefs from Rome enable the isolation of a distinctive commemorative genre datable to a relatively limited time period that can be attributed unequivocally to freedmen, a group that occupied a unique position in the social hierarchy of the capital in the late Republican era.3 A number of factors made the family motif an attractive choice of memorial for Republican freedmen. First, there is the centrality of the family in Roman life. A critical element in the formation of Roman identity and citizenship, the family provided protection, economic and emotional support, and was the institution through which wealth and property was protected and transmitted. In Roman thought the strength of the family reflected the stability of the state, making membership in a family membership in the polity of Rome itself writ small.4 Belonging to society was especially important for freedmen, who as slaves had been considered property and less than human, and who were eager to display their new status as Roman citizens. Even after manumission, however, libertini in the late Republic faced restrictions because of their social status that were not encountered by freeborn citizens, and the disgrace of former slave status continued to define the freedman’s life in numerous ways. In addition to owing work and allegiance (operae et obsequium) to their former masters, freedmen were hindered in the full engagement in the most prestigious activities of Roman public life, and their economic success could not be matched with a parallel strength in the gaining of honos, a major element in the biography of the elite. The formation of autonomous families with legitimate children, who were born with full Roman citizenship and 3 Over a third of the  portraits in Kockel’s catalogue are on complete (or almost complete) relief panels; these form the focus of this discussion. Of these, over half have inscriptions, and with two exceptions they indicate either libertine ownership or, as in  cases (e.g. Kockel : J, L, L, here Fig. .), a freeborn man married to a slave woman. The altars and urns that dominate funerary commemoration in the st cent.  were used not only by freedmen and their descendants, but also ingenui (with filiation), including even a few members of the senatorial and equestrian orders ( and  examples respectively; see Boschung : ). See also Kleiner , who focuses on altars with portraits. 4 e.g. Cicero says of the family (Off. . ): ‘id autem est principium urbis et quasi seminarium rei publicae’. See Treggiari in this volume.

Michele George
to whom they could leave an inheritance, was a crucial achievement for freedmen who had few ways of attaining public distinction. Moreover, the display of this accomplishment before the public gaze in family tombs was itself a form of participation in public life and one of the few permitted to late Republican freedmen. For them, the declaration of a family identity on funerary monuments was the assertion of an agency that was denied them in other spheres of Roman life.5 This is not to suggest that the family was more highly valued by freedmen than by freeborn Romans, but rather that the family was valued within a distinct set of realities. The heightened appreciation of family life by libertini must be viewed against the background of the slave condition from which they had been delivered, for behind these commemorations also lie the remnants of the slave experience which shaped the freedman mentality.6 Natal alienation, the denial of any claim to a personal family history and one of the basic premises of Roman slavery, meant that any family ties a slave might have had before enslavement were not recognized by Roman law. Domestic slaves, who were likely the majority among those manumitted, lived with the master and his family within the context of family life, and the structural and affective consequences of that experience shaped their attitudes and aspirations. Providing the slave with, in Pliny the Younger’s words, ‘a country and a kind of citizenship’, the domestic household was a critical locus for the socialization of slaves before manumission and served as a fulcrum for shaping the social values of servus and libertus. Within the master’s domus, family structure was both model to follow and, for many slaves, the mode of lived reality. Epigraphic commemorations between contubernales, slave partners in the informal, quasi-marital unions (contubernia) that slaves often formed, utilized conventional terms for ‘husband’ and ‘wife’, revealing slaves’ own view of their bond, despite the legal disqualification.7 5 On libertini generally, see Treggiari , Duff , Fabre . For the epigraphic evidence of freedman commemoration, see Rawson , Treggiari ; legal limitations on freedmen, Treggiari : ch. . 6 See Shaw . 7 Pliny, Ep. . . –. Natal alienation: Patterson ; slave marriages: Treggiari ; slave families: Bradley : –.
Family Imagery and Values in Roman Italy

Such relationships existed, however, only with the slaveowner’s permission and primarily to serve his needs. While a symbol of security and stability in Roman culture at large, family life for slaves represented precisely the opposite, a precarious arrangement that was utterly unstable, perpetually vulnerable, and under constant threat of disruption. Their informal marriages could be broken up at any time by the sale of one of the partners, and the practice of breeding slaves for profit meant that they could be separated from their children at the slave-owner’s convenience.8 The very real threat of family break-up and the psychological state of uncertainty in which such familial bonds developed in servitude must have made the attainment of a family after manumission a profound pleasure beyond the comprehension of most freeborn Romans. Family values Besides the public affirmation of legal and social legitimacy, family imagery resonated with a set of associated values and behaviours which were particularly attractive to former slaves. This included many of the core ideas in the Roman value system which were centred on the family, such as pietas, ‘affectionate devotion’, and archetypal social roles for men and women which were constructed around the family unit, such as the paterfamilias and the matrona. Furthermore, there existed the romantic myth of the archaic Roman family which, it has been argued, occupied an important place in the Roman mindset. Harking back to the early Republic, this idealized family consisted of stock members who exhibited the stereotypical characteristics on which the Romans prided themselves: the severe paterfamilias, guiding his brood with a firm but loving hand; the matrona, exemplar of feminine docility and faithfulness; the children, loving bulwarks for their parent’s old age and the promise of future family glory. Virtues with special significance in the family context—the auctoritas of the paterfamilias, the castitas of the matrona—are highlighted in this model, and furnished the moral backdrop against which these reliefs should be set. As a cultural reference point, this mythic Roman family was the notional standard, setting the bar 8
Slave breeding: Bradley a: –.

Michele George
for family conduct to which all, including freedmen, could aspire.9 The moral dimension and the stress on propriety in the iconography of these reliefs suggest that a broader moral legitimacy, beyond the public acknowledgement of their legal status, is also at issue in these commemorations. Implicit in this iconographic declaration of respectability was a claim to membership in the moral realm of Roman society, for the transformation to libertus meant a radical difference in cultural perceptions and expectations. Beyond the benefits of legal status there were critical conceptual distinctions between servus and civis. Slaves were believed to be morally degenerate and inconstant by nature, prey to irrational behaviour, and eminently corruptible; the two virtues proverbially considered most desirable in a slave, fides (loyalty) and obsequium (obedience), were also the scarcest.10 In the circular logic typical of ancient rationalizations for slavery, it was the slave’s inborn moral inferiority which justified his enslavement in the first place. The prejudices engendered by this view of the slave persisted in popular attitudes toward freedmen, to whom were applied the standards of servile behaviour, despite their legal status, rather than the higher moral behaviour expected of the freeborn. Cicero implies that good moral behaviour from libertini was worthy of comment simply because it could not be expected as a matter of course. Placing the greatest importance on conduct toward their patrons (and former masters), he uses the previous servile condition of libertini as the basis for evaluating their character, and writes approvingly of those who possessed probitas (modesty), who were frugi (honest), and above all who showed benevolentia (goodwill) and fides (fidelity) in fulfilling their obligations to their former masters. ‘Bad’ freedmen, by contrast, were distinguished by their superbia (arrogance) and improbitas (lack of modesty).11 In tarring them with the same brush as 9 On the concept of pietas in the family context, see Saller : , whose translation of the term I have used; for the sentimental ideal of the Roman family: Dixon . 10 A view Keith Bradley has called the ‘convention of the criminous slave’ (Bradley : –). For the theory of natural slavery, see Aristotle, Politics . b –; Garnsey : –. 11 Cf. Cicero’s comment (Pro Sex. Rosc. ) that every domus had a few dishonest slaves and freedmen, thereby grouping them together and viewing them
Family Imagery and Values in Roman Italy

slaves, Cicero reveals the fundamental liminality of the freedman, who occupied an ambiguous and temporary position somewhere between servus and dominus, the two opposed poles on the spectrum of status and morality in Roman thought. Despite enjoying the pleasures of freedom which most slaves never knew, libertini could not fully escape the stigma of slavery and the suspicion of their essential and irreparable inadequacy in the eyes of the elite. As the head of his own household, however, a libertus could fulfil on equal terms one of the few affirming social roles he shared with the freeborn, by wearing the toga, fathering freeborn children who had Roman citizenship, and being a good husband to a good wife. Part of the appeal of family imagery was therefore the moral validation it carried, and the public repudiation of the servile stereotype and all its pejorative associations that lingered in the popular perception of freedmen. Accordingly, family members on the reliefs are clothed in the standard attire of the Roman citizen, as men wear the toga, the garment of Roman public life which was forbidden by law to non-citizens, and women the tunic and palla of the Roman matron (Figs. .–). When the imagery expands to include offspring, it is initially only adult children who are shown (e.g. Fig. .), while young children do not appear until the Augustan age (Fig. .).12 They too display their social status through clothing and attributes, with children of both sexes attired in the toga praetexta while boys frequently sport the bulla, a circular locket of metal or leather that was another item restricted by law, in this case to freeborn youths.13 Demonstrations of affection between parent and child occur, but only in a minority of examples; in general, the children on these reliefs are portrayed with the same with the same low expectations. On the prejudices toward freedmen, see Treggiari : –; on the positive and negative stereotypes of freedmen, see Fabre : –. 12 Fig. .: relief of the Maelii, Kockel : K , dated to the mid-Augustan era. Inscription: Sex. Maelius Sex. l. Stabilio Vesinia (mulieris) l. Iucunda Sex. Maelius Sex. l. Faustus. Fig. .: relief of the Vettii, Kockel H , dated to s . Inscription: Antonia P. l. Rufa C. Vettius (mulieris) l. Nicephor C. Vettius C.f. Secundus Vettia C. l. Calybe. 13 e.g. Kockel : M . For the toga praetexta and bulla: Gabelmann ; bulla only: Palmer . The bulla could be worn only by the sons of freeborn men until the mid-nd cent. , when the right to wear it was extended to the freeborn sons of freedmen.

Michele George
formality and gravity as their elders.14 The recurrence of the dextrarum iunctio, the symbolic joining of right hands, conveys the attainment of legally recognized marriage, but also symbolizes concordia, marital harmony, the proverbial desideratum of Roman marriage, as well as related marital values such as fides, mutual loyalty and trust, and comitas, amiability.15 The Gratidii, for example (Fig. .), are inclined toward each other, the woman’s left hand resting lightly on her husband’s shoulder, their right hands clasped in the conventional gesture, thereby conveying not only the simple fact of their union’s legitimacy, but also the successful forging of the kind of marital bond to which freeborn Romans aspired in their marriages.16 The same concern for social respectability is expressed in the portrayal of the women on the reliefs through gesture and costume. Adopting the attitude and attire of the matrona, the cultural paradigm of virtue for Roman women, they signal their chastity through the so-called pudicitia gesture, in which the hand holds the palla as if about to veil the face and hide it from an intrusive gaze.17 With the palla sometimes covering their heads, they evoke an attitude of modesty befitting the most conservative view of female behaviour.18 Beneath the palla, some women on these reliefs wear the vitta, the woollen hair band which was closely identified with the Roman matron and her proverbial purity.19 The defining garment for the Roman 14 Gestures of affection, e.g. Kockel I , M, M. Kleiner  attributes the increased appearance of young children and affective gestures to the influence of Augustan art and social policy. See also Rawson a. 15 This was not the exclusive meaning of the gesture; for examples beyond married couples, see Davies , who does not, however, address the meaning of the gesture for this particular status group; see also Kockel :  n. . 16 For the effectiveness of these portraits in expressing Roman virtues, see Koortbojian : –. On the Gratidii, see Kockel : L . An inscription (CIL . , Gratidia M. l. Chrite M. Gratidius Libanus, now lost) identified them as Marcus Gratidius Libanus, a freeborn Roman probably descended from a former slave, and his wife, Charite, who was probably his own liberta. On Roman marital values: Treggiari : –. 17 e.g. Figs. .; .; ., central figure Fig. .. 18 e.g. Figs. ., ., ., second from left; cf. the story of C. Sulpicius Rufus, who divorced his wife for leaving the house with her head uncovered (Val. Max. . . ). 19 e.g. Fig. ., central figure; Fig. ., first woman on left. On the vitta and other forms of head adornment, see Sensi /.
Family Imagery and Values in Roman Italy
. .

Funerary relief of the Gratidii, Vatican
matrona, however, was the stola, a slip-like garment worn between tunic and mantle. In literature the stola goes hand in hand with the vitta as insignes pudoris, symbols of modesty, as Ovid puts it in his introductory admonition in the Ars Amatoria. Associated in funerary and honorific inscriptions with legitimate marriage, both items were synonymous with the respectability which marriage conferred on women, and with the moral rectitude it demanded of them.20 Although it constituted the closest sartorial equivalent for women to the toga for men, the stola was more exclusive still, for by law only freeborn Roman women had the right to wear it, a restriction that was amended to include freedwomen who were 20 e.g. the honorific expression matrona stolata or femina stolata (‘a stola’d woman’) in inscriptions; see B. Holtheide, ZPE  (), –. Vitta and stola: Ovid, Ars Amatoria . –; on freedman reliefs: Kockel : –; stola on honorific statuary: Scholz ; J. L. Sebesta, ‘Symbolism in the Costume of the Roman Woman’, in Bonfante and Sebesta : –.
. . Funerary relief of the Maelii, Raleigh, NC
. . Funerary relief of the Vettii, Rome
. . Funerary relief of the Furii, Vatican
Family Imagery and Values in Roman Italy

. . Funerary relief—Rome, Palazzo dei Conservatori
married to ingenui during the second century .21 If the scant ancient sources are correct, strictly speaking the women on these reliefs were excluded by law from wearing the stola because of their husbands’ freed status; in fact, the stola is absent except in rare cases (e.g. Fig. ., first woman from left).22 The same distinction does not seem to have applied to the vitta, however,
Festus .  L; Marcrobius Saturnalia . . –. Kockel points to the possible identification of the stola on a small number () of these reliefs (Kockel : –). It is difficult to identify stolae in all these cases, whether from photos or on-site autopsy; however, if accurate the identification can be explained in several ways: (i) there was a relaxation of the legal restriction in the late Republican period, and the stola was in fact worn by libertinae, which does not, however, explain its scant representation; (ii) noting a decline in the appearance of the stola on reliefs of the Augustan age in general, Kockel suggests that it had lost its force as a symbol of legal status in honorific and commemorative statuary, even if it kept its connotations in literary 21 22

Michele George
suggesting that the depiction of freedwomen on these monuments exploits as far as possible the elements of the matron’s clothing. The indictment of immorality imposed on slaves was an especially troubling burden for slave women, who were limited by their gender to the domestic sphere and whose worth was established by their good reputation. Believed to be promiscuous and morally depraved, slave women represented the precise opposite of the pristine matrona on the moral spectrum. Although credited with sexual licentiousness, in reality they had no control over their own bodies and were at risk of sexual exploitation by their masters. By adopting the attire and manner of the archetypal matrona, the Roman model of female propriety and sexual restraint, freedwomen could reform the moral universe they had inhabited as slaves and declare their willingness to be judged by the same standards of decency as freeborn women.23 Some reliefs represent several adults who appear to share some form of group identification but who do not conform to the patterns of lone married couple or nuclear family.24 The inscriptions attached to some of these reliefs suggest that the individuals depicted were probable conservi, slaves from the same household, sometimes with a sufficient resemblance to prove a blood relationship, as in the case of the Furii, another familiar example (Fig. .).25 The inscription suggests that the first four individuals (left to right) were slaves in the household of the Furii, while the fifth was previously owned by the Sulpicii. Even with the contexts; (iii) or, alternatively, in view of this decline in Augustan relief, that there was a stricter (if not absolute) enforcement of the regulation in response to the emperor’s concerns about status-orientated clothing. It is also possible that the tunic and stola were differentiated from each other on sculpture with paint which does not survive on existing examples. 23 The protection clothing brought was literal as well as figurative, for women who scorned the matron’s garments exposed themselves to the risk of harassment in public, a danger that was less likely for a woman who was easily identified as a matrona (Ulpian, Digest . . . ) On this passage, McGinn : –, George : –. For conceptual parallels between the matron and the slave, see Parker . 24 e.g. Kockel : E , F , G , I , L . 25 Kockel : G , dated to s . Inscription (CIL . ): Furia (mulieris) l. P. Furius P.l. Furia (mulieris) l. Furia (mulieris) l. C. Sulpicius C. l. Unique to this relief is the absence of cognomina; the most prominent and repeated name is that of the gens Furii.
Family Imagery and Values in Roman Italy

facial similarity and nomenclature, however, it is impossible to ascertain the relationships between these individuals. The three women might be sisters, and in fact they share the same physiognomy; it is possible that two of them are shown inclined toward their husbands, with an unmarried sibling on the far left.26 In examples with no extant inscriptions sorting out the possible permutations of the relationships between figures is even more difficult. On a relief in the Palazzo dei Conservatori (Fig. .), six figures are pictured in a row, with no obvious connection through gestures or facial resemblance.27 They might be two families, each a husband and wife with an adult son, or two married couples with an unmarried brother. The representation of non-nuclear groups on portrait reliefs points to the extended network of relationships beyond contubernia which arose among domestic slaves within the house and which were maintained after freedom was obtained. The impression is supported by epigraphic evidence in which the terms collibertus or conservus, rather than vir or coniunx, are used even when a marital relationship is implied, suggesting the particular importance of the shared slave experience to individuals even after manumission.28 The choice to be commemorated with siblings or colliberti might have been especially attractive to freedmen who were childless, a circumstance possibly common to many due to manumission later in life.29 Group reliefs might also reflect an early manifestation of burial collegia, organizations of individuals without families of their own who joined together to ensure proper burial and funerary ritual for themselves.30 By drawing on family 26 Alternatively, the high degree of similarity in their faces could be attributed to the generic output of a particular workshop rather than consanguinity. For further possible kin relationships, see Kockel : . More complex is the relief of the Servilii and Scaevii, whose panel depicts the portraits of three men and two women, but whose lengthy inscription names nine individuals whose connection to each other is far from clear (Kockel : J , CIL . ). 27 Kockel : F . 28 Flory , especially –. Although it is generally impossible to date the epigraphic evidence, and so tie it to the late Republic/early imperial era, it does indicate the significance of slave life to freedmen. 29 On the reduced fertility of slave women, see Treggiari : –; contra, Fabre : –. 30 The standard work on collegia is Waltzing; see also Joshel : –.

Michele George
imagery, individuals from the same (or even different) familia could claim for themselves the associated normative values of social respectability while commemorating relationships which fell outside the nuclear family model. In its conveyance of notions of propriety and stability, values especially associated with the family, such imagery endowed these relationships with the social worth of more conventional familial arrangements. Freedman self-representation Following the custom on most late Republican epitaphs, the dedicants on these reliefs are unspecified, and the inscriptions function as labels that assert the identity of each individual with no indication of the circumstances surrounding the erection of the tomb.31 In only a few instances are other forms of status or identity given visual expression. For example, the fasces, symbols of involvement in the cult of the Augustales, appear only twice on reliefs with family portrait groups, while in other cases titles of civic honour are included in the text but have no place in the imagery.32 References to work are also subordinated to the family motif, with tools appearing on only three family groups, while occupational title is given in two inscriptions, but without accompanying imagery.33 In a few instances, the portraits are of patrons, put up and paid for by their libertini.34 It is possible that 31 Also absent are the abbreviations such as ‘DM’ for Dis Manibus, which became standard in later funerary monuments. In only a few cases is reference made to which of the figures were deceased and which were living at the time of the monument’s erection, e.g. Kockel : H , here Fig. ., where there is a ‘v’ for vivit under the portraits of Nicephor and Calybe, indicating that the freedman father Nicephor and his daughter Calybe, herself a freedwoman who was probably born while her parents were still enslaved, survived the other two figures on the relief, Antonia Rufa, probably Nicephor’s wife and Calybe’s mother, and the young boy Secundus, the only ingenuus in the family. On others the epigraphic marker of the theta nigrum indicates that certain individuals were deceased. See also Kockel : A , B , D , E , F , F , H . 32 Fasces: Kockel : G , L . Kockel :  lists the handful of cases where engagement in the imperial administration and cult is included in the inscription. For libertini as staff to Roman magistrates, see N. Purcell, Papers of the British School at Rome,  (), –. 33 Kockel : . For work imagery on early imperial freedman reliefs, see Zimmer . 34 Kockel : C , K , L , L , and L.
Family Imagery and Values in Roman Italy

these patron portraits were rendered in this genre because they functioned not as funerary commemorations to patrons, but as honorific dedications, possibly built into the freedman’s tomb along with images of his own family. In constructing substantial family tombs situated in prominent public contexts, late Republican freedmen were participating in a form of self-representation that burgeoned in this era among a range of social and economic groups.35 Although borrowing aspects of gesture, costume, and portraiture from the conventions of elite self-representation, the genre presents a combination of these elements in a form that was innovative in Republican art. There is no single model for the genre, which shares characteristics with commemorative formats from both Hellenistic and Roman art.36 Relying on a rudimentary, schematic treatment of the external features to express individuality and family resemblance, these portraits of Republican freedmen employ a rough Italic verism and draw on standardized Republican portrait types, in contrast to some contemporary elite portraiture which often favoured the Hellenistic approach to the medium, with its interest in psychology and idealization.37 Similarly, the reliefs eschew heroic nudity, another trend favoured by the more Hellenophilic of the elite, giving preference instead to the traditional dress of commemorative sculpture, the garments of the Roman citizen— toga and tunic for men, palla and tunic for women—which fixed them firmly and unmistakably on Roman soil. Despite the contrast with contemporary elite monuments, the choice of this mode of representation should not be misconstrued as a lack of sophistication among freedmen, but rather as a preference for the stylistic idiom which presented them as Roman as possible and 35
.
On the rise of funerary commemoration: Zanker and von Hesberg
36 Possible inspiration can be found on one type of Hellenistic stela of the nd cent. , which differs from classical Greek models in the formality of its composition and the lack of interaction between the figures (see Zanker ). It has also been suggested that the format is a visual reference to the imagines, the ancestral portraits of the elite. See Kockel : –, with further bibliography; on the imagines and their exclusivity to the elite, see Flower . 37 On the style of the portraits, in addition to the works cited in n. , see also Zanker , , : –; Borg .

Michele George
in the most conventional and conservative terms. In creating a public persona of the most staid Roman kind they could hope to mitigate, although not hide, their servile origins. The remarkable consistency in theme and motifs on these reliefs suggests that on some level a collective identity had emerged for freedmen in the late Republic, or at least a common set of priorities which arose from a shared servile past and the realities of life after manumission. Relative to their blossoming strength in the early imperial era, libertini with considerable wealth were still somewhat new on the social scene of the late Republic. In attempting to assert a public identity at a time when their place in the tightly structured social order of Republican Rome was still ambiguous, freedmen were faced with an awkward problem. In contrast to the confidence and outrageous excess of Petronius’ infamous freedman Trimalchio, Republican freedmen were pioneers in formulating forms of commemoration and selfrepresentation that were appropriate to their unique status, that reflected the achievements they most valued, and that would also be acceptable to the wider society. Unlike the brash caricatures in the Neronian Satyricon, this generation of libertini was still engaged in the process of defining what it meant in Rome to be a successful ‘freed man’, a former slave who through his own efforts and his master’s good graces had navigated the often tortuous path to manumission. The relative homogeneity of the genre, therefore, might reflect not only common values, but also a lack of sufficient confidence to express a strong element of individuality in personal commemoration. Status could be asserted, but only within the safe confines of a genre based on the most traditional Roman ideals and presented in a highly formalized arrangement. The hesitancy to promote rugged individualism in these reliefs hints at the ambivalence of the social valuation of even wealthy freedmen and a degree of caution that influenced their initial commemorative choices. Much of this had changed by the end of the Julio-Claudian era, however, when freedman commemorations broadened considerably, moving from group to individual memorials and from the relative conservatism of family portrait reliefs to embrace a mix of eclectic themes. The most famous and most striking example of this new trend, the tomb of the baker Eurysaces beside the Porta Maggiore in Rome, best illustrates the willing-
Family Imagery and Values in Roman Italy

ness to minimize, if not depart entirely from, the family motif.38 Compared to the traditional, formal, and morally laden family portrait groups, the tomb of Eurysaces is a clear assertion of individual will and identity that transgresses the late Republican ideals of probitas in freedmen to fall squarely into the realm of superbia. Reflecting a radical shift in freedman commemoration, the tomb of Eurysaces anticipates the expanded and more idiosyncratic repertoire of funerary imagery that emerged in the libertine monuments of the first century , when mythological allegory becomes the pre-eminent choice.39 The strong presence of freedmen in Roman commemorative evidence in general has been characterized as an overrepresentation related to a heightened social pressure to strive for honour (honos) and to display their worldly success obtained against great odds.40 The overwhelming preference for family portrait reliefs among late Republican freedmen reflects a desire to display their embodiment of fundamental Roman moral values at the same time as they commemorated social status and affective bonds with family members. Their faith in the efficacy of the family motif to convey honos underlines the family’s primacy of place in the Roman social order while also revealing the lingering consequences of former servile status. Beyond Rome: Cisalpine Gaul Of the portrait reliefs from outside Rome, those from central and southern Italy are in the poorest condition and few have extant related inscriptions. In form, however, they are similar to the genre at Rome, with horizontal family groupings of two or more individuals wearing conventional Roman dress. Their chronological range is confined to the late Republic to Julio-Claudian period, and their dedicants were mostly libertini, although a
38 Ciancio Rossetto . Kockel has cast doubt on the conventional association between the tomb and the standing relief of a man and woman who have been identified as Eurysaces and his wife Atistia (Kockel : A , with bibliography). 39 Wrede . 40 Von Hesberg : .

Michele George
significant number of ingenui are also represented.41 Family portrait reliefs from Cisalpine Gaul (northern Italy) comprise a smaller corpus than the freedman reliefs at Rome but exist in sufficient quantity and condition to allow a brief comparative study.42 Appearing initially in the mid-first century , most are dated from the Augustan era to the end of the first century , with the majority falling within the Julio-Claudian to Flavian eras. The composition of the Cisalpine reliefs, which is more often vertical rather than horizontal, is the most obvious visible divergence from the Roman evidence. In many cases the vertical arrangement is used to reflect directly different forms of social hierarchy within the domestic context. In the relief of the Alennii from Bologna (Fig. .), the family group is ordered into vertical registers on the basis of age and rank: in the upper register, a freedman couple, below them their two freeborn sons, followed by their daughter, and beside her a freedwoman.43 The positioning of figures follows clear lines of status as dictated by both age and rank, as parents are placed above their children, male children above female children, and the freeborn daughter before a liberta. Not only does this hierarchical arrangement demonstrate the calibrated status of household members, it also puts visual emphasis on the successful formation of more than one generation of the family.
41 On portrait reliefs from central and southern Italy: Frenz , Eckert ; from northern Italy: Pflug , Mercando . Frenz () lists only  examples from the whole of southern and central Italy (excluding the city of Rome), compared with  extant (and  now lost) from Cisalpine Gaul (Pflug ). 42 Cisalpine Gaul: Pflug ; also Mercando . Outside Italy, funerary portrait reliefs most often depict single individuals, with a few married couples; family groupings are unusual, and occur with frequency only in Pannonia and Dacia, where they are dated to the late nd to rd cents. . For reliefs in Pannonia, see Boatwright, this volume; Dacia: Teposu Marinescu ; Bianchi ; for Spain: Edmondson, this volume; also Edmondson, Nogales Basarrate, and Trillmich ; for Gaul: Braemar ; for the Rhine region: Faust . 43 Pflug : no. , dated to mid-st cent. . Inscription (CIL . ): V(ivus) f(ecit) sibi et suis L(ucius) Alennius L(uci) l(ibertus) Stephanus sibi et Freiae M(arci) l(ibertae) Euphemi(ae) L(ucio) Alennio L(uci) f(ilio) Celeri T(ito) Alennio L(uci) f(ilio) IIIIII viro filis Stacte l(ibertae) Saturninae f(iliae) Q(uo) q(uo) v(ersus) p(edes) XVI.
Family Imagery and Values in Roman Italy
. .

Funerary relief of the Alennii, Bologna
In other examples the nuclear family is represented by a tightly grouped trio, as father and mother flank a small child as if enveloping him in a parental embrace (e.g. Figs. ., .).44 In general, however, the vertical format reflects critical distinctions in the relationship between the individuals included in the groupings, since the Cisalpine reliefs tend to use membership in a 44
Also Pflug : nos. , , .

Michele George
wider family household to illustrate their social status. Unlike the reliefs from Rome, dependent freedmen are sometimes included on these reliefs as if part of an extended family. For example, on a relief now in Ravenna a libertus commemorates in the top register his own nuclear family—himself, his wife, and his small son—as well as four of his male freedmen, who are arranged in two lower registers, thus illustrating both his role as a paterfamilias and as patron to other former slaves.45 In some cases, as on another stela from Ravenna (Fig. .), a complex set of household ties is suggested, not all of which can be untangled. At the top, there is the bust of a liberta who is probably patron to the others; beneath, a nuclear trio of father, mother, and child (the parents are freedmen, the child’s status is unspecified); beneath this, two freeborn men, whose nomenclature is unrelated to the others, but whose cognomina suggest servile ancestry; lastly, at the bottom, a small boy named Speratus, who is labelled a verna, a houseborn slave. Although the three adults in the upper half share the same praenomen and nomen, their relationship to the two freeborn men is not clear; it seems likely that the latter were attached to the household in some way, perhaps as employees in the family business. Their inclusion along with the verna, however, hints at a potential affective relationship between them and the dedicant, who was probably the woman holding the child.46 More significant than compositional differences between Cisalpine and Roman family reliefs is the status of the dedicants, who were not chiefly freedmen. On the contrary: of those with related inscriptions, fully a half belonged to ingenui, freeborn citizens who supply their filiation;  per cent belong to libertini, Pflug : no. , dated to  –. Pflug : no. , dated to c.  –. Inscription (CIL . ): P(edes) XX Firmia L(uci) l(iberta) Prima L(ucio) Firmio L(uci) l(iberto) Principi — Firmia L(uci) l(iberta) Apollonia Lezbiae filiae sibi et suis de pecun(ia) s(ua) v(iva) f(ecit) M(arco) Latronio Sal(vi) f(ilio) Secundo — Sal(vio) Latronio Sal(vi) f(ilio) Saturnino Sperato verna. Firmia Apollonia could be either patron or colliberta to the other libertini; Pflug prefers patron because she appears to be older than the others. The identity of Prima is also unclear; she might be the female bust at the top of the relief, and could be sister or colliberta to Apollonia. The two freeborn youths could be sons of Apollonia or Prima by a man who is not included, but other possible explanations exist. The verna could be a favoured juvenile slave who died young, or the illegitimate offspring of his master. 45 46
Family Imagery and Values in Roman Italy
. . Funerary relief of the Firmii, Ravenna


Michele George
who indicate libertination, and  per cent fall to incerti, those whose status is unspecified, but whose cognomina hint at a servile ancestry.47 This is a significant departure from the reliefs at Rome, on which ingenui appear in less than a handful of examples. Aside from indications of social status, however, there is little to distinguish between the Cisalpine stelae of ingenui and libertini in form and content. Much like libertini in the region, the ingenui from the north often memorialized an extended household, including dependent freedmen: so P. Arrius Montanus (Fig. .) commemorates not only his own nuclear family—himself, his wife, and his son—but also his sister-in-law and her husband (in the second register), and two of his own freedmen at the bottom.48 The relative distribution of freeborn and freed dedicants reflects the spread of urbanization in the region. There are more ingenui than libertini from the west Po plain and beyond (e.g. Lombardy and Piedmont), which in this period was generally more agricultural and had fewer towns, whereas freedmen dominate on the stelae from Emilia-Romagna, where were located some of the oldest urban foundations in Cisalpine Gaul, including Bologna, which was founded in the early third century , and the harbour towns of Ravenna and Rimini. It was in these urban centres in particular, with their thriving local economies and active commercial life, that freedmen flourished, whereas ingenui were more prominent in the social hierarchy of towns with a rural economic base.49 The greater focus on internal domestic hierarchies and on the representation of the extended household, rather than the nuclear family, can be partially attributed to the function and context of the reliefs. The family portrait reliefs from Rome were attached 47 Pflug : –. The remaining % is made up of peregrini (%) and servi (%). 48 Pflug : no. , dated to c.  –. Inscription (CIL . ): P(ublius) Arrius P(ubli) f(ilius) Montanus Mocazia Helpis uxor P(ublius) Arrius Pollux Q(uintus) Decimus Dacus, opt(io) de(triere) Pinnata Moca[z]ia Iucunda u(xor) P(ublius) Arrius P(ubli) l(ibertus) Primigenius P(ublius) Arrius P(ubli) l(ibertus) Castor. 49 On the history and culture of Cisalpine Gaul, see Chilver ; on the Romanization of the region, see Denti ; on the local middle class, see the assorted articles in Sartori and Valvo .
Family Imagery and Values in Roman Italy
. . Funerary relief of the Montani, Ravenna


Michele George
to larger monuments and might have been augmented by additional inscriptions which listed household slaves and dependent freedmen who possibly were included in the tombs. Conversely, in general the stelae from Cisalpine Gaul functioned for the most part as funerary stelae used to mark off family tombs of more modest scale and constituted a complete, self-contained commemoration, with no other opportunity for a secondary related inscription. This might also explain why both freeborn and libertine stelae often include substantives which indicate familial relationships (mater, pater, uxor, filia, etc.), an element only rarely seen on the reliefs from Rome itself.50 It is also likely that the presence of household freedmen (and in one case a slave) on these reliefs mandated the clarification of kin and status relationships in obvious terms, so that members of both the nuclear and extended family, as well as the domestic familia, were clearly distinguished from one another. The inscriptions on the Cisalpine reliefs are more typical of funerary epitaphs of the imperial era, naming a dedicant and the family members and dependants for whom the marker was erected. Most often the dedicant is the male head of the household, although sometimes it is his wife or children who set it up for themselves, deceased relatives, and dependants. Frequently, indications are given of which individuals among those depicted were alive and which were deceased when the monument was erected. On Fig. ., for example, the dedicant Stephanus is labelled vivus, there is a theta beside the portrait of his wife and the liberta Stacte, and the letter ‘v’ beside his two sons and daughter Saturnina. In unusual cases former slaves dedicated stelae to their former masters and his family, as well as to themselves, and in one case the patron and libertini shared the cost. On two examples (e.g. Fig. .) a bust of the patron sits at the top of the panel, isolated from the main family group, but still loosely connected to it and, by virtue of its position, superior in status.51 50 e.g. Pflug : no. . On kinship terms, see Pflug : –. At Rome, e.g. Kockel H . Kinship terms are also used on a handful of reliefs from south and central Italy, although the poor state of the inscriptions hinders conclusions about the degree of frequency. See Frenz : nos. , , , . 51 Dedicated to patrons: Pflug : nos. , ; shared cost: no. . Patron bust: Pflug : no.  (here, Fig. .); also no. .
Family Imagery and Values in Roman Italy

While incorporating many of the same elements as those from the capital, these reliefs also demonstrate a number of differences which are embedded in their distinct social milieux as well as artistic preferences which vary throughout the region. There is a greater proportion of examples with single portraits, mirroring a parallel trend toward individual representation on memorials in first-century Rome, and there is more frequent use of the portrait bust without a clothed torso, which puts less emphasis on status garments. Portraits which do include the upper body feature conventional Roman dress: tunic and toga for men and tunic and palla for women, the latter with their heads mostly uncovered.52 Children are mostly shown in bust form, but when the torso is shown they generally wear the toga praetexta, and few examples depict a child wearing a bulla. The dextrarum iunctio and the socalled pudicitia gesture are far less common than at Rome; instead, men and women tend to grasp their toga or palla, and in some cases men hold a scroll and women fruit or flowers.53 Overall, there is less subtlety and attention to detail in aspects of costume and gesture on these representations than on the reliefs from Rome. In part, this might be attributable to the skill of local sculpture workshops and the absence of a long-standing regional tradition in portrait sculpture. It could also indicate, however, that there was less concern among the dedicants for the finer distinctions of rank and status that such elements could express. Beyond the somewhat rarefied and status-conscious capital, with its highly refined set of social prejudices, the semiotic authority of the bulla or the vitta might simply have been less relevant. Moreover, the freedman’s symbols of honour were less appropriate, since in this region the genre was not his own, but was shared with freeborn Romans of a different background but similar economic means. 52 On occasion the palla covers the heads of older women to distinguish them from younger women on the same relief. There is only one example of a stola (Pflug : no. ), which Pflug interprets either as a sign that they were out of fashion or that de facto they were worn mostly by elite matronae, even though de iure others were allowed to wear them (Pflug : ). 53 Children: Pflug : ; bulla: nos. , ; possibly also ; gestures: Pflug : ; see also Frenz :  and . There are also references to a military career, reflecting the important role of the army in the region (Pflug : ).

Michele George
Other kinds of personal bonds and symbols of status are also in evidence, sometimes intersecting with one another. Three reliefs from Mutina (modern Modena) commemorate shared involvement in public service. On one, two couples are shown, and although all four were former slaves, they were not colliberti; instead, their mutual interest was in serving as Apollinares, members of a priesthood in the local imperial cult.54 On another example, a freedman, also an Apollinaris, appears with his wife and daughter, along with a portrait of his patron, who is identified as both a decurion and a fellow Apollinaris and who occupies the uppermost position at the top of the stela. Both men wear the toga praetexta, which is clearly indicated on the surface of the stone, and beneath the portraits are the insignia of public office (a chair, table, scrinium, scroll, and tablets), while the fasces decorate the frame.55 A stela from an elaborate monument now in Reggio Emilia suggests the network of kin and power relationships among local elites which involved libertini. Busts of two men are set side by side within a shell at the top, while portraits of their wives are contained in a single clipeus on a lower register. In fact, the two couples are apparently in-laws, the older man and woman the parents of the younger woman, who was probably married to the younger man, Rhenus.56 Both men served in the local sevirate; all, except for the daughter Festa, were freedmen. Rhenus was either the former slave of his father-in-law, who had become his patron, or his conservus; in either case the elder man probably assisted the younger in advancing his public profile in the town. The diminution in stark status distinctions between freedmen and the freeborn in Cisalpine Gaul compared to Rome can be attributed to another local factor. Competition for garnering honour at the municipal level seems to have operated differently in Cisalpine Gaul from other parts of Italy. It has been argued that in north Italian towns social mobility was limited since membership in the local ordo was less accessible to the freeborn than in other Italian cities.57 The existence of long-standing local 54 Pflug : no.  dated to  –. On the sevirate and social status in Cisalpine Gaul, see Duthoy  and Abramenko : – and –. 55 Pflug : no. ; see also no. . 56 Pflug : no. . 57 On the ceti medi of Cisalpine Gaul, see Abramenko ; Tassaux ; Sartori and Valvo .
Family Imagery and Values in Roman Italy

elites in most towns apparently reduced the opportunity for the entry of freeborn families with newly acquired wealth into municipal politics and forced them to find other means of gaining a public reputation. As a substitute, they turned to the sevirate and the cult of the Augustales, public priesthoods that in Rome were strongly identified with libertini, rather than ingenui, and that offered them a form of public service through which they might achieve some measure of social status. Moreover, some of the freeborn on these reliefs were quite possibly the first generation of their family to have full Roman citizenship, since it was only in   that Julius Caesar extended Roman citizenship to all the cities of Cisalpine Gaul. The ceti medi of Cisalpine Gaul, whose citizenship was freshly minted and who were excluded from local politics, shared the marginalization of successful freedmen and the necessity of circumventing the limitations to public recognition that were imposed by local circumstances. A form of social levelling between ingenui and libertini that did not pertain to the same extent in the rest of peninsular Italy (and certainly not at Rome itself ) might therefore be identified as a particular feature of urban northern Italy. The mutual exploitation of closely related genres of funerary commemoration by these two social groups indicates the shared concerns of social outsiders and their confidence in the family portrait group to communicate their worth as upstanding citizens. Conclusion A comparison of these two groups of monuments illustrates the value of family imagery to social groups from different parts of Italy with distinct cultural contexts, but not entirely dissimilar commemorative needs. Both sets of reliefs reflect the emergence of a social group, whether of freeborn or libertine origins, with middle to upper-middle means, which had no existing tradition of self-representation of its own. For freedmen in Republican Rome, the family was a particularly apt symbol of freedom, the social institution which best represented their rise in status, and a potent metaphor for their very particular personal achievements. By this iconographic sleight of hand, these former outcasts could display their adoption of Roman social values and insert themselves into the landscape of the idealized Roman

Michele George
family. While contact with Rome did not introduce family structure to Cisalpine Gaul, with Roman political hegemony came the high value put on the family in Roman culture and its attendant notions of status, stability, and morality. Representations of family portraits offered a means of identification with Rome and its traditions while simultaneously articulating aspects of status and local identity. The absence of refinements in gesture and costume on the Cisalpine reliefs and the strong representation of the freeborn among the dedicants suggest that, over the course of the century since its inception at Rome, the imagery lost its specific reference to former slaves and came instead to be related more generally to mainstream notions of stability and moral standing associated with the family in Roman thought. The robust representation of ingenui on the Cisalpine reliefs demonstrates the appeal the family motif had for the new urban gentry in the region who lacked other forms of social distinction. In this respect, the citizenry of Cisalpine Gaul, ingenui and libertini, shared the Roman freedman’s desire for social legitimacy and recognition of personal achievement. The genre of the family portrait and its visual conventions allowed both groups to clothe themselves literally and figuratively in the values of the dominant culture of the capital which they aspired to join.
 The Roman Child in Sickness and in Health Keith Bradley
O upon a time during a great pestilence the Sabine Valesius found his three children, two sons and a daughter, on the brink of death. As the doctors despaired of their lives Valesius prayed to his household gods and asked that his children’s illness might be transferred from them to him. Unexpectedly, however, the gods replied by saying that the children would be saved if they were taken down the Tiber to a place called ‘Tarentum’ and given water to drink from an altar there to Dis and Persephone. Where Tarentum was Valesius did not know, but sailing from Sabine Eretum he followed the gods’ directions and inadvertently cured his children when he gave them water from the Tiber at a spot in the Campus Martius which turned out to be the required place. In due course his slaves, digging the foundations of an altar of thanksgiving at the site, discovered another altar to none other than the gods of the underworld, and a little later his children were divinely instructed to offer sacrifice. For three successive nights therefore Valesius offered victims and held games and lectisternia to celebrate his children’s recovery from the terrible pestilence that had so severely threatened their lives. This story, summarized from the account of Valerius Maximus (. . ), provided an explanation for Romans of the historical age of how the Ludi Saeculares first came into existence and where they were celebrated: for Rome’s first consul Valerius Poplicola was believed to have founded the games in order to commemorate his ancestor Valesius’ discovery of the cult site of Dis and Persephone. The story is arresting, however, not so much for its aetiology as for what it implies about fathers and children in the historical age when it was circulating. It suggests, first, the desperation a Roman father might feel when his children were ravaged by uncontrollable disease of the sort which periodically

Keith Bradley
swept through the Roman world—the great plague under Marcus Aurelius being the most notorious example—and which over time must have carried off vast numbers of children; it connotes, secondly, an ideal of fatherhood according to which a parent was expected to undertake any action humanly possible, even give his own life, to protect his children; and it presupposes, thirdly, that children, of both sexes, were highly prized in Roman society.1 Romans of the historical epoch understood that in theory people should die in generational sequence. But they also knew that the reality was rather different: ‘non citamur ex censu’, as Seneca (Ep. . ) expressed it. So the story of Valesius also captures an anxiety with which Roman parents always had to live, the fear that their children might die untimely deaths—a fear that can be understood particularly well from the portents of child death catalogued in Artemidorus’ Interpretation of Dreams. To dream, for instance, of a woman losing her breasts (. ), of a small boy turning into a man (. ), of a father having sex with a son younger than  years old (. )—these were just some of the many dreams Artemidorus took to portend the death of dreamers’ children, the significant point being that he had collected his dream evidence from real people all across the Mediterranean world. The fear indeed knew no limits. It was said that in order to make their loss more bearable the Persians, well beyond the sphere of Graeco-Roman culture, did not even look at their children until the children had passed the age of .2 1 Zosimus (. –) also tells the story of Valesius, though with some variations: in his version Valesius offered the gods his wife’s life as well as his own in return for his children’s recovery, the number of whom is not specified. His text implies that Valesius’ celebration of the children’s return to health was the first celebration of the Ludi Saeculares, but Valerius Poplicola is shown offering sacrifice to Dis and Proserpina to free Rome from pestilence and to acknowledge the Roman people’s new freedom. See Taylor : ; Gagé : ; and for full sources, Pighi . On Tarentum see Platner and Ashby : –; Richardson : ; LTUR V. – (F. Coarelli). On the plague (smallpox) under Marcus Aurelius, Gilliam ; Littman and Littman ; cf. Sallares : ; for its social and economic ramifications, Duncan-Jones ; and for other outbreaks of pestilence under the Principate, Gilliam : . Translations are from the Loeb editions when available, or from the standard English translations in the case of, for example, Soranus. 2 Seneca: cf. Ep. . ; . ; . ; . ; Celsus, Med.  prooem. ;  prooem. . Many dreams: for further examples, Bradley : . Persians: Val. Max. . . ; cf. Hdts. . , where the age is given as .
The Roman Child in Sickness and Health

And die the children did. Any number of examples could be summoned by way of illustration—the sons of Aemilius Paullus succumbing at the ages of  and , Quintilian’s sons at the ages of  and , Plutarch’s daughter Timoxena at the age of —and their representativeness is confirmed by the results of modern demographic research which reveal clearly enough, even if experts are unable to reach consensus on matters of detail, that infant and child mortality at Rome was extremely high, with perhaps a third of newborns dying within the first year of life and half of all children dying by the age of —a regime scarcely imaginable in contemporary western societies. It means that the sight of children dying and the sight of children’s bodies being disposed of were sights that no one in the Roman world could really miss; and it also means that a distinctive mentality towards the loss of children emerged, which is reflected for instance in Marcus Aurelius’ choice (Med. . ) of the deaths of the very young and the deaths of the very old as natural points of comparison: from a modern perspective there is nothing natural about that at all.3 The story of Valesius suggests that one, perhaps the most important, factor that affected child mortality was the Romans’ inability to deal effectively with disease. This is an obvious demographic point but one worth demonstration. A famous passage in Soranus’ Gynecology (. –) offers an analysis from an observant and skilled practitioner of medicine of how bone deformities were thought to arise in infants, especially in the city of Rome itself: When the infant attempts to sit and to stand, one should help it in its movements. For if it is eager to sit up too early and for too long a period it usually becomes hunchbacked (the spine bending because the little body has as yet no strength). If, moreover, it is too prone to stand up and desirous of walking, the legs may become distorted in the region of the 3 Examples: Livy . ; Plut. Aem. . , . ; Quint. Inst.  pr. – (the older boy may have been more than  at his death but that is the last age mentioned for him); Plut. Cons. ad ux. ; for some imperial instances, Suet. Claud. . ; Ner. . ; Galb. . ; Vesp. , . ; Domit. . . Demographic research: Frier , for a summary account; cf. Parkin : – on the essentially fabricated nature of the evidence. Consensus: cf. Scheidel a, b; Sallares . Mortality: see references in Bradley :  n. . The emphasis placed by Sallares on regional and social variability in ancient demographic patterns does not affect the general point of relatively high infant and child mortality. For a select number of burials of girls and accompanying grave goods, from both Italian and provincial sites, see Martin-Kilcher .

Keith Bradley
thighs. () This is observed to happen particularly in Rome; as some people assume, because cold waters flow beneath the city and the bodies are easily chilled all over; as others say, because of the frequent sexual intercourse the women have or because they have intercourse after getting drunk—but in truth it is because they do not make themselves fully acquainted with child rearing. For the women in this city do not possess sufficient devotion to look after everything as the purely Grecian women do. if nobody looks after the movements of the infant the limbs of the majority become distorted, as the whole weight of the body rests on the legs, while the ground is solid and hard, being paved in most cases with stones. And whenever the ground upon which the child walks is rigid, the imposed weight heavy, and that which carries it tender—then of necessity the limbs give in a little, since the bones have not yet become strong.
There is evidence here of an intellectual effort to understand a medical condition—in this case not an inherently fatal condition—but there is also evidence of the massive ignorance that characterized all medical science before the modern era. The disease Soranus describes is almost certainly rickets, the essential cause of which is the absence of vitamin D which leads to softening of the bones, bowed legs, and hunched backs. Historically rickets in children has been associated with the growth of urbanization in Europe consequent upon the Industrial Revolution, a context in which huddled buildings blocked out sunlight—the chief source of vitamin D—and children were kept indoors for long periods of daytime by the demands of labour. Comparable conditions may have obtained at Rome in the late first and early second centuries  when Soranus was writing. In the absence of any ancient knowledge about vitamins, however, it was impossible for Soranus or for any other doctor to know the true cause of the deformities mentioned and any therapy prescribed for the problem therefore was unlikely to be effective.4 To what extent was disease regarded as a special characteristic of childhood in the Roman world, and, if it was, how did parents respond to it? What are the illnesses that were associated with childhood, and how visible in the historical record is the sick child? These are the questions with which this chapter is 4 Rickets: Grmek: : –; Sallares : ; Roberts and Manchester : –; cf. Garnsey : ; : –. The condition could presumably be treated with braces: Varro, Ling. . –.
The Roman Child in Sickness and Health

concerned. In treating them I survey briefly the evidence of select writers on medicine and medical remedies under the early Principate, a period from which a remarkable (but surely fortuitous) clustering of material survives. The emphasis falls only on Roman perceptions of childhood illnesses—the names and descriptions of illnesses used by ancient authorities—not on modern medical diagnosis and terminology, and largely does not go beyond the level of the descriptive. But by drawing attention to the abundant material available I hope to elicit something of the experience of childhood in Roman antiquity, to add some substance to the obvious demographic point mentioned a moment ago. It may be noted at once that Pliny the Elder provides reason for thinking that certain diseases were associated with childhood as a matter of course. ‘What can be found more marvellous than this,’ he states, ‘that some diseases should arise suddenly in a special part of the world, should attack the special limbs of human beings or special ages, or even people of a special position in life (just as if a plague chose its victims), one children (in pueris), another adults, one making the nobility especially liable, another the poor’ (HN . , my emphasis). Children emerge here as one of what could be called several demographic constituencies of illness, and as will become clear there is much to be filled in under the rubric.5 An incipient awareness of children’s illnesses emerges at an early date, the early second century , in the agricultural handbook of Cato. In the famous encomium of cabbage Cato says (Agr. . ) that barley-meal is to be added to cabbage as a remedy for ‘sores’ in boys and girls, and (Agr. . ) that weak children are to be washed in the urine of cabbage-eaters in order to make them grow strong. Cato also differentiates (Agr. . ) between children and adults in his recommendations about the use and application of cabbage for curing colic, even though prescribing the same treatment for men, women, and children. Elsewhere (Agr. . ) he distinguishes children from adults in the amount of a remedy to be administered for dyspepsia, strangury (painful passage of drops of urine), and worms. The 5 Writers on medicine: for full repositories of material from antiquity at large, see Bertier , ; Hummel . Terminology: cf. Nutton :  n. : ‘very few of the diseases named in antiquity . . . can be identified with anything like certainty.’

Keith Bradley
conditions mentioned are not of course unique to children, but Cato’s evidence is valuable for its recognition that children could form a medical category separate from that of adults. The implication that the paterfamilias who read Cato’s book ought to have first-hand knowledge of how to treat various ailments by which his dependants were affected is also important. The presumption must be that Cato’s evidence reflects widespread social conventions and attitudes, not the behaviour of a single individual.6 Much fuller information is provided by the encyclopaedist Celsus in De medicina. In a passage, first, relating health to stages of life and the seasons (Med. . . –), Celsus gives a long list of childhood conditions so that it becomes possible, despite the imprecision involved, to think in terms of predictable complaints that could appear at one stage of a child’s life after another: As regards the various times of life, children and adolescents enjoy the best health in spring, and are safest in early summer; old people are at their best during summer and the beginning of autumn; young and middle-aged adults in winter. Winter is the worst for the aged, summer for young adults. () At these periods should any indisposition arise, it is very probable that infants and children still of tender age should suffer from the creeping ulceration of the mouth which the Greeks call aphthas, vomiting, insomnia, discharges from the ear, and inflammations about the navel. Especially in those teething there arise ulcerations of the gums, slight fevers, sometimes spasms, diarrhoea; and they suffer as the canine teeth in particular are growing up; the most well-nourished children, and those constipated, are especially in danger. () In those somewhat older there occur affections of the tonsils, various spinal curvatures, swellings in the neck, the painful kinds of warts which the Greeks call acrochordones, and a number of other swellings. At the commencement of puberty, in addition to many of the above troubles, there occur chronic fevers and also nose-bleedings. () Throughout childhood there are special dangers, first about the fortieth day, then in the seventh month, next in the seventh year, and after that about puberty. The sorts of affections which occur in infancy, when not ended by the time of puberty, or of the first coitions, or of the first menstruations in the females, generally become chronic; more often, however, puerile affections, after persisting for a rather long while, come to an end.
6
Cabbage-eaters: cf. Plin. HN . .
The Roman Child in Sickness and Health

Adolescence is liable to acute diseases, such as fits, especially to consumption; those who spit blood are generally youths.7
From other passages more details emerge. There is the phyma, a kind of boil that affects children but not adults (Med. . . ) and ophis, a balding disease that generally occurs in young children, beginning at the back of the head and gradually creeping forward (Med. . . ). The mouth ulcers mentioned, apthae, are especially dangerous in children, if not unique to them, and lifethreatening: they begin from the gums, Celsus says (Med. . . –), invade the palate and the whole mouth, pass into the uvula and the throat, and make it difficult for the child to recover. For nursing infants they are a virtual death sentence. Celsus also reports a condition (Med. . . ) that generally occurs in children in which fatty and weighty cysts appear in the upper eyelids, preventing the eyes from being raised and producing a slight but persistent discharge; it was a problem that required surgery. There is a potentially fatal condition giving bladder pains, persistent fever, and no bowel discharge which was most dangerous in boys aged between  and  (Med. . . ); and a condition with diarrhoea, fever, inflammation of the liver or of the parts over the heart or stomach, great thirst, varied and painful stools, and dysenteries, which killed children ‘mostly . . . up to the age of ten’ (Med. . . ). Epilepsy is a disease which might be expected to affect boys until the age of puberty and girls until the onset of menstruation (Med. . . ). Children might also suffer from painful chilblains (Med. . . ) or cataracts (Med. . .  B: cataract surgery in childhood was not recommended). Boys, in particular, might need surgery for hydroceles (Med. . . ; . . ) or intestinal prolapse (Med. . . ). Surgery was also possible for bladder stones (Med. . . ), though only as a last resort. Dental surgery, including the extraction of roots, was recommended for children whose first
7 Stage: for critical moments of danger, Celsus, Med. . . –. For modern studies of Roman seasonal mortality which show that infectious diseases had their greatest impact in the late summer and early autumn, see Scheidel ; Shaw ; Sallares : –, –. For the same phenomenon in th-cent. Florence, with specific reference to gastrointestinal diseases in children, Herlihy and Klapisch-Zuber : –.

Keith Bradley
teeth were not falling out to make space for the second teeth (Med. . .  F).8 Celsus gives a good idea, therefore, of the range of illnesses and ailments that could be associated with childhood. They were not all necessarily fatal, but with conditions for which surgery was required there must always have been a likelihood of death from shock (in the general absence of anaesthesia) or from haemorrhage or infection: Celsus is in no doubt that surgery to correct the birth defect of deafness from a blocked ear canal was potentially lethal (Med. . . ) and that a procedure to separate the tongue from the lower mouth could lead to dangerous bleeding (Med. . . ). The illnesses he mentions were not all specific to childhood, but they were conditions that might be contracted in childhood, examples of what was possible. This is not to take account of accidental deaths. Celsus knew of a boy who was miraculously saved from the bite of an asp by drinking vinegar (Med. . . ), but presumably other victims were not so fortunate. He speaks too (Med. . . ) of dislocations of the bones resulting from accidents or the application of force, to which boys and youths were especially prone because their bodies were underdeveloped: physical deformity or even death could be the result, as with a dislocation of the head or spine (Med. . . ; . , ).9 8 Eyelids: observe that a certain Severus, cited by Aetius of Amida, wrote a now lost book on treating children’s eyes (RE s.v. ‘Severus’, no. ). Prolapse: ‘only admissible in boyhood and when the trouble is limited’ (Med. . . ). Stones: ‘This operation is not suitable for every season or at any age or for every lesion, but it must be used in the spring alone, in a boy who is not less than nine years of age and not more than fourteen, and if the disease is so bad that it cannot be relieved by medicaments, or endured by the patient without shortly bringing his life to a close’ (Med. . . ). Roots: ‘In children too if a second tooth is growing up before the first one has fallen out, the tooth which ought to come out must be freed all round and extracted; the tooth which has grown up in the place of the former one is to be pressed upwards with a finger every day until it has reached its proper height. And whenever, after extraction, a root has been left behind, this too must be at once removed by the forceps made for the purpose . . .’ (Med. . .  F). See also Med. . . ; . . , ; . . , ; . . ,  B,  E; . .  M; . . ; . . . 9 Likelihood: Galen’s account (AA . ; cf. Plac. Hipp. et Plat. . ) of the procedure performed on a slave wrestler with an infected sternum, apparently a case of osteomyelitis (Moraux : ), which involved exposure of the man’s heart, indicates how unexpected his recovery was. For the torment which surgery
The Roman Child in Sickness and Health

Pliny the Elder displays in the Natural History great interest in the medicinal properties of natural substances, and he mentions, strikingly so, many remedies pertinent to the subject of children’s illnesses. Childhood diseases per se he does not discuss, only the relief that various natural products will bring. His references to children’s complaints are generally oblique, therefore, the ailments themselves simply taken for granted. The following passage (HN . –, my emphasis), which contains certain points of overlap with Celsus, again gives some indication of the types of ailments Pliny could associate with childhood: For babies nothing is more beneficial than butter, either by itself or with honey, especially when they are troubled with teething, sore gums, or ulcerated mouth. The tooth of a wolf tied on as an amulet keeps away childish terrors and ailments due to teething, as does a piece of wolf’s skin . . . () Hare’s rennet applied to the mother’s breasts checks the diarrhoea of babies. Ass’s liver mixed with a moderate amount of panaces and let drip into the mouth protects babies from epilepsy and other diseases; the treatment, it is prescribed, should continue for forty days. Ass’s hide laid on babies keeps them free from fears. The first teeth of horses to fall out make the cutting of teeth easy for babies who wear them as an amulet, a more efficacious one if the teeth have not touched the ground. () Ox spleen in honey is administered internally and externally for painful spleen; for running sores with honey . . . a calf ’s spleen boiled in wine, beaten up, and applied to little sores in the mouth.10
From the Natural History as a whole, as with Celsus’ De medicina, a list can be made of children’s illnesses, equally imprecise brought, both physical and psychological, see August. De civ. D. . on Innocentius of Carthage; cf. Nutton : –. Anaesthesia: Pliny, HN . , indicates that the soporific mandrake (mandragora) was used for anaesthetic purposes in surgery (cf. Dioscorides, Mat. med. . , . , and see Zimmerman :  on Apul. Met. . ), as also, HN .  (cf. Dioscorides, Mat. med. . ), topically, the unidentifiable Memphis stone; so relief was not completely unknown. Defect: cf. Celsus, Med. . . . Accidental deaths: cf. Suet. Claud. . , Claudius’ son Drusus choked by a pear; Gal. De caus. procatarct. . , a boy accidentally killed by poison intended for his half-brother; August. De civ. D. . , a boy almost killed in a traffic accident, saved by his mother’s Christian piety. 10 The text of  is defective. On Pliny and medicine generally, see Nutton ; Beagon : – (stressing his belief in the efficacy of herbal remedies); French : –. Bonet  provides a repertory of references in the HN to children’s illnesses, noting () the special importance of HN . –, . –, and . –).

Keith Bradley
from a modern point of view, but helpful none the less. It includes teething problems, warts, various sores and ulcers (many remedies are prescribed for mouth ulcers, perhaps an indication of their incidence), various inflammations, coughs, worms, prolapse of the anus, heart disease, lichens and other skin rashes, intestinal hernia, umbilical rupture, stone and strangury, varicose veins, ophthalmia, and epilepsy (for which again many remedies are given). Treatments are recommended also for problems attributable to accidents such as snake bite and the consumption of poisons. For consumption by a snake there presumably was no remedy, but Pliny knew (HN . ) of such an event from the principate of Claudius: a child had been discovered when a large snake, a boa, was killed on the Vatican.11 All in all Pliny seems to confirm the picture given by Celsus and it could be assumed that their evidence, taken together, reflects circumstances that were widespread in Roman society. It is a natural inference after all that Celsus and Pliny wrote of medical conditions their audiences would recognize, that they did not describe anomalous situations. Celsus indeed was a man of practical medical experience, less a medicus than a paterfamilias in the tradition of Cato who expected to know a certain amount of medicine as a matter of course in order to meet the responsibility of taking care of his household—the same responsibility that in the Gospel of Matthew (: –), for example, underlies the centurion’s appeal to Jesus to heal the paralytic boy in his house and that was sadly not met by the Roman father Epictetus lectured (. ) when a sick daughter was abandoned. Pliny undoubtedly knew the tradition. An obstacle to this view, however, arises from the fact that parts of Celsus’ De medicina are translated or adapted from the sixty or so Greek treatises that make up the 11 List: HN . , . , .  (teething); .  (warts); . , . , . , . , . , . , . , , .  (sores); . , . , . , .  (inflammations); . , .  (coughs); .  (worms); .  (prolapse); .  (heart disease); . , .  (rashes); . , . , ,  (hernia); .  (rupture); .  (stone, strangury); .  (veins); .  (ophthalmia); . , . , . , . ,  (epilepsy); cf. . , . ,  (procidence and incontinence). Incidence: a high number of recorded remedies for a particular complaint may indicate a high incidence of the ailment, but also the ailment’s persistence or resistance to treatment (or both); further, the availability of herbal remedies may be seasonally affected; I am grateful to Patricia Clark for these observations. Accidents: HN . , .
The Roman Child in Sickness and Health

Hippocratic corpus of medical literature from the classical Greek and Hellenistic eras, and that the authorities on which Pliny drew for his medical information included not only, likewise, ‘Hippocrates’, but also Celsus. Such reliance on previous authorities was standard practice among medical writers in antiquity and it led to a rather rigid conceptualization of children’s illnesses. More importantly, however, the Roman representativeness of Celsus and Pliny is clearly brought into question. To give just one example, the passage on the relationship between diseases and stages of life earlier quoted at length (Med. . . –) can be traced directly to a section of the Hippocratic Aphorisms (. ; . –). How, therefore, is this obstacle to be confronted?12 The conversion of Greek knowledge into Latin form had a long history by the time Celsus and Pliny wrote their works. Cicero and Varro come easily to mind for instance as men who had once respectively transformed Greek knowledge of statecraft and household management into a Latin idiom. The process, moreover, continued well beyond the era of Celsus and Pliny, for as late as the second century Apuleius claimed (Apol. ) that he was the first to have rendered into Latin certain scientific terms from the Greek. The impulse behind such activity was to make available to Roman audiences useful knowledge of practical import, not to disseminate knowledge simply for its own sake, and it might be assumed in consequence that it was Celsus’ and Pliny’s purpose to convey Hippocratic material to Roman audiences because the material was of contemporary relevance. Practical application was undoubtedly the goal that lay behind the composition of the medical handbook of the first-century 12 Widespread: Dioscorides, the st-cent. writer on pharmacology, has relatively little to offer on children’s complaints but the information he provides is generally consistent with that in Celsus and Pliny; see Mat. med. .  (alopecia (cf. Riddle : ) ); .  (teething); .  (sialorrhea); . , .  (siriasis (cf. Riddle : ) ); .  (convulsions); .  (stone); .  (umbilical hernia); .  (intestinal hernia); .  (epilepsy). Medical experience: see the passages collected by W. G. Spencer, Loeb edn., vol. i, pp. xi–xii. On the Roman tradition of medical self-sufficiency, see Beagon : –. Responsibility: even on his deathbed an emperor could make enquiries about a sick granddaughter; Suet. Aug. . . Standard practice: Bertier : , . Aphorisms: the ‘most widely read, commented, and cited treatise of the entire [Hippocratic] collection’, which was not written before the th cent.  (Jouanna : –). For Pliny’s authorities, HN . –.

Keith Bradley
medical writer Scribonius Largus. In turn, this view could be supported by supposing that the medical problems of childhood, in the absence in antiquity of any truly revolutionary medical breakthroughs in knowledge, largely remained constant. It has to be acknowledged, however, that such a generalizing view covers a vast amount of time and space and that variations are to be expected, if difficult to specify, and, more tellingly, that diseases were sometimes recognized in antiquity as having their own histories: ‘The face of man’, Pliny wrote (HN . ), ‘has also been afflicted with new diseases, unknown in past years not only to Italy but also to almost the whole of Europe, and even then they did not spread all over Italy, or through Illyricum, the Gauls, and the Spains to any great extent, or in fact anywhere except in and around Rome.’ Rickets is one example of a disease that probably advanced dynamically over the course of antiquity. Osteoarchaeological evidence has revealed some isolated cases in classical Greece, but it was not apparently until a much later Roman epoch that the disorder became entrenched. It remains possible, therefore, that the apparent contemporary significance of Celsus’ and Pliny’s evidence is sometimes suspect because of their dependence on Greek sources, but it is unlikely that their evidence can be dismissed completely. No matter how abstruse or arcane the material sometimes seems, especially that of Pliny, the presumption behind their works is that the remedies they recommend would be effective for illnesses with which their readers were familiar on an everyday basis.13 The same is true of the children’s illnesses found in Soranus’ Gynecology, the work of a Greek physician who, as noted above, 13 Practical import: cf. E. Rawson : – on the assimilation at Rome of Greek medical knowledge. Scribonius Largus: Comp. pr. ; ; ; see Hamilton ; cf. Flemming : –. Own histories: note Celsus, Med.  prooem. : ‘Rarius sed aliquando morbus quoque ipse novus est’; cf. Nutton : –. Rickets: ‘Not until Roman times did urban density, changes of diet, and the impoverishment of one part of the population unite to bring about conditions favorable to rickets and to making its ravages apparent to medical practitioners’ (Grmek : –; by practitioners Grmek means Soranus and Galen (San. tuend. ), but note Plato, Laws . e: distortion from excessive pressure on children’s legs). Unlikely: on the independence of Celsus’ views, Nutton : –; cf. Flemming : –; and on Pliny, French : , ; cf. Nutton : ; Flemming : –.
The Roman Child in Sickness and Health

had personal experience of life in the city of Rome in the era of Trajan and beyond. As a book on midwifery and complaints specific to women, the Gynecology once more does not address children’s illnesses directly or comprehensively. But it contains incidental information on the afflictions of newborns which readers are expected to recognize, both genuine illnesses and problems resulting from accidents or the poor care provided by children’s nurses. The illnesses include inflammation of the gums, jaws, and tendons from teething (Gyn. . ), inflammation of the tonsils (Gyn. . ), mouth ulcers (Gyn. . ), blisters and itching (Gyn. . –), coughs (Gyn. . ), fever (Gyn. . ), and diarrhoea (Gyn. . –). The points of overlap with what has already been seen should be obvious. Other difficulties include accidental injury to the eyes as a possible outcome of nurses sleeping with their nurslings, a practice of which Soranus disapproved (Gyn. . ). The danger of strangulation from the nurse’s breast was also very real (Gyn. . ), while an ill-tempered woman, it appears, might simply drop an infant to the ground (Gyn. . ). Excessive crying or excessive movement of the child might harm the eyes and in boys endanger the testicles (Gyn. . –). A nursling might become comatose or apoplectic if a nurse drank too much wine (Gyn. . ). Ulcers might develop from dirty clothes if the infant were not kept clean (Gyn. . ) and if the nurse’s milk were poor, epilepsy or apoplexy might threaten the child’s life (Gyn. . ). Excessive bathing, excessive feeding, and abrupt weaning could also lead to illness (Gyn. . , . , . ).14 Soranus closes his remarks on children’s problems this way: ‘Having thus completely finished the discussion of things normal, we must go more deeply into the main topics which deal with things abnormal in women’ (Gyn. . , my emphasis)—a statement that implies that all that has been said about newborns’ complaints belongs very much to the everyday world of Rome at the time of writing. Soranus after all was a practising Methodist physician and his book was instructional in intent. Galen, the greatest of all medical authorities from antiquity, certainly supports what he has to say about the dangers of inadequate nursing, mentioning on the basis of autopsy the physical problems and 14 For other aspects of Soranus’ views of nurses, Bradley b: –. On Soranus in general, Drabkin .

Keith Bradley
disfigurements that occurred in girls from attempts nurses made to stretch their hips and thighs. It was ‘the fault and stupidity’ of such women, he said (De caus. morb. ), that resulted in broken backs and misaligned shoulder blades. He also knew (De alim. facult. . ) of a nurse who transmitted a disease due to inadequate diet, as he thought, to a nursling: the child became covered with the same kind of sores from which the woman was suffering herself. Throughout Soranus’ material on nurslings there is a consciousness of the fragility of the young child’s life, an awareness that death was never far away.15 A clear image, then, emerges from the authors resumed of the illnesses by which children in Roman antiquity were thought to be afflicted, of the medical dangers to which they were exposed, and of a consciousness among adults of children’s susceptibility or predisposition to illness. Perhaps an urban bias is detectable in the sources, all the work of elite men, but it can hardly be the case that the details they purvey pertain exclusively to urban contexts (especially if Cato is kept in mind). Many of the complaints concerned—coughs, fevers, sores—appear as symptoms rather than illnesses proper and their underlying causes remain generally unknown. In reality it is likely that Roman children, newborns in particular, suffered from overwhelming gastrointestinal diseases and respiratory tract infections such as cholera and dysentery, typhoid and tuberculosis, conditions which were often lethal because they were not properly understood. Poor standards of sanitation and hygiene, both public and private, together with contaminated supplies of water in fountains and baths contributed heavily to the high mortality rate, especially in cities. Chronic malnutrition, at all levels of society, may have been another relevant factor and malaria probably had ravaging effects, both as a cause of death directly and as a condition that rendered the body prone to other deadly dangers. But while illnesses cannot have been properly understood, society was very much aware of the vulnerability of children and of a tightly bound link
15 Transmitted: cf. Celsus, Med. . .  D; . . –. On criticisms of nurses, Bradley . Greatest: see Nutton  on Galen’s contemporary reputation.
The Roman Child in Sickness and Health

between sickness and death of a kind that is unimaginable in the contemporary western world.16 This awareness, moreover, was not restricted by geography. The literary sources that refer to children’s illnesses predictably tend to focus on Rome and Italy—Martial’s poem (Epig. . ) for instance about a girl who seemingly died from the mouth ulcers that the medical writers suggest were so common, or the correspondence of Fronto, in which an extensive catalogue can be found of the medical misfortunes suffered by the children of Marcus Aurelius. But occasionally a less Romanocentric source points to a broader, provincial perspective. In the Gospels, for example, the wonderworker Jesus is regularly called upon by concerned parents to heal sick children or even to restore to life the untimely dead: a Gentile woman asks for her daughter to be cleansed of an unclean spirit; a father brings a son who has suffered all his life from epilepsy to be healed; the daughter of the ruler Jairus and the older son of the widow of Nain are brought back from the dead; a fevered boy on the brink of death is dramatically restored to his nobleman father.17 How did parents respond? What might have been done to secure and to maintain the health of the child? To appeal for help from a wonderworker was one strategy by which an attempt might be made to overcome children’s vulnerability. To appeal to healing gods was another: among the patients cured long ago by 16 Likely: Grmek : ; Sallares : ; Garnsey : –. On typhoid, tuberculosis, and malaria, see the summary of Scheidel : –, and the fuller accounts of Brunt : – (,  on children), Grmek : –, –; Sallares : –, – (,  on children); : – (and – on endemic malaria especially in the city of Rome). On inadequate sanitation, Scobie  (especially , on the dangers to newly weaned infants of contracting infections from dirty dishes and contaminated food). Galen (De loc. aff. . ) knew of boys aged  and  who suffered from intestinal and bladder problems, one (he thought) as a result of fishing in a cold river. Chronic malnutrition: Garnsey : –; Sallares : –. 17 Correspondence: evidence collected in Wiedemann : –. Gospels: Matt. : –; Mark : –; Matt. : –; Mark : –; Luke : –; Mark : –, –; Luke : –, –; cf. Matt. : –; Luke : –; John : –. The disproportionate amount of attention the Gospels pay to miracles of healing is surely an indication of the general inefficacy of doctors in antiquity. D. . .  none the less implies that parents of decurial status in provincial communities were sensitive to the need to have doctors available for their children.

Keith Bradley
Asclepius at Epidaurus had been a boy who could not speak, another with bladder stone, a third who could not see, and a fourth who had a growth of some sort on his neck. Once installed at Rome the god could respond appropriately to local pleas for help. A fragment from a work by the elder Cato suggests that many women once relied on praecantrices, ‘singers of charms’, to heal their sick children, and in due course Christian martyrs developed the capacity to bring back to life young boys and girls who had died. The appeal of the supernatural, even magic, was evidently great. More prosaic attempts at cures, however, were also possible. From the passages already quoted it is clear that the medical authorities all prescribe remedies of one sort or another that were regularly administered, and some pharmacological therapies were certainly effective in at least treating symptoms. Galen (De praecog. ) describes how he treated a secondary fever in the young Commodus when the boy was suffering from tonsillitis. The pedagogue Peitholaus had given Commodus a mixture of honey and sumac which was too strong for a child and had caused sore spots to develop in the boy’s mouth; so Galen substituted a mixture of honey and rosewater and over the next day or two the boy’s fever and the inflammation gradually disappeared; Commodus was then bathed as the final part of his ‘cure’. Such treatment no doubt soothed and brought relief to the sufferer, but it could not of course have been a real cure, any more than Galen’s programme for ‘curing’ epilepsy: a purgation in the spring, followed by a daily regimen of gentle exercise, study, a carefully controlled diet, a drink made from honey and vinegar and a medicine made from honey and squills. Galen (Puero epilectico consil.) claimed that this could be effective within forty days and that he had cured many children with it. But the programme can only have brought symptomatic relief at most. The methodical procedure for variously adjusting the balance of the child’s body recommended in the following passage from Celsus (Med. . .  B–C), despite its apparent rationalism, is unlikely to have had much effect at all: But if a child is the sufferer, and not robust enough for blood-letting to be possible, thirst is to be used in his case, the bowels are to be moved by a clyster whether of water or of pearl-barley gruel; then and not before he is to be sustained by light food. Indeed in general children ought not to be treated like adults. Therefore, as in any other sort of
The Roman Child in Sickness and Health

disease, we must set to work with more caution in these cases; not let blood readily, not readily clyster, not torment by wakefulness and by hunger or excess of thirst, nor is a wine treatment very suitable. After the remission of the fever a vomit is to be elicited, then food of the lightest nature is to be given, after which let the child sleep; next day, if the fever persists, let the child be kept without food, and on the third day return to food as above. Our aim should be, as far as possible to sustain the child, by food when suitable, with abstinence in between when suitable, omitting all else.18
Whatever their efficacy, treatments of this sort were certainly dispensed. If not on the knowledge that a paterfamilias like a Cato or Celsus might command, their administration depended on the availability of doctors. Already in the last century of the Republic it had become conventional for elite Romans to have personal doctors, usually Greeks, at their beck and call and medici attached to the households of the great and glorious might treat the slaves and ex-slaves in the familia as well as the immediate family of the paterfamilias. (Other members of the lower classes had to seek help from the independent physicians who practised their profession, or rather plied their trade, in the streets of Rome.) Elite parents were also in a position to summon the most reputable of doctors to their aid as occasion required—so the solicitous Flavius Boethus summoned Galen to attend his son Cyrillus when the mischievous boy was sick for no apparent reason—and Galen even gave advice on how to do so. The social status of doctors, their training, and expertise all varied greatly. But they were the men, like Alexio, the doctor of Cicero, or the anonymous doctors who ministered to the dying Minicia Marcella, who knew something of the remedies listed by Celsus, Pliny, and Soranus that could be applied once their diagnoses had been made.19 18 Cf. Celsus, Med.  prooem. : ‘eademque saepe salutaria, saepe vana sint.’ Epidaurus: evidence in Edelstein and Edelstein : T. . Rome: Price : –. Fragment: Catus vel de liberis educandis fr.  Riese, cited by Dickie :  n.  (cf. ). Christian martyrs: August. De civ. D. . . For comparable strategies in medieval Europe, see Shahar : –; Orme : –. Effective: on drugs see Nutton :  n. ; Jackson : ; Riddle . Cf. Gal. De loc. aff. .  for a -year-old epileptic. 19 Dispensed: for all forms of therapy, see Bertier : –. Conventional: Treggiari : –; Jackson : . Independent physicians: Jackson : –. Flavius Boethus: Gal. De praecog. . Mischievous: cf. Ov. Fasti . –:  was the age a peasant boy might set a fox on fire. Advice: Galen, De

Keith Bradley
From the prescriptions recorded by medical writers an imaginative picture can indeed be painted of the Roman child’s medical career, as it were, an aspect recovered of the child’s experience of childhood. For instance, with Soranus’ help the infant suffering from teething inflammations can be imagined being treated by a nurse with ‘poultices of the finest meal, or fenugreek or linseed and fomentations with sea sponges, especially for the gums, and . . . honey boiled down to the right degree’ (Gyn. . ). For bodily itching, ‘a warm decoction of roses or lentils . . . with myrtle or mastich or bramble or pomegranate peel’ was the treatment (Gyn. . ), and for coughs, ‘lozenges prepared with small pine cones, roasted almonds, linseed, the juice of licorice, pine seed, tragacanth, and honey’ (Gyn. . ). An infant with mouth ulcers unfortunate enough to have a Syrian nurse might be imagined to have suffered more rather than less when the woman wrapped some hair around one of her fingers, coated it with olive oil or honey and wiped the ulcers, thereby removing their scabs and irritating them (Gyn. . ). These are just a few of Soranus’ remedies with which children in the Roman world must be understood to have been commonly treated.20 The older child, with Celsus’ help, can be imagined being subjected to bloodletting for almost any ailment as long as the treatment could be tolerated (Med. . . –). For a pestilential fever when a child could not afford to lose blood, however, a combination of water or barley-gruel enemas and a light diet without drinks might be the course to adopt (Med. . .  B). A child with worms might be given ‘pounded-up seeds of nettles or of cabbage or of cummin in water, or mint in the same or a decoction of wormwood or hyssop in hydromel or cress seeds pounded up in vinegar’ (Med. . . ). For chilblains, the treatment was ‘a hot decoction of turnips, or . . . some kind of repressant vervain’, or the application of hot copper if the ulcers were not yet open, or else, if they were, an application of ‘equal parts of alum and frankincense pounded together with the addition of wine, or pomegranate-rind optimo medico cognoscendo; see Iskandar  with Nutton . Social status: Nutton ; Mattern . Alexio: Cic. Att. . . ; .  (cf. Treggiari : ). On society doctors, see Gal. De praecog. . –. Minicia Marcella: Plin. Ep. .  (cf. Quint. Inst.  pr. ). 20 See also Gyn. . , , , –. Mouth ulcers: cf. the treatment prescribed by Celsus, Med. . . –.
The Roman Child in Sickness and Health

boiled in water and then pounded’ (Med. . .  B). In the much more serious case of bladder stones, from which children in antiquity appear to have suffered widely, surgery, although potentially fatal, might be practised as a last resort to deal with a painful condition that was itself potentially deadly. The procedure varied according to gender. But in the case of a boy, to produce an operable position a strong man sitting on a stool held the child on his lap, the boy facing outwards with his legs drawn up and his hands behind his knees, the man’s chest pressing down on the boy’s shoulders. Two other men stood on either side of the holder to prevent him from slipping. The doctor then set out to locate the stone in the boy’s bladder digitally—the fingers of one hand being inserted into the boy through the boy’s rear end, the fingers of the other hand being lightly placed on his lower abdomen—in order to manoeuvre the stone to the lowest part of the bladder. Then internal incisions were made to allow the stone to pass into the urethra, from where it could be extracted either manually or with an instrument. The operation was an exercise in one torment after another, the hope of complete success and recovery obviously minimal, and the child’s full experience in this particular instance perhaps utterly beyond imagination.21 With Pliny’s help the imagination can run riot. A gum from the vine, he says (HN . ), is good for children’s sores, the wild gourd colocynthis for inflammations of babies’ heads (HN . ), basil, especially with goose grease, for their ears (HN . ). Rocket soothes their coughs (HN . ), the mere touch of nettle restores prolapse of the anus (HN . ), a mixture of wild lupine, rue, and pepper treats worms (HN . ), glycyside is a cure for bladder stones (HN . ). Remedies of this sort can be regarded as folklore remedies, having a rational basis in generations of practice and, perhaps, observed benefits. But Pliny can often combine the 21 Bloodletting: strongly opposed by Galen: see references in Bertier : . Bladder stones: Siegel : –; Grmek : –,  (attributable to malnutrition of nurses and nurslings); Sallares : ; Garnsey : ; cf. Gal. De loc. aff. .  (ischuria as a frequent side-effect in children). Procedure: Celsus, Med. . . ; cf. Nutton : : ‘Celsus’ lithotomy, although perfectly acceptable in itself, must have led to many deaths simply through the infection of the wound, although . . . the Romans knew the antiseptic properties of wine and vinegar, and the belief that all operations were inevitably failures is a considerable exaggeration.’ Cf. also Celsus, Med. . . ; . . ; . . – (cosmetic surgery).

Keith Bradley
rational folkloristic remedy with preventative treatments in a way that leaps back to the miraculous and the magical:22 Babies that are troubled with curdled milk have a preventative in lamb’s rennet taken in water; or if the trouble has occurred with milk already curdled it is dispersed by this rennet given in vinegar. For dentition the brain of a sheep is very beneficial. The inflammation of babies called siriasis is cured by the bones found in dog’s dung worn as an amulet, and hernia in babies by bringing a green lizard to bite them while asleep. Afterwards they fasten the lizard to a reed and hang it in smoke, and they say that as it dies the baby recovers. The slime of snails applied to the eyes of babies straightens the eyelashes and makes them grow. Hernia is cured by the ash of snails applied for thirty days with frankincense in white of egg. There are found in the little horns of snails sandy grits; worn as an amulet these make dentition easy. The ash of snail shells mixed with wax checks procidence of the end of the bowel, but the ash should be mixed with the discharge that exudes when the snails are pricked. A viper’s brain tied on with a piece of his skin helps dentition. The same effect have also the largest teeth of serpents. The dung of a raven attached with wool as an amulet cures babies’ coughs. Certain details can scarcely be included as serious items, but I must not omit them, since they have been put on record. As a remedy for hernia in babies they recommend a lizard; there should be taken a male, which can be recognised by its having one vent beneath the tail. The necessary ritual is: that it must bite the lesion through a gold or silver barrier; then it must be fastened in an unused cup and placed in smoke. Incontinence of urine in babies is checked by giving in their food boiled mice. The tall, indented horns of the beetle, fastened to babies, serves as an amulet. In the head of the boa is said to be a little stone, which is spit out by it when in fear of violent death; they add that dentition is wonderfully aided if the creature’s head is cut off unawares, the stone extracted and worn as an amulet. The brain too of the same creature they recommend to be worn for the same purpose, or the stone or little bone found on the back of a slug. A splendid help also is the brain of a ewe rubbed on the gums, as for the ears is goose grease put in them with juice of ocimum. On prickly plants are grubs which are rough and downy. These worn by babies as an amulet are said to effect an immediate recovery when part of their food sticks in the throat. (HN . –)
Pliny is sometimes characterized as a credulous writer, but as this passage indicates he is not altogether uncritical and can dis22 Folklore remedies: Bonet : –. Glycyside: cf. Diosc. Mat. med. . . Preventative treatments: Bonet : , connects Pliny’s interest in preventing illness with the high mortality rate.
The Roman Child in Sickness and Health

tinguish (to his own satisfaction) between the serious and the silly. He has a particularly venomous attitude towards the Magi (the ‘they’ of this passage), and dismisses their medicinal remedies as utter nonsense, even though to a modern reader they seem no more outlandish than many of the treatments he himself recommends as beneficial. What Pliny writes, however, confirms the well-known fact that the dividing lines drawn in antiquity between medicine, magic, and the miraculous were very narrow, so that it is best to draw no lines at all but to think of all three categories as elements that merged into one another along a single continuum of restorative strategies. To Galen (De praecog. . ), notably, it was a source of pleasure that his medical rivals slandered him as a wonderworker and magician because they could not compete with his cures, while the beneficiaries of the wonderworker Jesus of Nazareth do not seem to have complained at all because they had been cured by a man who made no claim whatsoever to being a trained physician.23 Both as remedies and as protective devices Pliny has much to say about amulets, and it is on this form of the magical that I want to concentrate finally in order to develop the imaginative picture of the life of the Roman child as it was actually lived. Observe, for example, Pliny’s recommendation that teething infants or infants with coughs should have red iris tied to them (HN . ), that the pith of a branch of the wild fig, pulled from the tree and stripped of its bark by a pre-adolescent boy with his teeth, will keep the boy free from scrofulous sores if tied on before sunrise (HN . ), that a dolphin’s tooth or the tooth of a canicula worn as an amulet takes away a child’s sudden anxieties, that a frog tied backwards on an infant’s skull dampened with a cold sponge cures siriasis (HN . ). General medical protection was afforded by a stag-beetle hung from a child’s neck (HN . ), from branches of coral worn by babies (HN . ), from amulets of amber (HN . ), malachite (HN . ), and galaxias (HN . ). 23 Magi: see especially HN . ; cf. Nutton : ; French : –. Dividing lines: at D. . .  Ulpian draws a distinction between doctors and those who heal by means of incantations, imprecations, and exorcisms, acknowledging, however, that the latter were credited with genuine cures by grateful patients; cf. Sallares : – on the minimal use in antiquity of Hippocratic medicine (only on the part of the educated elite).

Keith Bradley
Pliny was not alone in recommending the use of amulets. Doctors did so as well. Soranus was categorical that they had no real healing value, but he was aware of the psychological relief they might none the less bring to those who used them: ‘even if the amulet has no direct effect, still through hope it will possibly make the patient more cheerful’ (Gyn. . ). The pharmacological writer Dioscorides was equally sceptical about the effectiveness of amulets but still recorded their use (Mat. med. . –); and Galen (De simpl. med. . ) wrote both of the general belief in their efficacy and his own use of one particular kind that he thought had therapeutic value. He had found the green jasper to have the kind of power others attributed to stones in general and that in the form of an amulet jasper helped the stomach and oesophagus. He conducted a personal experiment, making a necklace of small jasper stones which he wore from his neck, to the benefit of the area they touched. Even more to the point Galen also told of an amulet (De simpl. med. .) made from the root of the peony which cured children suffering from epilepsy. He knew of a boy who had been kept free from epileptic attacks for eight months by this means but who was afflicted with seizures once the amulet slipped off. When a new amulet was put around his neck, however, the boy recovered. To convince himself of the causal connection Galen in turn removed the new amulet, saw that the boy became convulsive once more, and so placed another piece of peony root around the child’s neck. The boy remained healthy and free from seizures thereafter. Galen believed that a logical explanation for what he had seen could be found in the drying nature of the peony root. But those who used and benefited from the peony may have been less interested in the logic of the matter than the amazing relief the peony amulet seemed to bring.24 The evidence on amulets does not apply to children exclusively, but the doctors’ sober statements help validate what at first seems the far less credible testimony of Pliny. In society at large opinions probably differed about the relative values of individual amulets. Where Galen is confident about green jasper and the root of the peony a much later source speaks of tying ‘bits of coloured thread round wrists, arms, and necks’ and fastening 24 Dioscorides: referring to selenite, jasper, and serpentine; see Riddle : –.
The Roman Child in Sickness and Health

‘moon-shaped plates of gold, silver, or cheaper material’ on infants as the work of ‘foolish old women’, a practice apparently as absurd as the incantations the women chanted to avert evil (‘at the same time licking the babies’ foreheads with their tongues and spitting, blowing to each side’). Whatever the degree of credibility placed in them, however, there can be no doubt about their use. The surviving physical objects themselves provide whatever confirmation might be needed. A standard catalogue of GraecoEgyptian amulets reveals an abundance of extant charms made from semi-precious stones and metal that were worn to avert all forms of evil, the evil of sickness included. The charms show protective deities or animals and are sometimes accompanied by inscriptions or images that indicate their purpose. Thus the common illustration of a bent-over grain reaper, sickle in hand, represents the sciatica or lumbago associated with the backbreaking toil that the wearer of the amulet wished to avoid. Other amulets combated stomach disorders, digestive tract ailments, colic, fever, eye diseases, consumption, and, in women, disorders of the uterus. In the case of children, therefore, the use of such devices should be taken seriously as a widespread response to dealing with their predisposition to sickness, and children themselves, from infancy on, must be pictured as commonly and noticeably wearing them. Amulets were an integral element of the culture of Roman childhood.25 Two types of charm especially associated with children were the phallic amulet and the bulla. Children sometimes wore the former 25 Pliny: Bonet : – associates Pliny’s amulets with the world of wise women. Later source: a scholium (Migne, PG .  BC) on Greg. Naz. Or. .  on baptism (0001 Migne, PG .  A), where Gregory remarks that the only amulet or talisman a child needs is the Trinity (cited by Bonner : ); cf. Leyerle : – on John Chrysostom, in whose works a recurring anxiety about children’s illnesses is detectable, recommending that Christian parents ‘apply the sign of the cross, or hang a small gospel text around their children’s necks for protection’, instead of other devices of dubious merit in common use. Standard catalogue: Bonner ; cf. also Zazoff : –. For the practice of accompanying the use of amulets with magical incantations, see Kotansky ; and on the complexity of the cure in folk-medicine, Gordon . Pictured: cf. the remarks of Montserrat : , on erotic amulets: ‘it is important to think of these objects being used by living people, and to imagine how individuals might have felt when they drew the gem onto the finger or hung it around the neck.’

Keith Bradley
in the form of finger rings but other amulets were worn around the neck. They brought the protection of the god of the divinized penis Fascinus, who fended off the influence of the sometimes strikingly visible Evil Eye. Listing reports of bewitchment from far-flung quarters of the ancient Mediterranean, Pliny (HN . –) records the ability of some African sorcerers to kill children with a look, while Plutarch (QC . ), speaking of a seemingly common belief in his day, tells how children, because of their physical immaturity, might be harmed by the gaze of certain people— in some instances even their fathers so that mothers took pains to prevent fathers’ eyes falling on them. The bulla, a golden locket, was in part a symbol of free birth but also an apotropaic device worn by boys from infancy on around the neck until it was discarded when they took the toga virilis. Both have to be understood as just two of many other comparable objects that were always to be seen on children. They were so common that at Caesar’s funeral many women tossed their children’s bullae onto his pyre.26 The widespread resort to amulets is a sign of the widespread awareness in Roman society of the frailty of childhood, and perhaps also of the well-recognized inadequacies of doctors in solving medical problems. It was easier to trust the protective powers of the superhuman than the restorative capacities of mortals: ‘medicus enim nihil aliud est quam animi consolatio’ (Petr. Sat. . ). From a modern demographic perspective, however, it is notable that the time at which the bulla was set aside was roughly the same as the age at which the Roman child’s prospects of greater life-expectancy had increased. Boys were usually between  and  when they assumed the toga of manhood, and girls, especially elite girls, were often about the same age when they first married. (Non-elite girls tended to be in their late teens.) These ages are approximately the same as those by which historical demographers suggest that children had emerged from the greatest period of deathly danger and could look forward to a relatively safe number of years before them until the next steep 26 Phallic amulet: Johns , with Varro, Ling. . . Strikingly visible: as on a Tripolitanian relief shown in Polidori et al. : ; see further Dunbabin and Dickie . Bulla: Palmer , at enormous length, stresses that it was the contents of the locket that afforded protection rather than the locket itself. How conscious Romans were in everyday life of this strict distinction it is difficult to tell. Caesar’s funeral: Suet. Jul. . .
The Roman Child in Sickness and Health

mortality gradient, that of old age, came into play. It is as if there were an understanding that the child’s survival to the mid-teens was a highly significant accomplishment, acknowledged, for example, in the way Ovid, telling mythological tales, can leap from the birth of a mythologically important infant to the child’s almost unbelievable emergence as a young adult, with the years of childhood left completely unmentioned. Thus in Fasti (. –) the bastard son of Jupiter by Callisto is suddenly a youth of  as he encounters a mother an irate and indignant Juno has transformed into a bear; and Romulus and Remus are infants one minute and -year-old men the next when the story of the City’s founding is told (. –). There is a certain correspondence here with Celsus’ incidental description (Med. . . ) of the period between the ages of  and  as ‘the most stable part of life’.27 One story Ovid tells (Fasti . –) is that of Ceres’ search for her lost daughter Persephone. In Eleusis Ceres encounters a peasant named Celeus, who informs the goddess that he has an infant son at home lying sick in his cradle unable to sleep. Ceres goes to his home and finds his wife and -year-old daughter stricken with grief, all hope of the son’s recovery lost. But the sadness caused by the boy’s impending death gives way to elation when the goddess miraculously heals the boy by performing a sort of mouth-tomouth resuscitation: she quite literally breathes new life into him, so that his pallor flees and his strength returns. ‘There was joy in the whole household’ as a result, Ovid says (. ).28 It is from literary sources that the quality of life in Roman antiquity, in all its varied detail, is most completely understood. From the material surveyed here a high consciousness in society of children’s susceptibility to illness, especially fatal illness, is readily admissible. Across time many parents are likely to have found themselves in the despairing position of Celeus and his wife when their children fell sick, as they are likely to have sympathized with Valesius’ fears and the threat to family continuity his children’s illness represented. Losing a child was something 27 Inadequacies: cf. Celsus, Med.  prooem. ; . . ; . . ; .  C; . .  B. For literary sources critical of doctors, see Amundsen : ; Ferngren ; Nutton . Toga of manhood: evidence collected in Dolansky . For the little that is known about the coming of age of girls, Sebesta : ; George : . On marriage ages, Treggiari : –. 28 For Ovid’s treatment of the story, Fantham : –.

Keith Bradley
pressingly real, almost calculable, to be anticipated. Parents accordingly had to find ways of coping with this reality, and when confidence in doctors was lacking or failed, recourse to the amulet was one culturally distinct method of trying to stall what must often have seemed the inevitable. Faith was put in the other-worldly in the broadly comparable way that a Victorian mother such as Catherine Booth, when the inevitable had actually occurred, made sense of the loss of her children by believing that they had entered upon a better, blissful state of existence in a Christian Heaven. By modern western standards Roman children suffered from appalling levels of sickness. Most of their illnesses could not be cured, and even trivial conditions could become fatal. When survival was all that mattered, every expedient had its place.29 Many children did of course survive. Death did not take them all. When that happened, however, as Celeus and Valesius knew, an element of the miraculous was involved, and perhaps an answer to prayer. In the process of recovering Roman family experience, a prime place might indeed be properly allowed for parents regularly praying to their household gods for their children’s safe entry into the adult world. Marcus Aurelius (Med. . ; cf. . ) was certainly aware that parents prayed not to lose their children, and recommended that their prayers should rather be not to fear their loss. I imagine that his advice was largely disregarded and that his was very much a minority attitude. When the age of adulthood was reached, the ceremony by which healthy Roman boys became men, the assumption of the toga virilis, was a presumably memorable and joyful event. It involved the laying aside at home of the bulla, the ritual exchange of the toga of childhood for the toga of manhood, the offering of prayers to the household gods, and a procession from the family home to the Forum and eventually the Capitol where sacrifices were duly performed. It was, that is to say, an occasion for thanksgiving, and at that moment in a family’s life there was much for which to be thankful.30 29 Catherine Booth: Hattersley : , . For a comparable classical view, Plut. Cons. ad ux. ; see also August. Conf. . : fear of a child’s death prompting immediate thoughts of baptism and the cleansing of sins; and Gregory of Nyssa, On Infants’ Early Deaths (the title itself is revelatory) with Gould : : a providential God ending the lives of evil children to prevent the growth of wickedness and to show mercy to their better parents. 30 Ceremony: Dolansky .
 Parent–Child Conflict in the Roman Family: The Evidence of the Code of Justinian Judith Evans Grubbs
T chapter looks at parent–child conflict in the third-century empire as seen in imperial rescripts, replies given by the emperors to petitioners who had written them for help or information on points of law. Our major source for these rescripts is the Codex Justinianus (Code of Justinian), which preserves nearly , rescripts from before the reign of Constantine. Most are subscriptiones (that is, replies to private individuals rather than to public officials), and date to the third century, more than half being from the reign of Diocletian, under whom the first collections of rescripts were made.1
I would like to express my thanks to Antti Arjava, Tom McGinn, Susan Treggiari, and the two readers for OUP, all of whom have read this chapter in some incarnation and have greatly helped to improve it. Needless to say, any errors or infelicities that remain are my own. 1 The th-cent. Cod. Just., our source for most surviving rescripts, drew on two earlier collections of rescripts made under Diocletian in the late rd cent., the Codex Gregorianus (containing rescripts from the reigns of Hadrian to Diocletian) and the Codex Hermogenianus (containing rescripts from Diocletian’s time, almost all dating to –). For the editing process, see Honoré  and Corcoran : esp. –. The Digest also cites dozens of rescripts of Hadrian and Antonine and Severan emperors, sometimes quoting them verbatim. Some rescripts in the Digest are subscriptiones, replies to the petitions of private subjects (including women), but many are actually epistulae, letters to imperial officials, whereas the great majority of Cod. Just. rescripts are subscriptiones. Additionally, some rescripts are preserved in late antique legal collections, such as the ‘Fragmenta Vaticana’, found in FIRA ii. Both subscriptiones and epistulae are rescripta, written imperial responses. This chapter looks at subscriptiones, which take their name from the fact that the emperor’s reply was subscripta, written underneath

Judith Evans Grubbs
Subscriptiones were the responsibility of the imperial secretary for petitions (a libellis), who was usually a jurist with the legal knowledge required to assess and comment on the petitioner’s situation. According to Tony Honoré, not only the composition, but sometimes the decision itself was that of the imperial secretary, not the emperor. This is probable for the mid-third century, when rescripts continue to set forth good classical Roman law despite a rapid turnover of military emperors who can have had little time or training for responding to private petitioners, but less likely for the earlier period. For the sake of conciseness, however, the rescripts will be referred to as the rulings of the emperors themselves, not of their more learned secretaries.2 As with all our ancient sources, using the rescripts as evidence for ‘lived reality’ presents problems. For one thing, the Code of Justinian preserves only the emperor’s replies, not the original petitions, and often the compilers abbreviated even the reply so that only the legal gist was retained. Sometimes, however, a rescript will recapitulate the original inquiry and enable us to reconstruct the situation, at least in part. Moreover, the Code does not include all, or even most, of the rescripts issued by second- and third-century emperors. It was intended to serve as a guide to the law of Justinian’s own day, and so preserves only the original petition. For the sake of conciseness, I will call them rescripts rather than use the more cumbersome term subscriptiones. Scholarship on the rescript system is extensive. In addition to Honoré  and Corcoran , see Coriat ; W. Turpin, ‘Imperial Subscriptions and the Administration of Justice’, JRS  (), –; and W. Williams, ‘The Libellus Procedure and the Severan Papyri’, JRS  (), –. There is much also in F. Millar, The Emperor in the Roman World (Ithaca, , repr. ). 2 Honoré () has distinguished the styles of a number of secretaries a libellis of the rd cent., identifying several of them with well-known jurists. ndcent. imperial constitutions reveal more of the individual emperor’s personality and are more likely to have been their own compositions (though few of those extant are subscriptiones); see W. Williams, ‘Individuality in the Imperial Constitutions: Hadrian and the Antonines’, JRS  (), –. For the continuation of classical Roman legal principles in rd-cent. rescripts, see A. Watson, ‘Private Law in the Rescripts of Carus, Carinus and Numerianus’, Tijdschrift voor Rechtsgeschiedenis,  (), –. This, and Watson’s ‘The Rescripts of the Emperor Probus (– A.D.)’, Tulane Law Review,  (), –, were pioneering in that they also looked at the rescripts as evidence for ‘real life’ of Roman provincials living in a world quite different from that of the classical jurists.
Parent–Child Conflict in the Roman Family

those texts that were still relevant and legally valid in the eastern empire of the sixth century. Nevertheless, the rescripts provide a valuable and underexploited source of information for our understanding of social relations in Roman society in a period for which we have few literary sources. The recipients of imperial rescripts represent a much broader spectrum of the population of the Roman empire than do the authors of classical literature. About  per cent of the rescripts are addressed to women, and the proportion rises to around  per cent in the reign of Diocletian. Soldiers, freedpeople, and sometimes even slaves also receive replies. Thus they allow us to hear, albeit indirectly, the voices of those who are otherwise largely silent. Many petitioners are from the provinces rather than Rome and Italy; indeed, almost all of the rescripts from the reign of Diocletian are from the eastern half of the empire. Most of the rescripts preserved in the Codex Justinianus postdate the Constitutio Antoniniana of , when the emperor Caracalla bestowed Roman citizenship on virtually all free subjects of the empire. Before this time the population of much of the empire, especially the east, had not been citizens and had by and large come under the native law of their province. After , Roman law was to apply throughout the empire, although it appears that often local law and especially custom continued. Consequently, third-century rescripts often respond to inquiries from recently enfranchised citizens who appear to have little knowledge or understanding of Roman law. The emperors repeatedly state what the law is and insist on its validity over non-Roman practices, and in so doing they provide evidence about local custom in the less ‘Romanized’ areas of the empire. Thus the rescript process itself was an important means of ‘Romanization’ in the provinces.3 This does, of course, raise questions about the ‘Roman-ness’ of the people of the rescripts and the relevance of their situations to the study of the ‘Roman family’—questions which I shall attempt to answer in the course of this chapter. It wasn’t easy to get a petition to the emperor, or indeed, to get a response. A petitioner would have to see to it that his or her petition was actually delivered to the emperor—not a light 3
Coriat :  and esp. –.

Judith Evans Grubbs
undertaking, especially in areas not regularly visited by the imperial court. Private subjects did not have access to the cursus publicus or other public means of transmitting requests to higher authorities. They would either have to travel to the place where the emperor was residing or send a trusted representative to deliver the petition—for instance, a freedman or son, or in the case of female petitioners, a husband. The petition, with the emperor’s reply written underneath, was then posted, apparently for a period of about thirty days, in a long sheet with other petitions answered at the same time, in a public place wherever the emperor was then residing. Petitioners were also responsible for getting the emperor’s reply while it was posted; rescripts were not delivered to the individuals to whom they were addressed, or even to the governor of the province where the petitioner lived. Second- and third-century emperors did travel a great deal throughout the empire and the rescript system illustrates the ‘hands-on’ style of ruling and the (theoretically) easy access to the imperial ear that Roman subjects were supposed to be able to enjoy. But the obstacles to delivering a petition to the emperor and receiving a reply were formidable, given ancient communication and transportation conditions, even in times of peace and prosperity. And frequently the emperor simply sent the petitioner back to his or her provincial governor, which might mean a delay of some months, as the petitioners would have to wait until the governor was near by on his annual assizes (conventus).4 Still, some petitioners managed to obtain not just one, but two or more, imperial responses to their petitions. ‘Since, not content with the rescripts which you had received in reply to your first petitions, you again wanted to make a plea, you will take back a rescript according to law.’ So wrote Diocletian and Maximian, with barely concealed annoyance, to a woman named Calpurnia Aristaeneta.5 Calpurnia was seeking help regarding her son, who 4 Already by the Antonine period, a standard phrase was being used in imperial rescripts: ‘You will be able to approach the man who is in charge of your province’ (D. . . –). On the conventus, see N. Lewis, ‘The Prefect’s Conventus: Proceedings and Procedures’, BASP  (), –. 5 Frag. Vat. , posted  Feb.  at Milan. This rescript is of interest because it illustrates the editing process of Justinian’s compilers. They split in two the imperial reply (which emanated from Diocletian’s western colleague Maximian, since it was posted at Milan), and put the pieces into two different
Parent–Child Conflict in the Roman Family

(she claimed) had exhausted his patrimony in lavish gift-giving; she wanted to know if she could recover the gifts in his name. Like so many other petitioners, she is told to see her provincial governor. The domestic nature of Calpurnia Aristaeneta’s petition is not unusual. Many rescripts, especially those to women, involve family affairs, often disputes over inheritance. It is these rescripts that interest me, for they give us a glimpse of the internal functioning—or dysfunctioning—of otherwise nameless families throughout the empire. Of course, given the nature of the rescript system, that is, as (in part) an imperial ‘advice bureau’ to deal with the legal problems of Roman subjects, what we see are the areas of tension and dispute in the society, the pressure points where harmony has broken down and at least one person has felt compelled to resort to the law. To assess family relations in the Roman empire solely on the evidence of imperial rescripts would be rather like assessing contemporary American society solely from letters to advice columnists like ‘Dear Abby’ or the late Ann Landers. Happy families do not usually write to the authorities for help; only when all other means have failed will they go to that much time and trouble. The vast majority of family disputes would be dealt with either within the household or by private arbitration, or, if the problem persisted, by petitioning local or provincial officials.6 But sometimes matters had come to such a pass that recourse to the highest authority in the empire was the only solution. In  Valerian and Gallienus received a petition from a woman books of the Code: Cod. Just. . .  (under the title de inofficiosis donationibus) and Cod. Just. . .  (under the title de donationibus). Neither of these actually preserve the rescript’s opening sentence quoted above, which was too ad hominem for Justinian’s compilers, as they were interested in the rescripts purely as statements of law. The entire rescript is preserved in one piece in a late antique western legal compilation known as the Vatican Fragments, with the exception of the last sentence of Cod. Just. . . , which does not appear in Frag. Vat.  (and therefore may be a Justinianic addition). On all this, see Corcoran : –, –, –. The Fragmenta Vaticana date to the early th cent.; see Corcoran :  and Robinson : –. 6 For settlement of family disputes on a local level, see D. W. Hobson, ‘The Impact of Law on Village Life in Roman Egypt’, in B. Halpern and D. W. Hobson (eds.), Law, Politics, and Society in the Ancient Mediterranean World (Sheffield, ), –.

Judith Evans Grubbs
named Galla, who was complaining about her children’s disrespectful behaviour and perhaps alleging physical abuse also. The emperors’ reply betrays a hint of exasperation at having to deal with such dirty family linen: It certainly seems more fitting that any controversies arising between you and your sons be brought to an end within the household. But if the matter is such that because of their injuries to you, you have proceeded to go to law and (to seek) retribution, the governor of your province, after you have approached him, will certainly order that the usual legal rule be employed concerning disputes over money. Moreover, he will force your sons to show the reverence owed to a mother, and if he discovers that their wickedness has advanced to unkinder injuries, he will punish more severely the insult to filial duty. (Cod. Just. . . )7
Surely it would have been better for such sordid family disputes to be dealt with in private, intra domum. But Galla has already started legal proceedings, and the emperors give her the imperial go-ahead to approach the governor, who will exact the proper filial reverentia from her undutiful sons. (Rather oddly, the Justinianic compilers put this rescript under the title on patria potestas, along with responses to indignant patresfamilias.) Similarly, Diocletian and Maximian reply to Apollinaris: If your daughter is living shamefully and with flagrant foulness, so that you think she should be excluded from inheriting from you, you will have the free choice of your final judgment—if you have been impelled to this hatred because of her deserts and not out of unthinking anger. (Cod. Just. . . )8 7 Cod. Just. . . , posted  June : ‘Impp. Valerianus et Gallienus AA. et Valerianus C. Gallae: Congruentius quidem videtur intra domum, inter te ac filios tuos si quae controversiae oriuntur, terminari. () Sed si ita res fuit, ut iniuriis eorum et ad ius experiundum et ad vindictam processeris, aditus praeses provinciae super disceptationibus quidem pecuniariis consuetum exerceri iubebit ordinem iuris: reverentiam autem debitam exhibere matri filios coget et, si provectam ad inclementiores iniurias improbitatem deprehenderit, laesam pietatem severius vindicabit. PP. XVI k. Iun. Aemiliano et Basso conss.’ 8 Cod. Just. . . , given  June  at Sirmium. ‘Idem (Diocletianus et Maximianus) AA. Apollinari: Si filia tua turpiter et cum flagitiosa foeditate vivit, ut a successione tua eam excludendam putes, si non inconsulto calore, sed ex meritis eius ad id odium incitatus es, postremi iudicii liberum arbitrium habebis. D. XV k. Iul. Sirmi ipsis V et IIII AA. conss.’
Parent–Child Conflict in the Roman Family

Apollinaris has the right to disinherit his daughter for her behaviour, but the emperors do not seem to be very pleased with his decision. Twice the rescript uses the conditional: if you are sure that you are not overreacting in the heat of the moment, and if your daughter really is living as shamefully as you claim (what did she do? have an illegitimate child? practise prostitution?). What the emperors do not say explicitly is that if Apollinaris’ reasons for disinheriting his daughter are not as good as he thinks, she may, after his death, bring a suit for ‘undutiful will’ (de inofficioso testamento) to claim her portion. If successful, such a suit would result in the will being either completely or partially voided, so that Apollinaris’ daughter would receive the ‘Falcidian fourth’ owed to a man’s sui heredes.9 Indeed, this rescript is found in the Code’s title on undutiful wills, which suggests that the Justinianic compilers at least thought that was a possibility. A number of rescripts testify to the popularity of this legal action, revealing how family conflicts could continue even after death. Rescripts concerning financial disputes, especially inheritance, have attracted the interest of several scholars, most recently Jane Gardner in her book on ‘family and familia’.10 This chapter focuses instead on two other issues which occasioned imperial rescripts: the choice of spouse for a young man or woman, particularly one under paternal power, and the ability of a paterfamilias to force his married child to divorce. Rescripts issued in response to these issues illustrate the tension between patria potestas, the theoretically all-encompassing power of the Roman paterfamilias, the demands of parental and filial pietas, and the personal desires of individuals.11 Consent to Marriage and Paternal Power The paterfamilias was responsible for arranging suitable marriages for his children, and his agreement was necessary for the marriage of any child under his power, male or female, to be legally valid. 9 That is, those who would have inherited if he had died intestate: any legitimate children he had (including emancipated children) and grandchildren by his sons. See Cod. Just. . . – (dating from  to ), all on the querela inofficiosi testamenti. For discussion, see Gardner : –; and : –; also B. Nicholas, An Introduction to Roman Law (Oxford, ), –. 10 Gardner . 11 On patria potestas and pietas, see Saller .

Judith Evans Grubbs
The validity of Roman marriage depended on the consent of all parties involved: the bride, the groom, and the paterfamilias of both, if they were still alive.12 But sometimes it was difficult to determine what precisely constituted consent, and at what point consent had to be given, and whether it could be revoked. The need for paternal consent and the power of the paterfamilias over his children’s actions set the stage for conflicts between fathers and their children over choice of spouse, a potential source of family disharmony in any age. The dynamics of consent as found in the Digest and in classical literature have been well studied, but less attention has been given to the third-century rescripts of the Code.13 The rescripts in the Codex Justinianus that concern familial disagreements over consent to marriage indicate how the competing powers and desires of children, fathers, mothers, and guardians led to family disputes which could only be resolved through recourse to legal authority. The rescripts are particularly valuable because they provide ‘case studies’ where the emperors apply legal principles to the actual situations brought to their attention. For instance, the reply of Diocletian and Maximian to Sabinus reveals the potential for conflict between father and son in the choice of marriage partner: The teaching of the laws does not permit even a son in the legal power of his paterfamilias to be forced to take a wife against his will. Therefore, just as you desire, after the precepts of the law have been observed, you are not prevented from joining in marriage to yourself a woman whom you wanted, in such a way, however, that you have your father’s consent in contracting the marriage. (Cod. Just. . . )14 D. . .  (Paulus); cf. D. . . .  (Paulus); D. . .  (Julian). See esp. Treggiari  and : –; also Corbett  and Matringe . R. P. Saller, ‘The Social Dynamics of Consent to Marriage and Sexual Relations: The Evidence of Roman Comedy’, in A. E. Laiou (ed.), Consent and Coercion to Sex and Marriage in Ancient and Medieval Societies (Dumbarton Oaks, ), – makes excellent use of Latin literature to show us ‘legal consent to marriage as an outcome of a dynamic process rather than the imposition of an arbitrary paternal will’ (). 14 Cod. Just. . . , given  Nov. : ‘Idem (Diocletianus et Maximianus) AA. Sabino: Ne filium quidem familias invitum ad ducendam uxorem cogi legum disciplina permittit. Igitur, sicut desideras, observatis iuris praeceptis sociare coniugio tuo quam volueris non impediris, ita tamen, ut in contrahendis nuptiis patris tui consensus accedat. D. non. Nov. Diocletiano A. II et Aristobulo conss.’ 12 13
Parent–Child Conflict in the Roman Family

Evidently Sabinus and his father have disagreed about whom Sabinus is to marry. Indeed, it appears that not only does Sabinus not want to marry the woman his father has picked out for him, but he may have someone else in mind. The emperors tell him he does not have to marry his father’s choice if he is unwilling, but that he does need paternal consent if the union he wants is to be considered legal marriage. This response is quite in keeping with classical law. The jurist Terentius Clemens, writing in the second century, had said that a son in power could not be forced to marry (D. . . ).15 But another second-century jurist, Celsus, stated: ‘If, when his father forces him, he does take a wife whom he would not marry of his own free will, he has nevertheless contracted marriage, which is not contracted between those who are unwilling. He appears to have preferred this course’ (D. . . ). It was also necessary to have the consent of a daughter under patria potestas, but, Ulpian says, ‘if she doesn’t fight against her father’s wishes, she is understood to consent. Moreover, the liberty to dissent from her father’s decision is only allowed her, if her father chooses a fiancé who is disgraceful or unworthy in his way of life’ (D. . . ). In other words, a fait accompli demonstrates consent: once marriage has taken place, the consent of both partners is assumed, even if it only occurred after intense paternal pressure. But tacit consent worked both ways: the famous Hadrianic jurist Julian (as cited by Paulus in the third century) said that a girl’s father was always understood to consent to her marriage, ‘unless he clearly refuses to consent’ (D. . . . ). It was, however, less problematic to interpret lack of objection as unspoken consent in the case of a daughter than that of a son. The children of a daughter in a legitimate marriage would come under the patria potestas of her husband, not her father, whereas the legitimate children of a son in power would come under his father’s potestas and therefore, as sui heredes, would have a claim on their paternal grandfather’s estate.16 15 D. . . ; this comes from the third book of Terentius Clemens’ commentaries on the lex Julia et Papia and so may reflect a provision in the Augustan legislation: see Matringe : . Cf. D. . .  (Paulus). 16 As Susan Treggiari has pointed out to me (on sui heredes, see n. ). After the senatusconsultum Orphitianum of  (which was passed after Julian was writing but before Paulus), however, children did have first claim to inherit from their mother if she died intestate, which would then put them in line to her

Judith Evans Grubbs
Jurists did admit the possibility that a person under paternal power might contract legal marriage without the consent of his or her paterfamilias. But the situations they discussed involve either situations where the father did not know of his child’s union, for understandable reasons (for instance, because the father was away, perhaps in captivity beyond the borders of the empire)17 or where he was unable to come to a rational decision about his child’s marriage due to mental incapacity. Even in the case of the son whose father had gone mad, jurists differed; the matter does not appear to have been definitively settled until , when Justinian extended the policy on daughters of mad fathers (who could marry, as long as their fathers did not actively object) to sons also.18 Papinian says that a son in power who is a soldier father’s estate. Even then, a father’s explicit consent to his daughter’s marriage does not seem to have been as essential as it was to his son’s. See Gardner : – regarding the consent of a mad father (cf. below, n. ). 17 When the paterfamilias was away: cf. Frag. Vat.  (on which see n.  below). In captivity: D. . . .  (Tryphoninus) says the son can marry, and a child of such a marriage will be in the potestas of his grandfather when he returns from captivity. Paulus (D. . . ) says that if a father has been absent for three years and it is unknown where he lives or whether he is still alive, his children (male and female) may contract legal marriage; cf. D. . . .  (Ulpian), mentioning sons only. D. . .  (Julian) says the children (male or female) may marry before the three years is up, as long as the spouse is someone whom the father would not repudiate. Although Corbett (: –) thought that references to a three-year waiting period were post-classical interpolations, this may instead have been an area where jurists disagreed (T. A. J. McGinn, in a personal communication). 18 D. . .  pr. (Ulpian) says that if a grandson in his grandfather’s power wishes to marry and the grandfather is mad (furens), the permission of his father (who would himself be a filiusfamilias to the grandfather) is sufficient, and if it is the father who is mad, then the grandfather’s consent is sufficient. Justinian’s law of  (Cod. Just. . . ; cf. his Institutes .  pr.) settled what Justinian says had long been a matter of debate. Evidently it had been generally agreed in classical law that the daughter of a madman (furiosus, as opposed to one who was captus mente) could marry, but the marriage of the filiusfamilias of a furiosus was more doubtful. The difference between a furiosus and someone who was captus mente seems to be that the former has ‘lucid intervals’ in which he can make reasoned decisions on matters such as his child’s marriage, whereas the latter is permanently mad and so incapable of any decisions: see Gardner : . Justinian extended a constitution of Marcus Aurelius which had allowed children (male and female) of fathers who were capti mente to marry to apply also to children whose fathers were furiosi. Cf. Treggiari : –; : –.
Parent–Child Conflict in the Roman Family

cannot legally marry without his father’s permission (sine patris voluntate).19 Probably such a situation would arise not because his paterfamilias actually refused to consent, but because (like the father held captive outside the empire), he did not know about the union, since his son was away from home. Nowhere in the classical jurists is there a clear statement regarding a filiusfamilias who deliberately defies his father and marries not merely without paternal consent, but against paternal wishes.20 Ancient legal writers appear to skirt the issue, perhaps because of its inherent unlikeliness to Roman legal minds and to Roman fathers. In Sabinus’ case, only the imperial reply is preserved, and so we do not know how the struggle of wills between Sabinus and his father was resolved. Did he manage to obtain his father’s agreement to his own choice of spouse, or was he forced to give in and take a woman he did not want, even against ‘the teaching of the laws’? And what if he had decided to defy his father and live with the woman he wanted without paternal consent? Such a union could not be iustum matrimonium, since that required the consent of all parties—especially that of the couple’s patresfamilias. But the actual legal status of a union contracted without the permission of the paterfamilias is a matter of dispute among scholars today; it may have been iniustum matrimonium, that is, the partners would not have conubium but the relationship would still have some of the legal consequences of marriage. Or it might 19 D. . .  (Papinian). Soldiers did not have the right to marry at all while on service until the reign of Septimius Severus. Thus the possibility that a soldier filiusfamilias might enter legal marriage without his father’s consent did not arise until the end of the nd cent. at the earliest. 20 By ‘classical’ jurists I mean those writing before the mid-rd cent. (, the death of Alexander Severus, is often taken as the cut-off date for ‘classical’ Roman law). A passage in the Sententiae Pauli does suggest that marriage contracted without paternal consent might still be considered legal: ‘The marriages of those who are in paternal power are not contracted legally without his agreement, but once contracted, they are not dissolved. For consideration of the public benefit is placed before convenience to private citizens’ (Sent. Pauli . . ). The expression ‘without his agreement’ (sine patris voluntate) could signify active objection, rather than simply lack of knowledge and thus of consent. But this passage has been interpreted differently by different scholars, and in any case is not by Paulus himself, as the Sent. Pauli date to the late rd or early th cent.; see Robinson : –. Thus this passage was written no earlier than the rescript to Sabinus, and probably later.

Judith Evans Grubbs
have had no validity whatever.21 At the very least, however, Sabinus could expect to be cut off by his father, losing both financial support now and his inheritance later. And any children of a marriage made by a filiusfamilias without paternal consent would have been considered illegitimate.22 Concerns about the legitimacy and inheritance rights of a grandchild appear to be behind another rescript from earlier in the third century. Alexander Severus assures a woman, Maxima: 21 Treggiari (:  and : ) considers it iniustum matrimonium, citing in support D. . .  (Paulus), on which see n.  below. But Corbett (: ), also citing D. . . , thought the union was null. Matringe (: –) says that the union’s validity ‘pouvait être contestée’, and even if a daughter got a magistrate to validate her marriage without paternal consent, she risked not getting a dowry or other financial help while her father was alive or after his death (sons who had a means of making a living or a peculium castrense or quasi-castrense would be less dependent on paternal resources). Matringe also cites D. . . .  (Ulpian), which says that a husband cannot prosecute his wife for adultery by a husband’s right (ius mariti) if the offence was committed before her father had consented to the union (he could still prosecute by ius extranei, however). Note that all these passages involve the marriage of a daughter in power, not that of a son, which would have been more problematical (see at n.  and cf. the later Sent. Pauli . . , cited n. ). This was really an issue to be decided on a case-by-case basis, when someone had an interest in either challenging or maintaining the marriage’s validity: ‘what is clear [is] that this was a very foggy area and that the situation might not be clarified until something happened, e.g. the birth of a child, [or] the death of a husband or father’ (Treggiari, personal communication). 22 Cf. D. . . .  (Ulpian), involving the marriage of an emancipated son against his father’s will. The son’s marriage was valid, as he was no longer under patria potestas and so did not need his father’s consent. Ulpian says that after the son’s death, his son was able to inherit from his grandfather, though he had not been mentioned in the grandfather’s will. Children of an emancipated son could inherit from their paternal grandfather under the succession rules put forth in the praetor’s edict, as could the emancipated son, if he had not been explicitly disinherited (though in this case he probably had been); see Gardner : –. This was possible through a legal fiction that the praetor had ‘cancelled’ (rescindit) the emancipation; see Gardner : , citing D. . . .  (Paulus). But, continues Ulpian, this does not mean that a legitimate (iustus) son of an emancipated father ceases to be legitimate because his father’s emancipation was (theoretically) cancelled. Thus even if a man had married a woman so ignominosa as to be a source of disgrace to himself and his father, his son by that woman is still able to inherit from his grandfather. The implication is that if the son had not been emancipated when he married against his father’s wishes, a child born of the marriage would have been illegitimate. See Gardner : – for explanation of this passage.
Parent–Child Conflict in the Roman Family

If, as you claim, the father of your former husband, in whose power he was, did not object to your marriage after he had learned of it, you ought not to fear that he will not acknowledge his own grandchild. (Cod. Just. . . ) 23
Maxima and her child’s father are no longer married, and although they may have been divorced, her anxiety about her father-in-law’s acknowledging his grandchild suggests that she was a widow. If her husband had died while still under patria potestas, then he would have owned no property in his own right, for everything he had possessed would legally belong to his paterfamilias. If Maxima’s marriage had been iustum matrimonium, her husband’s child would be under the patria potestas of his paternal grandfather, and therefore his heir. Maxima wanted to be sure that her son’s inheritance rights were safeguarded. It appears that Maxima’s father-in-law did not originally consent to the marriage, but when he learned about it, he didn’t object. The legal question of a son in power marrying without his father’s knowledge was mentioned above, as was the analogous situation regarding a daughter.24 A case similar to Maxima’s is discussed in a juristic commentary, which mentions a ruling of the jurist Paulus that a marriage made by a son in power in his paterfamilias’ absence was legal if the father had never raised an objection to it after he returned and that therefore after the son’s death, his wife had an action for the return of her dowry.25 (Maxima, however, does not seem to be concerned about reclaiming her dowry; perhaps she had not had one.) In Maxima’s case, her husband’s father may have been away for a long period of time and his son’s marriage may have been arranged without his knowledge; we might compare the third marriage of Cicero’s daughter Tullia.26 Or perhaps Maxima and her 23 Cod. Just. . . , not dated, but between  and : ‘Idem (Alexander) A. Maximae: Si, ut proponis, pater quondam mariti tui, in cuius fuit potestate, cognitis nuptiis vestris non contradixit, vereri non debes, nepotem suum ne non agnoscat.’ 24 See above, at nn. –. On the daughter’s marriage: D. . . .  (at n. ). 25 Frag. Vat. , a case involving ‘L. Titius’ and his wife ‘Septicia’ (a filiafamilias herself; her father must have consented to the marriage since we are told that he provided the dowry). See Corbett : –. The names suggest that this may be a hypothetical case, but the principle would stand. On the Frag. Vat., see n. . 26 On Tullia’s third marriage, see Treggiari : –.

Judith Evans Grubbs
husband simply eloped.27 In any case, all appears to have ended well: Maxima’s father-in-law implicitly approved his son’s marriage by not objecting to it, and her son was his grandfather’s heir. Possibly Sabinus’ situation, discussed above, had a similar outcome. Although Sabinus’ father had objected to his choice of wife in the beginning, if he had later relented and accepted the union, what began as a non-legal marriage would have become legal by virtue of the father’s belated consent. In that case, however, a child born before consent was given would not come under his grandfather’s potestas, and so would not inherit upon intestacy.28 Legally, both men and women who were no longer under paternal power did not require anyone’s consent to enter into marriage. But even if she was legally independent, a young woman marrying for the first time could expect the input of her mother and her guardian, and perhaps also of other interested relatives and friends, though she was not legally bound to follow their suggestions. Sometimes pressure was applied in the form of testamentary bequests promised if she married the man of the testator’s choice. The terms of such legacies were valid, as long as marriage with the designated spouse(s) was possible and not dishonourable.29 27 This would also explain the absence of any mention of dowry. In cases where a paterfamilias’ consent was unforthcoming, elopement presented an alternative. Elopement under the pretext of abduction (raptus) was harshly punished a century later by the Emperor Constantine (Cod. Theod. . ), who also penalized parents who had consented after the fact to a marriage brought about by abduction. See J. Evans Grubbs, ‘Abduction Marriage in Antiquity: A Law of Constantine and its Social Context’, JRS  (), –. 28 Cf. D. . .  (Paulus): if a paterfamilias died without ever knowing about his daughter’s union, her child could never be considered the legitimate son (iustus filius) of the man from whom she conceived him, even if the grandfather died before the child was born. The daughter’s union is characterized as coniunctio, not iustum matrimonium. This is cited as Paulus’ opinion, not necessarily shared by all. But it implies that if a man was unaware of his daughter’s union in the first place and died without ever knowing about it (let alone approving of it), it cannot be a legal marriage. 29 D. . .  (Gaius, in a commentary on the Augustan marriage legislation), says that a legacy left on condition that a daughter marry ‘Titius’ was perfectly legal, unless Titius was indignus. In that case, the woman’s father had essentially prevented her from making an honourable marriage, against the spirit of the Augustan law which said a father could not prevent his children marrying. (On the other hand, leaving a legacy on condition that she marry anyone but Titius was valid, since she still had plenty of men to choose from.) Cf. D. . .  (Terentius Clemens, also commenting on the lex Julia et Papia).
Parent–Child Conflict in the Roman Family

Several passages in the Digest refer to cases where a father left his daughter a legacy to take effect cum in familia nupserit /nupsisset— ‘when she has/had married in the familia’.30 Given the range of meaning embraced by the word familia, it is not clear exactly who these prospective husbands were, but they must have been connected in some way to the legatee’s household or family (that is, related by blood or through marriage). In all these cases, the legatee had a choice: if she wanted the inheritance or legacy, she had to marry the man designated by the testator, but she was free to refuse the marriage and lose the legacy. Therefore, jurists felt, the condition was reasonable.31 Some of the Digest cases may be hypothetical, but similar situations in third-century rescripts were certainly real, and it is significant that all involve the marriage of young women whose right to receive an inheritance or legacy depends on their marrying the testator’s choice of husband. Despite the legality of such conditions, the rescripts suggest that there were young women who tried to invalidate the condition and still accept the legacy. 30 D. . .  (Julian); D. . .  (Scaevola); D. . .  (Scaevola, evidently the same case); cf. D. .  pr. (Paulus), where a woman Paula makes a certain Callinicus (her husband?) part heir and leaves his daughter (her stepdaughter?) a legacy cum in familia nupsisset. Passing references to legacies given on this condition at D. . .  (Paulus) and D. . .  (Ulpian). All these examples listed in Matringe : . See Saller : – on the meanings of familia; he relies mainly on literary uses and does not cite these passages. He finds that when familia refers to kin (as opposed to slaves), it is used not of cognate relatives, but agnates only. I can find no discussion of these passages or the meaning of in familia in Gardner . Cf. D. . .  pr. (Papinian): a father in his will designated his daughter, Severiana Procula, for marriage with a cognatus, Aelius Philippus, and left her an estate if she married him. (Aelius Philippus was to get it if she didn’t marry him.) 31 But marriage, even with a designated spouse, was supposed to be a matter of consent. Thus Paulus ruled that a stipulation in a contract between a married couple that penalized anyone who stood in the way of the marriage of their children (who were step-siblings) to each other had not been made ‘according to good morals’ (secundum bonos mores), ‘because it did not seem honourable for marriages, either in the future or already contracted, to be constrained by the bond of punishment’ (D. . .  pr.). Cf. Cod. Just. . .  (Alexander Severus to Menophilus, ): pacts that say a couple may not divorce are invalid and stipulations to enforce such an agreement are of no effect. In the th cent., however, the western emperor Honorius declared that the betrothal pact a father had made for his daughter was to be enforced even after his death (Cod. Theod. . .  0001 Cod. Just. . . , dated ).

Judith Evans Grubbs
Alexander Severus informed a certain Licinia that she was mistaken if she thought she could still take the bequest left by her paternal uncle on the condition that she married his son (her cousin), when the son had died before the marriage could take place.32 Marriage with a cousin was in fact a perfectly reasonable prerequisite for inheritance, as Caracalla told Cassia in a rescript posted at Rome (which suggests that Cassia was probably from Rome or elsewhere in Italy): If you did not obey the condition under which you were instituted heir by your mother’s will, the appointment of a substitute heir is valid. For it cannot appear that she imposed widowhood (i.e. spinsterhood) on you under the pretext of (demanding) a shameful marriage, since she wished with justifiable reason to join you in marriage with her sister’s son, your cousin. () Nor do you need exceptional aid since, on the basis of what you included in your petition, it is revealed that it was not due to him that you did not satisfy the final judgment of your mother, the testator. (Cod. Just. . . )33
Marriage to relatives may not have been unusual, at least among those who had property to leave.34 The purpose was evidently to keep wealth and property within the kin group; thus Licinia’s uncle and Cassia’s mother made their bequests contingent on marriage between cousins. The reply to Cassia also 32 Cod. Just. . . , Emp. Alexander Severus to Licinia, posted  Dec. : ‘Legatum sive fideicommissum a patruo tuo relictum tibi sub condicione, si filio eius nupsisses, cum mortuo filio, priusquam matrimonium cum eo contraheres, condicio defecerit, nulla ratione tibi deberi existimas.’ Cf. D. . .  pr. (n.  above) for a similar case, where the daughter died before she was ready to wed (viripotens). 33 Cod. Just. . .  (posted at Rome  Mar. ): ‘Imp. Antoninus A. Cassiae: Condicioni, sub qua testamento matris tuae heres instituta es, si non paruisti, substitutio locum habet. Nec enim videri potest sub specie turpium nuptiarum viduitatem tibi indixisse, cum te filio sororis suae consobrino tuo probabili consilio matrimonio iungere voluerit. () Nec extraordinario auxilio indiges, cum ex his quae libello complexa es declaretur non per eum stetisse, quominus supremae voluntati matris tuae testatricis satisfieret. PP. VIII id. Mart. Romae Antonino A. IIII et Balbino conss.’ 34 The legal passages cited in nn.  and – would appear to agree with the view of Mireille Corbier (relying mainly on epigraphical evidence) that marriage between collateral relatives, especially cousins (whether first cousins or beyond) was not uncommon in the late nd and early rd cents. See Corbier : –, esp. –. I disagree with the thesis of Shaw and Saller : –.
Parent–Child Conflict in the Roman Family

implies that if her cousin had refused to marry her, she would have been heir, and other rescripts confirm that if the testator’s choice of husband rejected the marriage, the legatee could still inherit.35 In these cases the testator was not the woman’s paterfamilias. Mothers, uncles, and maternal grandfathers could all wield the power of the purse over their female relations, but could not actually force them into marriage. For girls marrying for the first time (when they could be as young as ),36 however, this freedom of choice could also create problems. Two rescripts refer to conflict over marriage plans between the relatives and guardians of a young person no longer under patria potestas. In  Septimius Severus and Caracalla told a man named Potitus: When a girl’s marriage is being sought and there is no agreement between her guardian (tutor) and her mother and relatives concerning the choice of a future husband, the judgment of the governor of the province is necessary. (Cod. Just. . . )37
Conflict has arisen between a girl’s guardian and her mother and other relatives regarding choice of husband. The guardian is a tutor impuberum, appointed for girls under  (and boys under ) whose paterfamilias was dead.38 At this period a tutor did not have the right to contract or break off a betrothal agreement on behalf of his ward, but he might well wish to have a say in the 35 Cod. Just. . .  (Septimius Severus and Caracalla to Alexander, posted  Oct. ); Cod. Just. . .  (Caracalla to Saturnina, posted  Dec. ). 36  was the age at which women could legally be married (D. . . , Pomponius), but betrothal pacts could be made when the prospective spouses were as young as  (D. . . , Modestinus). The rescripts do not provide much evidence for age at marriage, although if a tutor (impuberum) is mentioned, the situation involves a person under ; if a curator (minorum) is involved, the minor is over  but under . See below, at nn.  and . 37 Cod. Just. . . , given  May : ‘Impp. Severus et Antoninus AA. Potito: Cum de nuptiis puellae quaeritur nec inter tutorem et matrem et propinquos de eligendo futuro marito convenit, arbitrium praesidis provinciae necessarium est. D. non. Mai. Anullino et Frontone conss.’ 38 Not a tutor mulierum, who also would not have had the power to contract a betrothal in the name of the woman whose guardian he was. All references to tutela mulierum were edited out of both the Digest and the Codex Justinianus by Justinian’s compilers, as it had long ceased to be a living legal construct. See Arjava : – and Evans Grubbs : ch. , part .E.

Judith Evans Grubbs
arrangements.39 The girl in question is thus below the legal age of marriage, but plans are already being made for the future. The role of the petitioner Potitus in this family drama is unclear: he may be the guardian, a male relative, a thwarted suitor, or even a provincial official who had not known how to resolve the dispute and had sent the case on to the emperors. The emperors reply that in order to resolve such a family dispute, the provincial governor himself will have to decide! They do not even suggest that the girl herself be given the choice of whom she will marry. She was probably still a young child, perhaps not even aware of the conflict that had arisen in her name. This was an individual case, but the same situation, of a fatherless minor woman whose choice of husband is in dispute, is found in general laws of the fourth and early fifth centuries, which call for the same solution.40 In these cases, the nubile girls were probably from the wealthy elite, where social rank, property, and family honour were at stake to a far greater degree than in most marriage arrangements. The second rescript must involve a somewhat older minor, since it refers to a curator minorum, a guardian appointed for fatherless children above puberty (considered  for girls,  for boys) but under .41 In  Gordian III told a certain Romanus: 39 This appears to change in the th cent., when guardians are liable for their ward’s breaking a betrothal; see n.  below. In earlier times, also, when marriage with manus was the norm, a tutor would need to give his authority; see Corbett : –; cf. ibid. –. 40 Cod. Theod. . . , a law of Valentinian I (addressed to the Senate) dated , says that in controversies surrounding choice of spouse for a widow under  (whose father is dead; if he is still alive she must abide by his choice, even if she has been emancipated) a iudicaria cognitio is to be held, ‘as has been ordained in the marriage of pupillae’. Cod. Just. . . , a law of Honorius dated –, says that in a dispute over choice of spouse for a woman whose parents are both dead but who has a curator (and so is between  and ; see n. ), if the woman’s sense of shame (verecundia) prevents her from expressing her own opinion, a judge is to decide. Translation and discussion in Evans Grubbs : ch. , part .B. 41 On the curator minorum, see F. Schulz, Classical Roman Law (Oxford, ), –; and M. Lemosse, ‘L’Incapacité juridique comme protection de l’enfant en droit romain’, Recueils de la Société Jean Bodin,  (), –. By the time of Marcus Aurelius, appointment of a curator for fatherless minors past puberty but below  had become standard. On guardianship in general, see Saller : –, who rightly stresses how common it was for young people to have guardians, since so many would have lost their paterfamilias before they turned .
Parent–Child Conflict in the Roman Family

In joining marriage, neither the curator, who is responsible only for administering (his ward’s) property, nor the relatives by blood or marriage have any authority to intervene, but the wish of the person whose marriage is being discussed must be considered. (Cod. Just. . . )42
As it has been preserved in the Code of Justinian, this rescript gives only the legal rule and provides no information about the situation which prompted Romanus to petition the emperor. Indeed, the language of the rescript is completely gender-neutral: ‘the wish of (the person) whose marriage is being discussed’. Romanus himself may have been the young person in question, or, on the other hand, the suitor of a woman whose relatives and guardian disapproved of her choice, or a guardian who wished to determine his ward’s choice of spouse, or an interested relative ‘by blood or marriage’, or even a magistrate faced with settling a family dispute. Unlike the rescript of  to Potitus above, this one does not mention a judgment by the governor. This may be because the particular case it addresses involved a ward over puberty, or perhaps because the young person in question was a male, whose own wishes might be given greater weight. Another rescript of Septimius Severus and Caracalla, preserved in the Digest, suggests that the different solutions given in these two rescripts may be due to the age rather than the gender of the ward in question: ‘Management (of the business affairs) of a female ward is the curator’s duty; but she can marry by her own decision.’43 In the 42 Cod. Just. . . , given (data)  Feb. : ‘Idem (Imp. Gordianus A.) Romano: In copulandis nuptiis nec curatoris, qui solam rei familiaris sustinet administrationem, nec cognatorum vel adfinium ulla auctoritas potest intervenire, sed spectanda est eius voluntas, de cuius coniunctione tractatur. D. V k. Mart. Gordiano A. II et Pompeiano conss.’ Corbett (: –) says this must refer only to the marriage of a male under , not a female (despite the genderneutral language) because otherwise ‘it could not have been left standing in open contradiction with C(od. Just.) . . , . . , . . .’ But that does not mean that Gordian was referring only to males, only that Justinian’s compilers may have wanted the rescript to be understood that way. 43 D. . .  (Paulus): ‘. . . et ita Severus et Antoninus rescripserunt in haec verba: “Ad officium curatoris administratio (rerum) pupillae pertinet: nubere autem pupilla suo arbitrio potest.” ’ This excerpt is taken from Paulus’ book on the oration given to the Senate by Septimius Severus and Caracalla in , which introduced several measures intended to protect the interests of minors; see R. Talbert, The Senate of Imperial Rome (Princeton, ),  (no. ) for the legal sources.

Judith Evans Grubbs
second and third centuries, guardians were not responsible for arranging their wards’ marriages and did not have to give their permission for their ward to marry. This seems to have changed in late antiquity, however; legal and documentary evidence of the fourth and fifth centuries suggests that guardians of young women often did play an active role in determining their choice of husband.44 In his rescript to Romanus, Gordian upholds the freedom to choose his or her spouse of a person, male or female, who has no paterfamilias and is therefore legally independent. Paternal Power and Divorce Several rescripts involve another aspect of paternal power: the father’s right to break up the already existing marriage of a child under his power. Classical Roman law took essentially the same attitude in regard to ending a marriage as it did to beginning one. If either spouse no longer felt ‘marital intent’ and wished to dissolve the marriage, he or she had only to make this intention known, and the marriage would no longer exist; only in the fourth century did Roman emperors begin to restrict the freedom of either spouse to repudiate the other unilaterally. And since the consent of each spouse’s paterfamilias was necessary to make a marriage, it only stood to reason that he also had a say in his child’s divorce.45 Indeed, until the Antonine period, a paterfamil44 See Cod. Theod. . .  (a law of Constantine dated ), which refers to fathers, tutores, or curatores as equally responsible for keeping a betrothal contract; Cod. Theod. . .  (a law of Honorius dated ), which explicitly denies that a guardian can break the betrothal pact a woman’s father made for her before her death; Cod. Just. . .  (Honorius, –; cf. n.  above), which mentions the possibility of conflict between a woman’s curator and her relatives. Documentary evidence: P. Lips.  (0001 M. Chr. , later th cent.), where a woman’s curator took part in her marriage arrangements; cf. P. Vind. Bosw.  (dated ), where a young man’s curator is present at his marriage. Discussion of the guardian’s role in marriage arrangements in late antiquity in Evans Grubbs : –. 45 Whether a paterfamilias could prevent his daughter from divorcing is a matter of some dispute; there is nothing in the (extant) legal sources about a father actually trying to do this. Corbett (: ) cites a number of Digest passages to show that in the classical period daughters could divorce without paternal consent; Treggiari (: –) tends to agree. But Gardner (: ) disagrees, saying Corbett’s passages do not prove his point. This is true, but they and many other Digest passages show daughters initiating divorce with no hint
Parent–Child Conflict in the Roman Family

ias actually had the right to break up the marriage of any child under his legal power. This changed in the mid-second century, as a rescript of Diocletian and Maximian explains to a certain Scyrion: Our father, that most pious emperor the deified Marcus [Aurelius], decided that when a daughter under paternal power was a wife living in harmony with her husband, the objection of a father who consented to the union in the beginning is not valid, unless the father has done this (objected) because some great and just reason arose (after he had originally consented). () On the other hand, no constitution of law has ordered her to return to her husband if she is unwilling. () But a father does not have any choice over the divorce of a daughter he has emancipated from paternal power. (Cod. Just. . . )46
A father who originally agreed to his daughter’s marriage cannot step in to break it up later if she is living in harmony with her husband—unless he has a ‘great and just reason’. But neither of paternal opposition, even without paternal knowledge (D. . . , Papinian); see Arjava : –. The lack of a clear statement from the jurists could be due to changes in divorce law made by Justinian (e.g. Cod. Just. . .  of , which penalizes those who divorce collusively in order to defraud their paterfamilias of the dowry or pre-nuptial gifts he had given). Any juristic texts or rescripts which ran counter to Justinian’s policies would have been left out of his corpus or changed to reflect his law. See A. Arjava, ‘Divorce in Late Roman Law’, Arctos,  (), –, on Justinian’s editing policies and their effect on what we know of divorce in the classical period. I note one rescript regarding a case where (if I understand it correctly) a father left a fideicommissum to his daughter on condition that she guarantee not to divorce her husband; the emperors tell the recipient to proceed as though the condition had never been made: Cod. Just. . .  (Septimius Severus and Caracalla to Gallicanus, ) This may be an epistula to an official rather than a subscriptio to a private individual: see Honoré : . 46 Cod. Just. . . , given  Aug. (Nov.) : ‘Idem (Diocletianus et Maximianus) AA. et CC. Scyrioni: Dissentientis patris, qui initio consensit matrimonio, cum marito concordante uxore filia familias ratam non haberi voluntatem divus Marcus pater noster religiosissimus imperator constituit, nisi magna et iusta causa interveniente hoc pater fecerit. () Invitam autem ad maritum redire nulla iuris praecepit constitutio. () Emancipatae vero filiae pater divortium in arbitrio suo non habet. D. V K. Sept. (Mommsen: Dec.) Nicomediae CC. conss.’ Sent. Pauli . .  attributes the same policy to Antoninus Pius; evidently Roman emperors ruled more than once on the question. Ulpian (D. . . . ) refers to a rescript of Septimius Severus to the same effect, though this may have just been clarifying the Antonine ruling; see Corbett : .

Judith Evans Grubbs
should a daughter be forced to return unwillingly to her husband—even, it is implied, if her father (or her husband) wants her to. And if the daughter is no longer under patria potestas because her father emancipated her, he has no authority to bring about her divorce at all. Since the rescript as we have it retains only the legal rule, we do not know if Scyrion was the father or the husband, or perhaps a judge faced with a difficult family dispute. This decision was intended to discourage blatantly arbitrary paternal decisions without detracting too much from the concept of paternal power.47 Significantly, the rescripts in the Code of Justinian that concern parental attempts to break up children’s marriages all involve the marriage of daughters. And although the issue really revolves around the limits of paternal power, the emperors base their decision on the state of mind of the married daughter: does she want the divorce? Or is her father acting unilaterally to break up a harmonious union? For indeed, a father might have a selfish financial motive for taking back his daughter—the desire to get back the dowry he had given. A passage in a fourth-century legal manual describes a case which had come to the attention of the jurist Paulus in the early third century: A father sent a repudium to his son-in-law against his daughter’s wish; I ask whether he can get back the dowry which was provided from his own resources. Paulus replied that indeed the marriage seems to have been legally dissolved by the very act (of sending the repudium), but that it is not permitted for a father to take his daughter away from her husband against her will, nor is he able to get back the dowry unless his daughter consents.48
The divorce was valid after the repudium was sent, but the father could not forcibly separate his daughter from her husband. And (as other jurists and imperial rescripts also point out) after the
47 Cf. also D. . . .  (Ulpian): the rule that bene concordantia matrimonia should not be disturbed by right of patria potestas is to be followed in such a way ‘that the father should be persuaded not to exercise paternal power harshly’. 48 Frag. Vat. : ‘Pater invita filia repudium genero misit; quaero an profectam ex suis bonis dotem petere possit. Paulus respondit matrimonium quidem re ipsa iure solutum videri, sed patri filiam invitam a marito abducere non licere nec eum dotem repetere posse nisi filia consentiente.’
Parent–Child Conflict in the Roman Family

end of his daughter’s marriage, a father could bring an action for recovery of her dowry only with her consent.49 Paulus’ ruling may or may not have been in response to an actual situation; I suspect that it was. Certainly similar situations are found in the rescripts. Gordian replied to a certain Aper: If, as you assert, the marriage was broken up after a complaint concerning her husband was reported to you by your daughter, and she did not return to him with your consent, the union is not at all legal since the consent (voluntas) of the father, in whose power she is, has ceased. And therefore, as long as your daughter is not unhappy about it (non paenitente), you are not prevented from asking for her dowry back. (Cod. Just. . . )50
According to the rescript, Aper claimed that he had brought about his daughter’s divorce after she had complained to him about her husband, and that his daughter had left her husband and ‘did not return to him with your consent’ (nec te consentiente ad eundem regressa est). The bone of contention is the daughter’s dowry, still held by the husband, which Aper wants returned. 49 D. . .  (Ulpian); see also D. . .  (Africanus). At D. . . . , Ulpian suggests that a daughter might even hide from her father so as not to be forced to consent to an action for return of her dowry. A rescript of Caracalla cited by Ulpian (D. . . . ) said that a daughter seems to consent to her father bringing the action ‘unless she clearly opposes (him)’; cf. the policy on a daughter’s consent to marriage in D. . .  (also Ulpian), mentioned above. Rescripts also mention the rule that a father needed his daughter’s consent to get her dowry returned: Cod. Just. . .  (Septimius Severus and Caracalla to Aquilia, ); . .  (Diocletian and Maximian to Erotius, ) and . .  (same emperors to Epigonus, ). 50 Cod. Just. . . , posted  Oct. : ‘Idem (Gordianus) A. Apro: Si, ut proponis, post querellam de marito a filia ad te delatam dissociatum est matrimonium nec te consentiente ad eundem regressa est, minus legitima coniunctio est cessante patris voluntate, in cuius est potestate: atque ideo non paenitente filia petitionem dotis repetere non prohiberis. PP. IIII k. Nov. Sabino et Venusto conss.’ There is a textual problem in the last clause, which renders this rescript even more ambiguous. One manuscript (C) has non petente filia petitionem dotis (‘if your daughter does not ask for the return of her dowry’), and the th-cent. legal scholar Jacobus Cuiacius (Jacques Cujas) suggested non renitente filia petitioni dotis (‘if your daughter does not object to asking for the return of her dowry’). Note that I am assuming that the ablative absolute has a conditional force; there could be other translations. I confess that I am not sure of the meaning of this rescript.

Judith Evans Grubbs
Aper’s former son-in-law may be refusing to return the dowry because (he claims) the divorce was due to his wife’s fault, or perhaps he is denying the validity of the divorce altogether and claiming that her father acted against her will. The last clause of the rescript hints that Aper’s daughter may not agree with her father’s request for the dowry, in which case (as we saw above), he could not get it back. Indeed, the rescript could even be interpreted as suggesting that his daughter did return to her husband, but without Aper’s consent. Paternal consent was required even for the remarriage of a couple who had been married and then divorced, so, if Aper’s daughter went back to her former husband without her father’s approval, the remarriage would not be iustum matrimonium.51 Perhaps Aper had acted according to his daughter’s wishes when he brought about her divorce. But some daughters did not want to be divorced; they were living in a harmonious marriage. 51 D. . .  ( Julian): ‘Marriage renewed between the same persons is not considered valid unless their parents are willing’ (‘Nuptiae inter easdem personas nisi volentibus parentibus renovatae iustae non habentur’; note that the Watson edition of the Digest mistranslates this passage). There might be some question about whether a couple had divorced and then remarried or simply been temporarily separated and then reconciled; see D. . .  (Marcellus, in his third book on the lex Julia et Papia): ‘Many think, when the same woman returns to the same man, that it is the same marriage. I agree with them, if they have been reconciled after not much time has intervened and if, within the interval, she did not marry another nor did he take another in marriage, and especially if the husband has not returned the dowry.’ (Note, however, that Marcellus says nothing about paternal consent.) Cf. D. . .  0001 D. . .  (Paulus). Several passages describe cases where a woman divorced her husband and later remarried him (sometimes having been married to someone else in the interval), and the status of the dowry from the first marriage is in question; generally, it is assumed that the dowry too is renewed along with the marriage. See D. . .  (Ulpian, quoting a rescript of Septimius Severus to Pontius Lucrianus); D. . .  (Javolenus); D. . . .  (Scaevola); D. . .  (Modestinus); D. . .  (Ulpian); and D. . . .  (Javolenus). Could these be cases where a woman was pressured by her father to divorce, and then went back to her husband after her father’s death? On the other hand, doubt might be cast on the genuineness of a divorce if the couple remarried later, especially if the (ex)-husband had given his divorced wife a gift to induce her to return; since gifts between spouses were not valid, it might be suspected that the ‘divorce’ was a ruse to justify the gift. The marriage(s) of Augustus’ friend Maecenas and his wife Terentia involved just such a situation: see D. . .  (Javolenus) with Treggiari : –.
Parent–Child Conflict in the Roman Family

What we see is a conflict between the ideal of marital concordia, and the prerogatives of paternal power. The Antonine emperors, who gave public expresssion to conjugal concordia on their coinage,52 decided in favour of marriage. But it is clear from third-century rescripts that fathers continued to try to break up their daughter’s marriage when it suited them. Thus Diocletian and Maximian respond to a certain Faustina: Since you say that you did not violate the observance of pietas, but that you were unwilling to break up the marital union which you had been allotted, and for that reason your father, offended and angry, stooped to (punishing you with) the disgrace of disinheritance (ad exhereditationis notam prolapsum), you will not be prohibited from bringing a complaint of ‘undutiful will’. (Cod. Just. . . ) 53
Faustina’s father had disinherited her because she had refused to divorce her husband. Since it is assumed that her father had at an earlier time consented to the marriage, his demand that she now divorce is unreasonable. Therefore Faustina has grounds to bring a suit of inofficiosum testamentum to reclaim her share of the inheritance.54 Here the preservation of a concordant marriage is put clearly above the claims of filial pietas. Another rescript of Diocletian and Maximian is addressed to an abandoned husband, Alexander, rather than to a wife or her father: 52 Symbolized by the dextrarum iunctio, the handclasp of the married couple, found also on imperial sarcophagi. Originally the dextrarum iunctio was used to depict public concordia—agreement and loyalty among different political or social groups. Antonine emperors first made use of the image to portray on coins the marital concord between the imperial couple: see Reinsberg . Kampen  shows that the Antonine interest in depicting private familial ideology on public monuments continues into the Severan period also. See also at n.  below. Of course, the Roman ideal of marital concordia goes back much further; see M. B. Flory, ‘Sic Exempla Parantur: Livia’s Shrine to Concordia and the Porticus Liviae’, Historia,  (), –, and S. Dixon, ‘The Sentimental Ideal of the Roman Family’, in Rawson a: –. 53 Cod. Just. . . , posted at Nicomedia  Feb. : ‘Impp. Diocletianus et Maximianus AA. Faustinae: Cum te pietatis religionem non violasse, sed mariti coniugium quod fueras sortita distrahere noluisse ac propterea offensum atque iratum patrem ad exheredationis notam prolapsum esse dicas, inofficiosi testamenti querellam inferre non vetaberis. PP. Nicomediae XVI k. Mart. Maximo II et Aquilino conss.’ 54 See above, n.  on querela inofficiosi testamenti. The rescript to Faustina is found under the Cod. Just. title de inofficioso testamento.

Judith Evans Grubbs
If your wife is being held by her parents against her will, our friend the governor of your province—after you have approached him for help— will relieve your desire, following the woman’s own wish when she has been brought forth. (Cod. Just. . . )55
Alexander claims that his wife is being held by her relatives (parentes, which could be parents or more generally her family) against her will. Now the emperors cannot be sure that the petitioner is really telling the truth; maybe his wife left him and returned to her family willingly and now simply refuses to see him. So they tell Alexander to go back to the governor of his province, who will then use his official power to interview the woman herself and see what she wants. If she does want to go back to her husband, as he claims, then the governor will see to it that she does: ‘he will relieve your desire’ (desiderium), the emperors assure him. Two rescripts of Diocletian and Maximian refer to cases where mothers tried to force their daughters to divorce. A man named Piso is told: ‘A mother does not have the power to bring about her daughter’s divorce’ (Cod. Just. . . ).56 This gives no clue to the circumstances that gave rise to it, nor do we know what Piso’s role was. The other rescript, addressed to a certain Sabinianus, is more informative: A daughter left an orphan by her father, who is living in concord with the husband she married when her mother was willing, does not offer just cause for offence after the same mother’s change of mind, nor is she legally (iure) compelled to be married or unmarried at the momentary whims of her mother. (Cod. Just. . . )57 55 Cod. Just. . . , not dated, but between  and : ‘Idem (Diocletianus et Maximianus) AA. Alexandro: Si invita detinetur uxor tua a parentibus suis, interpellatus rector provinciae amicus noster exhibita muliere voluntatem eius secutus desiderio tuo medebitur.’ 56 Cod. Just. . . , given at Sirmium  Dec.  (): ‘Idem AA. et CC. Pisoni: Filiae divortium in potestate matris non est. D.III k. Ian. Sirmi CC.’ That would be  Dec.  but Mommsen (ad loc.) and Honoré () date it . This is clearly a case where the Justinianic compilers abbreviated the emperors’ reply, leaving only the statement of the legal principle—frustrating for the st-cent. social historian! Cf. Cod. Just. . . , on the same topic, dated six days later (at n. ). 57 Cod. Just. . . , given at Sirmium  Jan. : ‘Idem AA. et CC. Sabiniano: Filia in orbitate patris relicta cum marito, cui matre volente nupsit, colens concordiam iustas offensionis post eiusdem matris paenitentiam causas
Parent–Child Conflict in the Roman Family

No need to invoke the decisions of second-century emperors here, for Roman law had never granted mothers the power to dissolve their children’s marriages. A father’s authority to do so, limited though it was by second- and third-century emperors, derived purely from patria potestas. The daughter in question was now legally independent (sui iuris), and maternal claims to pietas did not include forsaking a harmonious marriage. Interestingly, the rescript considers relevant the mother’s earlier consent to her daughter’s marriage (since this makes her subsequent change of mind appear whimsical and ill-founded); in fact, a mother’s initial agreement to her child’s marriage was no more legally necessary than was her consent to its continuation. As it happens, we know more of the circumstances surrounding the reply to Sabinianus because another response, preserved in a different book of the Code under a different title, is part of the same original rescript.58 It appears that Sabinianus was the daughter’s husband (like Alexander above). His mother-in-law had provided the dowry, over which she had retained the usufruct and which she was now trying to sell. Evidently she wanted her daughter to divorce so that she could get control of the dowry— the same ploy attributed to a paterfamilias in the situation discussed by the jurist Paulus, cited above. The emperors say clearly that she (like Paulus’ paterfamilias) cannot do this. Sometimes, however, mothers could impose their will through their pocketbooks rather than legal authority.59 Maxima’s mother told her that if she wanted to be her mother’s heir, Maxima had to divorce her husband. Ever the dutiful daughter, Maxima obeyed. Later she had a change of heart and wrote to the emperors Valerian on praestat nec ex momentariis voluntatibus matris nupta atque vidua esse iure compellitur. D. non. Ian. Sirmi CC. conss.’ 58 Cod. Just. . . : ‘Idem AA. et CC. Sabiniano: Res, quarum usu fructu sibi deducto socrus venumdando auferre tibi nihil potest. D. nonis Iul. Sirmi CC. conss.’ Cod. Just. . .  and . .  both have the same addressee and place of enactment. Cod. Just. . .  is dated the nones of January ( Jan.), whereas the MS date of Cod. Just. . .  was the nones of July ( July). But confusion would be easy, and Mommsen (ad loc.) and Honoré () change Cod. Just. . .  to the nones of January also. 59 On the moral and economic powers of the Roman mother, see Dixon  and Gardner : – (esp.  on this rescript).

Judith Evans Grubbs
and Gallienus. They were highly displeased by Maxima’s mercenary behaviour: ‘You are more to be rebuked than your mother,’ they say: You are more to be rebuked than your mother. For if she wanted you to be her heir, she would not order something which is invalid, (that is) for you to break up your marriage with your husband. () You, moreover, approved her wish by divorcing. However, even if a condition of this type were allowed, you ought to have put marital concord before material gain. But indeed, since good morals (boni mores) forbid this (condition) to be observed, you would have been able to maintain your union without any loss. () Therefore, return to your husband, knowing that you will keep your mother’s inheritance even if you have returned, since indeed you would keep it, even if you had not gone ahead and left him. (Cod. Just. . . )60
Maxima had put her financial interests before concordia maritalis, reprehensible behaviour in the eyes of third-century emperors. ‘Therefore, return to your husband’—an excellent example of the crisp tone of third-century rescripts, combining legal ruling and moral declaration. These rescripts addressing a parent’s right to bring about a daughter’s divorce suggest that the issue came up fairly often and engendered a good deal of family conflict. Interesting evidence of the same conflict is found in Egyptian papyri of the Roman period, particularly the lengthy petition dated  which was sent to the prefect of Egypt from a certain Dionysia of Oxyrhynchus.61 60 Cod. Just. . . , posted  Nov. : ‘Impp. Valerianus et Gallienus AA. Maximae: Reprehendenda tu magis es quam mater tua. Illa enim si heredem te sibi esse vellet, id quod est inutile, matrimonium te dirimere cum viro non iuberet. () Tu porro voluntatem eius divortio comprobasti: oportuerat autem, etsi condicio huiusmodi admitteretur, praeferre lucro concordiam maritalem. Enimvero cum boni mores haec observari vetent, sine ullo damno coniunctionem retinere potuisti. () Redi igitur ad maritum, sciens hereditatem matris, etiamsi redieris, retenturam, quippe quam retineres, licet prorsus ab eo non recessises. PP. XII k. Dec. Valeriano IIII et Gallieno III AA. conss.’ 61 P. Oxy. . , first published in The Oxyrhynchus Papyri Part II, ed. B. P. Grenfell and A. S. Hunt (London: Egypt Exploration Fund, ), –. I give only a bald summary of this extremely complicated case, which involved two prefects (not including the prefect cited in a precedent) a strategos, and a deputy-strategos. A translation of the sections of P. Oxy. .  (including the precedents) that involve the issue of a father’s ability to break up his daughter’s marriage is found in J. Rowlandson (ed.), Women and Society in Greek and Roman Egypt (Cambridge, ), –. Discussed by Arjava : –.
Parent–Child Conflict in the Roman Family

Dionysia had been involved in a protracted dispute with her father Chaeremon over her claim to property from her mother which had been held and administered by Chaeremon. The case had been decided in Dionysia’s favour, which prompted Chaeremon to send another petition to the prefect asserting his right, under ‘the law (nomos) of the Egyptians’, to remove Dionysia from her husband against her will.62 In a counter-petition, Dionysia protested against Chaeremon’s assertion of his right to break up her marriage, rightly seeing this as another way for him to gain control of the property which had been the subject of their first dispute. She cited a number of earlier cases from Roman Egypt as precedents to back up her claim that her father’s attempt to bring about her divorce should be rejected.63 Among her many supporting documents are three official reports of legal proceedings before magistrates in Egypt (including one before an earlier prefect) and one jurist’s opinion, all on the question of whether a father could remove his daughter from her husband under Egyptian law. It appears that under local Egyptian (not Roman) law, a father could in some circumstances make his daughter divorce her husband.64 But in the three court cases cited by Dionysia, the judges, who were themselves Romans but were judging provincials who did not have Roman citizenship, all decided against the father. They did so not on the basis of legal principles, but because they 62 On the meaning of ‘the law of the Egyptians’ here, see J. Modrzejewski, ‘La Regle de droit dans l’Égypte romaine’, in Proceedings of the Twelfth International Congress of Papyrology (Toronto: A. M. Hakkert, ), –, at –. Such law could be Greek or Egyptian, but was recognized by Roman officials as being on the same level as custom, not law. 63 See R. Katzoff, ‘Precedents in the Courts of Roman Egypt’, ZSSR RA  (), –, especially – on P. Oxy. . . 64 Under the ‘law of the Egyptians’ a father could make a daughter who had been born to him in an ‘unwritten marriage’ divorce her husband against her will (whereas he could not if the daughter had been born in a ‘written marriage’). See H. J. Wolff, Written and Unwritten Marriages in Hellenistic and Postclassical Roman Law (Haverford, ), especially – on P. Oxy. . . Cf. N. Lewis, ‘On Paternal Authority in Roman Egypt’, RIDA rd ser.  (), –, who thinks that by the nd cent., an Egyptian father could no longer force a married daughter to divorce if she were unwilling (he could bring about her divorce if she wanted him to, however). In any case, the Roman officials who are judging these cases involving non-Roman citizens in Egypt base their decisions not on local law, but on what they perceive to be humane and just from a Roman point of view.

Judith Evans Grubbs
did not think it right for a father to break up his daughter’s marriage against her will. In a case dated , Flavius Titianus, the prefect of Egypt, decided: ‘It makes a difference with whom the married woman wishes to be.’65 The second case cited, before the epistrategos Paconius Felix five years later, used the decision of Flavius Titianus as a precedent. The epistrategos ordered that the daughter be asked (through an interpreter; evidently she was a native Egyptian and did not speak Greek) what she wanted; she said ‘to remain with my husband’, and so the epistrategos ruled.66 Here we see the same rationale being used by Roman provincial officials in cases involving non-Roman citizens as was later used by third-century emperors in rescripts to Roman citizens, even when that rationale runs contrary to the native law of the province where they hold office. It seems that by the second century, the prevailing sentiment among the Roman elite who were governing the empire was that even the claims of a paterfamilias should not be allowed to destroy a concordant marriage. In other words, maintaining marital bonds was more important than maintaining paternal control. The husband’s lawyers in the case before the epistrategos of Egypt in , using the decision of the prefect Flavius Titianus as a precedent, say that his decision ‘did not follow the inhumanity (ananthropia) of the [local] law but the intention of the child’.67 The cases cited by Dionysia actually occurred shortly before the reign of Antoninus Pius, and so apparently anticipated rather than followed the imperial rulings mentioned above. We do not know the outcome of her own case, which took place under Commodus, but if the prefect of Egypt in  followed his predecessor, he would have ruled in her favour. However, as the third-century rescripts indicate, these Antonine rulings on both the imperial and provincial level did not bring an end to family disputes over a father’s right to break up his daughter’s marriage. And later papyri show that the problem continued in the fourth century.68 P. Oxy. . , col. , line . P. Oxy. . , col. , lines –. 67 P. Oxy. . , col. , lines –. 68 See P. Sakaon  (0001 P. Flor. ), dated , where the father, Sakaon, attempted to take his daughter away from her husband Zoilus on the grounds that he had never received the hedna, the marriage gifts given by a husband to a wife or her parents. This was one stage of a long-standing family feud; 65 66
Parent–Child Conflict in the Roman Family

Conclusion Until quite recently, work on ‘the Roman family’ has centred on Italy of the late Republic and early Principate, and has drawn largely on literary and epigraphic sources from that period and from the juristic writings of the Digest, most of which antedate the death of Alexander Severus in . This focus is reflected in the papers given at the first three ‘Roman family’ conferences held in Canberra in the s and early s. In recent years, study of the ‘Roman’ family has shifted to the later, post-Severan empire and to the imperial provinces, and many of the papers given at the  conference in Canada reflect this change in focus. Very few of the rescripts preserved in the Code of Justinian are earlier than the reign of Septimius Severus, and most are considerably later; moreover, almost all of those dating to the tetrarchic period derive from the eastern half of the empire, under Diocletian, rather than the Latin-speaking west under his co-ruler Maximian.69 Are we in fact talking about the ‘Roman’ family any more? And even if the third-century petitioners to whom the rescripts respond are Romans, what relevance do their thirty-one years later, Sakaon was involved in another seizure of a married woman from the home of her husband (and of the husband’s father Zoilus, the deprived husband in the incident of !), this time at the instigation of the wife’s mother (P. Sakaon  0001 SB . ), dated . Texts and translation in G. M. Parassoglou (ed.), The Archive of Aurelius Sakaon: Papers of an Egyptian Farmer in the Last Century of Theadelphia (Bonn, ). See J. W. B. Barns, ‘A Fourth-Century Deacon’s Petition from Theadelphia’, Studia Patristica,  (Berlin, ), – on the feud. Cf. also P. Cairo Preis.  (dated , referring to an incident three years earlier), where a husband claims that while he was out of town, his mother-in-law, telling him that her daughter had a demon, removed her from his home and married her to someone else—a situation recalling Cod. Just. . .  and . .  above. Text in F. Preisigke, ed., Griechische Urkunden des Aegyptischen Museums zu Kairo (Strasbourg, ), –. 69 Out of a total of , pre-Constantinian rescripts in the Code, only twenty-three date before the reign of Septimius Severus. Of these, one is Hadrianic; ten are from Antoninus Pius; six are from the ‘deified brothers’ Marcus Aurelius and Lucius Verus, and four from the reign of Marcus Aurelius alone. Two are from the short rule of Pertinax (none are from Commodus). Some rescripts of Maximian are preserved outside of the Cod. Just., particularly in the Fragmenta Vaticana, such as Frag. Vat.  to Calpurnia Aristaeneta, cited at the beginning of this chapter. See Corcoran : – and .

Judith Evans Grubbs
family conflicts have to the family relationships of the people of Rome and Italy two or three centuries earlier? I, of course, would argue strongly that the people of the rescripts are Romans, both legally and socially, even if their lives, mores, and family structures differ significantly from those of the imperial family and Roman senatorial aristocracy, and their freedmen and women, who have been the focus of earlier ‘Roman family’ studies. After , all of our petitioners would have been Roman citizens (although the Code of Justinian does include rescripts to slaves, they are not among the sample I have discussed here),70 and even if those rescripts dated before  were addressed to non-citizens, their recipients must still have been sufficiently ‘Romanized’ and familiar with the imperial legal and administrative system to be able to get a petition to the emperor and make use of the reply. For the social historian, the lives and conflicts of otherwise unknown and undistinguished Roman subjects are as worthy of study as those of the elite inhabitants of late Republican or Augustan Rome. Moreover, many of the rescripts are contemporary in date with those jurists who are best represented in the Digest—Ulpian, Papinian, Modestinus, and Paulus—and indeed were in all probability often composed by those very same jurists.71 It is standard practice for those studying the Roman family in the early Principate to draw on the writings of these third-century jurists as evidence, and the rescripts themselves have been used to good effect to illustrate the legal situations posited in the Digest.72 How does the evidence of the rescripts, which shed light on family life in a later period and a more provincial milieu, compare to what we know of the ‘Roman family’ of the late Republic and early Principate? Can we detect differences in attitudes toward the marital bond and parent–child relations, or a change over time in ‘family feeling’? Or did individual emotions and desires, and the family conflicts which those emotions and desires produced, remain essentially the same? This question has been debated 70 On rescripts to slaves, see Corcoran : –; J. Evans Grubbs, ‘The Slave who Avenged her Master’s Death’, Ancient History Bulletin, / (), –. 71 As argued by Honoré (); see n.  above. 72 I am thinking particularly of the work of Jane Gardner, especially Gardner , but the rd-cent. jurists of the Digest provide essential evidence for the studies of Treggiari, Saller, and Dixon, among others.
Parent–Child Conflict in the Roman Family

recently by more than one scholar of the ancient family, and the answer is essentially unknowable, given the paucity of evidence for family life in all time periods and social strata (even among the elite about whom we know the most) and the subsequent impossibility of any sort of quantitative or comparative study.73 One of the most valuable aspects of the rescripts (inadequate and unsatisfactory though they are for answering the kinds of questions we would like to ask) is that they show us a far more diverse range of men and women, both geographically and socially, than do the authors of classical literature and even the legal writers of the Digest. The recipients of the rescripts were not for the most part members of the senatorial or equestrian ranks, or even of the provincial local elites.74 Clearly they had the means to get a petition to the emperors and enough property to occasion family disputes over its distribution, so we are not dealing with the lowest levels of humiliores whose lives are almost entirely unrecoverable. But the people of the rescripts do by and large derive from a social level not found in classical literature of the first centuries  and ; they are, perhaps, more like some of the provincials who appear in works of the late second and early third centuries, such as the Metamorphoses of the North African Apuleius.75 Now our evidence for the lives of individuals 73 On the problem of positing change in ‘family feelings’ in antiquity, and the relationship between legal change and change in family relations, see S. Dixon, ‘Continuity and Change in Roman Social History: Retrieving “Family Feeling(s)” from Roman Law and Literature’, in M. Golden and P. Toohey (eds.), Inventing Ancient Culture: Historicism, Periodization, and the Ancient World (London and New York: Routledge, ), –. See also M. Golden, ‘Continuity, Change, and the Study of Ancient Childhood’, EMC/CV ,   (), – on the pitfalls of looking for changes in parent–child relationships in antiquity and later. 74 Some were, however; for instance, senatorial women concerned with preserving their rank: Cod. Just. . .  (Alexander Severus to Severiana, no date); Cod. Just. . .  (Diocletian and Maximian to Paulina, no date). In some cases, rescripts that have been assumed to be addressed to ordinary subjects may have been to imperial officials: see M. Peachin, ‘Consultation with a Magistrate in Justinian’s Code’, Classical Quarterly  (), –; cf. Honoré : –. (This obviously would not be the case for the hundreds of rescripts addressed to women.) 75 Long recognized as a source for provincial society: see F. Millar, ‘The World of the Golden Ass’, JRS  (), –. Cf. also the early rd-cent. Passion of Perpetua, in which a North African (but Roman citizen) Christian convert provides a unique narrative of her conflicts with her father.

Judith Evans Grubbs
below the elite in the earlier empire is very limited, drawn from funerary inscriptions which are not likely to mention parent–child disputes over marriage and divorce. So if we try to locate changes or continuities in family sentiment between, say, the first and the third centuries, by juxtaposing the evidence of funerary epitaphs with the evidence of imperial rescripts, we are really comparing apples and oranges. Both sets of evidence are valuable and both have something to tell us, but they are not comparable. It is worth noting, however, that what evidence we do have for family conflict in the earlier period, such as the letters of Cicero and the lives of Augustus and the first-century emperors, reveal situations similar to those found in the rescripts I have been discussing. The circumstances surrounding the third marriage of Cicero’s daughter Tullia are well known and have often been discussed; they provide a clear illustration of the dynamics of paternal, maternal, and filial choice and consent, and show the limits of paternal authority.76 And as for a father’s attempt to break up his child’s marriage and the emotional upheaval it could cause, we need only recall Suetonius’ description of Tiberius’ behaviour after Augustus forced him to divorce Vipsania (who had already borne him one child and was pregnant with another) and marry Julia.77 Of course, that situation was exceptional— dynastic succession to the Empire was at stake, and in any case, Augustus did not at that time have paternal power over Tiberius, as he had not yet adopted him. And the cases known from the rescripts and the papyri all involve parental attempts to break up a daughter’s marriage, not a son’s. But the situations found in the rescripts will occur in any society where marriage arranged by the father is the norm, and where fathers wield extensive authority over their children, even their adult children, whether or not such power takes on the dimensions of Roman patria potestas.
76
time.
See n. . It is not certain that Tullia was still under patria potestas at the
77 Suetonius, Tiberius : ‘he grieved that he had driven [Vipsania] off even after the divorce, and when he saw her face to face just once, he followed her with eager and moist eyes, so much so that precautions were taken that she never again come into his sight’.
Parent–Child Conflict in the Roman Family

That said, the evidence of the rescripts may still reflect a change. More than one legal source refers to a decision by a second-century emperor that a father cannot break up the harmonious marriage of his child, and it may be that Antonine emperors departed from their predecessors in tilting the balance toward the wishes of the child. Certainly, both marital concordia and family pietas are hallmarks of second-century imperial iconography, on coins and monuments.78 Imperial interest in parent–child relations appears in other Antonine rescripts concerning the responsibility of parents to furnish physical support to their children and the duty of adult children to support their aged or disabled parents. Out of only twenty Antonine rescripts preserved in the Code of Justinian, three are found under the title de alendis liberis ac parentibus (‘on the support of children and parents’), and an excerpt from Ulpian’s second book ‘On the Duties of the Consul’ preserved in the Digest also cites a number of early rescripts on the topic of alimenta, at least four of which are Antonine.79 The legal and iconographic evidence both point to a real concern with children and with parent–child relations on the part of second-century emperors. But again, it is difficult to compare the evidence for the legislative agenda of second- and 78 See Reinsberg , Kampen , and Rawson . As with marital concordia (cf. n. ), idealization of family relationships in imperial iconography goes back to the Augustan era; see Dixon : – and Rawson a. There seems to be an intensification in the Antonine period, however. 79 Under the title de alendis liberis ac parentibus: Cod. Just. . .  (Antoninus Pius to Bassus, no date); Cod. Just. . .  (Marcus Aurelius and Lucius Verus to Celer, dated ); Cod. Just. . .  (Marcus Aurelius and Lucius Verus to Tatiana, dated ). The only other rescript under that title, Cod. Just. . .  is from Septimius Severus and Caracalla to Sabinus, dated . (See n.  for total number of pre-Severan rescripts.) Rescripts known from Ulpian’s book, de officio consulis: D. . . .  (citing a rescript of Antoninus Pius); D. . . .  (quoting verbatim another rescript of Antoninus Pius); D. . . .  (citing a rescript of Marcus Aurelius); D. . . .  (quoting another rescript of Marcus Aurelius, to Antonia Montana). In the same work, Ulpian mentioned several other rescripts without attributing them to a particular emperor: D. . . . ; D. . . . ; D. . . .  and . Most, if not all, of these are probably also Antonine; they cannot be later than , when Ulpian, then praetorian prefect, was killed by the praetorian guards, see Honoré : . Compare these imperial rescripts with Trajanic and Antonine monuments and coins that emphasize alimenta for needy children and familial pietas; on which see Rawson : –.

Judith Evans Grubbs
third-century emperors with that for earlier rulers. Even Augustus’ famous and extremely important legislation on marriage and adultery is known only through later writers, historical and legal. But the rescripts are well worth looking at for what they are— miniature case histories of families in conflict, tiny spotlights on interpersonal relations for a time and a place of which we know all too little. For that reason alone, we should listen to what they have to tell us, however brief and ambiguous.
 Searching for the Romano-Egyptian Family Richard Alston
I spite of the analytical energy devoted to family, often centred on its perceived ‘decline’, agreed definitions of family remain elusive.1 The issue is of considerable importance since much scholarship on the family focuses on the issue of change. The predominant historical model has been to relate industrialization to a move from the ‘traditional’ complex families to more simple nuclear family units.2 This shift was understood in economicfunctionalist terms as representing a substitution of traditional agricultural regimes with industrial or proto-industrial wage labour, though Saller and Shaw crucially modified the traditional model by suggesting that the development is to be associated with urbanism rather than with industrialism.3 Such an association has encouraged some to define family in economic terms as the group into which the economic resources of individuals are pooled, and the building block of the consuming and often the productive economy.4 Nevertheless, economic definitions of family are unsatisfactory since they stress only one of the many functions that 1 Roman moralists saw decline in the family as undermining the social order. See also Patlagean :  on Libanius’ linking of social and familial decline. For the problems of defining the Roman family see Saller ; : –; Hammel (: ) laments that a ‘major reason why talk goes on forever is often that those engaged in conversation do not know what they are talking about . . . The household is simply not a very good unit of observation . . . What we are really studying most of the time in investigating households . . . is the classification habits of census takers but without any Linnean system of our own against which to measure their decisions.’ 2 This idea can be traced back at least as far as Le Play . See discussions in Shaw  and Berkner ; cf. Yangisako . 3 Saller and Shaw ; Shaw ; Saller ; : –. 4 Netting, Wilk, Arnold . Carter () argues that economic functions map uneasily onto households.

Richard Alston
family performs. The family can be seen as the primary area for the socialization of the young and the subsequent reproduction and enforcement of social values that bind together a social group. The family is also a political unit. It is intermediate between the individual and broader social and political structures such as the polity, the village, the neighbourhood. Here, I avoid a restrictive definition: family is a social unit intermediate between the individual and wider social units such as neighbourhoods and polities. The family in Egypt appears most clearly attested in  census returns preserved from three centuries of Roman rule.5 The census declarations are submitted kat’ oikian, by house, and normally submitted by a single member of the household who registers all members of the unit and their relationship to each other.6 Bagnall and Frier demonstrate differences in composition between the oikiai of the villages and those of the cities (see Table .). Although all types of family structure could be found in both urban and rural settlements, oikiai in cities tended to be based on a single conjugal unit, while in the villages more complex oikiai, often based on multiple conjugal units, tended to be the norm. Nevertheless, the oikiai in the city were slightly larger than those in the villages, mainly because of the larger number of slaves in urban oikiai. The images of family derived from the census returns are, however, only partial depictions of the varied social systems that encompass family.7 To understand how the family operated, we need to animate those structures and to unpick the ideologies that lie behind their workings. In so doing, I emphasize the social practice that was generated by and generated social structures.8 In keeping with the permissive definition of family, Bagnall and Frier . The oikia was either the household or the inhabitants of a particular house: the Greek would allow either interpretation. See Bagnall and Frier : –; –. The issue is complicated by the practice of house-sharing in Roman Egypt. See Hobson ; Alston ; and : –. For the taxing system see Hanson . 7 Reay () argues (p. ) that ‘it is a giant step from structure to kin’, suggesting that the residential family easily reconstructable from various records does not account for all kin. Caftanzoglou () shows that there is a close correlation between structural type and the age of the head of the household. 8 Wilk and Netting () emphasize action rather than function as the key aspect of analysis and look to ascertain ‘the relative density of activity groups’ based on frequency of operation. Such analysis would undoubtedly be useful, but fails to take into account the significance of the action in establishing 5 6
Searching for the Romano-Egyptian Family

T .. Families and households in cities and villages (after Bagnall and Frier : tables . and .) Typology
Solitary No family Conjugal family Extended family Multiple family All types
%age of households
Average size of families
Average size of households
Urban
Rural
Urban
Rural
Urban
Rural
. . . . . 
. . . . . 
  . . . .
 . . . . .
. . . . . .
.  . . . .
though not as a result of it, the image of the family in Egypt that emerges is complex and varied. I start at the centre, the conjugal unit and offspring, and proceed by examining the way in which relations with other elements of the family are constituted, from the nuclear to the extended and beyond, to the point at which family merges into community. Descent and Conjugal Relations Egyptians placed considerable emphasis on the nuclear family and the direct line of descent. Both matrilineal and patrilineal lines were valued. This is seen at its most obvious in issues of inheritance where although testamentary bequest allowed a certain freedom of action to the testator, the normal assumption appears to have been of a more or less equal division of goods between children or the nearest relatives.9 As such, the will appears not to have been quite so controversial a document in Roman Egypt as it was in Italy.10 Such patterns could be complicated if, for any reason, the lines of succession became community. Hammel () shows how varied in operation the Serbian ‘zadruga’ is, making issues of classification problematic. Instead of classifying households, one should attempt to understand how they behave in individual cases. For a combination of traditional cultural and economic factors shaping households see Douglass : –. 9 Champlin : –. 10 Frier (: –) argues that wills and the division of estates was one of the activities most likely to lead to litigation in Italy. Champlin (: –) suggests that the will was an opportunity to speak without restraint. See also Pliny, Ep. . .

Richard Alston
confused by the formation of two conjugal relationships or through the provision of use of property to a wife or to another party, but this does not invalidate the general principle.11 Wills often made specific provisions for the divisions of estates, but in numerous cases, the property was divided simply.12 In these circumstances, the rights of children were far more significant than those of any extended family. Status also followed descent in Roman Egypt. This can be seen most spectacularly in the Roman creation of a gymnasial metropolite (urban) status group.13 By the second century, applicants for the status were expected to trace their descent on both maternal and paternal lines back to a mass registration in the mid-first century (the date varies slightly depending on which city the applicant came from), or an Augustan registration. This mirrored the provisions for Alexandrian citizenship, the examination for which required proof of Alexandrian descent on both sides of the family.14 The same rules applied to the lesser metropolite status, though the status examinations were less rigorous. Descent was also crucial in establishing priestly status and even in determin11 See, for example, P. Oxy. . , P. Meyer , P. Mil. Vogl. . , and P. Fam. Tebt. . On wills in general see Montevecchi ; Arangio-Ruiz ; Kreller  and for editions and discussions of individual wills see n. . 12 See, for example, SB . , P. Oxy. . , P. Mil. Vogl. . , P. Oxy. . , P. Stras. . , P. Köln . , SB . , ChLA . . For more detailed divisions, see for example P. Mich. . –, P. Tebt. . , P. Oxy. . , , P. Amh. . , P. Tebt. . , CPR . , P. Ryl. . , P. Mich. .  at least some of which appear to strive for equality between the children. BGU .  is an unequal division of property between the son and grandson of the deceased and P .Tebt. .  is a similarly unequal division between brothers, cf. PSI .  0001 Sel. Pap. . . For examples of other divisions of property see P. Cair Isid. , P. Mert. . , P. Oxy. . , P. Wisc. . . An equal division is sometimes confused by the inclusion of provision for the spouse in the will: P. Oxy. . , FIRA 2 iii. , P. Mich. .  0001 ChLA . , P. Oxy. . . For the laws of inheritance see Taubenschlag : –; Amelotti ; and Migliardi Zingale , . The will was also used to adjust the flow of goods in the light of unusual family circumstances: BGU .  is a veteran’s will in favour of his freedwomen (see Keenan ), P. Princ. .  records a daughter disinheriting her sons to benefit her mother. The holding of property in common could remove the need for elaborate testamentary provisions but create problems later: Youtie a, b. 13 Nelson : –. 14 Delia . For more liberal qualifications for citizenship of Antinoopolis, see Bell , ; Zahrnt .
Searching for the Romano-Egyptian Family

ing eligibility for particular priestly offices.15 Patterns of nomenclature which show both local and family traits demonstrate a concern with descent and ancestry.16 The economic character of Egyptian society, with a high population practising intensive farming and land being at a premium, would probably also work to emphasize descent since the family was presumably the major source of wealth.17 In such circumstances, social power could concentrate on the senior generation of landowners, who had power to dispose of property and thus control younger members of the family.18 Nevertheless, although there are numerous examples of fathers and sons in partnership, and powerful patriarchal figures may lurk behind these seemingly co-operative and indeed educational relationships, it is not easy to find figures who might correspond to the type.19 One example, and rather late, might be Sakaon, whose influence is shown in an extensive fourth-century archive. In  Aurelius Melas petitioned the prefect about Sakaon (P. Sakaon ). Melas had betrothed his son Zoilos to his aunt’s daughter Taues. Taues’ mother had died some years earlier and when Taues’ father, Sakaon, married Kamoution, Taues was expelled from her house. Melas took Taues in and brought her up until she reached an age when she could be married to his son. This was accomplished by a public ceremony, but Sakaon abducted Taues and laid claim to wedding gifts. After arbitration, Melas paid the gifts, but Sakaon refused to provide the girl, for whom Melas now petitions. The Nelson : –; Gilliam : esp. pp. –; Otto : i. –. Hobson . 17 There is a crucial difference between the stem family, as outlined by Berkner (), in which one child inherited the family farm and the other siblings either stayed on without marrying or moved, and the Egyptian situation in that there were only limited opportunities for alternative economic activity or acquiring land. 18 See Champlin :  on the power over children that came with the will. Pliny (Ep. . ) became involved in a very complex case in which a son was disinherited by his mother, an example of matriarchal power. See also Pliny, Ep. . , on a ‘good will’. Lane Fox () argues that partible inheritance systems are most likely to discourage emigration. Veyne () argues, esp. pp. –, that the absence of power for the young can be connected to the Roman concern with parricide, but see Saller . 19 One example is the Roman veteran Bellenus Gemellus, see Hohlwein . One can compare the early nd-cent. farmer Sarapion (P. Sarap.) and his son Eutychis. 15 16

Richard Alston
family politics here are intriguing. The marriage itself is endogamous and the fact that Taues was brought up within Melas’ household renders the relationship closer. Although, if we believe Melas, Sakaon had taken no role in the education of his daughter, he still claimed an interest that had to be reflected in marriage gifts. We may also be seeing here Sakaon defending his honour which had not been adequately acknowledged or perhaps had even been slighted by Melas. A generation later Sakaon acted against Zoilos’ son in abducting his wife, though the context of this dispute is less clear.20 Key to control of a family is the influence that can be exerted over the formation of marriage alliances and, perhaps subsequently, if the emphasis on direct descent is correct, on the sexual behaviour of women. Since women could hold property both as part of the dowry and under their own authority, there was at least some economic motive for ensuring that women remained within the family and, therefore, for practising endogamous marriage.21 Retaining women in the family would also mean that they, and their spouses, would be protected against the unknown of other families and minimize the disturbance caused by the new conjugal relationship. Patlagean has argued that what she perceived as an increasing tendency towards endogamous marriage in late antiquity was in part a reaction to the political and social turbulence of the period, with families seeking marriage partners of whom they could be certain.22 In this scheme, preference for P. Sakaon . For similar cases, see P. Oxy. . . I know of no correlation between societies that practise endogamy and societies which bestow significant property holdings on women, though the Athenian epiklerate would seem to encourage close-kin marriage. In Egypt the economic advantage of sibling marriage was negligible, given that the exporting of dowries would have been compensated for with the corresponding importation of wives. Assets bestowed on women, and thus potentially removed from the family, were not controlled any more tightly than property held by men. The Egyptian kurios appears to have been a rather more liberal institution than the Roman tutor. See, for example P. Oxy. . , . , Modrzejewski  and Arjava . At times, women could even act as representatives of men; see P. Oxy. . , . , . ; SB . ; P. Coll. Youtie . ; P. Mert. . . For female property-holding, see Hobson . 22 Patlagean (: ) argues that individuals in the high empire showed indifference to kin in the choice of marriage partners, operating ‘un régime de panmixie’ (compare Shaw and Saller ) but later adopted close-kin marriage because ‘la structure familiale reprend la force et l’importance que la structure 20 21
Searching for the Romano-Egyptian Family

endogamy relates to social insecurity and families turning away from an unattractive and potentially hostile wider social world. Roman Egypt provides examples of extraordinarily close kin marriage in the sibling marriages common in some areas in the first two centuries .23 Lévi-Strauss (: ) described the prohibition against incest as a meeting point, ‘a union’ between biological and social drives. He nuances his argument by characterizing this meeting point as ‘less a union than a transformation or transition. Before it culture is still non-existent; with it, nature’s sovereignty over man is ended. The prohibition of incest is where nature transcends itself. It speaks of the formation of a new and more complex type of structure and is superimposed upon the simpler structures of physical life through integration, just as these themselves are superimposed upon the simpler structures of animal life. It brings about and is in itself the advent of a new order.’ For Lévi-Strauss, incest regulation is at the birth of culture and society which explains that while there are many different marital practices and forms of incest regulation, the regulation of marital and sexual partnerships and the limitation of possible spouses is universal and all societies have some form of incest prohibition (pp. –). Lévi-Strauss asserts that this does not derive from an instinctual repugnance, in contrast to Scheidel’s recent reassertion of this position, since there is a tremendous variety of incest regulation, the taboo would be pointless if incest prohibition was genetic (pp. –), and ‘incest does exist and is no doubt even more frequent by far than a collective conspiracy of silence would lead us to believe’ (p. ).24 The reason for the prevalence of regulation against close-kin marriage is, in Lévi-Strauss’s view, to ‘bring the civile est en train de perdre’, linking the issue to the seeming decline of the city from the th cent.  onwards. Patlagean acknowledges (p. ) that her analysis is based on inadequate data, but argues that legislative activity defining incestuous partnerships was a response to an increased propensity for close-kin unions (pp. –). Nathan (: –), argues that Christian concern with close-kin marriage relates to early accusations of incestuous activity levelled at Christian communities. O’Roark () argues that John Chrysostom thought exogamy was the norm. 23 Hopkins ; Scheidel : –; Scheidel () argues for notable regional and social differences in the prevalence of sibling marriage. 24 Scheidel : –.

Richard Alston
women out of the family and offer them to the group’ (p. ), the group in endogamous marriages being a close-knit band of fellows, and, in exogamous marriage, being a rather more amorphous, loose social unit (even society perhaps).25 In this scheme, women are a commodity which men exchange and this allows Lévi-Strauss to reuse the methodologies of Marcel Mauss in studying gift exchange.26 The comparison explains Lévi-Strauss’s reading of the regulation of incest as being central to the formation of a social system since the gift was fundamental to binding together different groups through reciprocity. Some of Lévi-Strauss’s argument, especially the commodification of women, seems a little strange for modern western societies (to which the argument should similarly apply) and is perhaps too mechanical a response to the delicately negotiated relationship of marriage in many societies.27 Zonabend (: ) argues that ‘the diversity of forms in which marriage occurs shows that its purpose is not the satisfaction of sexual instincts, nor even the socialization of children. Its primary function is to establish a union between two individuals, to ensure that a marriage contract ties them to each other from that time forth. The terms of the contract do not really matter: the important thing is that the contract should actually exist . . .’ ‘This legal agreement does not exist for reasons of sexual morality or because of any other preoccupa25 The argument bears some similarity to John Chrysostom, I Cor. hom.  (PG . ): ‘And He devised another pretext of arrangement. For having forbidden the marriage of natural kin, he led us out among strangers and in that place drew them again to us. For since on account of this natural order of kinship, it was not possible that they should be united with us. He bound us newly by marriage, uniting together entire households through the single person of the bride, and mingling entire peoples . . . by taking a wife from outside the family, and through a chain of kinsmen, both mother and father and brothers and their connections’, as discussed by O’Roark . See also Augustine, City of God . . 26 Lévi-Strauss : : ‘The relationship of reciprocity which is the basis of marriage is not established between men and women, but between men by means of women, who are merely the occasion of this relationship’. Lévi-Strauss developed his theory from Mauss . 27 The notion that the key factor in establishing a marriage was the consent of the conjugal couple irrespective of the wishes of their families or any feudal lord was a fundamental tenet of both popular and learned Church teachings on marriage as early as the th cent. in the west: see Sheehan . Hiatt () relates the complex regulation of sexual behaviour in aboriginal societies.
Searching for the Romano-Egyptian Family

tions with sensual pleasure; rather it reflects considerations of an economic nature . . . The common element in all types of marriage is the establishment of mutual services between men and women. Marriage lays the foundation for a sexual division of labour which has the effect of making both sexes dependent on each other: co-operation is necessary to survive’ (p. ). This again follows a strongly materialistic line, relating marriage to economic structures, but a line which would not in itself explain incest and certain patterns of avoidance of marital partners. To explain these regulations, Zonabend turns again to the giftexchange model of Lévi-Strauss, emphasizing the social functionalism of forcing individuals to marry outside their immediate family for fear of social dislocation and isolation, turning society into a series of hostile families. This is a model with which I have some difficulty, since it seems to accept that the rules are imposed on social groups by societies fully aware of what is good for the functioning of that society, whereas it seems to me that incest taboos are too common and fundamental to be the imposition of some higher communal authority.28 If we were to assume, for a moment, that the incest taboo was not generated by a normative power in society, but emerged within the basic family and social structure, then we would have to find a material interest for individuals in seeking their sexual and marital partners outside their immediate kin: that outsiders brought something extra. Biologically, outsiders bring different genes, but it seems very unlikely that individuals would be sufficiently aware of the deleterious effects of close-kin marriage for it to determine their sexual and marital behaviour to the extent that the incest taboo does.29 In some societies, marrying 28 In most cases, dominant discourses within societies defend exogamous relationships, arguably reflecting perhaps the interests of the group in promoting exogamy. In pharaonic Egypt, however, according to Forgeau , from the XVIIIth dynasty onwards, there was a tradition in Egyptian literature which idealized loving spouses as sibling equivalents, creating a discourse that would seem to represent the harmonious and therefore socially ideal couple as being siblings. Similarly, in middle Persian texts, marital relations between mothers and sons and other extreme close relations are treated as being morally ideal (see Lee ). 29 I find the socio-biological arguments forwarded by Scheidel (: –), unconvincing. Scheidel argues that there is an incest taboo programmed into human behaviour because of the genetic implications of such relationships and adduces support for this from the reluctance of those educated together from an

Richard Alston
out brings an additional family, perhaps particularly important when large groups are required from time to time to pool labour or economic resources. Thus, the relationship with the brotherin-law or the father-in-law becomes crucial and this can be transferred a generation to help explain the much-puzzled-about close relationship of some children with the maternal uncle. While for the woman, exogamous marriage spreads the possible resources that she or her children might eventually have call upon, possibly also producing another set of family relationships that can be manipulated. In some ways this seems an absurdly functional explanation for a predominant social pattern. Incest, in this regard, would be a foolish marital strategy, but of little concern to wider society. Similarly, the rejection of the other, the outside, inherent in close-kin marriage might give rise to feelings of isolation, but would again not seem to be worthy of the depth of feeling relating to incest, and would not explain why some societies allow close-kin marriage. We need an explanation that is rather more fundamental. early age to form sexual relationships. Yet the examples Scheidel uses are drawn from societies which have incest prohibitions so that the quasi-sibling relationships he discusses would inevitably lead to feelings of uncertainty about the sexualization of the relationships. This is further discussed by Willner (), who argues that in many societies children growing up together often engage in sexual play, that the cultural and psychological damage caused by an incestuous and abusive relationship can be mistaken for genetic abnormality, and that when that incestuous relationship is socially approved (which is only in very unusual circumstances), there appears to be no psychological damage. du Boulay () shows that relationships between cousins were allowed in law but were regarded as being dangerously close to incest and thus caused social worry, while Merzario () argues that families regularly allowed incestuous marriages in certain parts of Italy without obvious reluctance, justified by the absence of other feasible sexual partners. For a restatement of the original proposition see Wolf  but the case studies may point in the opposite direction. A woman from Lu-kang is reported as having an unsatisfactory sexual relationship with her husband: ‘People say it was this way because they were raised the same as they would have been if they were brother and sister. They weren’t really brother and sister, but they were raised the same way’ (p. ). In another description of sim-pua marriage, the local informant said ‘Often the boy and girl will say they don’t want to marry because they are like brother and sister and think it is embarrassing to marry . . .’ (p. ) and a sim-pua notes herself ‘I just couldn’t do it. I was shamed by it. At the time I thought to myself, “It’s just because I don’t want to marry yet,” but that wasn’t the real reason. I think that it was because he was my brother. Marrying your brother! I felt embarrassed about it . . .’ (p. ). See also Wolf .
Searching for the Romano-Egyptian Family

In most societies, incest regulations create two groups: those with whom sexual relations are allowed and those with whom it is forbidden.30 The latter represent a quite small selection of people who are defined as special in part because of the asexual nature of the relationship. The breach of that relationship through sexual relations of an incestuous nature is destructive of close personal relations since it annuls one of the basic elements by which the relationship is understood. The understanding that a particular relationship must be asexual defines an inner core of the family, but that inner core may be defined in very different ways. The incest taboo may also be extended to quasi-kin, which may explain reluctance to enter into marriages, such as the Chinese sim-pua marriage, in which the prospective partners were educated as if brother and sister. Thus in societies which practise endogamy, the inner core may exist with a separate outer core with whom sexual and especially marital relations are preferred: cross-cousins for instance. For societies that practise sibling marriage, all it means is that the brother or sister stands outside that inner core, though perhaps in the outer element of the family with whom sexual-marital relations are preferred. Although papyri provide numerous documents relating to marriages from contracting to dissolution, the nature of the documentation is such that it provides little insight into the psychological drives of the individuals concerned. Also, it is very difficult to establish whether there was any prior relationship between the marital couple. Sibling marriage may be detected through patronymic and metronymics, but more distant endogamous marriages may escape our attention, especially given the restricted range of names in use in any particular community. Endogamous marriage limited the exchange of property within the community and also potentially limited the changes in residence that occurred on marriage, Egypt being a predominantly virilocal society.31 Bringing a new female into the family may not have been traumatic, but exporting a female to another family may have been of greater concern.
30 Possible marital partnerships are often a smaller group than possible sexual partners. See Augustine, Confessions .  and Saller . 31 Barker .

Richard Alston
Endogamy strengthened the family bond and the parental couple’s relationship extended for another generation. Endogamous marriage means that the partners were known to each other and this hopefully reduced the trauma of the foundation of a new conjugal unit. For the male, this new conjugal relationship may not have been any great change from his pre-marital life but a woman in her spouse’s house was more vulnerable to abuse and there may have been some desire to protect the woman. Interestingly, although men often mention property in divorce disputes, only women mention the behaviour of errant spouses.32 Such patterns perhaps suggest a hostility or suspicion between the family and wider society and a tendency of the family to place more emphasis on internal relations than on external links.33 Residence and the Co-habiting Family The introversion of the family might explain a reluctance on the part of some families for conjugal units to break from pre-existing paternal or sometimes fraternal residences. Nevertheless, if the complexity of family structure can be seen as a response to similar social pressures that produced sibling marriage, we need to explain why village oikiai appear more complex than urban oikiai while the incidence of sibling marriage was higher in cities. The answer to this apparent contradiction lies in residence patterns. Although households might occupy only a portion of a house, there was a parallel tendency among Romano-Egyptians to conArnaoutoglou . This argument is similar to that proffered by Shaw (), who argued that sibling marriage in Roman Egypt was a response to the Greek colonial situation. Greeks, wishing to maintain their integral Greekness, avoided marriage with Egyptians with the result that their choice of possible partner was almost impossibly restricted. This argument cannot hold since the Greeks settled in communities not isolated farmsteads, there is evidence for Greek intermarriage with Egyptians, and sibling marriage was not an exclusively Greek phenomenon. Forgeau (: ) suggests that ‘the tender relationship between brother and sister served as a reference point for both amorous and conjugal relationships and was a sign of their perfect harmonization’, an observation that suggests that siblings were preferred marital partners in the pharaonic period. In spite of these problems, Shaw’s fundamental observation, that the reluctance to engage with the wider society may lie at the root of the custom, has considerable weight. 32 33
Searching for the Romano-Egyptian Family

centrate house property in limited areas. This can be seen in topographical registers from third-century Oxyrhynchus (P. Oslo. . ) and fourth-century Panopolis (P. Berl. Bork.) which show that although owners did not necessarily look to acquire contingent properties, there was a premium on the close concentration of housing property.34 This can also be seen in P. Oxy. .  of   in which the weaver Tryphon of Dionysios acquired a share in property belonging to Pnepheros, his maternal cousin. That property had been inherited by Pnepheros from his mother, Tryphon’s aunt. West of this property was a house belonging to Pnepheros’ sister and to the north was a house belonging to Thamounis, Tryphon’s mother. What we seem to have is three adjacent houses which were divided between Thamounis and her sister. The sister’s houses were inherited by her children. In , Tryphon appears to be reconstituting the original holding, probably in the expectation of inheriting his mother’s house.35 In such a case, it seems that there were discrete units of at least two households but the family extended beyond the boundaries of each house to encompass the group of houses and the combined family continued to have a stake in all the property at the site. The same pattern can be observed in a small number of status examination returns from families of the gymnasial group. These, as already mentioned, traced back descent over several generations. For most entries, the place of residence of the applicant or his antecedents at the point of their applications is noted. There is only sufficient of these texts to produce an analysis for the gymnasial group of the city of Oxyrhynchus, but the texts show that even down the male line, families moved from generation to generation across the city. Given the tendency of families to acquire adjacent properties, it would seem probable that families that change residence over a generation would be predisposed to remain within the same district. The movement itself, then, is notable, but there is also a clear tendency of secondary movements (so of a third or fourth generation from the founder of the line) to return to the original district. This cannot be a random pattern and must reflect a drawing back of families to the Alston : –. See below for further discussion on Tryphon and family. For a similar situation see P. Mich. .  with P. Mich. . , as discussed in Alston : –. 34 35

Richard Alston
original property, perhaps caused by continued powerful cultural and social links to the original residence.36 Instead of looking at the house as the significant container of family, the co-residential group appears to be just one manifestation of family and a family might encompass residents of several houses.37 Searching for the Extended Family: Archives and Letters Finding this extended family is not, however, straightforward. As they were not co-resident they do not appear in the census returns and tax lists. As a social group that was not legally recognized and had no obvious financial function, they did not generate much documentation. Instead of relying on such ‘hard’ evidence, we are forced to rely on the essentially anecdotal information that can be derived from archives and private letters. The archival material is at first sight the most promising. The collections of families’ papers sometimes allow papyrologists to construct huge, often awe-inspiring family trees. When based on legal documents, one can have a certain confidence in many of the relationships discovered, though epistolary data in archives present more problems. One of the most impressive of these family trees is that For further discussion, see Alston . Segalen  shows that in Brittany it was common for conjugal couples to establish their own residence, but there was still a regular, often daily, commensality at a parental residence, and virtually all productive activity and social care was based on the extended family to the extent that the residences of junior conjugal couples were often empty. Kertzer, Hogan, Karweit () argue that the continued concentration on elucidating the structures of households fails to reflect the great importance of non-residential kin and although most family historians are aware of this, the quality of the data for that non-residential kin network is so poor that historians return repeatedly to census and similar data which concentrate on households. Nevertheless, several painstaking attempts have been made to reconstruct the links between non-resident kin in a variety of communities. Reay () argues that the nuclear family was one of a series of family cells within a community which co-operated in matters such as social care. For similar arguments for Italy, see Levi . Cooper and Donald () argue that the ‘servants’ in houses were often kin, especially maternal kin, which makes them very difficult to relate to the householder. This observation probably means that the now conventional view of the dominance of the nuclear family in English late medieval and modern demography is somewhat oversimplified. For a similar discussion with regard to the Roman epigraphic record, see Martin  and Rawson b. 36 37
Searching for the Romano-Egyptian Family

produced by the editors of P. Fam. Tebt. This family tree contains  named individuals, spread over several generations and attested in documentation dating from   to  . Nevertheless, not all these relationships could be described as active. Many persons on the family tree appear merely as patronyms, papponyms, or metronyms. There is also a generational problem since, obviously, family members died and would thus effectively disappear from the social world.38 More nuanced patterns emerge if we only use chronologically close texts. P. Fam. Tebt.  of – relates to the repayment of a loan. Dioskoros alias Kastor, son of Kronion and Apollonarion (maternal brother of Herakleides), acknowledge receipt from Herakleides of Maron in partial repayment of a loan contracted between Thaubarion (mother of Herakleides) and Apollonarion. The contract is made in conformity with an earlier contract between Herakleides and his maternal brothers Apollonios and Lysimachos. The active family tree is displayed in Fig. ..39 The initial loan was contracted between mother and daughter, but when the debt was extended to the next generation the loan was recalled, Dioskoros alias Kastor being quite remote from Herakleides, though the social relationship with the half-brothers
Maron = Thaubarion = Lysimachos
Herakleides Apollonios Lysimachos Apollonarion = Kronion alias Gaius Dioskoros alias Kastor
. .
The family of Thaubarion I
38 Although I use stemmata with short chronological frames here, I acknowledge that this is not always a suitable tactic since when a family constructs itself around a remote ancestor death does not remove the important individual from the social world. The distinction I operate is between genealogical kin and affective kin, as suggested by Plakans . 39 Throughout the figures that follow, active family members appear in bold.

Richard Alston
appears to have been maintained. Further documents, P. Fam. Tebt.  ( –) and  () fill in elements of the same family. Fig. . provides evidence for more extended relations at the same degree of separation within the family and one could further extend the stemmata by inclusion of more of Didyme’s family, as in P. Fam. Tebt.  of   and  of  , and the children of Didyme and Herakleides appear by   (P. Fam. Tebt. ), – ( ) and  (P. Fam. Tebt. ).40 By Fig. ., the stemma is becoming scholarly fantasy. P. Fam. Tebt.  shows Didumarion acting with her cousin, Kronios, as her kurios. Since the contract concerned her dowry and was with her brothers, neither Lysimachos nor Herakleides could represent her. There was thus a social relationship between these cousins, but there is no evidence for a similar relationship between Didumarion and her other cousins, still less between Kronios and the sons of Apollonarion and Kronion. Moreover, several of the dealings attested deal with the recovery of loans contracted by parties now dead, some of which are within the family.41 Loans made within the family or shared by the members of the family are also attested in P. Fam. Tebt.  and . The transfer of resources within the family, even from son to parents (P. Fam. Maron = Thaubarion = Lysimachos
Didyme = Herakleides Apollonious Lysimachos Apollonarion = Kronion alias Gaius Kastor
Lysimachos
Dioskoros alias Kastor
. . The family of Thaubarion II
40 This section of the family tree is particularly interesting for the patterns of nomenclature. Didyme’s children had names derived from her father, from her husband (with the addition of the mysterious Valerius), and from herself. The names Lysimachos and Kronios/Kronion (both common in the region) appear on both sides of the family. 41 P. Fam. Tebt. ; ; ; .
Searching for the Romano-Egyptian Family Lysimachos
Lysinias
Kronios

Maron = Thaubarion = Lysimachos
Didyme = Herakleides Apollonios Lysimachos Apollonarion = Kronion alias Gaius Herakleides Lysimachos Didumarion Kastor alias Valerius
Lysimachos
Dioskoros alias Kastor
. . Family of Thaubarion and Didyme
Tebt. ), may be seen as reinforcing or establishing a social link. Nevertheless, the financial dealings which dominate the archive do not show any great use of an extended family network apart from in resolving issues of inheritance. Apart from the cases already mentioned, only in P. Fam. Tebt.  (of  ) do we see an extended family at work. In this case, a man represents his female paternal cousin.42 Another extensive archive involves the weaver Tryphon son of Dionysios.43 Tryphon was born into a family of weavers, the male members of which (excepting his younger brothers Abaros and Onnophris) are registered in a census or status return of  –, submitted when Tryphon was about  years old.44 All the adults (those other than the sons of Dionysios) were weavers. Approximately twenty years later, Tryphon complains that his wife Demetrous had deserted him, taking some of his property.45 In  , Tryphon entered into a contract with Saraeus daughter of Apion who deposited the equivalent of a dowry with him.46 42 In P. Fam. Tebt.  Herakleia of Hermes rather oddly gave permission to her husband Herakleides alias Valerius to sell a slave using as kurios her relative Lysimachos. It is not evident why Herakleides did not sell the slave while acting as kurios and this rather more elaborate process of appointment of a representative had to be undertaken. 43 See on this archive Brewster . 44 The texts were re-edited by M. V. Biscottini (). The abbreviation P. Tryphon will be used for this edition. The census return is P. Tryphon  0001 P. Oxy. . . 45 P. Tryphon  0001 P. Oxy. .  0001 M. Chr. . 46 Gagos, Koenen, McNellan () argue that such deposits were common practice in early Roman Egypt, perhaps representing a native tradition of dowry payment. Whitehorne () posits a psychological explanation, suggesting that the damaging break-up with Demetrous had left Tryphon reluctant to commit

Richard Alston Tryphon
Didymos Dionysios = Thamounis Thoonis
Tryphon
Thoonis
Abaros
Onnophris
. . The birth family of Tryphon of Dionysios
They were cohabiting without contract.47 In the same year, Thamounis apprenticed Abaros to a fellow-weaver with Tryphon acting as kurios. Sometime after that, another contract appears to allude to the birth of a daughter to Saraeus and Tryphon.48 In , Tryphon complained about a violent attack on Saraeus by Demetrous and her mother Thenamounis.49 In , Thamounis, acting with the kurios of Sarapion son of Sarapion registered Thoonis, her son, among those who had gone missing.50 In  –, Saraeus and Tryphon became embroiled in a dispute with a certain Pesouris over a child whom Pesouris claimed to have been placed with Saraeus for wet-nursing and Saraeus and Tryphon claimed to be born to them.51 In , Tryphon contracted a loan of  drachmae with a Thoonis son of Thoonis whom one might assume to be either his cousin or his nephew, and in the same year he bought a half-share in a family house.52 This share was bought from his maternal nephew, Pnepheros, who had inherited the property from his mother. Alongside this property to his new relationship and Saraeus worked round this with this dowry equivalent. Yet, as Allam () points out, marriage was essentially a private matter, and it would seem unlikely that even without a formal dowry, Saraeus’ children by Tryphon could have been declared apatros. On illegitimacy in Roman Egypt see Youtie . 47 P. Tryphon  0001 P. Oxy. .  0001 M. Chr. . 48 P. Tryphon  0001 P. Oxy. . . 49 P. Tryphon  0001 P. Oxy. . . 50 P. Tryphon  0001 P. Oxy. . . 51 P. Tryphon  0001 P. Oxy. .  0001 M. Chr. ; P. Tryphon  0001 P. Oxy. .  0001 M. Chr. . 52 P. Tryphon  0001 P. Oxy. .  0001 SB . ; P. Tryphon  0001 P. Oxy. . .

Searching for the Romano-Egyptian Family
was a house belonging to Thamounis and a house belonging to Pnepheros’ sister Tausiris. In , Tryphon, Onnophris, and Saraeus stood security for a loan from a certain Tryphaina (the name perhaps suggesting a familial relationship), whose son-inlaw acted as kurios.53 Tryphon’s sons, Apion and Thoonis, both followed their father into the weaving trade.54 The picture of the active family in the s and s is rather different from that of the earlier family. Tryphon had contact with an extended family, one that includes his maternal cousins and possibly also paternal cousins. There is no guarantee that the sample of documentation within the archive accurately reflects either the social or business dealings of Tryphon, but Thamounis and her family emerge as at least as influential as Dionysios and his brothers and, after the presumed death of Dionysios, Thamounis took an active part in running the family, both apprenticing Abaros and registering Thoonis as having left the community. Thamounis’ prominent role is probably reflected in Tryphon’s interest in securing her property. Thamounis acted with a range of kurioi, including Tryphon, which is perhaps indicative of her independence. Female influence may also be reflected in the names of Tryphon’s children: Apion being derived from Saraeus’ line.
Tryphon
Papontos =
Thamounis = Dionysios Thoonis
Pnepheros Tausiris
Didymos
Onnophris Thoonis
Abaros
Apion
Tryphon = Saraeus
Apion Thoonis
. . The family of Tryphon in the s and s
AD
P. Tryphon  0001 P. Oxy. . . P. Tryphon  0001 P. Oxy. .  0001 SB . ; P. Tryphon  0001 P. Oxy. . . 53 54

Richard Alston
There are relatively few archives that allow this detailed analysis over time. More often we can only deal with fragments of family. The small archive of Pekebkis from second-century Tebtunis contains a complaint from Kronion the elder, son of Pakebkis.55 Kronion and his brother, Kronion the younger, and their wives had been living with their father’s wife Taarmiusis, presumably after the death of their father. The property (that of Pakebkis?) was shared with her and her sons and contracts were made to that effect. On the death of Taarmiusis, the brothers came into conflict with Taarmiusis’ children. The family unit appears to survive the death of the father, but cannot survive the death of his wife. The other texts in the archive are fragmentary letters or rental agreements and involve the brothers Kronion with a Louris and an Isidora who may be relatives.56 The archive of Kronion is described by Foraboschi (: xxii) as providing an example of a patriarchal family with a close nucleus co-operating continually in economic matters. Much of the archive centres on those economic matters and such documentation only rarely sheds light on the family. Nevertheless, in P. Kronion  four members of the family, Kronion, his wife Thenapunchis, their son, Kronion, and their daughter Taorsenouphis contracted a loan. Taorsenouphis and the younger Kronion were married to each other. Thenapunchis’ brother may appear in P. Kronion  in which an Orsenouphis, who carried the same patronym as Thenapunchis, partnered the elder Kronion in borrowing grain. In , the elder Kronion rented some of his and Patunis
Kronion = Thenapunchis
Orsenouphis
Kronion = Taorsenouphis
Harphaesis
. . 55 56
The active family of Kronion
P. Mil. Vogl. . . P. Mil. Vogl. . –. Texts  and  are to  – and  respectively.
Searching for the Romano-Egyptian Family

his daughter’s lands to a Helene. Taorsenouphis’ role in this transaction seems passive. In , the elder Kronion repayed a loan contracted with Helene to her daughter-in-law who appears to have inherited her estate. This extremely fragmentary portrait of family life emphasizes the economic interdependence of the nuclear family of the elder Kronion and Thenapunchis and their children. Nevertheless, the probable appearance of Thenapunchis’ brother in the archive suggests that the extended family may have played a significant role, and, as with the Tryphon archive, the maternal connections retain their importance. Another family, associated with Soter, appears in inscriptions on coffin lids and from papyri probably found in a tomb in Thebes, though the archaeological context is not completely clear.57 The tomb contains three generations of a family, most and possibly all of whom were direct descendants of Soter and Kleopatra. It contains a married daughter of Soter and her child and also the children of Herakleios suggesting that in death, Soter’s family continued to control both male and female members of the family, irrespective of their marital status. There are considerable problems reconstructing this family (Fig. .).58 The Greek texts, as is normal, give the paternal line of the individual and sometimes the maternal line while the Egyptian texts often only give the maternal line. In addition, it is possible some individuals used different Greek and Egyptian names. This leads to problems reconstructing the family tree, most notably in placing a Kleopatra and a Petronios, both born of Kandake. These two are either Kleopatra alias Kandake and her brother or further children of Soter and Kleopatra and I would, on balance, favour the first option.59 It also seems possible that Taloulou and See Kakosky  for the archaeology of the tomb. The reconstruction here is based on the collection and identification of texts carried out by Landuyt (). 59 Kleopatra alias Kandake normally appears in both Greek and Egyptian as Kleopatra. It is not unusual for the ‘alias’ name to be derived from a parent, which would suggest Kleopatra was the daughter of Kandake. Of course, names follow family lines and it would be normal for Kleopatra to name a child after herself or her mother. The name Petronius adds to the difficulties by bringing a second unexplained Roman name into the archive. Soter was probably born c.  , which puts his father’s birth c.  . The name could derive from a military connection or be a reward to an influential individual and Soter’s boast of his importance may reflect the status of his family. In such circumstances, a 57 58

Richard Alston
Kornelios = Phimois
Ammonios = Kandake
Soter = Kleopatra alias Kandake
Petronios
Apollonides Petemnouphis Sensaos Tkouthi = Pebos Taloulou = Herakleios = Sarapous alias Ammonios Petemenophis
Phaminis
Tphous
. . The probable family of Soter and Kleopatra alias Kandake
Sarapous are the same person. If the reconstruction in Fig. . is correct and Kleopatra’s brother Petronios was buried in the tomb as well, then we may be examining the family of Kleopatra alias Kandake rather than that of Soter.60 Further evidence of familial and domestic social groups can be found in letters.61 Very often letters include greetings from people close to the writer and to people close to the recipient. Such greetings reinforced social links to a remote group of friends and family and had a symbolic value in maintaining social relationships that distance had put under threat. The culture of letters was less private than we are perhaps used to and we might expect letters to preserve a relatively complete listing of local social circles.62 The letters of Claudii Terentianus and Tiberianus mostly relate to the social circle of the army, though there is some connection with another family which had previously been rewarded by the Roman rulers seems less implausible. 60 The archive has a mix of Roman, Egyptian, and Greek names, and, although the burials are in Egyptian fashion, they use a mixture of Greek and Egyptian in the funerary documentation. The most interesting name is Kleopatra alias Kandake, combining the nomenclature of Ptolemaic and Meröetic queens. 61 One of the earliest family archives is that of Athenodoros and family which dates to the first three decades of Roman rule and consists mainly of letters between seeming siblings. Even though family terms could be used as marks of honour for non-kin (explaining Athenodoros’ multiple parentage), the nomenclature suggests biological family. For the texts, see BGU . ; ; ; ; ; ; cf. ; . ; ; ; ; . 62 For these archives, see Alston ; : –; Pighi ; Adams .
Searching for the Romano-Egyptian Family

mention of non-military connections.63 For instance, P. Mich. .  includes greetings sent to fifteen military figures by Terentianus. In P. Mich. . , however, Terentianus writes to his father Tiberianus about a series of apparently unforeseen events that had delayed Terentianus, including his mother giving birth, and his father Ptolemaios leaving for Alexandria and being involved in a fracas. This other father also appears in P. Mich. .  in which Ptolemaios and Terentianus’ brothers send greetings and is referred to without being named in P. Mich. . . Females with Tiberianus include Aphrodisia (P. Mich. . ) and Isidora (P. Mich. . ). Also a Tabatheus addresses Terentianus as ‘brother’.64 As far as can be judged, Terentianus resides with his mother and father Ptolemaios, and his brothers, but writes to his father Tiberianus, who both shares his nomenclature and holds some authority over him.65 The archive of Iulii Sabinus and Apollonarius also provides us with a great range of people, many of whom were also military, with whom the central pair are connected. Nevertheless, Apollonarius had a sister, Iulia Sarapia, and the letters send greetings to Abaskantos, Ptolemaios, Thermouthis, Tasoucharion, Demetrous, and Aphrodisia, who appear to be located at ‘home’.66 A Sabinianus wrote to ‘his brother’ Apollonarius thanking him for a kindness towards Sabinianus’ mother and sister, and Sabinianus’ name would suggest a connection with Iulius Sabinus.67 Letters which are not part of an archive are more difficult to exploit yet they provide insight into one moment in the life of a family. In the third century Ischyrion wrote to Cornelius, his father. Ischyrion had received a letter, to which he was replying, from Cornelius through Serenus. Ischyrion had news of his mother. He discussed sending clothes. He sent greetings to his brother Horion, his mother Thaesis, and ‘all those in the house 63 One could see the army as providing some sort of alternative family since fellow soldiers may provide a primary or basic social group intermediate between the individual and the community. It is possible that the community of soldiers overlay or existed alongside family communities: see Alston . 64 P. Mich. . . 65 P. Mich. . . 66 P. Mich. . ; . In P. Mich. .  Theon addresses Apollonarios as brother and mentions another brother Apollonios, but it seems very likely that these were honorary titles. 67 P. Mich. . .

Richard Alston
by name’.68 This final phrase, or its reverse, ‘all in the house greet you’ is frequently repeated in various letters.69 The letters seem to operate a two-stage distinction between those who are named and those who are subsumed into the unnamed categories. Quite frequently, siblings are included in the named category, though, equally frequently, those named have no stated familial relationship. In – Thermouthis wrote to her brother Isidoros. The letter covered many items of business, such as the movements of the governor, problems with the grain supply and clothing, then we have greetings from Taesis, Kopreas, Amois, and ‘each and every man’.70 Arsinoe wrote to her sister saying that she would collect rents. Her sister was married to Polykrates, to whom Arsinoe sent greetings and also forwarded greetings from Poleta and Demetrous.71 The two sisters, at least one of whom resided with her husband, maintained financial and emotional ties. P. Oxy. .  provides an example of a different sort of letter. Valerius wrote to his brother Ophellianus asking him to look after and support the son of Diogenes, the father being an Epicurean philosopher, a striking example of social and familial networks in co-operation. Letters are often formulaic and business-like, but correspondents sometimes express strong emotions, though the papyri have not, unfortunately, produced an extensive corpus of amatory epistles, an exception being P. Oxy. .  in which Didyme wrote to Apollonios, her ‘brother and sun’: ‘Know that I cannot see the sun because you cannot be seen by me for I have no other sun but you’. She goes on to send thanks to Apollonios’ brother Theonas and to his father. Didyme and Apollonios were lovers, and the ‘brother’ appears not to reflect consanguinity. A less romantic, though perhaps more typical letter was sent by P. Oxy. . . See, for example, and with minor variations, such as ‘look after’ rather than greet BGU . ; ; . ; . ; CPR . ; P. Brem. ; P. Col. . ; P. Fay. ; ; P. Graux ; P. Haun. .  0001 SB . ; P. Iand. . ; P. Lond. . ; P. Mert. . ; P. Oslo. . ; . ; P. Oxy. . ; . ; . ; . ; . ; P. Wisc. . ; ; P. Würzb. ; SB . ; . ; . ; O. Florida . The reverse is less common. See BGU . ; P. Hamb. . ; P. Mert. . ; P. Mich. . ; P. Michael. . ; P. Oxy. . ; . ; PSI . ; P. Wash. Univ. . ; P. Wisc. . ; SB . ; . See also UPZ .  for a similar expression used in a papyrus of  . 70 P. Oxy. . . 71 P. Oxy. . . 68 69
Searching for the Romano-Egyptian Family

Dionysios to his sister.72 Various goods were sent along with the letter and greetings to the children ‘and those in the house’ and greetings were sent from Chaeremon and Agathos. Filial devotion is shown by Harpokras who wrote to his father ‘knowing that you will be most pleased, I am obliged to write to you that there is nothing wrong with me . . . I took great pleasure in your letter in which I found you were in the best of health, lord father, and my spirit was raised on receiving your letter and at that hour I thought it must be an oracle of a god and my health improved.’ Harpokras sent a gift of sandals and greetings to Thatres, his mother, and three females, Thaisous, Nike, and Eutyche, and all those in the house.73 The recent excavations at Kellis have produced a number of letters from closely defined archaeological contexts (house  or house  at the site). There are reasons to be suspicious of these letters, since they are both later than the period on which I have focused and deal with an unusual oasis community. Moreover, that community was religiously atypical, having Manichaean elements. It is difficult to disentangle elements of the religious community from the familial communities and, anyhow, it is not always clear that this is proper.74 These letters, however, suggest extensive familial or quasi-familial groups. A letter from house  Titoue to his son Shamoun informs Shamoun that his mother and brethren are well. Greetings come from Shamoun’s wife and children, his son Titoue who has gone to the monastery, and from his siblings Pshemnoute, Kuria, Tapshai, and Taoushai and from a seemingly unrelated person, Tbeke. Shamoun appears to be away from his entire family and in the delineation of that family, 73 P. Oxy. . . P. Oxy. . . For instance, P. Kell. Copt.  is from Matthaios to his mother Maria. He tells her he sent his fathers Pseke and Pishai with medicines to her and his teacher had not yet met his father because ‘they are in mourning in the city for the blessed soul of my great mother. We are remembering her very much. And I was distressed that she died when we were not with her, and that she died without finding the brotherhood gathered around her. Do not neglect to write to us about your health.’ The letter opens with an extraordinary expression of affection: ‘To my mother, my loved lady, very precious to me, the beloved of my heart; the one whose memory and worthy motherhood are sealed in my heart every hour; the one whose kindness and goodness that she performs for me at all times are sealed in my innermost thought, my mother, very precious to me, Maria.’ 72 74

Richard Alston
we find father, mother, siblings, wife, and children.75 A more nuclear family is attested in another letter from house  in which a mother writes to her son sending greetings from his sister and his children and mentioning his father.76 Perhaps the most interesting of the letters, however, is P. Kell. Copt.  which, at face value, would reconstruct the family in Fig. .. A similarly extensive group is registered in P. Kell. Copt.  where Matthaios sends greetings to or mentions five fathers, five mothers, and eight other nuclear family groups. The Kellis Coptic papyri are at one end of an extreme in the depictions of family, an end in which family disappears into community. Far from there being a nuclear family structure, family becomes a concept extended far beyond any possible genetic or even economic link. This is not to say that family is not deployed as a useful concept here or that family has no meaning. The extension of the terminology of family suggests, on the contrary, that family terms retained a meaning which Egyptians sought to apply to a range of relationships. Motherhood had extended beyond the progenetrix to the collectivity of mothers. Although the Kellis Coptic papyri may provide extreme examples, multiple fathers are found elsewhere and the terms ‘brother’ and ‘sister’ appear to have been widely applied. Although one would not expect there to be real confusion over paternity, there is inherent in these terminological slippages a blurring of definition: we
Talaphanti
Tamougenia
Partheni
Tskmshai
children Choris Philammon Mo Isis Makarios Maria ]fnoute Kouria
children
Mattheos
Andreas
Drousiane
Tshemnoute
Kame
. . The family of Makarios: P. Kell. Copt.  P. Kell. Copt. . P. Kell. Copt. . There is no mention in this letter of the recipient’s wife and the ‘sister’ carries the same name as the author of the letter, adding to the supposition that she was the ‘sister’s’ mother. 75 76
Searching for the Romano-Egyptian Family

should recall that for a writer sometimes a ‘brother’ who was a husband was also a brother.77 The letters addressed and stemmed from populous local communities comprising different groups of people. Some received familial epithets, some are given names alone, and others seemingly based around the house, perhaps the household, greet and are greeted by a ‘catch-all’ phrase. If there is a gender bias in the naming of those seemingly ‘at home’ it is towards women. The studied family archives present a rather less diverse family, emphasizing the nuclear or close to nuclear structure, with family connections not being extended laterally. Where we do see connections beyond the nuclear, these appear often to be maintained through the surviving member of an older generation who stops the fragmentation of families. It is notable that although families form around a conjugal couple, maternal relations often appear to be as or more important than paternal relations. The archives sometimes give the impression that women formed the centre of families in spite of male dominance of public and economic life. Conclusions At the heart of most extended or complex families are one or more nuclear families and these elements of families figure heavily in our evidence. The nuclear family was important for the transmission of property and status and, at least in some urban areas, appears to have been the basis of residential groups. This is hardly surprising and is common to many societies. What is more difficult and interesting to measure is how those structures were animated in social life, how they provided support for individuals, how they interacted with social institutions and forces outside the immediate family group, how power was felt and expressed within the family. The assembled evidence in this study points to another not very radical conclusion: that there was considerable variation in family structure and probable experience in Egypt and there is no reason to postulate even a relatively 77 Familial terms are used for those outside the family in other cultures. In colloquial Arabic ‘maternal uncle’ or ‘paternal uncle’ can be applied to respected older men and ‘sister’ to women of roughly equivalent age to the speaker and we would not expect there to be any confusion as to the familial status of the individuals addressed.

Richard Alston
homogeneous experience of the ‘Egyptian family’. Although I have not elaborated on this theme here, there is every reason to believe, partly because of the differences noted in urban and village census returns, that families of different cultural leanings and of differing economic status might have disparate family experiences, and, no doubt, some more Roman patterns of behaviour were introduced. Families appear to have had a fairly tight-knit centre, often concentrating on a single conjugal relationship. Family extended from that conjugal couple to include children, sometimes the siblings of the conjugal couple, incorporated the community of the house, and a wider, more amorphous community, reflected especially in the letters. Most of our texts emphasize the legal and economic functions of social entities but with the extended family, we have a grouping of people the functions of which appear to have been social, and were ill-defined. We have good evidence that this grouping included non-co-resident kin. There are plenty of parallels in the comparative literature to suggest how such relationships might work. Further, we have the greeters and greeted of the letters, most of whom are just names to us, some of whom have familial designations, but many of whom might not have been kin. At this stage, family begins to dissolve into community. Yet, there are boundaries, boundaries powerfully reinforced by endogamy. The practice of endogamy reflects a certain insecurity about those outside the group. Endogamy enhances the sense that family was a focal point, a place around which other social relations were negotiated. If my reasoning is right, brothers and sisters were not seen as having the same proximity of familial relationship as parents and children, but since the incidence of sibling marriage was so high, we know that siblings were preferred marital partners, in a special and possibly idealized loving relationship. Siblings existed in a secondary core, a step closer to the extended familial community.78 The family’s functions, social welfare, social reproduction and socialization, might thus be spread across a spectrum of familial feeling, from the family’s conjugal heart to the quasi-familial fringes. We may speculate that in matters of law and status, the conjugal couple and 78 The distinctions here are obviously different from the modern nuclearextended dichotomies.
Searching for the Romano-Egyptian Family

children might be the crucial element, there might be a greater diversity of structures for economic purposes, and for functions such as the socialization of children or political activities, the wider family became important. The value invested in the wider group should not be seen as detracting from the importance of the core groups. Instead, a variety of structures coexisted and reflected the importance placed on the family by Egyptians in the face of the hostility of the wider world.
This page intentionally left blank
 The Jewish Family in Judaea from Pompey to Hadrian—the Limits of Romanization Margaret Williams
As for marriage and the burial of the dead, Moses saw to it that the Jews’ customs should differ widely from those of other men. But later, when they became subject to foreign rule, as a result of their mingling with men of other nations . . . many of their traditional practices were disturbed. (Hecataeus of Abdera, as cited in Diodorus Siculus . )
H, as the full text of Diodorus makes clear,1 was speaking of the impact of Persian and Macedonian rule upon the Jews. But foreign domination of the Jews did not stop with the early Macedonians or even with the Ptolemies and Seleucids. With Pompey’s conquest of Syria in  , Judaea came under Roman control. From then until the seventh century , it was Roman power that was paramount there, at first exercised indirectly through the Hasmonaean and Herodian families and subsequently by governors sent out from Rome. The nature of Roman political power in Judaea, particularly in the period covered in this essay ( – ), has been extensively studied.2 What has received far less attention is the social impact of Roman rule, particularly upon the Jewish family. That is not to say that there are no accounts of the Jewish family in Judaea in the early Roman period. The most accessible, S. Safrai’s chapter, ‘Home and Family’, in volume ii of CRINT, provides a substantial Thanks are due to Professor J. A. Crook of St John’s College Cambridge, for reading a draft version of this chapter and suggesting a number of improvements. The views expressed here are my own. 1 Stern –: no. . 2 See, for instance, Schürer i ; Smallwood ; Millar : chs.  and .

Margaret Williams
discussion of the subject.3 But the problem with this and with works drawing heavily upon it is that, being based mainly on rabbinic evidence, they are essentially ahistorical and so present the Jewish family as if it were homogeneous and unchanging.4 But if, as Hecataeus claims, Persian and Macedonian rule changed Jewish family practices, it surely is no less likely that Roman rule will have done so too and, what is more, impacted upon different families in different ways. That is the hypothesis that will be tested here through an examination of contemporary source material for the Judaean Jewish family in the early Roman period.5 While the main focus will be on the writings of Josephus and papyri from the Judaean desert, attention will also be paid to archaeological evidence for family habitation and burial, ossuary inscriptions, and relevant New Testament texts. Questions that I will be asking are these: Is there any evidence for change in the Jewish family as a result of the establishment of Roman hegemony over Judaea, particularly after  , the year when the rule of Herod, otherwise known as C. Julius Herodes, effectively started? If there is, does that change intensify (i.e. become greater) over time? What was its nature—was it fundamental, affecting the very structure of the Jewish family, or did it amount to no more than the adoption of the outward signs of romanitas— for example, Latin personal names, Graeco-Roman style houses, and Graeco-Roman burial practices?6 And finally, who was affected by it—only elite families, such as that of Herod, whose members hobnobbed with Roman officialdom, visited Rome, and networked and intermarried with other aristocratic families in the Roman east, or were lower status families, through imitation of the elite, in due course affected as well? 3 S. Safrai a: –. Useful material is also to be found in S. J. D. Cohen  and Ilan . 4 This is especially the case with Archer (), who treats the rabbinic evidence, much of it rather late, as though it has a virtually timeless validity. Although Collins () considers a wider range of evidence, there is very little sense of movement during the six hundred years covered by his study. 5 Unlike Collins (n.  above), I shall not include the documentary evidence from Elephantine in my discussion, as it is too early (th cent. ) and not Judaean. I shall also exclude rabbinic material because of its lateness and dubious historicity. 6 On the fusion of Greek and Roman elements in contemporary Roman culture, see Huskinson  and, more briefly, Goodman : –.
The Jewish Family in Judaea

I begin with a brief discussion of the shape of Jewish families in the period between Pompey and Hadrian. I then investigate what changes Jewish families underwent at that time. Since it is in the archaeological evidence for habitation that change is most obvious, I start by considering that. Then, using my findings as a template, I will see whether analogous developments occurred in other areas of family life. Taking the family life-cycle as a broad framework, I will look first at the circumcision and the naming of sons, secondly at education and the arrangement of marriages and, finally, at divorce and burial practices. My reason for selecting these topics is that they provide the most fruitful evidence for attempting to answer the questions posed above. Ideally, I would have liked to discuss also other aspects of family life such as the exposure of infants (ekthesis), the adoption of Roman social rituals (e.g. mixed-sex banqueting and the celebration of birthdays), and Jewish use of Roman legal practice in, for instance, the area of testamentary disposition. The reason that I have not done so is that the evidence is either too slight7 or too restricted in its social range.8 The Shape of the Jewish Family in Judaea: Monogamy and Polygamy Although in the period with which we are concerned a minority of Jews (e.g. the Essenes) rejected conventional family life altogether in favour of a celibate, ‘monastic’ existence,9 most Jews took literally the divine injunction to ‘be fruitful and multiply’ (Genesis : ). As to how that multiplication could best be achieved, the Almighty himself offered no prescription. Genesis and Exodus, however, did supply two models—one through 7 The tiny amount of evidence suggesting that Jews, as a result of Gentile influence, may have practised infanticide mostly relates to the Diaspora, not to Judaea. For a useful discussion of the subject, see Reinhartz . 8 i.e. it relates solely to the Herodian family. For their mixed-sex banqueting, see Osiek and Balch : –; for their birthday celebrations, see n.  below; for the testamentary dispositions of Herod the Great, see Josephus, Jewish War . – and . –; Antiquities . – and –. For a modern discussion of the Jewish and Roman features of Herod’s will, see Hanson : –. 9 For celibacy in the Second Temple period and the extent to which it was practised at Qumran, see Baumgarten .

Margaret Williams
monogamy, the paradigm here being the relationship between Adam and Eve,10 the other through polygamy, the most conspicuous and influential example of this being provided by the Patriarch Jacob. A further reason for polygamy was subsequently provided by the levirate law (Deuteronomy : –), under which a married man was required to take as an additional wife the childless widow of his brother and have children by her in order that, fictitiously, that brother’s name and line might be kept alive: the first son she bears shall perpetuate the dead brother’s name so that it may not be blotted out from Israel (Deuteronomy : ).11 Given this biblical background, it should come as no surprise to learn that two types of families, both patriarchal, are attested in Judaea in our period—one, monogamous, the other, polygamous. With regard to their relative prevalence, the former is by far the more frequently attested in the literary sources—the Gospels and early chapters of Acts, for example, texts full of references to Jewish families, elite and other, contain not a single instance of polygamy. If we had only such sources to go by, the conclusion would be inescapable that by the first century  bigamy and, even more so, polygamy had become rare practices, restricted to a tiny minority of the elite.12 Papyrological finds in Judaea over the last few decades, however, have forced a revision of that conclusion. The key text here is P. Yadin , a document from the Babatha archive dating from  , relating to a property dispute between the two widows of Judah, son of Eleazar Khthousion, namely Babatha herself and Miriam, daughter of Beianos. I cite only the relevant part of the text: Before the attending witnesses who also affixed their signatures, Babathas [sic], a Maozene woman, daughter of Simon, summoned Miriam, an En-gedian woman, daughter of Beianos, to accompany her in person before Haterius Nepos, legatus Augusti pro praetore, wherever his venue may be, [to answer] why you seized everything in 10 Note here the Damascus Doc. . , where the Creation narrative is used to denounce polygamy as fornication. For a translation of this text, see Vermes : . 11 For a brief discussion of levirate marriage, see S. Safrai a: –. 12 Apart from Herod the Great’s grand-scale polygamy (n.  below), the practice is little attested, even in the Herodian family. The few cases known, some of them arising out of the levirate law, all relate to aristocratic families, some of them High Priestly. For documentation and discussion, see Lowy : –.
The Jewish Family in Judaea

the house of Judah son of Eleazar Khthousion my and your late husband . . . and, equally important, to attend before the said Nepos until judgement. Miriam replied, saying: Before this I summoned you not to go near the possessions of my and your late husband . . .13
Although R. Katzoff () contends that the situation here is one of serial monogamy, N. Lewis () is surely right to argue that ‘the simple, unforced sense’ of the phrase, my and your late husband, points to polygamy. For while it is understandable that Miriam, even if a divorcee, would use those words to put herself on a par with Babatha, it is, to quote Lewis, ‘inconceivable that Babatha, if Judah’s sole wife at the time of death, would tolerate such language . . . blurring the distinction between herself, the lawful wife, and a divorced former wife’. Assuming, then, that these two women had been married to Judah concurrently, it follows that polygamy was more prevalent than had hitherto been thought and also practised further down the social scale: Babatha and Miriam, though belonging to wealthy, property-owning families in the border region between Judaea and Nabataean Arabia, were not in the same socio-economic league as the other families hitherto known to have practised polygamy.14 P. Yadin , however, is not the only documentary evidence pointing to the ongoing practice of polygamy in the social stratum occupied by Babatha. Several of the broadly contemporaneous marriage contracts from the Judaean Desert (six out of the eight certain examples) contain a clause making special provision for the sons of the marriage in the event of the mother dying before the father—namely, they are to be the sole inheritors of her dowry.15 This stipulation is seen by H. M. Cotton as a means of protecting the ‘sons of polygamous marriages against the loss of part of their mother’s property to the sons of another woman’.16 Assuming that she is right, this clause provides good, For the full text, translation, and commentary, see Lewis : no. . Lewis (: –), followed by, inter alios, Goodman (: ) and Collins (: ). 15 See P. Mur. , , , and  in Benoit, Milik, and de Vaux  0001 DJD , nos. , , , and ; P. Yadin , for which see Yadin, Greenfield, and Yardeni ; XH . ev./Se. Gr. , for which see Cotton and Yardeni : no.  0001 DJD , no. . For the two marriage contracts without this clause (P. Yadin  and ), see Lewis : nos.  and . 16 See DJD .  and Cotton : –. 13 14

Margaret Williams
unwitting testimony to the likely continuation of polygamy at least in the rural parts of Judaea in the late first and early second centuries .17 Family Habitation in Judaea (st century BCE–st century CE) Compared with the Byzantine era, the number of excavated Jewish dwellings dating from the early Roman period in Judaea is fairly small.18 Despite this comparative lack of evidence, there are still enough examples from across the social spectrum to enable us to form some idea of the impact of Rome and its culture upon the various strata of Jewish society. To illustrate how the very rich lived we have Herod the Great’s palaces, most notably those at Masada and Jericho, built to house not only his large extended family, the consequence of his grand-scale polygamy (ten wives) but his even larger court.19 Rich Jewish families, like their Roman counterparts, owned multiple dwellings, some in the city, others in the country. Likely urban dwellings of Judaea’s priestly aristocracy are thought to have been the substantial first-century () houses, most of them equipped with ritual baths (miqvaot), that have been excavated in the Upper City of Jerusalem.20 The best known and most frequently illustrated of these is the so-called Palatial Mansion.21 A particularly fine example of an elite rural dwelling of the same period, possibly also belonging to a priestly family,22 is the so-called H . ilkiya Palace, a magnificent Roman-style villa situated at Khirbet el-Muraq on the western slopes of the Hebron hills.23 Although non-elite housing is comparatively poorly represented in the archaeological record, examples exist of both medium and small town houses24 and farms.25 Most typical, 17 For the dates of the marriage contracts listed in n.  above, see Cotton :  n. . 18 Hirschfeld : –. 19 The most easily accessible accounts of Herod’s palaces are those of Nielsen  and ; Roller ; Broshi . Note also Netzer . For his court, see Kokkinos, forthcoming. 20 The fundamental treatment here Avigad . 21 Avigad : –; Hirschfeld : –. 22 Z. Safrai : . 23 Hirschfeld : –. 24 Hirschfeld : –, the so-called Wall House at Gamala. 25 Hirschfeld : –, the farm building at Umm Rih.an.
The Jewish Family in Judaea

though, of non-elite housing are the complexes made up of small dwelling units (usually of one or two rooms) built around the sides of a central courtyard, generally onto its perimeter wall, if it had one.26 A good example of this is the so-called House of Peter at Capernaum, once the home to one or more fisher families.27 To what extent does Roman influence display itself in these buildings? In non-elite dwellings it is not to be seen at all—we have to wait until the Byzantine era before mosaics, for instance, become widely used in ‘middle range’ homes.28 However, in many of the dwellings of the rich we find examples of the decorative and the structural features most commonly associated with elite housing in the early Roman empire—mosaics, frescoes, carved stucco, peristyle courtyards, triclinia, and heated bath complexes. Nowhere are these more in evidence than in those palaces of Herod the Great which were built at the height of his power when he was enjoying the close friendship and support of both the emperor Augustus and his right-hand man, Marcus Agrippa. For detailed information about the palaces, reference can be made to the works cited above (n. ). All that is required here is a brief demonstration of how these buildings became increasingly Romanized as Herod’s reign progressed. This can be achieved most easily by comparing the earliest and the latest of them, namely the so-called Western Palace at Masada, dating from the s , and the Third Winter Palace at Jericho, built around  . In the former, thought to have been constructed by local craftsmen, there are no specifically Roman features—the internal courtyard, for example, is in the local ‘columnless’ style (i.e. it is non-peristylic) and the baths (all non-heated) are of either Greek or Jewish design (miqvaot).29 By contrast, the Third Winter Palace at Jericho contains so many western features that scholars generally have concluded that Herod probably hired for its design and construction architects and craftsmen from Italy.30 Z. Safrai : – and Hirschfeld : –. Snyder : –; Blue : – and –; Osiek and Balch : ; Guijarro : . 28 Hirschfeld : . 29 Nielsen  and : –. On the ‘essentially Oriental’ character of this palace, see Foerster : –. 30 Nielsen  and : ; Netzer : , citing Kelso and Baramki, the original excavators of the palace. 26 27

Margaret Williams
Among the most obvious of those features are opus sectile flooring, opus reticulatum wall facings, wall paintings in the Second Style, and, as a complement to the obligatory ritual baths (miqvaot), a state-of-the-art Roman bath suite right out of the pages of Vitruvius. Roman influence on Jewish dwellings, though most powerfully in evidence in Herod’s palaces, is not restricted to them. We can also see it in some of the homes of the Jewish aristocracy, some of whose families, priestly and lay, were connected to the Herods by marriage.31 N. Avigad (: ) has observed that ‘Herod’s winter and summer palaces established norms which deeply permeated the material aspect of the lives of the upper crust in Judaea’. So it is not unexpected to find mosaic flooring becoming increasingly popular in aristocratic houses in Jerusalem in the first century .32 Nor is it surprising to learn that, while the Palatial Mansion included elaborately carved stucco and fine wall paintings in the Second Style among its various expensive Roman effects,33 the villa at Khirbet el-Muraq contained, suitably decorated, a tetrastyle atrium, a peristyle courtyard, two triclinia, and a heated bathhouse.34 But these palaces and houses, while providing indisputable evidence of Romanization, do something else as well—they reveal with absolute clarity the limits of that Romanization. Herod was so rich and domestically so powerful that he could have built and decorated his palaces in any way he liked. It is therefore highly significant that, in obedience to the Second Commandment, the mosaics and wall paintings contain no depictions of humans, birds, or animals. That is also the case with nearly all the other elite dwellings of the period which show signs of Roman cultural influence. Rare exceptions are the house on Mt. Zion, containing a depiction of birds in one of its frescoes,35 31 Josephus, Jewish War . , , and  (daughters of rich Jerusalemites); Antiquities . – (Mariamme II, daughter of Simon, son of Boethus, one of Herod’s appointees to the High Priesthood). 32 Hirschfeld : . 33 Avigad :  and , noting parallels with Pompeii. 34 Plan and description in Hirschfeld : –. 35 Broshi : . The only other depiction of birds in Jewish art of the st cent.  is to be found in the tomb of the Goliath family at Jericho. See Hachlili and Killebrew : –, who claim that the inspiration for the painting came from ‘the Graeco-Roman tomb frescoes prevalent at this time’.
The Jewish Family in Judaea

and Herod Antipas’ palace at Tiberias which, shockingly, had representations of animals on its walls and was in consequence burned to the ground by the Jews in the rebellion of   ( Josephus, Vita –). What this avoidance of the figural reveals is that while there was an openness on the part of at least some aristocratic Jewish families, priestly as well as lay, towards the elite culture of the Roman world and a willingness to appropriate some of its status symbols, such Romanization was to be on Jewish terms, the principal one being that the commands of God, as set out in the Law of Moses, had to be observed. With this in mind, we will now turn to other areas of Jewish family life and examine what changes, if any, took place in them as a result of the establishment of Roman hegemony over Judaea. Circumcision and the Naming of Sons These two practices, as the Gospel accounts of the circumcision and naming of John the Baptist and Jesus of Nazareth make clear, both took place in the family home on the eighth day after birth. Of the former, Luke (: ) writes: Then on the eighth day they came to circumcise the child; and they were going to name him Zechariah after his father. But his mother spoke up and said, ‘No! he is to be called John.’ 36
Although circumcision and name-giving both took place at the same time, in status and significance the two acts were quite different. Circumcision, the outward sign of the covenant between God and Israel, was a divine injunction: Every male among you in every generation shall be circumcised on the eighth day (Genesis : ). By contrast, the name-giving which happened to take place on the same occasion was merely a matter of custom, the Torah being totally unprescriptive in the area of onomastic practice. Such being the situation, one would not expect to find much evidence for the abandonment of circumcision in the early Roman period, notwithstanding the mixture of contempt and abhorrence with which elite Romans tended to regard the practice.37 With name-giving, however, innovation is only to be 36 37
For the circumcision and naming of Jesus, see Luke : . Schäfer : –.

Margaret Williams
expected—historically, change of ruling power had always brought in its wake change in Jewish onomastic practice.38 So what do our sources reveal about circumcision and namegiving in the early Roman period? With regard to circumcision, the considerable evidence supplied by Josephus about Herodian practice is extremely illuminating. The Herods, if anyone, might have been expected to abandon the practice, given their enormous political ambition and keen desire to find acceptance at the highest levels of Roman society. Josephus’ narrative shows that they did no such thing. Not only did they observe the ritual themselves but they insisted that Gentiles who wanted to marry into their family underwent circumcision too.39 (The one recorded instance when this did not happen will be dealt with below under marriage.) On two occasions highly desirable alliances with other Roman client kingdoms are known to have foundered upon this Herodian insistence on circumcision.40 The reason for this tough stance of the Herods is not hard to deduce: they recognized that their Jewish subjects would not tolerate the rule of those who ‘forsook the covenant’, as those who abandoned the practice of circumcision were described ( Maccabees : ; Jubilees : –). The uprising of the Maccabees, which had humbled the Seleucids and swept away their aristocratic non-circumcising Jewish collaborators, provided all too clear a warning of the likely fate of those who flouted popular feeling on this issue. If the Jews remained uncompromising on the issue of circumcision, in the area of onomastics they were prepared to show a modicum of flexibility. Here, once again, it was Herod the Great who led the way, by starting the practice of using Latin personal names. Unlike his fashion-setting in architecture and interior decoration, however, Herod’s innovations in the sphere of onomastics were extremely modest and rather late. It was only with his grandson, Agrippa, born in   and named in memory of Herod’s late friend and patron, Marcus Agrippa (died  ),41 that Roman names entered the family name-pool. For his children, Herod had generally chosen the Greek names that had N. Cohen , . Josephus, Antiquities . ; .  (two cases); . . 40 Josephus, Antiquities .  (Nabataean Arabia); .  (Commagene). 41 For Herod’s friendship with Agrippa, see Josephus, Antiquities . – and –. 38 39
The Jewish Family in Judaea

become popular in the Hasmonaean period—for example, Alexander, Aristoboulos, and Philip.42 Given the slowness with which even the Herods adopted Latin names, it is not surprising to find that their take-up by other Jewish families in Judaea was also limited and late. The evidence from ossuaries is particularly illuminating here. These were the small, rectangular, (mainly lime)stone chests, used for secondary burial,43 that enjoyed great popularity with wealthy Jewish families from the middle of Herod’s reign (c.– ) down to the fall of Jerusalem in  .44 Often decorated with geometric designs and inscribed with the name of the deceased,45 they provide us with a rich supply of data for aristocratic naming practices. Of the one hundred and forty-seven names found in the largest assemblage of ossuary evidence, L. V. Rahmani’s catalogue,46 only seven are Latin.47 Other sources for the same period tell the same story—in the writings of Josephus, hardly any Jews bear Latin names apart from some of the later Herods (e.g. Agrippa I’s daughter, Drusilla, named after the sister of the emperor Gaius), a handful of pro-Roman aristocrats in the city of Tiberias (e.g. Julius Capella), and the sons born to Josephus himself after his exile to Rome (e.g. Agrippa);48 in the Gospels and the Acts of the Apostles we do not meet a single Judaean Jew 42 For the overwhelmingly Greek character of Herodian prosopography, especially of Herod the Great’s children, see Kokkinos : Appendix . 43 This involved placing in a special container, in a special way (Hachlili and Killebrew : , fig. ), the disarticulated bones of corpses from which the flesh had been allowed to decay. Previously, in their primary burial, these bodies had lain in coffins or on shelves in the family tomb. The reason for the sudden and short-lived use of ossuaries is much debated. On this, more below in the section on burial practices. 44 Hachlili and Killebrew : –; Rahmani b: . 45 Hachlili and Killebrew : –; Rahmani b: –. 46 Rahmani a. 47 Rahmani a: nos. , , , , and  (both instances of Gaius), , , . 48 For these and other Jews with Roman names, see L. H. Feldman’s index to the Loeb edition of Josephus. Confirmatory evidence for increased Herodian use of Latin names in the mid-st cent.  has recently come to light through excavations at Jatt, an area that in Roman times was part of the territory of Caesarea Maritima. See Porat, Yannai, and Kasher . Of particular note is the occurrence there of the names Tiberius, Marcus, and Paulus and several more instances of Agrippa.

Margaret Williams
with a Latin name. As to the names which were popular, on this point our sources are unanimous: they were largely those borne by the Maccabees and their royal descendants—for men, Judah, Simon, John, and Matthew and, for women, Mariamme/Miriam and the various forms of Salome(zion) (0001 Peace over Zion).49 For the period after   we have much less evidence— ossuaries ceased to be as fashionable as they had been and narrative sources dry up. From documents found in the Judaean Desert, however, we can see that Latin names had at last begun to trickle down to the non-elite. The Babatha archive, for instance, provides two good examples of this process— Germanus, son of Judah, a one-time clerk with the Roman army in Judaea (P. Yadin – and ) and a certain Judah Cimber, husband of Babatha’s stepdaughter, Shelamzion (P. Yadin ), whom we shall be meeting again below in connection with marriage and divorce. Such instances, however, were rare, the majority of Jewish families continuing to opt for the patriotic favourites listed above. Education and the Arrangement of Marriages Should children survive infancy, the prime duties of the father (food and clothing apart)50 were their education and the arrangement of their marriages. Education, as prescribed by the Torah, involved only sons (Deuteronomy : ). It was a father’s duty to teach them the statutes and ordinances, that is, the Law and commandments.51 In the Hellenistic period the practice had grown up in some aristocratic families of supplementing this with Greek paideia. Greek had become the language of government and international diplomacy. So, for the politically ambitious, a sound knowledge of Greek culture was essential. Hence the appearance in Ptolemaic Judaea of private Greek tutors52 and in 49 On this, the most striking, feature of the Palestinian Jewish onomasticon in the st and early nd cents., see Ilan : ; Williams : –. 50 These are stipulated in several of the surviving marriage contracts, e.g. P. Yadin  and . 51 The familial and son-centred character of Jewish education in the GraecoRoman period is well brought out by Archer : –. 52 For their use by Joseph the Tobiad in the education of his sons, see Josephus, Antiquities . .
The Jewish Family in Judaea

Seleucid Jerusalem of the gymnasium and palaestra, so abominated by ‘fundamentalist’ Jews.53 The establishment of Roman hegemony over Judaea in   made very little difference to this area of family life. Greek remained the language of government and diplomacy in Judaea, as in the rest of the Roman east. So the acquisition of Latin and knowledge of Roman culture did not become pressing requirements for Judaean Jews. Apart from the Herodian princes, who from the s  were routinely sent to Rome to complete their education54 and presumably did include a knowledge of Latin among their accomplishments, most aristocratic Jewish youths appear to have studied only the Torah and ‘Greek Wisdom’ in the course of their higher education. The amount of time devoted to each depended upon the proclivities of the families concerned: while in the palace ‘schools’ of the Herods the emphasis was mainly on the Greek,55 in priestly families, such as that of Josephus, the opposite was the case. The formal education of Josephus and his brother, conducted privately within the family home, appears to have consisted largely of Torah study.56 If Josephus is to be believed, it was only after his move to Rome that he started on his intensive study of Greek literature and acquisition of basic Latin.57 The minimal use of Latin by the elite is confirmed by the ossuary inscriptions—of the two hundred and thirty inscribed ossuaries appearing in Rahmani’s collection, only two bear texts written in the Latin script.58 Of the education of the non-elite little is known—the degree to which the system of
53 Despite being so detested, these pre-Maccabean institutions were probably never abolished. So H. Thackeray in the Loeb edition of Josephus, Jewish War .  and Gruen : . 54 For Herod’s three Hasmonaean sons, see Josephus, Jewish War . ; Antiquities .  and . ; for Archelaus, Antipas, and Philip, see Jewish War . ; Antiquities . –; for Agrippa I, see Antiquities .  and ; for Agrippa II, Antiquities .  and . . 55 Josephus, Jewish War . ; Antiquities .  and . On the Greek paideia/sophia of Agrippa II, see Josephus, Vita  and Contra Apionem . . 56 Josephus, Vita . For discussion, see Rajak : –. 57 For the suspicion in which Judaean Jews held Greek paideia, see Rajak : . 58 No.  () and no.  (). Given the rarity of the use of Latin in Judaea, Rahmani suggests a Roman origin for both these Jews.

Margaret Williams
primary education attributed to Joshua ben Gamala (first century ) was operative in Judaea is disputed.59 In the area of marriage arrangement, too, the establishment of Roman hegemony appears to have had no major impact. One reason for this lack of change will have been the prescriptions of the Torah about whom Jews could marry. Intermarriage with all Gentiles, for instance, though not explicitly banned in the Pentateuch, by implication was divinely forbidden.60 If anyone was likely to ignore the Torah on this point, it would be the Herods, given their own ambitions abroad and the desire of the Romans to integrate them fully into their network of eastern ‘client’ dynasties (Suetonius, Augustus ). However, as we saw above, even the Herods were careful, when negotiating exogamous unions, to stipulate that the prospective bridegroom be circumcised—that is, become a Jew. Only one member of the family is attested flouting the law—Drusilla, daughter of the late Agrippa I, who saw marriage with the then Roman governor of Judaea, the uncircumcised Antonius Felix, as the best way out of her personal problems ( Josephus, Antiquities . –). Josephus’ heavy criticism of her ‘transgression of the ancestral laws’ is indicative of the disapproval with which intermarriage was viewed and helps explain why we do not hear more of it. Although many of the documents found in the Judaean Desert reveal easy relations between the Jews, their Gentile neighbours, and the occupying Romans, they contain no reference to intermarriage.61 Outside the Herodian dynasty, it has yet to be attested in Judaea in the early Roman period. But there is another reason for the absence of change in the marriage arrangements of the Jews during the early Roman period: their nuptial practices will already have been brought broadly into line with those of the Romans through the ‘disturbance’ to their marital customs in the Persian and Hellenistic periods, mentioned by Hecataeus in the passage quoted at the 59 For a cautious estimate see Rajak : ; for an uncritical acceptance of the rabbinic evidence, see S. Safrai b: –. 60 See Exod. : –; Deut. : – for the divine ban on marriage with specified neighbouring peoples to prevent the Israelites falling into idolatrous ways. After the return from Babylon, Jewish leaders, for reasons of purity, extended the ban to all Gentiles. See Archer : –. 61 Goodman : .
The Jewish Family in Judaea

start of this survey. Although Hecataeus does not specify what that ‘disturbance’ was, it can be deduced fairly easily by comparing the marriage arrangements described in the narrative sections of the Bible with those attested in post-biblical literary and documentary sources. While in the Bible marriages are established through the payment by the suitor or his father to the prospective bride’s father of a bride-price and the dowry system does not appear at all,62 in post-biblical sources the bride-price is unmentioned—indeed, the word for it, mohar, drops out of use63—and it is the dowry that comes to the fore. Whether this change came about as a result of oriental influence in the Persian period64 or from western influence in the Hellenistic period65 is a matter of dispute which need not concern us here. It is sufficient to note that by the early Roman period the dowry system had become predominant. Thus, the numerous marriages arranged by the Herods for the women of their family were all dowry-based.66 Essentially dotal also were all those that figure in the marriage contracts, whether Aramaic or Greek, found in the Judaean Desert. In the five Greek documents, this is made obvious by the use of such words as proix, prosphora, and pherne, standard Greek terms for dowry.67 Although in the three Aramaic, ketubba-type contracts, precise words for dowry are not used,68 the arrangements there described are to all intents and purposes dotal, in that the only property that actually moves on the establishment of a marriage is the bride’s contribution.69 Given that Roman marriage was also dowry-based, one would not expect to see any fundamental change in this area of Jewish 62 Wasserstein : , citing Gen. .  and Exod. . ; Archer : –. 63 Cotton : . 64 Wasserstein :  n. . 65 DJD .  (comm. on P. Mur. ). 66 See, for instance, Josephus, Jewish War .  and ; Antiquities .  and . . 67 P. Mur.  and ; P. Yadin  and ; XHev./Se. Gr.  0001 DJD , no. . . 68 In P. Mur. , the expression ‘the money of your ketubba’ occurs instead. Translated there by the editors as ‘l’argent de ta dot’, the same expression is also to be found in P. Mur.  (restored) and P. Yadin . 69 So Cotton (: ). In P. Yadin , for example, Babatha’s ketubbamoney is treated in exactly the same way as the dowry in the Greek marriage contracts.

Margaret Williams
nuptial practice in the Roman period. But is no change at all to be detected? The evidence of the Babatha archive shows that Jews in the border area between the Roman provinces of Judaea and Arabia had not the slightest qualm about using Roman courts and adopting Roman legal practices when it suited them.70 Did their marriage arrangements remain immune from the influence that can be detected in other family matters where resort to the law was taken? Some scholars think not. A. Wasserstein, for instance, has seen Roman influence in the dotal arrangements of P. Yadin , the marriage contract between Judah Cimber, mentioned above, and Judah, the father of Shelamzion, Babatha’s stepdaughter. Drawn up in   by a Roman clerk in a Roman provincial court, this document is a fascinating mélange of diverse cultural elements—for example, Greek dotal terminology, Aramaic signatures, and Roman legal language.71 To the elements universally accepted as Roman, namely the opening dating formula and the closing stipulatio clause, Wasserstein would add another: the groom’s undertaking (lines –) to add to the bride’s prosphora, mutually agreed to be worth  denarii of silver, a further  denarii, the total of the two amounts together to be counted as her dowry ( panta eis logon proikos). Hitherto seen as ‘the most uniquely Jewish’ feature of the document,72 for Wasserstein it is ‘primarily reminiscent of donatio ante nuptias in dotem redacta’.73 Were his view to be accepted, then here at least we would have one piece of evidence for Roman influence in the area of Jewish marriage arrangement. However, the late date of that procedure—S. Treggiari, following P. Corbett, assigns it to the fifth–sixth centuries74—suggests anachronism on Wasserstein’s part. Better, therefore, with Katzoff (n.  above), to see the groom’s undertaking here as essentially Jewish after all. But if the documents yield little in the way of Roman influence on Jewish marriage arrangements, Josephus is not so uninformative. From the detailed information he provides about Herodian 70 As Babatha did in her property-dispute with her co-wife, Miriam (P. Yadin , discussed above) and in her quarrels with the guardians of her son. On this, see Cotton , especially –. 71 For a succinct description, see Feldman and Reinhold : . 72 So Katzoff in Lewis, Katzoff, and Greenfield : . 73 Wasserstein : –. 74 Treggiari : – and , citing Corbett :  ff.
The Jewish Family in Judaea

marriage arrangements over five generations, it seems fair to deduce that the Herods came to abandon polygamy as a result of their contact with Rome. Although this form of marriage was practised in the early days of the dynasty, most notably by Herod the Great,75 no Roman-educated Herodian prince (n.  above) is mentioned as having more than one wife—perhaps polygamy smacked too much of the desert and the east.76 Divorce Before we can assess Roman influence on the dissolution of marriages, we need to discuss briefly Roman and Jewish divorce procedure. With regard to Roman divorce procedure, it is sufficient to point out that it was basically simple and, in gender terms, equitable: either party to a marriage (or the paterfamilias of each) could initiate divorce proceedings and no bill of divorce was required.77 With the Jews, the situation was more complicated, largely on account of the fact that the Torah has remarkably little to say about divorce. It does not legislate on the subject78 and in the one brief passage devoted to the topic, Deuteronomy : –, the focus of the discussion is on the circumstances in which a divorced woman may or may not remarry the husband who has issued her with a written notice of divorce—in later rabbinical parlance, a get. Whether a wife may also initiate divorce proceedings is not addressed.79 Given the limited guidance of the Torah material on this important social issue, it is not surprising that divorce was both intensely debated80 and variously practised. Many Jews, particularly those inclined towards Pharisaism and rabbinism, took the line that only the husband could issue a bill 75 Jewish War .  and –. For the marriages of his bigamous eldest son, Antipater, see Jewish War .  (Loeb edn. with Thackeray’s note ad loc.) and Ilan : . 76 On the harems of the kings of Adiabene and Charax Spasini (mid-st cent. ), see Josephus, Antiquities .  and . Note also Lewis’s suggestion (: ) that the polygamy implied in P. Yadin  was the result of Nabataean influence. 77 78 Crook : ; Treggiari : –. Collins : . 79 On the narrow scope of the Deuteronomy passage, see Brooten , followed by Cotton and Qimron :  n. . 80 Shown most clearly by Mark : –.

Margaret Williams
of divorce and thus free his wife to remarry another Jew, a view succinctly set out by Josephus, himself a Pharisee, at Antiquities . : It is only the husband who is permitted by us to do this [i.e. issue a bill of divorce], and not even a divorced woman may marry again on her own initiative unless her former husband consents.81
The categorical nature of Josephus’ comments here, intended primarily to point up for his Graeco-Roman readers the main difference between Jewish practice and their own, should not be allowed to obscure the fact that not all Jews followed this Pharisaic/rabbinic line. Jewish papyri from Egypt point to a variety of divorce practices there,82 and Mark : , if taken at face value, suggests that even in Judaea there were Jews prepared to countenance the idea of women as well as men initiating divorce proceedings. That complicates the situation for us: where we find evidence of women taking such a step or being legally empowered to do so, are we to see in it native custom83 or Roman influence? To decide, we will consider carefully each case in turn, starting with the most contentious, XH . ev./Se. Ar. , a document most easily accessed at DJD , no. .84 Known since  but receiving preliminary publication only in 85 and final publication two years after that,86 this controversial Aramaic document has been variously interpreted. While J. T. Milik, who first made its existence known, thought it was a bill of divorce issued by a woman,87 others subsequently categor81 For a documentary example of this, see P. Mur. —the so-called Masada get, probably to be dated to / . So Goodblatt . 82 Note, for instance, CPJ ii, no.  ( ), in which a regular, Greekstyle divorce is enacted. For Jewish use of Hellenistic law in Egypt, see V. A. Tcherikover in CPJ i. –. 83 For arguments that there were two Jewish traditions of divorce, the rabbinic and another, traceable from the Elephantine papyri onwards, which accorded women the same status in the matter as men, see most recently Ilan : –. For earlier enunciations of this view, see the authorities cited by Cotton and Qimron : –. 84 XHev./Se. Ar. or Gr.: documents from the Seiyal collection, written in . either Aramaic or Greek; believed to have come from one of the Nahal H . ever caves in the Judaean desert. 85 86 Details in Cotton and Qimron : . As DJD , no. . 87 DJD ,  (comm. on P. Mur. ).
The Jewish Family in Judaea

ized it as (i) a receipt by a woman of a get issued by a man,88 (ii) a receipt by a woman of a ketubba,89 and (iii) a possible (?) waiver of claims by a wife.90 The position adopted here is that recently argued for by H. Cotton and E. Qimron (), who translate the disputed central section of the text thus: I, Shelamzion, daughter of Yehosef Qb˘sn from Ein-gedi, have no claim against you, Eleazar, son of H . ananiah, who previously were my husband and who had (have) a deed of abandoning and expulsion from me. You, Eleazar, owe me nothing concerning anything whatsoever. And I accept as binding on me, I, Shelamzion, daughter of Yehosef, all (the obligations) written above.
For Cotton and Qimron (: ), the document is thus a waiver of claims, which mentions ‘en passant as a background to the wife’s renunciation’ the writ of divorce she had previously issued. Assuming that they are right and we therefore do have evidence here of a wife divorcing a husband, is Shelamzion following Roman practice or some ancient, non-Pharisaic, non-rabbinical Jewish tradition? Total certainty in the matter is not possible. However, the overall Jewish character of the document makes Roman influence extremely unlikely. In addition to the fact that it is written in Aramaic, its method of dating could not be more emphatically Jewish: On the twentieth of Sivan, year three of Israel’s freedom, in the name of Simon bar Kosibah, the Nasi of Israel (0001 / ). With T. Ilan (: ), it seems fair to surmise that the document was drawn up ‘in a full Jewish court of law’. Less problematic is the case of Shelamzion, wife of Judah Cimber, whose marriage contract, P. Yadin , was discussed above. Here we need mention only the part relevant to divorce, namely, the clause immediately following those dealing with the dotal arrangements: Judah called Cimber shall redeem this contract for his wife Shelamzion, whenever she may demand it of him, in silver secured in due form, at his own expense interposing no objection. (lines –)
88 89 90
J. C. Greenfield, ‘in his lectures’. Ilan : . Yardeni and Greenfield . Yardeni’s revised view at DJD , no. .

Margaret Williams
The implication of this would appear to be clear: Shelamzion may, at any time she likes, effectively terminate the marriage by requiring that her dowry be paid back.91 But does this reflect Jewish tradition or Roman influence? Given that the contract was drawn up by a Roman official and its Greek and Roman characteristics far outnumber the Jewish ones, Roman influence, though not absolutely certain, seems quite probable. Finally, we come to the Josephan evidence for divorces instigated by women in the Herodian family. The earliest relates to Herod the Great’s sister, Salome, who c./  sent her husband Costobar a bill of divorce and thus terminated their marriage (Antiquities . –). Everything here points to Roman influence. Besides the general Romanizing tendencies of the Herods, already in evidence at this relatively early date, we should note Salome’s personal friendship with Augustus’ wife, Livia. This, according to Antiquities . , was sufficiently close for Salome to request and receive advice from her about whom she should marry next. Important also in assessing the Salome–Costobar divorce is the language used by Josephus to describe Salome’s behaviour. She chose, so he tells us, not to follow her native law (ton engene nomon) but, acting on her own authority (ep/ap exousias), repudiated her marriage, telling her brother Herod that she had separated from her husband out of loyalty to Herod himself. This is the comportment of a legally independent Roman matron, which is precisely what (Iulia) Salome was. Salome’s divorce of Costobar, though the first case recorded by Josephus of a Herodian female taking the initiative in ending her marriage, is not the last. The contempt with which Herodian women came to regard ‘the native law (on marriage)’ reached its peak with Agrippa I’s three daughters: while Drusilla, as mentioned above, married an uncircumcised Gentile, her two sisters, Mariamme and Berenice, abandoned their spouses, Iulius Archelaus and M. Antonius Polemo, and embarked on new marriages with a casualness that Jews such as Josephus assumed was a male prerogative.92 Behaviour of this type was very common in Wasserstein , with supporting evidence from other Greek papyri. Josephus, Antiquities . – (Drusilla);  (Berenice);  (Mariamme). For an earlier example of similarly ‘lawless’ behaviour in the matter of divorce by their aunt, Herodias, see Josephus, Antiquities . . 91 92
The Jewish Family in Judaea

the aristocratic Graeco-Roman circles in which the Herods had been moving with increasing frequency and confidence since the reign of Augustus. It will surely have been the major influence on (Iulia) Mariamme and (Iulia) Berenice here. Funerary Practices Finally a few words on changes in Jewish funerary practice in the Roman period. Hecataeus, as we saw at the beginning, observed that as a result of mingling with foreigners, Jewish burial practices had undergone some modification in the Persian and Hellenistic periods. As with marriage arrangements, that change had been made possible by the lack of prescription in the Torah, which barely concerns itself with funerary matters. Certain practices, however, even though not prescribed by the Law, had become conventional, as the narrative sections of the Bible (OT) show. Corpses were buried, not burned; inhumation normally took place in a family tomb, whence arose the common biblical metaphors for death and burial—being gathered unto one’s people93 and sleeping with one’s fathers.94 When burial space in the family tomb, often a cave, ran out, old skeletons were exhumed and the disarticulated bones consigned to a charnel pit or pile.95 Individuated tombs bearing personalized epitaphs, such as we regularly meet in the Graeco-Roman world, are rarely found.96 During the Hellenistic period, however, a number of changes are to be seen, at least in the area of elite burial, to which most of our evidence relates. The most striking of these is the appearance in the second century  of monumental family tombs, such as that of the Maccabees at Modin ( Macc. : –) and the Tomb of Jason in Jerusalem.97 Also to be noted is the more modest practice, now seen in Judaea for the first time, of the burial within family tombs of individuals in wooden coffins—a borrowing, according to Hachlili and Killebrew (: ), from the Greek world. See, for instance, Gen. :  and  (burial of Jacob). Of the numerous examples in Kings and Chronicles, note  Kings :  (David) and :  (Solomon). 95 Hachlili and Killebrew : ; Rahmani b:  and . 96 Hachlili and Killebrew : —only  in the whole First Temple period. 97 Rahmani b:  and . 93 94

Margaret Williams
Such was the situation at the end of the Hellenistic period. What changes are to be seen in Roman Judaea? Of the burial of the non-elite, there is little to say—for many of them, a shallow, unmarked grave was probably still as much as they could hope for.98 In the burials of the elite, certain new practices are to be observed. A minor innovation was the placing of a coin (Charon’s fare) in the mouth of the corpse, examples of which have been found both at Jericho and in the tomb of the Caiaphas family at Jerusalem.99 Although Greek in origin, the custom had been adopted by the Romans and through their influence disseminated widely through the empire, including the province of Britain.100 We may suspect such influence in these first-century  Judaean examples of the practice, too. A far more substantial change in which Roman influence has also been discerned at least by some scholars is the fashion for using stone ossuaries, many of them ornately carved, that started with a puzzling suddenness in Jerusalem around the middle of Herod’s reign (c.– ). The traditional explanation for this—that it reflects a change in views about the afterlife101—is rather implausible, not least because there is no evidence for any abrupt change in Jewish eschatological beliefs at that time. Far more convincing, in my opinion, are the hypotheses recently advanced by S. Fine and Y. Tsafrir. While the former connects the new funerary fashion closely with the rapid development of the stone-carving industry under Herod,102 the latter sees in the use of ossuaries the importation and adaptation of a Roman funerary custom.103 Italian craftsmen are a conspicuous element 98 See Zissu :  and  for details of some  of these graves from  sites around Jerusalem. 99 Hachlili and Killebrew :  and —Jericho; Greenhut : — Jerusalem. 100 Toynbee : , , , and . 101 i.e. a more widespread belief in bodily resurrection. For a brief exposition of this popular theory, see Rahmani b: –. For pertinent criticisms of this ‘nineteenth century scholarly trope’, see Fine : –. 102 Fine (: ) sees great significance in the fact that the construction of the Temple, Herod’s biggest building project, began around   (i.e. just before ossuaries became fashionable). 103 Y. Tsafrir in ‘Herodian Building Projects and the Romanisation of Judaea’, a paper delivered at ‘The World of the Herods and of Nabataeans’ conference at the British Museum, – April, .
The Jewish Family in Judaea

in Herod’s workforce from   onwards.104 Since carved stone chests for funerary purposes had long been common in Italy, the idea for their use in Jewish funerary rituals could well have come from those workers. Even if Roman influence remains conjectural here, it surely cannot be doubted in the widespread occurrence from Herod’s reign onwards of personalized epitaphs. Before the Roman period, epitaphs of any kind were exceedingly rare. From around the middle of Herod’s reign, coinciding with the new fashion of decorated ossuaries, it became quite common in elite families not only to inscribe the name of the deceased on their final resting place but to give some indication of their status in life also—a very Roman practice. While occasionally the individual’s standing in the wider community was given prominence,105 mostly status was expressed solely in familial terms—for example, son of X, daughter of Y, wife of Z. The recently published ‘James, son of Joseph, brother of Jesus’ ossuary is a good example of this.106 Conclusion I started this chapter by quoting Hecataeus on the changes that Jewish family customs had undergone in the Persian and Hellenistic periods. I suggested that change to the Jewish family in Judaea in the Roman period was no less likely, and formulated a series of questions to enable us to identify and evaluate that change. It is now time to provide specific answers to those questions. To the question, Is change to be observed, particularly after  ?, the answer clearly is yes. Did it intensify over time? For families other than the Herods, we have insufficient data to answer this question. As far as the Herodian family itself is concerned, the answer has to be in the affirmative. The increasingly Roman character of Herod’s palaces is testimony to this, as is the increasingly careless attitude of his descendants towards the 104 For their probable employment at Jericho, for instance, see the authorities cited in n.  above. 105 e.g. teacher (CIJ  and –), scribe (CIJ ), priest (CIJ ), builder of the sanctuary—Naveh (: –); freedman of Queen Agrippina— Hachlili (:  and –). 106 Lemaire . Other examples of the practice are provided by the epitaphs of the Goliath family. See Hachlili : –.

Margaret Williams
ancestral laws of the Jews107 and their growing preference for conducting their lives in an overtly Roman way.108 Were the changes we have seen superficial or fundamental? Undoubtedly the former. For the most part change occurred only where the Torah was unprescriptive, as in the areas of marriage arrangements and burial. Where the Torah issued precise injunctions, very little change is to be observed. Circumcision, a divine command, continued to be regarded by virtually everyone as the sine qua non of being a genuine Jew ( Josephus, Antiquities . ) and breaches of the Second Commandment are exceedingly rare— decoration, whether of tombs, ossuaries, houses or palaces, remained for the most part non-figural. And what of the social diffusion of change? The evidence presented above shows that it was largely restricted to the elite. To some extent, the limited nature of Romanization is consequent on wealth—only the very rich could afford such expensive status symbols as mosaic floors, stuccoed and frescoed walls, and Roman-style baths. But choice also played a large part, as the onomastic evidence, important for relating to all social strata, makes abundantly clear. It would not have cost non-elite Jews anything financially to give their children Roman names. That they chose not to do so but continued thoughout the period from Pompey to Hadrian to display a very strong preference for the Hebrew names used by the Maccabees and the Hasmonaean dynasty speaks volumes not just about their cultural identity but about their political attitudes too and helps explain the intensity of their opposition to Rome in both  and  . 107 In addition to the evidence discussed above, note also Josephus, Antiquities .  and Mark : – (trangression of the levirate laws by two of Herod’s sons). 108 Besides the topics discussed in the text, note also the lavish celebration of birthdays by the Herods, for which see Matt. : –, Mark : –, and Josephus, Antiquities . . For discussion, see Hoehner : –. For Jewish disapproval of celebrating birthdays, regarded as a pagan custom and contrary to the Law, see Josephus, Contra Apionem .  and S. Safrai a: . For the Roman celebration of birthdays, see Balsdon : –.
 Family Relations in Roman Lusitania: Social Change in a Roman Province? Jonathan Edmondson
Introduction I the later second century  the handsome tombstone shown in Fig. . was set up in a suburban necropolis of Augusta Emerita (modern Mérida), Roman colony and capital of the province of Lusitania. Asellia Claudia, who describes herself as a ‘most devoted daughter’ (filia piissima), set it up for her mother, Asellia Hygia, and father, M. Publicius Felix.1 Typical of a number at this Roman colony, the tombstone is classically Roman in style, with fine portraits of the deceased couple shown within an aedicula, which has Cupids adorning its sides.2 Reminiscent of some of the best funerary monuments from the city of Rome, it provides a snapshot of a simple nuclear family comprising a mother, father, and daughter, caught at a moment when the surviving daughter sought to present a striking image of this ‘most beloved’ I am grateful to the Director, Prof. Dr J. M. Álvarez Martínez, and all the staff of the Museo Nacional de Arte Romano, Mérida, to Dr A. U. Stylow and Dra. H. Gimeno Pascual of the Centro CIL II, Alcalá de Henares, and to Prof. J. d’Encarnação, Director, Instituto de Arqueologia, Universidade de Coimbra, for assisting my work on the inscriptions of Lusitania, to audiences at the Roman Family IV conference at McMaster University, Hamilton, at the Centre Ausonius, Université Michel de Montaigne—Bordeaux III, and at York University for their comments on earlier versions of this chapter, and to the Social Sciences and Humanities Research Council of Canada for its continued support of my research on Lusitania. 1 HEp. .  0001 Edmondson, Nogales Basarrate, and Trillmich : –, no.  with plates A–C and figs. .–: Asellia Hygia an(norum) XXXIX et M. Publicius Felix amantissimi h.s.s.s.v.t.l. [ex testame]nto matri et patri et Asell(ia) [Cl]audia fil(ia) piiss(ima). 2 For a full study, see Edmondson, Nogales Basarrate, and Trillmich .

Jonathan Edmondson
. .. Funerary monument with portraits of Asellia Hygia and M. Publicius Felix, from Emerita
married couple (amantissimi) to her fellow Emeritenses. In perpetuating their memory, she was fulfilling her duty as their daughter and heir, as she had been instructed to do in the will (ex testamento) of one of her parents. At the urban centre of the Igaeditani (modern Idanha-a-Velha, distr. Castelo Branco), originally a peregrine civitas stipendiaria promoted to municipal status under the Flavians,  km. northwest of Emerita, a rather different act of funerary commemoration took place in the later first century . Claudia Tangina, whose bicultural name with its Roman nomen (gentilicium) and distinctly Lusitanian cognomen suggests that her father had been enfranchised during the reign of Claudius ahead of the general grant of the Latin rights of citizenship by Vespasian to the Hispanic provinces in  –, designed an epitaph on a well-cut granite block with a moulded edge to be inserted, we may pre-
Family Relations in Roman Lusitania

sume, into the façade of a tomb (Fig. .).3 On it she commemorated her paternal grandparents, Lubaecus Antae lib. and Binarea Triti f., her paternal uncle, Boutius Lubaeci f., and his wife, Cilia Caenonis f. In the final clause of the epitaph, she had it recorded that she was setting up the monument ‘for her own people’: suis f(aciendum) c(uravit).4 Her notion of sui, those she considered related to her, was rather broader, it appears, than that of Asellia Claudia at Emerita. The difference may have something to do with the different type of monument used in each case: a block to be set into a mausoleum rather than a free-standing aedicula, several of which could have been set up for various members of the same family in the same funerary enclosure or within the same mausoleum.5 But
. . Funerary monument of Lubaecus Antae lib., Binarea Triti f., Boutius Lubaeci f., and Cilia Caenonis f., set up by Claudia Tangina, from the civitas capital of the Igaeditani 3 For Tanginus as a distinctly Lusitanian name, see Untermann : –, map ; Grupo Mérida : map ‘Tancinus, -a’. For Vespasian’s grant of the ius Latii, see Plin. HN . : universae Hispaniae Vespasianus imperator Augustus iactatum procellis rei publicae Latium tribuit. For discussion, Richardson : –; Le Roux . For this type of funerary monument, cf. Almeida : figs. , , , , , , –, , –, –, –, , , –, ; for discussion, Encarnação : –. For viritim grants of Roman citizenship in Lusitania under Claudius, cf. CIL .  0001 IRCP , rev. Encarnação : . 4 HAE  0001 ILER  0001 Almeida : no. : Lubaeco Antae lib. avo Binareae Triti f. avitae Boutio Lubaeci f. paterno Ciliae Caenonis f. amitae Claudia Tangina suis f.c. 5 For a brief discussion of how questions of genre can affect the nature of the evidence that epitaphs provide for Roman family history, see Dixon b: .

Jonathan Edmondson
we do appear to be dealing with something more significant than just the genre conventions of our evidence. For at Emerita of more than  acts of funerary commemoration where the relationship between the deceased and the commemorator is expressed, only four involved grandparents and just one an uncle or an aunt.6 At the civitas capital of the Igaeditani, on the other hand, of the  cases where the relationship is clear, four others involved grandparents (three of whom were maternal grandparents), one a maternal aunt, and another a maternal uncle.7 The participation of more extended kin in funerary commemoration, especially on the maternal side, does seem to have been more marked here than at Emerita. In this chapter I want to argue that there were subtle differences in the ways in which ‘the family’ was conceived in different parts of a Roman province. Family relationships were not monolithic across the entire province, let alone across the entire Iberian peninsula. A combination of quantitative and qualitative approaches is needed to bring out this regional cultural and social variation, which, I would argue, probably existed in many other provinces of the Roman empire as well. Ideally one would like to conduct such an inquiry diachronically, to see how the shape of family relations developed as Lusitania became more integrated into the Roman empire after its creation as a separate province when Augustus split Hispania Ulterior into Hispania Ulterior Baetica and Hispania Ulterior Lusitania (sometimes called, more fully, Hispania Ulterior Lusitania et Vettonia), perhaps in  .8 But the main type of evidence, private funerary commemorations, makes this difficult, as they are never explicitly dated. Broadly speaking, the large majority of surviving Lusitanian epitaphs range in date from the early first century  to the midthird century , but much more detailed epigraphic work on the 6 Emerita: grandparents: CMBad. ; EE .  0001 ILER ; EE .  0001 CMBad. ; AE , ; a maternal aunt (matertera): AE ,  0001 Edmondson, Nogales Basarrate, and Trillmich : –, no. . 7 Grandparents: CIL .  (maternal grandmother); HAE  0001 AE ,  (maternal grandfather and maternal grandmother); HAE  0001 ILER  (grandson, unclear if paternal or maternal); maternal uncle (avunculus): AE ,  0001 HAE ; maternal aunt (matertera): HAE . 8 For the division into Baetica and Lusitania, see Plin. HN . ; for discussion regarding the date, see Étienne : –; Le Roux : –; Richardson : –.
Family Relations in Roman Lusitania

chronology of the epitaphs from the different regions of the province is needed before a diachronic analysis is possible. In the last twenty years, Roman social historians have placed the family very firmly at the centre of our scholarly attention. Although a variety of approaches and evidence has been deployed, funerary inscriptions such as the ones just discussed have featured prominently in a number of the most important studies. In particular, it was epitaphs from Italy and the western provinces that provided Richard Saller and Brent Shaw with their data for an important series of interrelated articles on family relations in the Roman west published since .9 For Saller and Shaw, acts of funerary commemoration are revealing because they ‘flowed from previous social relationships with the deceased’.10 In particular, it was usually the heirs of the deceased who felt the obligation to set up the tombstone or the family member or members ‘thought to be tied by the strongest bond of duty’.11 As a result, epitaphs do provide some useful orientation as to how the various commemorators wished to advertise their family connections within their respective communities. The rituals of burying and commemorating the dead provided families with one of the few opportunities they had to act in public as a family unit and make a powerful statement to their fellow-citizens about the emotional ties that bound the unit together. Although their method has elicited some criticism, it still retains validity if we remember that its aim was to elucidate family relations rather than family structures.12 Some of Saller and Shaw’s critics have misrepresented the goal of their study. They were clear that their figures did not allow them to reconstruct precise family structures, still less the shape and size of households; for them funerary commemorations reflected in a necessarily approximate manner the most important emotional ties that lay at the heart of family relations. Saller and Shaw’s main conclusion was that in civilian families in the Roman west funerary commemorations were made in the overwhelming majority of 9 Saller and Shaw ; Shaw and Saller ; Saller , a, ; Shaw , a, . 10 Saller : –. 11 Saller and Shaw : . See also Meyer : –; Saller : –. 12 For criticism, see Krause ; Martin ; Corbier : ; for a recent defence, see Rawson b.

Jonathan Edmondson
cases by members of the nuclear family: most commonly, wives commemorating husbands or vice versa, parents setting up memorials for children, or children for one or both their parents. These were the emotional bonds that for the most part lay at the heart of family relations and family obligations. More recent studies have provided more nuance by emphasizing that while conjugal ties and bonds between parents and children were important, the family was a flexible unit into which in-laws, stepchildren, and half-brothers and sisters sometimes entered as the result of marriage, adoption, or remarriage following a death or divorce.13 Slaves and dependent freed-slaves were important members of the family too, both conceptually and in actual social practice.14 Family relations evolved over the course of the life-cycle, expanding as a result of the birth of new offspring or through the need to care for an aged grandparent, contracting after death or divorce or once children departed to form conjugal units of their own. Saller and Shaw also looked in more detail at patterns of commemoration within the nuclear family category. Here they highlighted some interesting local variations across the western empire. Particularly striking was the marked prominence of women (both wives and mothers) as commemorators in Roman Spain. This, they suggested, might hint at some vestigial matrilineality in kinship organization and transmission of property.15 It was this ‘Spanish peculiarity’ that first attracted my attention to Saller and Shaw’s method, but the fact that they treated the three Spanish provinciae (Baetica, Tarraconensis, and Lusitania) as a single unit raised suspicions. For one of the most important features of Roman Spain was the high degree of local diversity discernible in many aspects of its society, economy, and culture.16 And while Saller and Shaw did note some regional differences in passing, their main aim was to emphasize the general patterns that they encountered across the Latin-speaking west, since they were keen to establish ‘Roman family norms’ broadly conceived. There is now a growing awareness that we need to look more 13 Bradley : chs. –; Corbier a, c, b; Dixon : ch. ; Saller : . 14 15 Bradley : chs. –; Saller b. Saller and Shaw : –. 16 Étienne et al. ; Keay  (with a full bibliography).
Family Relations in Roman Lusitania

closely to see if there were significant differences in the ways in which the family was conceived and family relations operated across the various regions of Italy and the Roman empire.17 A few regional studies are beginning to emerge, and this volume, it is hoped, will represent an important step forward in this direction.18 This increasing body of work on family relations in the Roman provinces is a timely development. Studies of Roman provinces have traditionally focused on the cultural and social changes brought about by their incorporation within the political system of the Roman empire and recently there has been much stimulating debate concerning the difficult concept of ‘Romanization’ in various provincial contexts. The development of Roman-style public and private architecture in town and country, the adoption of Roman consumer goods, the nature of personal names, the spread of epigraphic culture, and changes in the local religious universe have all received attention, as have the complex legal issues of personal and civic status and the spread of citizenship.19 The family, however, has remained largely on the sidelines of this discussion. But since the family represented the basic unit of social organization and social reproduction in Rome, it is a fundamental question to investigate to what extent the ‘Roman’ model of family was adopted in various provincial contexts and whether any alternative, perhaps indigenous, forms of social organization persisted or indeed developed in a province under Roman rule. It is the aim, therefore, of this chapter to investigate funerary commemoration and family relations in just one part of the Iberian peninsula, the Roman province of Lusitania, to see whether any variations can be discerned within a single province. A more regionally based study should add nuance to the general conclusions reached by Saller and Shaw. Furthermore, their 17 Important regional distinctions are made in Shaw ; for the need for more regional studies, see Dixon : –; Bradley : . 18 Martin  (on Asia Minor); Gallivan and Wilkins  (on Italy); Cherry  (on Algeria); Bradley a (on North Africa, as represented in Apuleius’ Apologia); Hope : esp. chs. – (comparing Aquileia, Mainz, and Nîmes); cf. Bradley b. 19 For example, Blagg and Millett ; Alcock ; Metzler et al. ; Laurence and Berry ; Woolf ; Keay and Terrenato .

Jonathan Edmondson
computations were based just on CIL  and its supplement, published in  and  respectively. The number of inscriptions now available has increased at least fivefold, thanks in no small part to the project to produce a completely new edition of CIL , and so by using all known inscriptions, one can test whether the balance has been altered to any significant degree by the large number of new inscriptions now available.20 But more importantly, this larger quantity of data means that it is now becoming feasible to analyse family relations in more restricted geographical zones, since larger and hence more meaningful samples can be assembled for individual towns or regions of the province than was possible when just CIL  and its supplement were used. While this study was in preparation, two articles have appeared that examine family relations in Lusitania using Saller and Shaw’s methodology. A very detailed study of women in the conventus Scallabitanus makes a number of useful remarks about family relations in the north-western third of Lusitania. Although some comparisons are made with the southern part of modern Portugal, the Spanish part of the province is excluded entirely from the discussion.21 In a briefer survey of the ‘Lusitano-Roman family’, Curchin has analysed the funerary commemorations from Lusitania according to the three juridical conventus into which the Roman province was divided: the conventus Pacensis, Scallabitanus, and Emeritensis. His analysis clearly reveals that the province as a whole does not differ substantially from the picture drawn for the whole of the Iberian peninsula by Saller and Shaw. However, he chose not to concentrate on regional variations within the province, restricting himself to a mere twelve lines on this in his concluding remarks.22 This is the subject that I want to treat in much more detail in the rest of this chapter.23 Patterns of Funerary Commemoration in Lusitania As a starting-point for this study, I analysed all funerary inscriptions from seven different towns or regions in Lusitania (see Map .). 20 On the project to prepare a new edition of CIL , see Stylow ; 21 Fernandes – []. Edmondson . 22 Curchin a, with regional differences discussed at –. 23 This chapter will amplify a number of points made in my earlier study of family relations at the Roman colony of Emerita: Edmondson .
Family Relations in Roman Lusitania
 . The Roman Province of Lusitania


Jonathan Edmondson
. Augusta Emerita (modern Mérida, prov. Badajoz), a Roman colony and capital of the province; . the south-western third of the province, the conventus Pacensis, which contained one Roman colony, Pax Iulia (modern Beja), three native communities granted the Latin rights of citizenship under Augustus at Salacia (modern Alcácer do Sal, distr. Setúbal), Ebora (modern Évora) and Myrtilis (Mértola, distr. Beja), and at least another four communities promoted to municipal status in the Flavian period: Ossonoba (Faro) and Balsa (Tavira, distr. Faro) in the Algarve, Ammaia (São Salvador de Aramenha, distr. Portalegre), and Mirobriga Celtica (Santiago do Cacém, distr. Setúbal); . the civitas capital of the Igaeditani (modern Idanha-a-Velha, distr. Castelo Branco), in the north-central part of the province, a native community (civitas stipendiaria) promoted to Roman municipal status following the grant of the ius Latii in  – (its precise Roman name is as yet unattested); . Olisipo (modern Lisbon), already a municipium civium Romanorum in the Augustan period; . Conimbriga (modern Condeixa-a-Velha, distr. Coimbra), in the north-west of the province, another native community that became a municipium under the Flavians;  and . the modern Spanish provinces of Salamanca and Ávila in the north and east of Lusitania; these areas contained relatively few urban centres: no Roman colonies and only one certain municipium, Salmantica (modern Salamanca), and three possible ones: at Mirobriga (modern Ciudad Rodrigo, prov. Salamanca), Bletisa (modern Ledesma, prov. Salamanca), and Ávila (whose Roman name was perhaps Avela). All commemorations involving soldiers have been excluded, since the legal ban on soldiers’ marriage in force until the reign of Septimius Severus requires that they be treated separately.24 All told, these seven regions provided a sample of , funerary commemorations, which compares favourably with the , used by Saller and Shaw for their study of the whole of the Iberian peninsula. I have collated the data using precisely the same method as Saller and Shaw (see Table .). The samples from Conimbriga and 24 Saller and Shaw : –, –; for soldiers’ marriage, see recently Phang .
Family Relations in Roman Lusitania

from the modern province of Ávila are too small (N 0001  and  respectively) to provide the basis for any firm conclusions, and so they have been largely excluded from the discussion that follows. The sample from the modern province of Salamanca is sizeable enough (N 0001 ), but the fact that only  per cent of these epitaphs included the name of a commemorator renders the sample useless for this type of analysis. This leaves us with four sizeable samples: () Augusta Emerita;25 () the conventus Pacensis; () the civitas capital of the Igaeditani; and () Olisipo. In addition, I have provided a total for all seven areas of Lusitania sampled and—for comparative purposes—included Fernandes’ figures for the conventus Scallabitanus (which includes Olisipo) and Saller and Shaw’s figures for the Iberian peninsula as a whole. In Table . three columns of figures are given for each city or region. The first column gives the total number of acts of funerary commemoration of each specified type. The second expresses this figure as a percentage of the total number of commemorations from the city or region from which the relationship between the commemorator and the deceased can be extracted. The third column is restricted to those commemorations involving members of the nuclear family and expresses the number of each type of commemoration as a percentage of all commemorations made between members of the nuclear family. Thus from Emerita  commemorations set up by husbands to their wives have survived (Column ). These amount to  per cent of all the commemorations from Emerita where a relationship is expressed or can be securely inferred (Column ) or  per cent of all commemorations set up by members of the nuclear family (Column ). To put the results into a clearer perspective, I have prepared a comparative table (Table .) which draws together the statistics for the various broad types of commemorative relationship attested in each of the seven regions of Lusitania, as well as Fernandes’ figures for the conventus Scallabitanus and Saller and Shaw’s figures for Spain as a whole and for various other parts of the western empire.
25 For a more detailed study of family relations from Emerita, see Edmondson .
T .. Funerary commemoration in various regions of Lusitania Dedicator
Deceased
. Emerita N
%
          – – – – – – – –       –  –    – –   – – –          AMICI    Libertus/a Patron  Master   Slave –  Patron Libertus/a Slave Master — — Conservus/libertus Conservus/libertus –      ()    No dedicator  Se vivo sibi    Husband Wife Husband Wife   Parents Son Daughter Father Son Daughter Mother Son Daughter   Father Son Mother Daughter Father Mother   Brother Brother Sister Sister Brother Sister       
. conv. Pacensis
. civ. Igaedit.
(%)
N
(%)
N
() () () () () (–) (–) (–) (–) () () () () () () (–) () (–) () () ()
      –  – . –  – . –        –    –                        — —   — —     ()    
() () () () () () () () () () () () () () () () () () () () ()
 ()    ()    ()    ()    ()    – – (–)  (–) –   – – (–)  ()    ()   – ()   – – (–) – – ()   – – (–) –  ()   – – – () – ()   – () –  — – — —  ()    ()         — — — — — —  —     — — — — — —     ()          
  
%
  
%
. Olisipo (%)
N
. Conimbriga
%
(%)
N
           .  . 
(.) (.) () () (.) (.) () () () () () () () () ()
            – –   – –     — — — —       — —   — —             — —  — — — —    ()   
(.)  ()    — — —  — —  ()   .
%
(%)
() () () () (–) () (–) ()
()
() ()
  
Dedicator
Deceased
. Salamanca (prov.)
. Ávila (prov.)
. Lusitania (total)
N
N
N
%
(%)
N
%
(%)
N
%
(%)
   – – – – – –  – – – –  – – – –        –  –     
     –  –      –                 ()
() (.) () () () (–) (–) (–) () () () () () (–) () (–) () (–) () (.) ()
   – –   – –  – – – –  – – – –             n/a  n/a
.  . . . . . . . . . . . . . . . . . . . . . . .  . .  . ()
(.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) ()
                                 
                         .  .   ()
() () () () () () () () () () () () () () () () () () () () ()
     – –    —  —   – — –      AMICI  — Libertus/a Patron Master — Slave Patron Libertus/a — Slave Master — Conservi/libertus Conservi/libertus —       No dedicator  Se vivo sibi —   Husband Wife Husband Wife   Parents Son Daughter Father Son Daughter Mother Son Daughter   Father Son Mother Daughter Father Mother   Brother Brother Sister Sister Brother Sister       
%
(%)

()

()

()
    
() ()
 ()
  —
    —            — — —      — — — — —    — 
%
(%)

()

()

()
    
() ()
 ()
  —
  
. conv. Scallabitanus . Hispania (Fernandes) (Saller/Shaw)
  
T .. Comparison of commemorative practices in Lusitania and other parts of the western empire City/Region
Emerita conv. Pacensis civ. Igaedit. Olisipo Conimbriga Salamanca prov. Ávila prov. Lusitania conv. Scallabitanusa Iberian Peninsulab Rome: lower ordersb Ostia/Portusb Italia: Latiumb Italia: Regio XIb Gallia Narbonensisb Britanniab Germania Inferiorb Germania Superiorb Noricumb Africa: Lambaesisb Africa: Auziab Africa: Caesareab a b
A. Conjugal
B. Descending
C. Ascending
D. Nuclear family relationships/ all relationships
N
%
(%)
N
%
(%)
N
%
(%)
nucl. family
all
%
                     
        .             
() () () () () () () () (.) () () () () () () () () () () () () ()
                     
        .             
() () () () () () () () (.) () () () () () () () () () () () () ()
                     
        .             
() () () () () () () () (.) () () () () () () () () () () () () ()
                     
                     
        .             
Figures derived from Fernandes –: , table . Figures derived from Saller and Shaw : –, Tables –.
Family Relations in Roman Lusitania

The main conclusion from these figures is that in Lusitania members of the nuclear family massively predominated as commemorators of the dead over extended kin, non-kin heirs, friends, and dependent or fellow-slaves and freedmen (Table .D). The commemorator is drawn from the nuclear family group in  per cent of the cases. This is very much in line with what Saller and Shaw found for Spain in general (where  per cent of all commemorations were made within the nuclear family) and everywhere else in the western empire, where nuclear family members account for between  and  per cent of all acts of commemoration. It is also consistent with what Fernandes and Curchin found for the conventus Scallabitanus, where members of the nuclear family were responsible for . per cent (Fernandes) or  per cent (Curchin) of the commemorations where a commemorator was named.26 Of all the regions analysed in my study and by Saller and Shaw, Emerita, the provincial capital of Lusitania, has one of the lowest proportions of nuclear family commemorators (N 0001  of ; or  per cent). This must be connected to the much higher slave and freedman population in this the most thriving urban centre of the province, where slaves and freedmen were responsible for  per cent of all commemorations. This is significantly higher than the Spanish average for commemorations by slaves and/or freedmen ( per cent), and only comparable with the lower orders at Rome ( per cent) and Ostia/Portus ( per cent), thriving urban centres where slaves and ex-slaves were active within the local economy and had better chances of social mobility than in other more agrarian regions of the western empire. In contrast, extended kin form a small proportion of commemorators in all the Lusitanian samples. With one exception, this proportion ranges between  and  per cent, very much in line with what Saller and Shaw found in general for the Iberian peninsula ( per cent: see Table .). Two towns slightly diverge from the norm: Emerita, where  per cent of all commemorations were set up by extended kin, and the civitas capital of the Igaeditani, where as many as  per cent were. At Emerita many of these extended kin were in fact in-laws or step-relations rather than grandparents, aunts and uncles, or cousins.27 But at the civitas capital of the 26 27
Fernandes – []: , table ; Curchin a: , table I. Two commemorations were set up by a father-in-law or mother-in-law

Jonathan Edmondson
Igaeditani we find a rather different pattern: two cases of a fatherin-law commemorating his son’s wife (HAE ); one grandmother commemorating a grandson along with her own son (HAE : the grandson is certainly not this son’s son, although it is unclear whether he was another son’s son or her daughter’s son); two cases of a niece commemorating a paternal aunt or paternal uncle (HAE ); while two more nieces commemorated a maternal aunt and maternal uncle respectively (HAE , ); two granddaughters commemorated maternal grandfathers (HAE , ), while three commemorated maternal grandmothers (HAE , ; CIL . ). All in all, more than just a hint of an emphasis on matrikin in this native community which was promoted to municipal status in the Flavian period and which displays a marked indigenous streak in its local pattern of personal names (a point to which I shall return). As regards patterns of funerary commemoration within the nuclear family, spouses, parents, and children were the most active categories of commemorator, effecting , , and  per cent respectively of all the acts of commemoration on record from the province. Commemoration by siblings was much rarer in Lusitania (. per cent), as in the rest of Hispania ( per cent) and throughout the western empire, presumably only being practised when other close kin were not available.28 However, if we compare the relative balance between ‘conjugal’, ‘descending’ (i.e. father and/or mother to a son and/or daughter), and ‘ascending’ (i.e. a son and/or daughter commemorating a father and/or mother) commemoration within the nuclear family, considerable (CIL . ; AE , ); three by a son-in-law (AE , :  cases; CIL . ); one by a daughter-in-law (AE , ); two by a brother-in-law (CIL . ; AE , ); one by a stepmother (HEp. . ); one by a stepson (AE , ); two involving an alumna (HAE  0001 ILER  0001 Edmondson, Nogales Basarrate, and Trillmich : –, no. ; CMBad.  0001 HAE b); cf. four by grandparents (CMBad. ; EE . , ; AE , ); one by a nephew to a maternal aunt (AE , ); five by maternal cousins (sobrini) (AE , ; ERAE ; HEp. . ; EE . ; Álvarez Sáenz de Buruaga : , no.  0001 Gamer , , BA  and plate d). One commemoration was set up by Aponia Serana for Petronia Agilis, who is described as familiaris sua (AE , ). For further discussion, see Edmondson : . 28 Siblings were responsible for between  and % of the commemorations set up within civilian families in the regions studied by Saller and Shaw (: –, tables –), with  to % being the most common range.

Family Relations in Roman Lusitania
Conjugal 102
Conjugal 51
43.2%
37.0%
24.6%
38.4%
8.5%
23.7%
Descending 58
8.7%
Siblings 12
Siblings 20 15.9% Descending 53
Ascending 22
Ascending 56
2. CONVENTVS PACENSIS
1. AVGVSTA EMERITA
Descending 27 Conjugal 22 27.6%
Conjugal 13
Descending 58
22.4%
61.1%
13.7%
7.1%
8.4% Siblings 8
Siblings 7
42.9%
16.8%
Ascending 16
Ascending 42
4. OLISIPO
3. CIVITAS IGAEDITANORVM
Conjugal 199
Descending 242
Conjugal 210 Descending 257 29.0%
30.8% 35.5% 37.4% 8.7% 23.2%
Ascending 150 5. LUSITANIA (total)
10.0% Siblings 56
25.4%
Siblings 72
Ascending 184 6. HISPANIA (Saller/Shaw)
. . Types of funerary commemoration within the nuclear family in Lusitania

Jonathan Edmondson
local variation within Lusitania can once again be discerned (see Fig. .). Emerita has a very high degree of conjugal commemoration:  per cent of all commemorations within the nuclear family were made by husbands for their wives or vice versa, which is much higher than the Spanish norm of  per cent. This, I would argue, may again be explained by the much higher proportion of slaves and freedmen in the population of the provincial capital compared to that found in other cities of the province.29 Recently manumitted slaves were by definition kinless. They had no recognizable parents to commemorate them in death and so were especially keen to advertise their emergence within local society with as eye-catching a funerary monument as possible. At Emerita a number of libertini (and even some slaves) chose to commemorate themselves, their spouses, or their children with the most elaborate form of funerary monument found in the colony: those in the form of an aedicula which incorporated a portrait of the deceased or, especially relevant here, portraits of a deceased couple, emphasizing the strength of the conjugal bond (see Fig. .).30 The parallel with late Republican–early imperial Rome and Italy is striking. For here too there was a high proportion of conjugal commemoration and a number of tombstones of freedmen included portraits of this newly established family.31 On the other hand, Olisipo (modern Lisbon) and its territory has a far higher proportion of parent–child (‘descending’) commemoration than the norm ( per cent, compared to  per cent for Lusitania, and  per cent for Hispania in general).32 This suggests strong nuclear family bonds, with children remaining 29 For further discussion, see Edmondson : –. Saller and Shaw (: ) note that in Republican Rome and Latium, where there is a strikingly high proportion of conjugal commemoration (%), funerary inscriptions tended to be set up from among the middling and lower echelons of society. 30 See Edmondson, Nogales Basarrate, and Trillmich : esp. –. In some cases they set these up when one of the partners was still alive (se vivo/viva): e.g. Edmondson, Nogales Basarrate, and Trillmich : nos. , ,  0001 García y Bellido : –, no. , EE . , CIL .  respectively. 31 Shaw : –; Zanker ; Kockel ; George (this volume). 32 This is confirmed in Fernandes’ study of the conventus Scallabitanus as a whole: that is, the region of Portugal stretching north from Olisipo to the Duero, where .% (or  of ) of all commemorations within the nuclear family were set up by one or both parents to a son or daughter: see Fernandes –:  and , table .
Family Relations in Roman Lusitania

closely in touch with parents and natal kin even after marriage (a point to which I shall return). Although Olisipo was a thriving port-city with a significant population of slaves and freedmen, its territory had some of the most agriculturally promising land of the entire province.33 The agrarian nature of local society may well have encouraged this strong maintenance of nuclear family ties. It is particularly striking that mothers played a very significant role as commemorators. They were responsible for thirtyeight of all the acts of commemoration (that is,  per cent), almost double the rate found in Lusitania as a whole, and sometimes they commemorated adult sons and daughters. In nine of the fifteen cases from the urban centre, an age-at-death was recorded. These range from  to , while at least three of those commemorated by mothers were married at the time of their death.34 From the territory, there are fourteen cases where mothers commemorated sons, whose ages range from  to , and another nine in which mothers buried daughters who had died between the ages of  and .35 This would all suggest that parents maintained strong emotional and social bonds with their children even after the latter had married. Finally, at the civitas capital of the Igaeditani it is the high proportion of child–parent (‘ascending’) commemorations that is striking; these account for  per cent of all commemorations within the nuclear family. This would also suggest strong family ties, with children remaining close to their natal family even after marriage. This was probably connected with the desire to inherit landed property from their parents. At a larger, more flourishing urban For Olisipo in general, see Mantas : –; Ribeiro . Mothers commemorating sons: CIL . ,  (in both cases son aged ),  (aged ),  (aged , married). Mothers commemorating daughters: CIL . b (daughter aged , married),  (aged ),  (aged ); da Silva : no. -C (aged ); CIL .  (aged ). Age unspecified or not fully preserved: CIL .  (married), , ; da Silva : nos. ,  (0001 CIL . , from Olisipo, not from Elvas). 35 Mothers commemorating sons: AE ,  (son aged ); CIL .  (aged ); AE ,  (aged ); CIL .  (aged ),  (aged ),  (aged ),  (aged ),  (aged ),  (aged ),  (aged ); AE ,  (aged ); age not specified or not fully preserved: HAE ; Pereira : –. Mothers commemorating daughters: AE ,  (daughter aged ); CIL .  (aged ),  (aged ),  (aged ); HAE  (aged ); CIL .  (daughter aged ); age not specified or not fully preserved: CIL . ; HAE ; Camacho, Calais, and Nunes : –. 33 34

Jonathan Edmondson
centre such as Emerita with its strong urban economy families functioned as working units, as at Rome, for example, and children’s desire to maintain links with their natal kin was commensurately less strong than in smaller centres where agriculture dominated within the local economy. The prominence of child–parent commemoration could also denote that there was a strong respect for the aged in this community located in the ethnic heartland of the Lusitanians proper. Strabo believed (. . ) that the Lusitanians sat to dine ‘according to age and rank’ and singled out their particularly harsh punishment of parricides, who were stoned to death way beyond the traditional territory of the community. For Strabo these were two clear indices of their respect for the aged, a point to which we shall return (see below, pp. –). Gender and Age as Factors in Funerary Commemoration Commemorators One of the peculiarities of Spanish commemorative practices noted by Saller and Shaw, as we have seen, was the striking prominence of women (especially wives and mothers) as commemorators of the dead. In their samples of commemorations involving spouses, husbands almost always outnumber wives, but the gap is narrowest in Spain (a ratio of . husbands to every wife).36 In my sample from Lusitania (see Table .A), the balance between husbands and wives closely approximates that found by Saller and Shaw for the whole peninsula (again, a ratio of .). Within Lusitania, one area seems to diverge: the civitas capital of the Igaeditani, where a high predominance of wives over husbands as commemorators is found ( per cent wives to  per cent husbands, or a ratio of approximately two wives to every husband). Admittedly, the sample of conjugal commemorations is small (N 0001 ), but the importance of wives as commemorators is consistent with the emphasis on matrikin already noted in extended family commemorations (see above, p. ). Saller and Shaw also found that mothers played a more important role than fathers as commemorators in certain areas of the 36 Saller and Shaw : – and table , with the summary table C, p. . For further discussion of the important role played by women in funerary commemorations in the conventus Scallabitanus, see Fernandes –: –.

Family Relations in Roman Lusitania T .. Gender balance between commemorators A. husbands : wives
Emerita conv. Pacensis civ. Igaedit. Olisipo Conimbriga Salamanca prov. Ávila prov. Lusitania conv. Scallabitanusa Hispaniab Rome: lower ordersb Ostia/Portusb Italia: Latiumb Italia: Regio XIb Gallia Narbonensisb Britanniab Germania Inferiorb Germania Superiorb Africa: Lambaesisb Africa: Caesareab a b
B. C. fathers : mothers sons : daughters
N
Ratio
N
Ratio
N
Ratio
: : : : : : : : : : : : : : : : : : : :
. . . .
: : : : : : : : : : : : : : : : : : : :
. . . . .
: : : : : : : : 6: : : : : : : :
. . . .
{ :
.
. . . . . . . . . . . . .
. . . . . . . . . . . . .
: :
. . . . . . . .  . .
Figures derived from Fernandes –: , table . Figures derived from Saller and Shaw : –, tables –.
Roman empire (for example, among the lower orders of imperial Rome, in Latium, and in Gallia Narbonensis they outnumbered fathers), but nowhere to the same degree as they did in Spain, where mothers outnumbered fathers by  per cent to  per cent: that is, a ratio of . father-to-child commemorations to every mother-to-child commemoration. In all my samples from Lusitania mothers also outnumber fathers, but it is only at Olisipo (ratio .), the civitas capital of the Igaeditani (ratio .) and Conimbriga (.) that they do so to the same degree as in Hispania as a whole (see Table .B). In the conventus Pacensis and at Emerita fathers played a much greater role than was usual for Hispania: here they were responsible for  per cent and  per cent of all commemorations set up by parents (that is, ratios respectively of . and . father-to-child commemorations

Jonathan Edmondson
to every mother-to-child commemoration). This may hint that Emerita as a Roman colony and provincial capital and the southwestern third of the province, which could boast a Roman colony (Pax Iulia) and three native communities granted the Latin rights of citizenship under Augustus (Ebora, Salacia, and Myrtilis), constituted a rather different cultural milieu from the more northerly and interior zones of the conventus Scallabitanus and the conventus Emeritensis to the north of Emerita. Again the predominance of women seems to be more marked in less Romanized parts of the province. When it comes to children commemorating their parents, sons predominate over daughters in every area of the western empire studied by Saller and Shaw, except among senators and equestrians from Rome (see Table .C).37 They found that the balance was more even in the Iberian peninsula than elsewhere (a ratio of . commemorations by sons for every commemoration set up by a daughter) and my sample from Lusitania would confirm this general picture: sons outnumber daughters as commemorators by a ratio of . :  (see Table .C). There are variations between different regions within Lusitania. Sons are most predominant in the conventus Pacensis with . commemorations set up by sons to every one by a daughter, while daughters narrowly outnumber sons at Emerita and Olisipo, where the ratios of commemorations set up by sons as opposed to by daughters stand at . and . respectively. But the samples from each city or region are rather too small to draw meaningful contrasts between different regions of Lusitania. Those commemorated It is now widely recognized that cultural factors played an important role in determining whether an individual deserved to be commemorated with an epitaph.38 In particular, Brent Shaw () has shown that the chances of receiving commemoration varied considerably depending on the gender and age of the deceased. In most regions of the western empire sons and fathers 37 For senators and equestrians at Rome, see Saller and Shaw : , tables  and :  commemorations by sons, compared to  by daughters: a ratio of .. 38 See Meyer ; Woolf ; Hope ; cf. Cannon .

Family Relations in Roman Lusitania
were more likely to receive a tombstone than daughters and mothers. This is the case in Spain, but as usual the gap between sons and daughters is narrower than in many other parts of the western empire (see Table .B: a ratio of . sons commemorated for every daughter). In other words, women (and especially daughters and mothers) appear to have been more highly valued culturally in Spain than in many other regions of the empire and so in general stood a better chance of receiving an epitaph than women did elsewhere in the western provinces. In this respect, the province of Lusitania does not appear to have differed substantially from the rest of the Iberian peninsula. Here, as in other parts of Spain, women (and especially daughters and mothers) were also more likely to receive an epitaph than their counterparts from most other areas in the western empire T .. Gender balance between those commemorated A. husbands : wives N Emerita : conventus Pacensis : civ. Igaedit. : Olisipo : Conimbriga : Salamanca prov. : Ávila prov. : Lusitania : conv. Scallabitanusa : Hispaniab : Rome: lower ordersb : Rome: Ostia/Portusb : Latiumb : Italy: Regio XIb : Gallia Narbonensisb : Britanniab : Germania Inferior : & Superiorb Africa: Lambaesisb : Africa: Caesareab : a b
B. fathers : mothers
C. sons : daughters
Ratio
N
Ratio
N
. . . . .   . . . . . . . . . .
: : : : : : : : : : : : : : : : :
. . . . . .  . . . . . . . . . .
: : : : : : : : : : : : : : : : :
. .   . — . . . . . . . .  . .
. .
: :
. .
: :
. .
Figures derived from Fernandes –: , table . Figures derived from Saller and Shaw : –, tables –.
Ratio

Jonathan Edmondson
(see Table .). However, there are some regional distinctions within Lusitania. In the conventus Pacensis, for example, sons predominate to a slightly greater extent (. sons per daughter commemorated) than they do elsewhere in the province (. sons per daughter), while conversely at the civitas capital of the Igaeditani daughters even outnumber sons (. sons per daughter). These indices are in line with the marked prominence of males as commemorators in the conventus Pacensis (Table .) and the greater prominence of wives and mothers as commemorators at the civitas capital of the Igaeditani (Table .A–B). In the commemoration of mothers as opposed to fathers (Table .C), Lusitania is again very close to what Saller and Shaw found for Spain as a whole in that fathers were commemorated slightly more frequently than mothers. In one region of Lusitania, the conventus Pacensis, there appears to be some divergence from the Spanish norm, with mothers receiving more commemorations than fathers (.). It is not clear why this should have been the case, and one might not want to press the distinction too much since the sample is disappointingly small (N 0001 ). In addition to gender preferences, certain age-groups in the population were more likely to receive an epitaph than others, although this varied considerably not only between regions of the western empire, but also within regions between urban and rural contexts. In urban and rural Spain Shaw found a very low rate of commemoration of children under , while in the countryside the aged (people in their th decile: i.e. those aged –) received among the highest degree of commemoration anywhere in the western empire.39 Such analysis of age preferences is more suggestive than proven, given the much smaller number of inscriptions available for study that provide an age at death, but the epitaphs from Emerita, the conventus Pacensis, and Olisipo allow some insights for Lusitania. In the commemoration of children by one or both parents (see Table .), the high proportion of children over the age of  who were still commemorated by their parents rather than by spouses is striking:  per cent of all children at Emerita,  per cent in the conventus Pacensis, and  per cent at Olisipo. 39 Shaw : – and table . (children under ),  and table . (those aged –).
Family Relations in Roman Lusitania

T .. Children commemorated by parents according to age cohorts Age
Emerita
conv. Pacensis
–
 (%)  (%)  (%)  (%)  (%)  (%)  (%) 
 (%)  (%)  (%)  (%)  (%)  (%)  (%) 
– – – – – 0002 Total
Olisipo
Total
 (%) 
 (%)  (%)  (%)  (%)  (%)  (%)  (%) 
 (%)  (%)  (%)  (%)  (%) 
This Lusitanian sample confirms a general feature of commemorative practices in Spain as a whole, already noted by Saller and Shaw in their respective studies of the age of Roman men and women at marriage.40 It must reflect the continued role of natal kin in the affairs of their children even after marriage. This desire to commemorate children is further reinforced by those commemorations set up by parents-in-law to sons-in-law and daughters-in-law. So, for example, at the civitas capital of the Igaeditani Rufus Triti f. commemorated his daughter-in-law, Prisca Frontonis f. along with his two sons, Niger, Prisca’s husband, and Rufinus, his daughter, Vitalis, and Camala Docquiri f., perhaps another daughter-in-law (HAE  0001 ILER ), while at Olisipo Sempronia Rufina looked after the burial of her year-old son and -year-old daughter-in-law (CIL . ).41 These parents-in-law presumably took on the responsibility only when parents were not available to look after the burial of their children, either because they had died or because their children had moved from their home region to settle elsewhere, as in an 40 Saller a:  and n.; Shaw a: ; cf. Saller : –, esp. –,  and table .h. 41 For an unnamed father-in-law commemorating his daughter-in-law at Emerita, see AE , .

Jonathan Edmondson
example from Emerita, where an immigrant from Aeminium (modern Coimbra) was commemorated jointly by his daughter and his mother-in-law (CIL . ). In some cases, this could extend a further generation with grandparents looking after the burial of grandchildren. So in the conventus Pacensis at Rio de Moinhos (Aljustrel) Agria Rufina was responsible for commemorating her own husband, M. Valerius Rufus, her son, M. Valerius Marcellus, and her grandson, also called M. Valerius Marcellus (IRCP ). In the western Algarve Pompeia Exoce looked after the burial of her -year-old grandson, C(a)epio Sempronianus (IRCP , Mexilhoeira Grande, Portimão), while at Olisipo Annia Corinthia was responsible for the burial of her grandson, M. Cassius Marcianus, who had died aged  (CIL . ). Similarly at Conimbriga, two examples survive of grandmothers commemorating their granddaughters, who had died aged  and  respectively, with the deceased’s mother also involved in the first case (CIL . ,  0001 Fouilles de Conimbriga, , nos. , ). Clearly marriage and the formation of a new conjugal unit did not, in Lusitania, necessarily sever the bonds, both social and emotional, with one’s natal family. In the commemoration of children by their parents (Table .), no obvious gender preference seems to have operated. This is again something of a Spanish anomaly, since boys tend to predominate in Italy and the other western provinces.42 This anomaly should certainly be connected to the higher cultural valuation of women and girls in the Iberian peninsula, a feature apparent in other types of commemorative practice. Furthermore, there is very little variation between males and females at each of the various stages of the life-cycle. One possible exception is at Emerita where girls in their third decile (i.e. aged between  and ) outnumber boys by a proportion of more than  to . This is the decile in which women were culturally most valued, since this was when many women had recently married and were at their most fertile for child-bearing. The loss of a woman at such a moment was especially distressing and represented an event particularly worthy of commemoration. 42 See Shaw : –; for Italy, see also McWilliam : –, with tables . and ..
Family Relations in Roman Lusitania

T .. Gender balance in the commemoration of children at various stages of the life cycle at Emerita, in the conventus Pacensis, and at Olisipo Age
Sex
Emerita
conv. Pacensis
Olisipo
–
M F ? M F ? M F ? M F ? M F ? All
  —   —   —   —   — 
        —   —    
 — —   —   —   —   — 
– – 0002 
Total         —   —    
In commemorations involving spouses (Table .), some interesting gender distinctions can be discerned at various stages of the life-cycle. For at Emerita and in the conventus Pacensis husbands dramatically outnumbered wives as commemorators of those who died between the ages of  and ; but as the age of couples increased, wives took over as the dominant commemorators. This adds a considerable nuance to the apparent predominance of wives in funerary rituals (see above pp. ‒). They only emerged as dominant commemorators in Lusitania, it now appears, in middle- to old-age. It might be argued that this relates to the demographic profile of the local populations of Emerita and the conventus Pacensis; that is, that women tended to outlive their spouses and so became more visible in funerary commemorations as married couples aged.43 This may well be the case, but cultural factors were surely also at play. The distribution of ages shows a shifting cultural valuation of women and men over the life-cycle. Women during their child-bearing years were 43 On the dangers of using Roman tombstones to reconstruct demographic profiles, see Hopkins  against older studies such as McDonnell , R. [sic] , or Szilágyi .

Jonathan Edmondson
thought especially worthy of a marked commemoration; hence the much higher ratio of tombstones set up by husbands to wives who died between the ages of  and  ( :  0001 a ratio of almost four tombstones for a wife to every one for a husband). But as women grew older past menopause, their centrality to the family unit became less important and the male head of the household emerged as the most honoured member of the family. Hence for those aged over , three and a half times more tombstones were set up for husbands by their wives than for wives by their husbands ( : ; that is, a ratio of . husbands to every  wife). The epitaphs from the modern Spanish province of Salamanca at last come into play here. For although they usually fail to name a commemorator, they very frequently include the deceased’s age at death. This occurs in  of the  epitaphs (i.e. in just over threequarters of the sample). Again the spread of ages cannot be used to provide a demographic profile of the population, but they can provide useful data about changing patterns of commemoration of men and women over the life-cycle (Table .). The representation of males and females in the sample is fairly equal, with a total of  males and  females. (In  cases the gender of the deceased is T .. Balance of commemorator gender and age spread in conjugal commemoration Age at death – – – – – – : – – – – – – 0002 : 0002 
Emerita
conv. Pacensis
Total
Male
Female
Male
Female
Male
Female
—           — —  
—  —            
          — —   
— —          —   
           —   
—              
Family Relations in Roman Lusitania

T .. Age at death for men and women: modern province of Salamanca Age
M
F
– –
                        
                        
– – – – – – – – – – – – – – 0002
– – – – – – – –
Uncertain                       
Total
 (%)  (%)  (%)  (%)  (%)  (%)  (%)  (%)  (.%)
unclear.) But in contrast to what we have just observed at Emerita or in the conventus Pacensis (see Table .), there is a generally even gender balance for each decade through the life-cycle. From the particular ages at death recorded, it is clear that agerounding was practised in this region—to a much more significant degree than elsewhere in the province.44 Almost half ( per cent) of those attested were thought to have died on a round decile (four deaths at , fourteen deaths at , fifteen at , six at , twenty-one deaths at , twenty-seven at , ten at , four 44 See briefly also Curchin b and, for the Portuguese part of the province, Encarnação .

Jonathan Edmondson
at , and one at ), while another . per cent were recorded as having died at the mid-point of a decade (three at , seven at , nine at , six at , three at , four at , seven at , one at , and one at ). It is only up to the age of  that any attempt was made to estimate precisely the age of the deceased, with deaths recorded at age ,  (two occurrences),  (two occurrences), ,  (three occurrences), , ,  (three occurrences),  (three occurrences), , ,  (two occurrences), , ,  (two occurrences),  (three occurrences), , and . The only ages attested above  that do not coincide with a round decile or a mid-decile point are single occurrences of , , , , , , , ,  (i.e.  out of a total of  or just . per cent). As for the spread of ages, there is a marked proclivity to commemorate the older members of local society. Of those whose ages were recorded on their tombstones (N 0001 ),  (or  per cent) were aged over , while  (or  per cent) were older than . Conversely, the very young were thinly represented: so, for example, only  (or  per cent) were aged  or younger. In the middle decades of life there is a more or less even distribution:  per cent in their teens,  per cent in their twenties,  per cent in their thirties, and  per cent in their forties. This stress on the aged must reflect the honoured position that the aged held within local society. Shaw has noted the tendency for rural regions to put special emphasis on the elders of the community, especially in rural North Africa,45 and this seems to be the case too in the province of Salamanca, a predominantly rural society, as we have noted, where only a handful of cities developed under Roman rule. However, among those aged over , women were almost as prominent as men ( women as opposed to  men). So here at least stress was not just placed on the male heads of patrilineal descent groups; the almost equal appearance of older women hints that kinship relations were organized along essentially bilateral, or cognatic, lines. I do not wish to place any emphasis on the actual ages recorded. As Duncan-Jones has shown, agerounding involved not only a fair degree of approximation, but often also the addition of extra years to the ages recorded on tombstones, especially among the more elderly in society.46 But 45 46
Shaw : . Duncan-Jones , revised at Duncan-Jones : –.
Family Relations in Roman Lusitania

the very urge to exaggerate and round up ages underlines still further the special social prestige accorded the most senior members of local society. The region around modern Salamanca, therefore, seems to have retained the marked respect for the aged that Strabo (. . ) noted was so strong among the Lusitanian peoples who inhabited this general area of Spain in the later Iron Age. It is no coincidence that this was one of the areas of Lusitania in which native culture seems to have retained more of its force than occurred elsewhere in the province. A number of late Iron-Age hillforts remained in occupation right through the Roman period, while native deities and indigenous personal names continued to flourish into the third century , when the epigraphic record dies out. The inhabitants of the region set up tombstones that were highly distinctive in style, aggressively local with little in the way of Roman stylistic influence, such as the funerary stela from Salmantica (Salamanca) for Cloutia Ambini f. shown in Fig. . (HAE  0001 ILER ).47 It would be entirely consistent, therefore, if some elements at least of pre-Roman kinship and family relations had survived under Roman rule. The patterns of funerary commemoration that we can discern for Lusitania, therefore, suggest that it was members of the immediate family who were in the overwhelming majority of cases responsible for the commemoration of the dead: in the main, surviving spouses, parents, or children; only in the absence of one of these groups were siblings drawn in, and only in the latter’s absence did it then devolve to members of the wider kin group of the deceased, whether in the paternal or maternal line. However, subtle differences can be discerned across the province. Augusta Emerita, a Roman colony implanted ex novo on the landscape in   and within ten years chosen as the capital of the new Roman province of Lusitania, developed along rather different lines compared with other communities whose social structures had their roots in the pre-Augustan period. At Emerita there was less emphasis on long-established lines of descent. In a 47 For Iron-Age hillforts, note, for example, Yecla de Yeltes: Maluquer de Motes : –, with plan; Martín Valls . For native deities, note, for example, AE , , correcting AE ,  (Martiago), HEp. .  (Alba de Tormes); CIL .  (Ciudad Rodrigo); in general, Blázquez Martínez ; for onomastics, see Salinas de Frías ; for local tombstones, de Navascués .

Jonathan Edmondson
. . Funerary stela of Cloutia Ambini f., from Salmantica
community with a relatively high proportion of slaves and freedmen, new conjugal units were more frequently being formed, especially as slaves won their freedom. Although house-bred slaves were very occasionally allowed to maintain contact with their natural slave-mothers and their natal kin, much more often than not these units had no ascendant nor collateral kin.48 Emerita also boasted a higher proportion than elsewhere of freeborn immigrants, who came to settle not just from the other towns of Lusitania, but from the rest of Spain and overseas as 48
See further Edmondson : –, –.
Family Relations in Roman Lusitania

well.49 The possibility that these immigrants could have had complex kin-networks was severely limited. On the other hand, in the conventus Pacensis, at the civitas capital of the Igaeditani, and at Olisipo there was greater stress on maintaining bonds with one’s natal family and as a result more frequent commemoration by parents and by children. In a more traditional, agrarian society bonds with one’s natal family were that much stronger, as children were keen to stake their claim to inherit some of their parents’ property. Finally, in the more rural interior of the province women appear to have had a more predominant role in funerary commemoration than in more urbanized parts of the province and certainly than was the norm in all but Rome’s Spanish provinces. These significant fluctuations between the various cities and regions of Lusitania suggest that a globalizing analysis such as that of Saller and Shaw for Spain can obscure important local differences.50 Joint Burials and Joint Acts of Commemoration One criticism that has been levelled at Saller and Shaw’s method is that their focus on individual acts of commemoration required them to split up all acts of joint or multiple commemoration into a series of individual relationships. Thus an epitaph that was originally set up by a man for his wife, son, daughter, son-in-law, and cousin was separated by Saller and Shaw into five separate acts of commemoration: three within the nuclear family (one of conjugal type, two of descending) and two involving extended kin. Dale Martin () in particular has argued that this skewed their resulting computations and led inevitably to the conclusion that Saller and Shaw favoured: namely, that the nuclear family was overwhelmingly dominant in acts of funerary commemoration in the western Roman empire. Martin instead proposed a different set of categories that do not divide a multiple act of commemoration into its individual components. Thus, for Martin the hypothetical example I have just mentioned would count as a single act of commemoration undertaken by an extended family. 49 See García y Bellido : esp. , , –, ; Haley : –; Edmondson forthcoming. 50 Saller and Shaw were willing to admit differences between various towns in Africa (: –); the same seems to apply in Lusitania.

Jonathan Edmondson
His case-study of funerary commemoration in various cities of Roman Asia Minor seems to show a much greater emphasis on extended kin than Saller and Shaw had argued was normal in the western provinces. Martin’s point has some validity and led me to reanalyse the funerary commemorations from Emerita and the civitas capital of the Igaeditani according to his divisions, to see if this resulted in radically different results from those obtained using Saller and Shaw’s method.51 The results are summarized in Table .. The results of this reanalysis of the material from two regions of Lusitania do not diverge to any significant degree from those obtained using Saller and Shaw’s method (see above, Table .). Commemorations involving extended kin are still relatively insignificant at  per cent of all commemorations at Emerita and  per cent at the civitas capital of the Igaeditani and broadly commensurate with what we found using Saller and Shaw’s method:  per cent at Emerita,  per cent at the civitas capital of the Igaeditani (see above, Table .). What Martin’s study does underline, I would argue, is that in Roman Asia Minor family relations operated in rather different ways than was the case in Rome’s western provinces; there was perhaps a broader emphasis T .. Funerary commemorations from Emerita and the Civitas Igaeditanorum according to the method of D. B. Martin Type
A B C D E
Fragmentary inscriptions Epitaphs of one person: no commemorator Husband and wife only Nuclear family (complete or partial) Persons outside immediate family included (extended) F Multi-person epitaph, but relations unclear G Slave relations 
Emerita
civ. Igaedit.
N
%
N
%
   
   
   
   
   
  
   
  
51 To Martin’s categories (: –, –), I have added an extra one (‘G. Slave relations’), to take account of commemorations set up by slaves to their fellow-slaves, slaves to their masters, ex-slaves to fellow conliberti, or ex-slaves to their former masters or patrons.
Family Relations in Roman Lusitania

on maintaining links and emotional ties with more distant kin. It does not seriously undermine the validity of Saller and Shaw’s methodology. However, Martin’s more general point that it is profitable to look at epitaphs in their entirety reminds us that qualitative analysis of epitaphs can be just as revealing as quantitative. In particular, joint acts of commemoration where several different individuals collaborated in commemorating a deceased person and epitaphs which reveal that various individuals were buried within the same tomb and commemorated jointly on a common epitaph are particularly valuable types of documents to focus on.52 There is not time here to undertake a detailed analysis of all these joint-acts, but a brief survey of the main types of jointburials and joint acts of commemoration is revealing (see Tables . and .). In the four regions surveyed, joint or multiple burials or multiple acts of commemoration usually involved various members of the nuclear family.  of the  joint or multiple burials (or  per cent) (Table ., lines , , ) and  of the  multiple acts of commemoration (or  per cent) involved the nuclear family (Table ., lines –). At Emerita it is impossible to establish securely any kinship bonds between the deceased in a relatively high number ( per cent) of the joint burials (Table ., line ). In some of these cases the deceased may have been related, but chose not to express their relationship on their epitaph. However, in other cases they may have been individuals who had joined a funerary collegium and were buried alongside non-relatives in a columbarium-style tomb. For example, a large plaque, more than  metres wide, that was reused as a grave-cover in a fourthcentury necropolis, contained at least six separate epitaphs inscribed—on the basis of the letter-forms and funerary formulae—in the first and second centuries .53 Three of the epitaphs 52 Ideally one would like to go further and analyse the epitaphs in the physical context of the tombs for which they were originally designed, but in Lusitania very few epitaphs have been discovered in situ: for some examples from Emerita, see Edmondson : –; see also Nogales Basarrate and Márquez Pérez ; Caetano . For studies from elsewhere that reveal the benefits of this approach, see, for example, Hope a; Nielsen  []; Heinzelmann ; Feraudi-Gruénais . 53 Ramírez Sádaba and Gijón Gabriel : –, no.  with photos 0001 AE ,  0001 HEp. . .

Jonathan Edmondson T .. Joint burials in various regions of Lusitania by type
 Married couple  Two or more members of a nuclear family (parent 0002 child; two siblings)  Members of nuclear family 0002 one or more extended kin  Members of nuclear family 0002 slave(s) or ex-slave(s)  conservi/conliberti  sibi et suis  Multiple burial, with unclear relation between deceased 
Emerita
conv. civ. Pacensis Igaedit.
Olisipo
N
%
N
%
N
%
N
% N
%
 
 
 
   
 
 
 
 
 


















  

  
 
  
  
  
  



 

Total



T .. Joint acts of commemoration in various regions of Lusitania Emerita
conv. civ. Pacensis Igaedit.
Olisipo
Total
N
N
N
N
N
%
 Parents   Siblings   Members of nuclear  family  Member(s) of nuclear  family 0002 extended kin  Extended kin   Members of slave  familia  Coheredes   Amici   
  
  
  
  
  





 
 
 
 
 
  
  
  
  
 
Family Relations in Roman Lusitania

were set up by members of separate nuclear families: the doctor C. Domitius Pylades was commemorated by his wife, Cornelia M.l. Maurilla, and freedman, [C.] Domitius C.l. Restitutus (AE , a 0001 HEp. . a); L. Valerius Reburrus, a retired soldier (missicius), set up an epitaph for himself and his wife Attia Libas (AE , b 0001 HEp. . b), while C. Nonius Batullus had one inscribed for himself and for Fausta Coelia and Atlia Specia, his first and second wives (AE , c 0001 HEp. . c: meas uxores). But the fourth epitaph on the plaque simply lists two pairs of names with a gap between: L. Iustulenus Chrestus and Helvia Aucta and then below L. Peticius Felix and Iulia Apana (AE , d 0001 HEp. . d). These may each have been married couples, but even if this is correct, the relationship between the two pairs was left unclear. The fifth epitaph was set up by P. Curtius Phoebus for himself and for Scantia Bella, Iulia Maura, Ascanius, Pastorilla, and Echion (AE , e 0001 HEp. . e), but again the relationship between the individuals was not specified. The final epitaph simply commemorated Naevidia Optata, daughter of C. Naevidius Optatus (AE , f 0001 HEp. . f ). Most of all, the fact that these six disparate epitaphs were all carved on the same plaque would strongly suggest that they had joined a funerary club and were buried in the college’s multiple tomb, on the façade of which the plaque was originally displayed.54 However, occasionally more complex groupings drawn from natal and conjugal kin were buried together or were active in joint acts of commemoration. This was especially so at the civitas capital of the Igaeditani, from where two examples survive of an individual who buried his father, mother, brother, and his own son or daughter—all, it appears, in the same tomb (HAE , ). Extended kin were particularly prominent, buried alongside members of a nuclear family in almost one-quarter of the joint burials (Table ., line ) and implicated in almost half the joint acts of commemoration (Table ., line ). Several generations within the same family sometimes joined together to commemorate the dead: so, for example, two daughters and a granddaughter set up an epitaph for Chresumus and Amoena, their 54
.
For funerary collegia, see further Hopkins : –; Patterson  and

Jonathan Edmondson
father/grandfather and mother/grandmother respectively (HAE ). The joint acts of commemoration or burial involved members of an extended patrilineal descent group. However, there are also examples from this community where blood-kin joined with affines in commemorating the dead, as, for example, when Modestus Proculi f. was commemorated alongside his mother-inlaw, Dutia Puci f., by his wife, Rufina Rufi Tongetami f., and his own daughter, Modestina (CIL . ). Sometimes the remains of even more extended kin were laid to rest within the same tomb, as in the case discussed at the start of this chapter, where Claudia Tangina commemorated her paternal grandparents, her paternal uncle, and her paternal aunt (HAE  0001 ILER ; see above, Fig. .).55 These more complex relationships at the civitas capital of the Igaeditani point to more complex family relationships in this community in the heartland of the Lusitanians—very different on the whole from what we find at the Roman colony of Emerita or in the conventus Pacensis.56 Onomastics and Family Relations Funerary commemorations, however, can be used in other ways to throw light on family relations in a province such as Lusitania. Quantitative analysis is not the only avenue of approach. Much can be gleaned, for example, from patterns of nomenclature, not least because personal names were bestowed within a familial context. The quintessentially Roman type of name (the tria nomina) underlined the importance of patrilineality in Roman family organization and the transmission of property.57 This was used by Roman citizens and perhaps also by provincial non-citizens (peregrini) who wished thereby to advertise their emerging Romanness, whether or not they realized that it was strictly See above, n. . For a fuller analysis of multiple burials and commemorations from the civitas Igaeditanorum, see González-Conde Puente –; Edmondson 2004. 57 On Roman onomastics, see the various studies in Duval and Pflaum ; Salomies ; Lassère ; for a brief survey, with bibliography, see Salway ; for more specific studies on Lusitania, see Dias ; Salinas de Frías ; Stanley ; and (especially) Grupo Mérida . On patrilineal or agnatic descent, see Saller : ch. . 55 56
Family Relations in Roman Lusitania

illegal.58 The family name (nomen gentilicium) was handed down from father to legitimate children, both male and female, who in turn expressed their filiation by means of their father’s praenomen. Thus from Emerita, for example, the family of C. Voconius C.f. Pap[iria tribu] comprised himself, his wife Caecilia Anus, their son C. Voconius C.f. Proculus, and daughter Voconia C.f. Maria (HAE  0001 ILER  0002 ). However, more indigenous forms of personal names also emphasized important family relationships. So among the Igaeditani, for example, children often received a single name related to that of their father, but in adjectival or sometimes in diminutive form: Rufinus Rufi f. (HAE ); Rufina Rufi f. (CIL . ); Modestina daughter of Modestus Proculi f. (CIL . ); Probina daughter of Probus (HAE ); Severina Severi f. (HAE ); Longinianus Longini (f.) (HAE ); Liguria Liguris f. (HAE ). The recently enfranchised also displayed this same process in their names: for example, C. Valerius Flacci f. Q(uirina tribu) Flaccinus (HAE ) and Iulia Severi f. Severina (CIL . ). There are also a few cases of grandsons who bore the same name as their patrilineal grandfather: for example, Clemens Lupi (f.), whose father was Lupus Clementis (f.) (HAE ) or Vegetus Vegetini f., whose patronymic would suggest that Vegetinus’ father was a Vegetus (HAE ).59 This all suggests a strong system of agnatic descent, even in contexts where the full Roman onomastic system had not taken root. However, to balance this, in the territory of the Civitas Igaeditanorum two cases survive where granddaughters were named after one of their maternal grandparents: Prisca Reburri f., whose mother was Cabrula Prisci f., and Sunua Elavi (f.), whose 58 On the usurpation of the Roman tria nomina by non-citizens, note CIL .  0001 ILS  (edict of Claudius granting citizenship to various peoples in the Italian Alps in  ), esp. lines –: nominaque ea, quae habuerunt antea tanquam cives Romani, ita habere is permittam; see further Alföldy . 59 For further examples from the territory of the civitas, note FE  0001 AE ,  (Orjais, conc. Covilhã): Silo Tranquilli f. whose father was Tranquillus Silonis f.; AE ,  (from near Fundão): Tongius M(a)elonis f. whose father was Maelonius Tongi f.; AE ,  0001 HAE  0001 ILER  (Telhado, conc. Fundão): Caeno Lovi f. whose father was Lovius Caenonis f.; note also Nepos Arconis f. (AE , , near Fundão). From further west in Lusitania, the sons of Florus Valentis f. were named Florinus and Florentinus: Vaz : –, no.  (Quinta dos Matos, Rãs, Sátão, distr. Viseu).

Jonathan Edmondson
grandmother was Sunua Pisiri f.60 This would confirm the emphasis on maternal kin found earlier in our analysis of funerary commemoration in this community (see above pp. ‒, ‒). Further examples can be found to the north-east in the modern province of Salamanca.61 Furthermore, there are a few isolated cases from Lusitania where a matronymic was used as part of a personal name in addition to, or sometimes instead of, a patronymic. At Mirobriga (modern Ciudad Rodrigo, prov. Salamanca) a deceased pair (perhaps a married couple) each bore a matronymic as well as a patronymic: Eligius Flavius was described as the ‘son of Flavinus and Flaccilla’ and Allia Avita the ‘daughter of Serenus and Varilla’ (CIL . ). Again at the urban centre of the Igaeditani, the year-old Lubana was described as ‘the daughter of Amoena’, when her mother commemorated her with a distinctively tall granite stela topped with a triangular pediment decorated with a crescent moon (HAE  0001 ILER  0001 Almeida : no. ). On a funerary monument that in style is very similar to those found in more northerly districts of Lusitania and the neighbouring zones of Hispania Tarraconensis further to the north, we may be witnessing the vestigial survival of pre-Roman matrilineal patterns of descent and familial organization. Two further examples survive from the territory.62 Quantitative analysis suggested that maternal kin in this community played a more important role than in more fully Roman contexts; the appearance of matronymics provides vivid confirmation of this. 60 Prisca Reburri f.: AE ,  (region of Fundão, distr. Castelo Branco); Sunua Elavi (f.): Garcia : –, no. v, with photo (Castelo Branco). 61 For example, AE ,  (Villar de la Yegua): Amoena, married to Cloutius, sets up a monument for her mother Apana Triti (f.) and her own daughter Apana Clouti f. 62 For matronymics from the territory of the Igaeditani, note Gallus Amoen(a)e (f.) (HAE  0001 AE , , Vale de Senhora de Póvoa, conc. Penamacor) and Amminus Andaitiae f. (CIL . , Capinha, conc. Fundão). From elsewhere in Lusitania, note also Culua Paugendiae f. (HAE  0001 AE , , Lamego, distr. Viseu). But caution is needed with indigenous names: the name Decumus Placentiae filius (Encarnação and da Silva : no. , Mouriscas, conc. Abrantes, distr. Santarém) might appear to contain another matronymic, but the full text of the epitaph makes it clear that Placentia was Decumus’ father: Decumus Placentiae filius ann(orum) XII h.s.e.s.t.t.l. Placentia et Iulia filio f(aciendum) c(uraverunt).
Family Relations in Roman Lusitania

However, onomastics also reveal that the nuclear family was not the only organizing principle of kinship in the interior of Lusitania during the Roman imperial period. For in the modern Spanish provinces of Salamanca, Ávila, Cáceres, and Toledo (i.e. the north-eastern and eastern third of the province, or the region inhabited by the ethnic group known by Roman sources as the Vettones), more than fifty men and women were identified not just by a patronymic, but also by their affiliation to what appears to be a wider kinship grouping, or cognatio. For example, at the town of Ávila we find Curundus Aelcecum Cadani f. (LICS ) or Dobiterus Caburoniq(um) Equasi f., who was buried alongside Arena Mentovieq(um) Aelci f. (LICS b).63 Similar kinship groupings are found outside Lusitania to the north in Asturia and Cantabria, to the north-west in Gallaecia and to the north-east and east in Celtiberia.64 Because the evidence is purely onomastic, it is difficult to reconstruct how these kinship groupings functioned in actual social practice, but it is clear that their titles derived from indigenous personal names. So from Lusitania, for example, we have an Aravus Araviaq(um) Turani f. (LICS ) or Matugenus Matugen[i]q(um) Tancini f. (LICS ), both from Ávila, or Caurunius Ambati (f.) Caurunicum from Yecla de Yeltes, prov. Salamanca (Albertos Firmat : no.  0001 González Rodríguez : no. ). It is impossible to tell whether those individuals whose personal names matched the title of a wider kinship grouping were the heads of these kinship groups, but it does permit the inference that they were descent groups, tracing their existence back to a common ancestor, real or mythical. Not all homonymous individuals, however, were members of the same kinship grouping: at Yecla de Yeltes Alaesus Triti f. was a member not of the kinship-group Tritecum (which is attested at Yecla), but of the group Ubonicum. Similarly Erguena
63 For full studies, with lists of all known occurrences, see Albertos Firmat , , ; González Rodríguez , ; for analysis, cf. Sánchez Moreno . 64 González Rodríguez (: –) lists  of such wider kinship groupings in total, of which  come from Lusitania, the other  from neighbouring regions. Her work needs to be updated in light of recent discoveries: to her examples from Ávila, add LICS , , , , , bis.

Jonathan Edmondson
daughter of Boutius belonged to the kinship grouping Ammaricum, not Boutiecum.65 Some kinship-groupings are found in more than one region of the province: for example, Calaetiqum is attested at Ávila and Guisando (prov. Toledo); Arreinicum at Ávila and Villar del Pedroso (prov. Cáceres); Matugeniqum at Ávila and Yecla de Yeltes.66 Some of those found in Lusitania are also attested outside the province: Crastuniqum at Ávila (to which an immigrant from Uxama, modern Osma, prov. Soria, belonged), and also at Cuevas de Amaya (prov. Burgos) and Langosto (prov. Soria); Magilan(i)cum at Alconétar (prov. Cáceres) and at Montealegre de Campos (prov. Valladolid).67 This feature ought not to be taken to suggest that these kinship groupings had wide geographical boundaries, cutting across Roman political limits. It is much more likely that they were very localized kinship groupings, with the replication of their titles due to the relatively restricted quantity of indigenous personal names.68 In some cases children appear to have been members of the same cognatio as their fathers: at Caesarobriga (modern Talavera de la Reina, prov. 65 Alaesus Triti f. [U]bon[ic]um: Albertos Firmat : no.  0001 González Rodríguez : no. ; cf. Amaenia Vironi f. Tritecu(m): Albertos Firmat : no.  0001 González Rodríguez : no. . Erguena Bouti (f.) Ammaricum: Albertos Firmat : no.  0001 González Rodríguez : no. ; cf. Talaus Toncetami f. Boutie(cum): Albertos Firmat : no.  0001 González Rodríguez : no. . 66 Calaeticum: Albertos Firmat : no.  0001 González Rodríguez : no.  0001 LICS – (Ávila); CIL .  0001 Albertos Firmat : no.  0001 González Rodríguez : no.  0001 LICS  (Guisando). Areinicum: Albertos Firmat : no.  0001 González Rodríguez : no.  0001 LICS ; LICS  (Ávila); CPILC  0001 Albertos Firmat : no.  0001 González Rodríguez : no.  (Villar del Pedroso); Matugeniqum: González Rodríguez : no.  0001 LICS  (Ávila); Albertos Firmat : no.  0001 González Rodríguez : no.  (Yecla). 67 Crastuniqum: CIL .  0001 Albertos Firmat : no.  0001 González Rodríguez : no.  0001 LICS  (Ávila); Albertos Firmat : no.  bis 0001 González Rodríguez : no.  (Cuevas de Amaya); Albertos Firmat : no.  0001 González Rodríguez : no.  (Langosto). Magilan(i)cum: Albertos Firmat : no.  0001 González Rodríguez : no.  (Alconétar); AE ,  0001 ,  0001 ,  0001 ,  0001 ,  0001 HEp. .  0001 .  0001 .  0001 .  0001 .  (tessera hospitalis from Montealegre de Campos, attesting a cognatio Magilancum). 68 On indigenous personal names from Lusitania, note Palomar Lopesa ; Untermann ; Grupo Mérida .
Family Relations in Roman Lusitania

Toledo) Ammia daughter of Pistirus belonged to the cognatio Pistiricum, to which her father Pistirus was presumably also attached.69 Similarly outside Lusitania at Argovejo (prov. León), Turennus son of Boddus was a member of the cognatio Boddegun.70 Clearer proof that membership could be inherited patrilineally is provided by an epitaph from Ventosilla y Tejadilla (prov. Segovia), where two sons explicitly belonged to the same cognatio (Abinicum) as their father.71 On the other hand, there is at least one case from just beyond the eastern limit of Lusitania where a daughter belonged to a different cognatio from that of her father.72 There is also one example from Barcebalejo (prov. Soria) of a son, C. Iulius Barbarus Medutticorum C.f., who shared the same cognatio as his mother, Aemilia Acca Medutticorum, but here it seems likely that the man who appears to be his father, C. Iulius Labeo Crastunonis f. Medutticum, was also a member of the same kinship grouping, in which case we would have another confirmation of patrilineal descent.73 It would also be useful to know how the nuclear family related to these wider kinship groups. However, there are no attestations from Lusitania of married couples both of whose cognationes appear on their epitaphs. From outside the province there are two (or possibly three) cases of a husband and wife belonging to the same kinship grouping, but two where they belong to different ones.74 So in the current state of knowledge we cannot determine whether exogamy or endogamy was the norm. The examples of husband and wife in the same clan might also be the result of wives changing their cognatio after marriage, on the analogy of Roman marriage cum manu. Albertos Firmat : no.  0001 González Rodríguez : no. a. Albertos Firmat : no.  0001 González Rodríguez : no. . 71 HAE  0001 González Rodríguez : no. . 72 Albertos Firmat : nos. – 0001 González Rodríguez : no.  (Malamoneda, La Puebla de Montalbán, prov. Toledo). 73 Albertos Firmat : nos. – 0001 González Rodríguez : no. . 74 Same cognatio: Albertos Firmat : no.  0001 González Rodríguez : no.  (Beleño, prov. Asturias); CIL .  0001 Albertos Firmat : no.  0001 González Rodríguez : no.  (Uclés, prov. Cuenca); for another probable married couple from the same cognatio, see previous note. Different cognatio: CIL .  0001 Albertos Firmat : no.  0001 González Rodríguez : no.  (Almadrones, prov. Guadalajara); CIL .  0001 Albertos Firmat : nos. ,  0001 González Rodríguez : no. a–b (Medinaceli, prov. Soria). 69 70

Jonathan Edmondson
In sum, the evidence is too patchy to allow us to discern how these wider kinship groupings related to the nuclear family, but they do at least provide striking evidence for the existence in certain parts of Lusitania (and in other northern regions of the Iberian peninsula) of family structures more extensive than the two-generational Roman nuclear family. Moreover, these extended kinship groups were still functioning in the third century and perhaps even into the fourth century .75 They are often assumed to provide evidence for the preservation of indigenous, pre-Roman social structures under Roman rule, and this may well be so.76 But it might also have been the appearance of Roman armies in the area in the second century  that forced the locals to form these wider kinships units out of concern for property or defence, a situation for which there are anthropological parallels.77 Their existence in the less Romanized parts of the province of Lusitania may also help to explain the relative prominence of extended kin as commemorators in centres such as the civitas capital of the Igaeditani, where no cognationes are yet attested, but where indigenous traditions (observable in personal names) seem to have remained quite strong even after the community had been promoted to municipal status following Vespasian’s grant of the Latin rights of citizenship to all the Spanish provinces in  –.78 Conclusion Quantitative analysis of funerary commemoration does reveal some interesting features of family relations among the ‘inscription-erecting’ classes in various towns of Lusitania. Nuclear families clearly predominated, but there were some subtle variations between different regions of the province. The most typically ‘metropolitan Roman’ patterns were found—not surprisingly—in Roman colonies such as Augusta Emerita, and in the more Romanized zones of the provinces (as, for example, the conventus Pacensis). Emotional attachment to the simplest form of family unit (based on the mother-father-children triad) was most preva75 76 78
González Rodríguez : . For this in Galicia, see Tranoy : –. For personal names, see Dias .
77
Cf. Fox : .
Family Relations in Roman Lusitania

lent in flourishing cities with strong urban economies, such as Emerita, in particular among those ex-slaves who were able to establish new family units after manumission or to solidify quasifamilial unions that their masters had permitted while they were slaves.79 But in areas where Roman settlers and Roman culture penetrated to a lesser degree, a greater reliance on wider kin is discernible and women (whether in their capacity as mothers, wives, or daughters) appear to have played a more prominent role as commemorators and were commemorated more often than in more Romanized parts of the western empire. This might suggest that ties of matrilineal kinship were stronger here than was normal elsewhere. But at the same time there is no doubt that patrilineal descent predominated. What we, therefore, seem to have is a stronger emotional emphasis being placed on bilateral kinship relations in less Romanized parts of Lusitania and of Spain in general than in other parts of the Roman west, and as a result a greater social valuation of women. Rather than matrilineality, it might be better to think of ‘matrifocal death rituals within a patrilineal descent system’.80 It is going too far, however, to think in terms of matrilineal descent, still less vestigial matriarchy, as has sometimes been suggested.81 After all, even Strabo commented of the Lusitanians that they ‘married in the same way as the Greeks’ (. . ), restricting his hostile comments about ‘gynaecocracy’ (or ‘women on top’) for the Cantabrians in the far north, where, he claimed, husbands paid dowries (or better, bride-price) to their wives, daughters were left as heirs, sisters were responsible for marrying off their brothers (. . ), and women worked the fields (. . ) and in the mines (. . ). Here too husbands took to their beds after their wives had given birth, happy to tend their sick husbands and to work the fields: the practice known to anthropologists as the couvade (. . ).82 Appian’s report (Iber. ) on the physical bravery of Lusitanian women who fought alongside their menfolk against D. Junius Brutus’ army in   needs also On which see in general Rawson ; Flory . To quote Morris : . 81 For example, Étienne et al. : . 82 On the couvade, see Rivière ; Broude , Rival . (I am grateful to my York colleague Thomas Gallant for these references.) 79 80

Jonathan Edmondson
to be read as an instance of a familiar ethnographic topos rather than necessarily relating to any real social practice.83 However, the continued existence into the second and third centuries  of wider kinship groupings in several, more remote parts of the province shows at the very least that a completely Roman model of family organization did not spread evenly across the entire province. Or even if it did, it did not necessarily obliterate all elements of pre-existing kinship and family systems. In Roman colonies, where Roman family law pertained, the development of Roman-type families is to be expected. In towns promoted to municipal status Roman legal norms came to influence how marriage and family relations were regulated here too, as is clear from the Flavian municipal charters from Baetica and especially from Domitian’s subscript of   to the charter for Irni, in which the emperor made it clear that certain forms of marriage previously recognized under local law were no longer valid now that the new municipium had been granted a Roman charter.84 But even in these communities, there was still a peregrine, native stratum, which did not necessarily have to fall into line with Roman norms regarding the manner in which they organized and conducted their relations with kin. In less urbanized regions of the province, there was less incentive to adopt a completely Roman model of family relations. It is in the distinctive patterns of personal names (determined within the family or at least the kin-group), in the survival of wider kinship groupings, and in the slightly divergent patterns of funerary commemoration that social historians can catch flickering glimpses of the fact that the inhabitants of Lusitania did not necessarily conceive of, and practise, ‘family’ relations in an identical manner across the entire province. The later history of family relations in Spain and Portugal is also marked by significant regional variation.85 This would suggest that the elements of diversity that we have been 83 Cf. Strabo . .  on the Cimbri, Tac. Agr. .  on the Britons, and Germ. .  on the Germans, with the comments of Rives  ad loc. 84 Note especially, lex Irnitana, ch.  (guardians), ch.  (assuming the existence of patria potestas in non-Roman family relations). On Domitian’s subscript regarding types of marriage, see Mourges . 85 See the various articles in the Journal of Family History, / () and / (), special issues devoted to family history in Iberia and in Portugal respectively. In general, see Douglass .
Family Relations in Roman Lusitania

able to discern, however imperfectly, in family relations in Roman Lusitania should not cause surprise. Although the spread of Roman power did bring a good deal of uniformity across its provinces, it did not obliterate, still less seek to obliterate, all local traditions and particularities. The manner in which family relations were organized in Roman Lusitania, it seems, provides no exception to this general rule.
This page intentionally left blank
 Family History in the Roman North-West Greg Woolf
The Limits of the Knowable I an ideal world, the study of the Roman family in the provinces would mean an investigation of the diversity of family life across the empire. What were the peculiarities of Dacian divorce and Mauretanian marriage, Turdetanian testation and Pannonian patriarchy? And so on. These questions have more than local significance. Much written on the Roman Family leaves the geographical scope of the institution unclear. How similar were the families on which Roman-style states were founded? How deeply into provincial societies did Roman institutions penetrate? Was family life left untouched by imperialism, the empire only a game for the males of the family, or was it transformed by it? Sadly the world is not ideal, and there are fundamental obstacles to the practices of family history in relation to Rome’s northern and western provinces. Put simply, there is too little evidence, and no reliable means have been found of extrapolating general truths from the few cases we can document. Classical ethnographies say virtually nothing (and nothing reliable) about family This chapter has been improved enormously by the friendly and acute criticism of the Press’s readers and of other participants at the conference at McMaster. It was an honour on that occasion for me to present a paper at the university of Edith Wightman, whose tragic death deprived Roman Gaul of one of its most expert investigators. Wightman was an alumna of St Andrews University and I write this surrounded by her own books which were bequeathed to the university. Edith Wightman never wrote directly on the Roman family, but several of her studies of Gallo-Roman society touched on the subject (Wightman , , and ). Her work was a model of how to combine literary, epigraphic, and archaeological data with caution and imagination. Its continuing importance for those of us working on Gallo-Roman topics of all kinds is impossible to overstate.

Greg Woolf
life before the conquest. Even in prehistory there was probably little uniformity among all the peoples of the region, and the law codes of medieval peoples speaking Celtic languages are no longer regarded as useful guides to Iron Age realities.1 Equally, this is not a region within which social history can be written from epigraphy.2 There are too few usable inscriptions to reach conclusions that are statistically significant.3 The north and west is poorer in epigraphy than most regions of the empire: not only are there fewer texts but they are rarely lengthy or detailed, and we have almost no documentary sources. Worse still, as historians have become more and more alert to regional variation in family structure and to the possibility of quite dramatic change over time, the cruder statistical approaches are shown to be even more deficient. In order to identify such variability, investigators must subdivide the gathered epigraphic evidence again and again, producing samples so tiny that even the most tempting correlations are unsafe. The epitaphs from these provinces do yield some local peculiarities of commemoration. They were tabulated, and compared to those of other provinces, nearly twenty years ago.4 Behind those variations in commemorative practice, peculiarities of family life may lie, at least for the inscribing classes. But even this is uncertain. Death rituals often vary considerably even when we have no reason to suspect other differences in family structure. A few monuments and grand inscriptions offer vignettes of family life. The Secundinii from Igel is one example, another is the family of Caius Julius Rufus which dominated early Roman 1 James  is the latest of a series of critiques that have demonstrated the enormous problems facing this approach. 2 For imaginative attempts to do this for other regions of the empire see Cherry  and Edmondson, this volume, both rather more optimistic than this author about the prospects for this line of inquiry. 3 Anyone who doubts the limitations of the Gallo-Roman epigraphic evidence need only look at Keith Hopkins’s () comments on Yves Burnand’s () learned contribution to an earlier colloquium on the Roman family in . Burnand’s epigraphic and prosopographic knowledge of Roman Gaul is without rival, yet even so the very paucity of the evidence means it is impossible to move from the anecdotal and particular to the general. 4 Saller and Shaw . Among their findings are that ante-mortem commemorations were relatively common in eastern Gallia Narbonensis, as in Noricum, Raetia, and Cisalpina (p. ); and that there, as in all civilian populations in the western empire, the vast majority of named commemorators were closely related to the deceased.
Family History in the Roman North-West

Saintes for three generations.5 Those families, certainly among the wealthiest and best connected of their communities, represented themselves as the very image of The Roman Family. But it is impossible to generalize from the tiny number of cases of this sort to broader social patterns and trends. At best, their monuments may be used to illustrate propositions made plausible by other means. It is no surprise, then, that evidence from these provinces has played little part in the debates over the Roman family that have flourished during the last two decades. Other chapters in this volume show the potential of Egyptian papyri and Greek epigraphy, of the rich and varied evidence for the Jewish family and the still unexhausted potential of the iconography, epigraphy, and literature of Roman Italy.6 Rome’s northern and western provinces, however, offer nothing similar. The focus of this chapter, then, is on plausibilities, not certainties. It is also, necessarily, a hunt for the familiar, that is for features that recall Roman families as we know them from elsewhere. The argument is as follows. Although it is not possible to demonstrate that large numbers of families were organized on Roman lines in the region, it is certain that many members of local communities were well acquainted with Roman notions of the family. Many heads of families had the power as well as the knowledge needed to reproduce them if they wished to do so. Since it will often have served their interests to do so, we must suspect many did. Yet families, as we know from our own lives, are rarely driven wholly by considerations of interest. The irrational dimensions of family life are beyond our reach, in this part of the empire at least. Arguments from Analogy If we are to deal with plausibilities it is a good idea to narrow down the range of what we might consider. Comparative evidence often provides a good starting point, and many of those who 5 On the Secundinii, Drinkwater  and . On C. Julius Rufus and his relatives, Woolf . 6 To which might be added the papers gathered in Kertzer and Saller  and Rawson and Weaver  on Italy and the important study of van Bremen  on Asia Minor.

Greg Woolf
have written the social history of barbarian Europe’s encounter with the Roman world have been inspired by the better documented histories of more recent expansions and imperialisms. For Roman Gaul, the Scottish Enclosures and the Atlantic Slave Trade have both been invoked. For some, the well-attested trade in wine into Iron Age Britain and Gaul has suggested the working of economic forces similar to those manifested in modern world-systems.7 Since analogies have provoked imaginative reconstructions and suggested novel interpretations, can they help with family history? Family structure and kinship were high on the agenda of those anthropologists who, for much of the twentieth century, recorded the social systems of so-called indigenous peoples on the verge of their incorporation into nation-states and the world economy. Such societies were often envisaged as essentially stable entities subjected to dramatic change only by the arrival of European colonialists. When, in Chinua Achebe’s () phrase, ‘Things Fall Apart’, the family, it was commonly held, was often one of the first things to go. Ethnography does not, however, supply general rules about what happens to ‘traditional families’ when drawn into larger political and economic systems. It is easy enough to find cases where new pressures (or new opportunities) have split up large family groups. The growth of migrant labour in the wake of industrialization and urbanization is a common direct cause. Changes in demography linked to medical developments and the disease and diet regimes characteristic of underdevelopment are important indirect causes. Not all pressures lead to social disintegration. Missionary activity, legal reforms, and some social policies have strengthened some familial ties, if often at the expense of others. Gender politics in particular have often been transformed. Younger men may acquire greater opportunities for gaining wealth outside familybased activities; the state may determine to deal only with male heads of family; previously communal property may be registered in the names of individuals; foreign aid or health care may be 7 e.g. among many others, Stevens ; Nash , the latter drawing on the anthropology of Meillassoux and Terray. The technique is best exemplified in the work of prehistorians such as Michael Dietler and Michael Rowlands. On the use of world-systems theory in relation to Rome, Woolf  and . See also Webster .
Family History in the Roman North-West

directed primarily to mothers of young children and so on. In many parts of the modern world urbanization and globalization together have led to the emergence of new employment opportunities open to young women from secretarial work to domestic cleaning and the sex industry. Education policies can even alter the relations between children and their parents, for example by empowering children to speak, read, and write ‘official languages’ better than their parents can. On the other hand this can lead to new burdens being placed on them. It is clear enough that understanding the precise impact of change involves in each case a careful analysis of a wide range of variables. The local effects of incorporation are enormously variable. If in some societies the social and economic power of older males has actually been strengthened, in others younger males or even women and children seem to be the ‘winners’. But none of this is easy to predict. Modern experience cannot provide a single pattern that can be applied to the Roman world. Nor is it completely obvious that imperialism always brings change, or that pre-colonial situations have always been static. Arguably the entire project of trying to gauge the impact on family structures of incorporation into a wider world is fundamentally flawed. The first objection to this approach is that families are not passive objects on which social forces act. ‘Traditional’ families were in reality no more genuinely stable or well balanced than ‘indigenous’ societies. It is metropolitan historians of empire who imagine empire’s victims in this way, when they allocate agency and responsibility only to colonizers and government. The pace and direction of change may well be dramatically altered by the colonial encounter, but all societies have their own historical trajectories. Pre-Roman Europe was no exception. The mortuary archaeology of northern Europe allows us to trace a slow growth in the size of human communities, the emergence of social ranking at least by the late Bronze Age and the appearance of prominent families, ones who buried even their female and infant members with extravagant investment of goods and time. The European Iron Age saw further changes, in particular enormous increases in social and economic power. There is some reason to believe in dynastic marriages between powerful chiefs and in the development of fictive kinship alongside clientage. Whatever else

Greg Woolf
archaeologists of late La Tène Europe disagree on, there is a consensus that it was a period of rapid social change. It is difficult to imagine family life being unaffected by all of this. The second objection is that ‘family structure’ is itself an artefact of anthropological and sociological analysis, the product of a particular style of imperial historiography and ethnography. Like ‘society’ and ‘culture’, ‘family structure’ and ‘kinship systems’ were rendered apparently stable and orderly by the very functionalist and normative paradigms with which they were investigated. The effect was reinforced when they were described through conventions like ‘the ethnographic present’ which notoriously stressed statics over dynamics and edited traces of ‘modernity’ out of the picture. Family structure, in this sense, has been produced by the way family relations were studied. It is preferable to treat each family as a transient expression of the changing balance of power among a group of more or less closely related individuals differentiated by gender, by age, by access to resources, by social skills and much else, but united by affective bonds and some common interests. Families are transient corporations for several reasons. Some change is often manifested in kinds of conflict that are perfectly normal and have been well studied for Roman society.8 Change deriving from patterns of social replacement can be thought of as cyclical. Chayanov showed that many peasant families underwent a cycle from poverty through prosperity to poverty as children moved from being dependants to becoming a source of productive labour before moving out to establish their own families. Indeed this kind of change is shared with the family groups of other primates. But if all change were like this it would be easy to adjust analyses of family structure in terms of norms to take the cycle into account. In fact, many if not most families are composed of individuals whose interests do not always coincide, and whose cooperation is provisional. New opportunities may suddenly make it possible for individuals to strike out on their own, or to alter their position within the family. The key point is that family structure has no inertia, and it can be transformed very rapidly when circumstances change. 8 Dixon  for insightful remarks on conflict and also on collective expressions of family solidarity.
Family History in the Roman North-West

All this is a familiar scenario today, when effects of this kind are produced at an unprecedented level, owing to the complexities of western division of labour and the power of capitalist enterprise. But even Roman society was more complex than most of its predecessors in the west.9 ‘Stadtluft macht frei’ went the medieval proverb. There is no Latin equivalent and there are few stories of Roman Dick Whittingtons seeking their fortunes in the rapidly growing metropolises of the west. But it is difficult to see how cities like Rome and Carthage can have grown or remained large without a constant influx of population, not all of whom can have been slaves.10 Even the much smaller cities of Rome’s European empire recruited their craftsmen and labourers from among former peasants.11 The army offered chances for individual males to move outside their natal societies, either permanently or, as seems to have been the case among the Batavii, simply for a portion of their lives.12 Conceivably commerce and some labour-intensive industries, such as pottery production, promoted the mobility which is only partly visible in the epigraphic record.13 Finally, the Roman peace, and the existence of empirewide media of communication including the Latin language, Roman coinage, and Roman law, expanded the horizons of many westerners. Put otherwise, there were now many more routes out of the family or ways available for family members to enrich themselves individually. Perhaps the best indication that Roman families were far from stable is paradoxically provided by repeated claims for the authority of ancestral custom (mos maiorum). Ancient historians for a long time behaved like those ethnographers who investigated family structure through questioning ‘expert’ local informants (usually high-status males) about social ‘norms’ and conventions. 9 The argument here owes much to Runciman  pointing out how increased social differentiation leads to higher rates of social mobility. 10 Hopkins  is the classic statement of this, and the only real doubt is over how voracious these cities were as consumers of population. 11 Woolf  for further discussion. 12 On the Batavii see Roymans . 13 Wierschowski  is the fullest study, but see also Horden and Purcell  on the phenomenon of mobility in general, with arguments that apply outside the Mediterranean world, however defined. On mobility among potters see Hartley : the example is important because it indicates mobility well below the social level of those normally commemorated epigraphically.

Greg Woolf
Anthropologists are now very aware that the supposedly normative orders extracted are often interested statements, usually heavily gendered. Social and religious rituals, themselves controlled by wealthier members of society, mystified the power of the dominant members of family groups. Roman historians too need to consider quite how complex the relationship is between reported norms and actual practice. Romanists think naturally of the paterfamilias in this context. It is now reasonably clear that a combination of high age at first marriage for males and low life expectancy would have made the large extended family dominated by a paterfamilias relatively rare in practice.14 So why did Valerius Maximus, Livy, and so many other authors rehearse such extraordinary claims about the prerogatives of the male head of the Roman family? I suggest we are dealing here with a ‘norm’ of family life promulgated by the powerful, perhaps promulgated all the more energetically in the face of social change. These images are ideological, then, attempts at reinforcing differentials of power within Roman families that were already under threat. Ethnographic analogies suggest that dramatic changes in family structure—including assertions of traditional values— often manifest wider changes in the balance of social power. In times of rapid social change, we should expect family structure to be susceptible to sudden transformations, even when it seems (or claims) to be deeply rooted. There are some brakes on change. Rational choice analysis applied to the interests and strategies of individual actors explains only so much. Quite apart from the success with which rituals and symbols reproduce images of the family that suit those who are powerful in any society, there is affection, love, piety, possessiveness, and the rest, emotional states conceptualized differently from one culture to another, but to some degree universal. The affective dimensions of family life derive their power from deep-seated priorities we shared with our nearest non-human relatives. Virtually all humans live in families of one kind or another, and few seek to emancipate themselves from family life for very long, preferring to remodel their birthfamilies or create new ones. All that said, we should be prepared for the possibility of major changes in the family lives of provin14
Saller  for the definitive demonstration of this.
Family History in the Roman North-West

cial Romans. Predicting the precise nature of those changes is less straightforward. Learning the Roman Family What access did individual provincials have to new and different ideals of family life? Here, at least, it is easy to provide an answer . . . at least for the elite. That qualification applies to the rest of what follows, since family structure is notoriously variable according to social status and wealth. Here at least it is less plausible that innovations piloted by the local elites would be disseminated more widely in society through imitation or other mechanisms. Most western elite males learned Roman family values at school. It is certain that the standard Roman educational system became widespread in the west from early in the Principate,15 not long, in other words, after it reached the form it would have until late antiquity. Tacitus describes a school of the children of GalloRoman nobles in Tiberius’ reign, while orators and other writers are known from Gaul, Spain, and Africa in the late first century . Elementary education involved becoming familiar with a series of canonical texts. By the middle of the first century  this canon included many of Cicero’s speeches and Virgil’s Aeneid, and perhaps Terence and Horace as well. Virgilian quotations are frequent in the graffiti and epigraphy of the western provinces. It is easy to see how Roman family values might be learned by those working with this canon.16 Cicero’s speeches provided many exemplifications, positive and negative, of Roman family norms.17 The Aeneid too mythologizes familial relations, idealizing Aeneas’ relations with his father and his son, and offering competing models of matrimonial relations and of male and female virtue. The next stage of education was a series of rhetorical 15 On education in Gaul see Haarhoff  and Marrou : –. Suetonius’ Lives of the Grammarians along with the pedagogic works of the elder Seneca and Quintilian allows this process to be followed throughout the west. For discussion of its role in relation to the urbanistic and civilizing projects in the st cent. see Woolf . 16 Morgan : – for a discussion of this mostly in relation to Greek education in the Roman period. 17 Treggiari’s chapter in this volume explores the subtleties of the Ciceronian discourse of the family.

Greg Woolf
exercises designed to teach young men not only to speak, but also to act, as Roman males. Education inculcated normative gender roles and embodied them in gesture and posture as well as speech.18 Making practice speeches, students took on the roles of historical and mythical figures.19 Family relationships, especially those of fathers and sons, were common in both kinds of exercise, to judge from the examples provided by the elder Seneca. The same is true of the exemplary anecdotes gathered by Valerius Maximus, ostensibly for the use of orators. By the end of their formal education, then, the male children of the western nobilities had learned how (some) Romans imagined the family, and were practised impersonators of them. These exercises did not compel them to accept these values, any more than they enforced moralizing norms in Rome itself. Education might also have empowered them to break the rules, to play against type and adopt and create minority and counter-cultural models of behaviour.20 Educated provincial males knew, however, how Roman fathers and sons were expected to behave. Equally they could measure their female relatives’ behaviour against the norms made patent in these late Republican and Augustan texts. An authoritative image of Roman family life, in other words, was available to them. Families in Law Epic poetry and rhetorical exercises were not the only media through which Romans theorized about the family. Of all media, law made the clearest statements and the most detailed. Modern historians of the Roman family have mined Roman legal texts for details of family life, and have also disentangled from them more fundamental norms, constructs, and ideals such as patria potestas and familia itself.21 Only a few westerners will ever have come to understand the Roman family in this way, and then only very late in the principate. Nevertheless, individual laws and civic charters communicated an ethic of the family, both when promulgated and also when used by enfranchised western aristocrats to further their own and their families’ interests. 18 20 21
19 Beard . Gleason . A major theme of Gleason . This section owes a great deal to Gardner .
Family History in the Roman North-West

Enfranchisement is the easiest stage of this process to map out. The coloniae of Gaul and later of Germany and Britain were the nearest thing in the north-west to Italian municipia, entire communities of Roman citizens living by Roman laws. Can we imagine un-Roman families within societies like this? Certainly we can, given the vast cultural differences between Roman colonies in this period. Nevertheless the citizens of Orange and Lyon, Nyon and Narbonne and all the others were legally bound by Roman laws on marriage and adultery, inheritance, property, and status including the moral legislation of Augustus and Tiberius. More westerners lived in communities that were formally Latin or peregrine in status. Only from Spain have charters survived for Latin communities, although they were certainly widespread. Peregrine communities must have varied considerably, despite some indications of a convergence on Roman institutional forms. As far as family life went, however, conditions were probably broadly similar throughout the western provinces. To begin with, Roman families—in the sense of families, the heads of household of which were full citizens—must have been rare in such communities. They will have coexisted alongside families of many other kinds, families governed and defined by local laws and customs. Perhaps Roman families formed something of a closed circle within some communities: the institution of the conventus of Roman citizens remained important in Gaul, even after most Gallo-Romans were enfranchised.22 But there would also be local pressures not to allow the chance or timing of enfranchisements to disrupt existing relations of amity. The situation eased over time, since during the two and a half centuries between the Augustan reorganization and Caracalla’s edict the Roman family in this strict sense must have expanded at the expense of competing forms. Grants of colonial status to communities like the Treveri converted hundreds of families at a stroke, while magistracy in Latin municipia, immigration, the return of auxiliary veterans, and (perhaps especially importantly) emancipation by Roman citizens promoted gradual but presumably exponential growth. Family politics in Latin and peregrine communities must have been rather complex. Even if the emperors had intended to 22
Audin et al. .

Greg Woolf
create communities in which citizen aristocracies ruled over noncitizen masses, the actual effect would have been rather different. The propertied classes would for a long time have been divided between citizens and non-citizens. Likewise, citizenship would also have been found among the masses, notably among ex-slaves of local citizen nobles. In these circumstances, we might expect Roman conventions and practice to become widespread, even among those not legally entitled or obliged to behave in accordance with Roman law. It has recently been argued that charters of the kind issued to Latin municipia like Irni encouraged many to live as if they were Roman citizens.23 Chapter  of the lex Irnitana gives explicit instruction to the Irnitani to deal with each other as if they were Roman citizens in any matters not covered by the law itself. But this fictio does not show a lack of concern with legal status, or with the effects of municipalization. The Spanish charters include some indications that the Roman authorities believed that the grant of municipal status, with the consequent changes in personal status for their inhabitants, risked undermining ancestral institutions including the family.24 Domitian’s letter, appended to the lex Irnitana, indicates that marriages must in future conform to the requirements of the law. Special problems arose in families which had been incompletely enfranchised. Crudely simplified, although a magistrate’s agnatic ascendants and descendants were enfranchised at the same moment as he was, some significant relatives were omitted. Siblings were not included making them in principle ineligible to inherit from their newly enfranchised fathers or to receive legacies from their citizen brothers. Perhaps, in practice, few magistrates had living fathers who had not already obtained citizenship.25 No doubt in many such families all the brothers took Gardner : , cf. Gardner . The idea that these clauses of the charters provide evidence for the existence of institutions such as patria potestas in Roman Spain prior to the Flavian period seems unsafe. Rather we may simply be seeing the limits of the Roman imagination about alternative kinds of family life and an indication of the kind of institutions they were concerned to promote rather than undermine. 25 To the arguments of Saller  about the likely effects of demographic conditions on family structure may be added the likelihood that in many Latin communities a ruling class with Roman citizenship would have emerged in the course of the nd cent. with the result that many candidates for magistracy would already be citizens. 23 24
Family History in the Roman North-West

their turn at high office. But there would have been exceptions, and sisters remained vulnerable. Analogous problems would arise in some marriages. A magistrate might, in principle, find that he had condemned his future grandchildren to illegitimacy by invalidating his children’s marriages.26 What solution the elites of Latin communities found to these problems is unknown. When all parties agreed, no doubt the letter of the law might safely be ignored, local laws might be applied, or the fictio that all were Roman citizens might be resorted to. But in disputes, for example over inheritances, it might be less easy to reach resolution between parties of different statuses, one of them claiming the exclusive rights of the Roman citizen. However conflicts were resolved, the elite would be drawn into a debate about the nature of the family in which Roman normative models would be a constant point of reference. In debates of this kind, there might be distinct advantages to both sides in conforming to Roman norms, at least temporarily. Similar questions occurred in some peregrine communities. It is probable that many of the Julii prominent in the historical accounts and epigraphy of first-century  Gaul were descended from aristocratic warband leaders enfranchised and enriched after service in triumviral campaigns and the Augustan wars of conquest.27 Well before municipalization, then, an influential group of citizen families was emerging in Gallo-Roman communities. The social mix of communities with only a few dozen Roman families offered several choices to these citizen elites. They might choose to form closed circles of the kind already mentioned, bound together by webs of intermarriage and testation.28 The grandest families might begin to seek husbands and wives for their children among the richer citizen families of neighbouring communities. It is not possible to show this strategy was widespread, but analogy with the better attested situation in Achaea and Asia makes it likely. Yet many would presumably wish to marry within local elite circles which were still part-citizen, partperegrine. The obvious solutions would be either to marry under 26 On mixed marriages in the provinces and the disabilities imposed on children arising from them cf. Cherry : –. 27 Drinkwater . 28 The family of Julius Indus is a good example, on which see Wightman : –.

Greg Woolf
local law or to marry as if both parties were Roman citizens. Given goodwill on all sides, any resultant inheritance problems could be sorted out amicably. This was risky, but no more risky than the decision many people today make to remain intestate or to live together outside matrimony. Ancient practice may have been no more rational than our own. Despite all the uncertainties, it is clear that a degree of choice existed in the Roman provinces over how far to conform to what they supposed to be the norms of Roman family life. Roman law was not so much a straitjacket as a way to think about and discuss the family, as well as support for families of a particular type. Disputes of various kinds brought this Roman legal theory of the family out into the open. It does not follow that all members of western elites accepted Roman notions. But they were certainly aware of the Roman model. Equally, what we call the Roman law of family had itself been developed over centuries as a means of extending and facilitating the control exercised by adult males over their property and dependent relatives. Unsurprisingly it was often useful to provincials. While we cannot measure the extent to which it was used to engineer choice,29 there is no doubt that it enabled provincials to choose to constitute Roman-style families if it suited or attracted them to do so. Un-Roman Families in Provincial Houses? One drawback with the method pursued so far is that the hunt for the familiar is inevitably one-sided. Roman law and Roman will always illuminate the Roman at the expense of the other perspectives. Yet there are no local texts to answer, correct, or supplement Roman municipal laws and pedagogic handbooks. The only evidence that offers a chance of identifying peculiarly local species of family life is archaeological, specifically the traces of houses, urban and rural. Evidence for house design is accumulating in the provinces, and modern excavations no longer routinely interpret all features of these structures in terms of Italian models and agronomical texts. Yet, in the provinces as in Italy itself, it is not a straightforward matter to excavate the family. 29
For example through testation, cf. Champlin .
Family History in the Roman North-West

The point may be illustrated in relation to Gaul, where some of the best traditional and also some of the most radical interpretation has been conducted. Recent years have seen a substantial increase in the excavation and publication of Gallo-Roman domestic sites.30 The acts of two conferences held in the mids offer new syntheses on Gallo-Roman rural residence and urban housing respectively.31 Not only is the scale and elaboration of the larger residences clearer than ever, but so is the diversity in house plans. Some examples are difficult to interpret except in terms of the translation onto provincial soil of building types developed in Italy. The phenomenon was long ago recognized for public buildings.32 The first Gallo-Roman peristyle houses and atrium houses too are difficult to explain except in terms of the work of Italian-trained architects and craftsmen. Other types seem rather to be an application of Roman constructional techniques, materials, and decorative elements in the rebuilding of massive rural residences first constructed in the late Iron Age.33 Controversy really focuses on how these various house types should be peopled. Should we imagine Roman families going with Roman architecture, or can traces be found of un-Roman families in local peculiarities of structure and construction?34 Social analysis of urban housing in terms of the family has focused mainly on the Vesuvian cities.35 Initial research sought to identify normative house plans, or features so common they could be said to structure Roman houses, and then to relate them to features of Roman society known from literary sources. 30 I am grateful to Simon Esmonde-Cleary for bibliographical help with this part of the chapter. 31 Bayard and Collart  and Balmelle et al. . The conferences themselves took place in  and , and their publication marked a new high standard in documentation. 32 Ward-Perkins . 33 Woolf : –, a discussion indebted to those of Percival  and Millett . 34 Smith  for the most trenchant arguments that non-Roman families can be detected in the villa plans of Britain and northern Gaul. For discussion see below. Reece (no date) offers limited support for the hypothesis. For thoughtful discussions of house types and their social correlates, cf. Hingley  on Romano-British material, and on the Lower Rhine the publications of the Pionier-project ‘Power and Elite’ to which Slofstra  is a good introduction. 35 Wallace-Hadrill ; Bon and Jones ; Laurence and Wallace-Hadrill  give an idea of the range of the debates.

Greg Woolf
Subsequent studies have examined the differentiation of Vesuvian houses, again in relation to categories important in Roman discourse such as luxury and wealth, leisure and work, and ‘Greek’ as opposed to ‘Roman’ elements. House designs were held to reflect customary patterns of usage, from ritual such as the worship of the Lares and Penates, through formal social routines such as dining and bathing, to the mundane organization of sleeping and cooking. The various valuations of these repetitive activities were manifested in the design choices of those for whom these houses were built, modified, and decorated. Houses could be thought of as the exoskeletons of long dead family units, another Pompeian plaster-cast that revealed the shape of what once had been living social tissue. Various nuances have since been added to these pioneering accounts, and others might be. First, there is a distinction to be drawn between what was normal in the sense of common, and what reflected norms in the sense of publically articulated notions of proper and normative behaviour. Some patterning relates to this first, uncharged, sense of normal. That storage rooms are often close to kitchens, for example, need carry no ideological charge. Second, houses do not simply reflect social structure, but play some part in shaping the lives of all who inhabit them. So some aspects of house design may be seen as providing ideologically charged visions of family life that, in some respects, served the interests of the powerful, men rather than women, adults rather than children, the free rather than their slaves, and so on.36 One might even see houses as part of the means by which the powerful invested their wealth so as to influence the way social inequalities were reinforced and replicated, so long as it is allowed that the rich will often have been conformists rather than innovators in this respect. It is easy to exaggerate the social force and significance of architecture. If atria were highly distinctive stages on which were performed a range of family rituals from common cult to the reception of clients, the function of many other rooms could be modified with a simple change of furniture.37 The continua between public and private, between grand and less grand spaces, 36 37
Alston ; George ; Flower . e.g. Allison  and Berry .
Family History in the Roman North-West

and the like account for only a small part of the elaboration and diversity of known houses. In other contexts, anthropologists have focused less on the normative aspects of housing than on the potential that dwellings offer their inhabitants for self-fashioning and individuation.38 The approaches are complementary. If normative analysis tells us why atria are so important, looking at differences helps explain the attractions of those decorative media, such as wall-painting and mosaics, which can use a small range of materials and stylistic variants to make every room in every house distinctive. House architecture and decoration offered their Roman owners ways of differentiating themselves as well as ways of conforming to particular social or cultural norms. Part of being Roman involved knowing precisely how to balance between social status and tasteful individualism. The modern analogy would be the poise needed to situate oneself between banality of taste and eccentricity. When we turn to housing in Roman Gaul, then, the appropriate question is not, Did Gallo-Roman house owners conform to Roman styles? but rather, Did they know how to manipulate these complex assemblages to signal novelty and tradition, conformity and individualism in an appropriate way? Put otherwise, does house architecture suggest that Gallo-Roman nobles could play at Roman Families if they wished to do so? The start of an answer is provided by the massive domus PC re-excavated by the University of Lausanne working within the international project of excavations on Mont Beuvray in Burgundy.39 The house closely resembled contemporary Italian housing. Its size—well over , square metres in surface area— amazes. In its fifth phase, constructed around  , the house incorporated an atrium, equipped with an impluvium, and a large peristyle. Some traces of wall-paintings have been found, 38 e.g. Miller . Admittedly his case study deals with a very different social context, modern housing schemes in which the architecture itself is uncompromisingly uniform and personalizing spaces with decor assume a new importance. 39 In advance of the full publication of this structure the best account is included in the very full Rapports annuels d’activité scientifique produced by the Centre archéologique européen du Mont Beuvray. I am grateful to Vincent Guichard for providing me with copies of these reports and for his hospitality at Glux-en-Glenne.

Greg Woolf
apparently in early Pompeian third style.40 PC is not an isolated example. A series of large houses with atrium and/or peristyle is known from this area of the hillfort, surrounded by a number of much smaller houses organized in a systematic fashion.41 The complex as a whole dates from the last half of the last century , between the Caesarian conquest and the Augustan reorganization of Gaul, and some parts of it remained occupied until   or , some time after the foundation of Augustodunum, presentday Autun. Here then is proof that some Gallic notables at least could, from the very beginning, acquire perfectly good Roman houses to live in. The dossier of urban housing from southern Gaul provides many more examples of relatively large urban houses, with surface areas of several thousand square metres, many equipped with wallpaintings, mosaic floors, hypocausts, and the like. More peristyle houses have been identified than houses with atria. Entire quarters have been excavated in a few cities: those of Glanum, Orange, St Romain-en-Gal, and Vaison are the best documented. Large and small houses exist side by side, and porticoes with shops line major streets. There is little sign of specialised quarters outside the monumental centre. The organization of space, in other words, is broadly comparable to that in the Vesuvian cities. A number of northern cities have also produced evidence of large houses equipped with similar amenities. Internal courtyards feature in the larger houses known from Amiens and Reims. The house plans show considerable variety, and not just in size. A few are simple structures with just a few rooms, usually interpreted as in the tradition of small rural structures. But the larger houses are better known. There is no simple trend in size or elaboration over time. A few very early houses, in both the north and south, were very large and elaborate, with decor equivalent to that of contemporary Italian domus. Large houses appear all over Gaul in the later first century  but small, irregular houses never disappear entirely from Gallo-Roman cityscapes. The variety of house types at the upper end of the size range does not include any types that are unique to Roman Gaul, but then the larger houses of Italy 40 This style appeared in Lyon and Vienne from around  , within a few years of its first appearance in Campania, and is widely attested in southern and central Gaul. 41 Meylan .
Family History in the Roman North-West

and Africa are also fairly variable in detail. There is no gradual convergence on a canonical Roman house-type that might accommodate a canonical Roman family. This is no surprise, given the way the Vesuvian houses are now interpreted. The strongest case for locally specific house types that might reflect Gallo-Roman family structures has been made in relation to the grand, rural residences of the northern provinces. Earlier generations termed these villae and saw in them signs of the spread of Italian systems of agriculture and residence of the kind described by the agronomists. The revelation that most were built and inhabited by provincials rather than settlers led to a search for local peculiarities. What was spreading was a set of constructional techniques and associated building materials. Typical features included roofs of tegulae, walls built on stone socles, carpentry based on a variety of Roman joints and extensive rectilinear plans. The larger examples had mosaic floors and bathhouses, and sometimes wall-paintings, towers, and second storeys. Some were located in imposing locations, on hill tops, in the bend of a river, even within an Iron Age hillfort at Villejoubert in the Limousin, and might be associated with tombs and shrines. There is no doubt that in some regions of Gaul at least, the building of enormous rural residences, with extravagant façades and approaches, reflected social competition among the landowning classes. How far can the differences between the commonest GalloRoman villa types and those of Roman Italy be explained in terms of differences in social or family structure?42 It has been suggested that the plans of some of the largest farms show traces of a ‘unit system’, each unit comprising accommodation and other facilities for a single family. If so, the several families resident in a villa might then form a kin-group who together worked an estate owned in common, or one which had been divided among the heirs of a single head of household. Rural residences would provide some hard evidence, at last, for the existence of nonRoman styles of family in Gaul. Unfortunately the evidence for this reconstruction is ambiguous. While it is true that house plans are one respect in which 42 Smith  is the culmination of a number of influential conference papers and short articles.

Greg Woolf
these structures least resemble Italian farms,43 it is difficult to identify the functions of rooms from plans alone, especially when these plans have not been the result of extensive excavation. Even where separate residential units have been identified, there is no real way of deciding who inhabited them. The labour systems used on farms of this kind are uncertain.44 There is no reason to believe that local family structure closely resembled that described in the early medieval Welsh law codes.45 Finally, if students of the Vesuvian cities are now less sure that Roman houses there always reflect the standard Roman family, then deductions from the much-less-well-preserved remains of GalloRoman farmhouses must be treated with caution. None of this means that Gallo-Roman farms were the residences of families identical to those that inhabited Pompeian houses. It simply means that these structures have little to tell us about family life in Roman Gaul. What the archaeology of the Roman house in Gaul shows is that those who built and commissioned houses had their choice of virtually every component used elsewhere in Italy and the western provinces. There are many rooms, decorated with mosaics and wall-paintings, that would have served as perfect venues for social rituals such as cenae just as the great town houses on Mont Beuvray could have played host to salutationes and the like. How far these spaces were exploited in this way, of course, is another matter. The Seductions of Roman Patriarchy The leading men of the Gallo-Roman communities, through their education and citizen status, in their towns and in their residences, possessed all they needed to join the circle of Roman families. Their plate, pottery, and wine cellars, like their triclinia, opened up the possibility of joining the social rituals through Percival  remains the best single-volume account of the subject. Whittaker ; Wightman : –. 45 An idea suggested by Stevens  and developed by Smith . Whatever view is taken of the existence of ‘Celts’ in antiquity, a matter of considerable debate, it is certain that there were enormous variations in the social institutions of the various areas that have been considered ‘Celtic’ at one period or another. 43 44
Family History in the Roman North-West

which family life was put on display. The rights conferred on them by the ius civile empowered them to organize their families and property very much as they wished. The investment must have been considerable. The only question that remains is whether or not they chose to make Roman families for themselves. It was pointed out above that the number of Roman families, in a formal sense, increased as a consequence of successive enfranchisements. Today, we usually think of the spread of Roman citizenship in terms of the recruitment of individuals to the Roman citizen body. Perhaps this is so partly because citizenship, at least since the French and American revolutions, has been most important as a way of describing the relationship of individuals to nation-states. Roman perspectives were perhaps different. When they offered enfranchisement to magistrates of Latin communities, to auxiliary veterans on discharge, or to Roman allies as commemorated in the tabula Banasitana grants were often made explicitly to families. Only manumitted slaves seem to be exceptions to this rule, and even then their enfranchisement could be viewed as bringing them closer within the family that had owned them. Enfranchisement, among other things, was about entering an interconnected network of families. The ancient right of conubium, along with the capacity to inherit from other Roman citizens, makes it clear that in some contexts the populus Romanus was still thought of as bound together by endogamy. Romans might not celebrate common ancestry, but through citizenship they celebrated an alternative style of kinship. Historical analogy helps us envisage one way that this might have been conceptualized. Peter Laslett, the founding father of family history, once castigated historians of early modern England for their ‘failure to look realistically at the familial texture of society. What does the word England mean, for the year , shall we say?’ He answered in the following terms: England was an association between the heads of such families, but an association largely confined to those who were literate, who had wealth and status, those, in fact, who belonged, with their families as part of them, to what we have already called the ruling minority. Almost no woman belonged to England as an individual, except it be a queen regnant—scarcely a woman in the ordinary sense—or a noble widow and heiress or two, a scattering of widows of successful merchants and

Greg Woolf
yeomen. No individual under the age of thirty was likely to be a member, except in the very highest reaches of society, and very few men who had never been married.46
By ‘such families’ Peter Laslett meant a set of relationships that had at its heart a nuclear family but was also a business concern, and might in certain senses include servants and apprentices alongside blood relatives. There are clear resemblances between this conception of the early modern family, and the way the Roman familia functioned as a central component of the legal institutionalization of the Roman economy.47 Rome, naturally, represents a different cultural matrix and one within which family life was institutionalized slightly differently. It has been insisted, for instance, that historians recognize the ‘slave context’ within which the Roman family and the normative roles of all its members were defined.48 Adapting Laslett’s characterization of patriarchal society to suit the Roman state would involve some other changes: deletion of the possibility of queens regnant and translation of ‘successful merchants and yeomen’ into more disciplinary correct terminology. But, with those slight modifications, we would perhaps have an authentic representation of how some Romans understood what we call the Roman state.49 Authentic, that is, for some of its members. This vision too is ideological, an interested view, and it was never the only way that Rome was defined. But the widespread metaphorical usage of familial terms shows their centrality in Roman conceptions of the state. Senators were addressed as patres conscripti and each emperor adopted the title pater patriae. If the public cults of 46 Laslett (: ) unchanged in Laslett (: ) except that he expands his initial question to read ‘What does the word “England” mean, for the year , shall we say, the England of politics and political history, the England of the older textbooks?’ For England read Rome, for  read  , for ‘the older’ read ‘modern’? 47 Wallace-Hadrill (), and Aubert () have begun to show some of the ways the familia, including slaves and ex-slaves, managed a wide range of economic ventures. 48 Saller : . 49 This is not the same claim as that which Saller () challenges, that Roman families were dominated by fathers to an extraordinary degree, or that Roman fathers, especially in early Rome, had virtually unlimited power over all their dependants.
Family History in the Roman North-West

Rome were managed by the patres conscripti much private cult was managed by the heads of households. Roman law was a powerful medium for communicating these ideas.50 Roman citizens who were not patres were not fully independent agents, and patres mediated between society at large and the other members of their households.51 Members of the familia could get their rights only via the pater, and he was responsible to some extent for their actions. The myth of the all-powerful paterfamilias tells us something about Roman society, even if it does not describe reality. This myth of patriarchy maybe suggests one attraction Roman family life might have had for those elite males able to bring it about around them. Gaulish chieftains would certainly not have had their power over their dependants diminished by adopting the Roman family. Roman patriarchy offered power to old over young and to males over females. When the enfranchised were young, citizenship and the spoils of Roman warfare may have helped them evade the power of their own elders. But their children, in a Roman family, would never have the same opportunity. The possibility of living family life the Roman way emerges as a potential incentive for adult males to sign up to the Roman order. The family, and the socialization entailed in growing up within one, becomes a key institution for the integration of new Romans. This should be no surprise. All expansionist societies need to reproduce their core institutions in new territories. During the Middle Ages, fiefdoms and bishoprics were rolled out across Europe in all directions. European empires in the nineteenth century set up law courts and companies, schools and churches in the same way. We should expect institutions that were equally central at Rome, like the Roman family and Roman slavery, to play similar roles in the expansion of Roman society.52 Perhaps the key difference is that the Roman family was never compulsory in the Roman provinces. Even when mass enfranchisements like those associated with colonization suddenly created hundreds of ‘Roman families’ it would have been easy to evade Roman norms in favour of diverse local practices. In the end, it is consideration 50 51 52
So for Edwards (: –) law was ‘a moral discourse’. Gardner  passim. Purcell, forthcoming, for a compelling vision of how this was achieved.

Greg Woolf
of the interests of adult males, those with citizenship and those who hoped to acquire it, that makes it likely that many chose to organize their families in the Roman way even when they did not have to do so. I am well aware, of course, that much of this chapter has been concerned with working out how things must have been, rather than in documenting how they were. It would be much more comfortable to be able to demonstrate the truth of what I have proposed, but for the social historian of antiquity good guesses are often the best we can do, and they are always preferable to bad arguments.
 Family and Kinship in Roman Africa Mireille Corbier
T objective of this chapter is to consider our current understanding of the family in the Roman era in a regional setting, specifically the provinces of North Africa, with a view to enabling a comparison with other provinces in the empire. This task presents a number of difficulties, but also affords an opportunity for some general reflection on several important methodological problems of ancient history. The difficulties that arise are both spatial and temporal. Roman Africa covers a vast territory and the extant information is not only very scattered, but also unevenly scattered, throughout the region. The evidence covers a multitude of cultural contexts and deals with people of very different juridical status and social levels, furnishing a series of examples that should be examined as individual cases before any attempt at generalization is made. Furthermore, this evidence extends over half a millennium, from the first century to the fifth century , and bears the mark of multiple changes which are far from unilinear. Although the important work of African authors such as Apuleius, Tertullian, or Augustine on this topic cannot be neglected, the fact remains that the evidence is primarily epigraphic, thereby introducing two additional distortions. First, there are the social and cultural biases inherent in a reliance on epigraphic sources. A concentration on inscriptions privileges not only the elite, but also those, often the same people, who wanted to express their Romanitas more than their ‘difference’, even if it is far from clear that such differences can in fact be indirectly revealed through the use of Roman vocabulary. Large strata of the population therefore remain hidden. One revealing sign is the Many thanks to Elena Silvestri for her translation and to Magali Cullin for her help in the layout of the documents.

Mireille Corbier
occurrence in late antique sources of African-sounding names that were less represented than Latin cognomina in funerary inscriptions during the first three centuries . This should not be read as a ‘return to the roots’, but rather that these names had probably never stopped being in use by sections of the populace who had only limited access to epigraphic commemoration.1 In a charming letter to Maximus of Madauros, a grammarian who was still pagan and who disdained the African names common among Christians probably because they were characteristic of the poor and uneducated, Augustine suggests that, since both of them were Africans, they were hardly in a position to deride Punic names.2 Secondly, there are the biases related to the history and constitution of the field of ancient history. Traditionally, the discipline has been built on the accumulation of epigraphic texts that were more often than not entirely cut off from their context, about which we therefore know nothing, except when the editors of the standard collections (e.g. CIL , ILTun., ILAlg.) were content to give the briefest description reduced to its bare essentials. In the same rhythm and direction as archaeology, however, ancient history has evolved over the past few decades, with the result that the most recent publications of our Tunisian, Algerian, Moroccan, and Libyan colleagues (and of colleagues, notably French and Italian, who work with them) closely associate text and context, providing the placement of inscriptions, funerary and honorific, on the monuments on which they were found. These publications also show us the extent of the work that would have to be done, through systematic verification in the field (and obviously beyond the scope of this study), in order to reproduce the same context for previously discovered and published inscriptions. When a monumental mausoleum is studied in a monograph, such as the mausoleum of the Flavii at Cillium, and the mausoleum of Quintus Apuleus Maxsimus at El Amrouni, or when we can reconstruct the order in which epitaphs were engraved, thanks to the photographing of a funerary stela with multiple epitaphs, such as in two recently published 1 Optatus, De schism. donat. , ; Augustine, Ep.  (letter of Maximus); see Lassère : –. 2 Ep. –.
Family and Kinship in Roman Africa

collections from Dougga, the set of problems for consideration expands significantly.3 Biases aside, no inquiry concerning the family in Roman North Africa can be meaningful without asking an array of questions related to cultural anthropology. Rome took lasting control over this vast territory, bringing along with its colonists its language, its urban models, its administration, its civilization; in short, its own modes of cultural expression. But Rome was not the first power in the region. It was preceded by Carthage, whose cultural influence radiated much more widely than its political control, and by Hellenistic influence, as has recently been suggested for Numidia.4 As strong as Roman influence must have been, at least among the elite, it is unlikely that it eliminated everything that preceded it. What we are looking for, beneath the translations and adaptations, are elements of continuity and some evidence of the unique features of local societies and cultures. This is all the more important since, after being one of the most Christianized regions of the empire, with all the profound modifications in the structures and conceptions of family, marriage, and kinship conventionally attributed to Christianization, North Africa was later Islamicized, a change which, however progressive, was nonetheless long-lasting and irreversible. It is precisely this Islamicized North Africa that some twentieth-century anthropologists have used as a field of study to identify specific kinds of relationships, the best known being ‘Arab marriage’, preferential marriage to the parallel patrilateral first cousin, that is, the daughter of the father’s brother.5 New biases can emerge, leading us to try above all to identify pre-Roman origins for these regional peculiarities, whether Libyan or Punic, and thus to minimize the deep influence Rome might have had. These then are the hypotheses—and calls for caution—that directed this inquiry. Adopting several different angles and combining a critical presentation of the evidence with some reflection on the structures of family and kinship, I will focus my attention on marriage, with particular regard to the following issues: 3 Les Flavii de Cillium, ; see also Vattioni –, Ferchiou , Khanoussi and Maurin  and . 4 Coarelli and Thébert . 5 On family relationships in Islamicized North Africa, see most notably Tillion  and Bourdieu .

Mireille Corbier
—burial and family relations in connection with the corresponding vocabulary of kinship; —the inscribed text, and especially the epitaph, as ‘biography’; —most especially, the question of the choice of spouse. Rather than presenting a statistical exploitation of a vast epigraphic corpus,6 I have instead chosen an approach that is more sensitive to individual lives and to the particular context of the specific families, an approach in which micro-historians have led the way. Burial and Family Structure Funerary stelae decorated with images of married couples are far from rare.7 Exceptional examples, the product of local workshops, have been found in the Bou Arada region as well as at Mactar in Tunisia.8 Sometimes the couple is accompanied by a child or a servant (Figs. .–.).9 Many funerary stelae with portraits, however, say very little about the deceased, some even being anepigraphic, and, conversely, lengthy epitaphs that give more information than others are not necessarily accompanied by images of the individuals who are mentioned. Married couples often share the same tomb as well as the same stela, with the epigraphic field divided vertically in half. Sometimes only one of the epitaphs has been inscribed (Fig. .). 6 Especially Saller and Shaw , and Shaw and Saller ; more recently other types of statistical analysis have been conducted by D. Cherry ( and ). 7 The representation of the married couples is found not only on funerary monuments, but is also characteristic of votive stelae, in particular stelae dedicated to Saturn, the Latin name given to an African divinity which was especially venerated. See Le Glay a and b. 8 See notably Ferchiou  (Bou Arada); Picard et al. , and M’Charek  (Mactar). For other published funerary monuments, see Ferchiou  and M’Charek . 9 Fig. .: stela with married couple, Mactar (Tunisia) (photo Y. Thébert). CIL . , ; Picard, Le Bonniec, and Mallon : , no. , fig. ; M’Charek, : – (fig. ). Fig. .: stela with banquet scene from Tlil Bou Eukla, region of Bou Arada, Bardo Museum, Tunis; Benzina Ben Abdallah : no. . Fig. .: cippus with couple and daughter (imperial freedman and slave); exact provenance unknown. Bardo Museum, Benzina Ben Abdallah,  (see AE : ).
Family and Kinship in Roman Africa
. .
Stela with married couple, Mactar


Mireille Corbier
The metrical inscription on the funerary cippus of Hr Ben Ismaïl near Dougga declares that the spouses ‘are reunited in a common tomb, a witness to their shared lives’.10 The identification of those deceased who share collective tombs is more relevant to the topic at hand. At Mactar, the founder of a mausoleum had foreseen the presence, with him and his wife, of his parents, of his children and descendants, and of his brother and sister-in-law.11 At Dougga, the cippus of the Licinii connects the founder, P. Licinius Secundianus, another man of the same name, and a Licinia Nina, who could be either a sister, as proposed by the editors, or the wife of one of the two men; these three who died in very old age are commemorated along with two very young children, a telling combination in a society with a high rate of infant mortality.12 These family mausolea and cippi, however, do not include an individual’s entire circle of loved ones. The real-life experience of Fronto, as we know it through his correspondence, was similarly not limited to his wife, his daughter and son-in-law, and his grandsons, who were nonetheless very dear to him. In the case of Augustine, an assertion by Brent Shaw can be more finely nuanced: ‘His own experience of family relations was concentrated overwhelmingly on the rather narrow circle of his mother and father, his siblings and his own child. Notices of persons outside this group are rather rare; they include the chance mention of some nieces, a nephew and two cousins, each case being alluded to in passing only once.’ In fact, his mother aside, Augustine is no more communicative about his close family, either his son (whom he names, in recounting his death in his Confessions, as Adeodatus), or his brother and sister, less still about his concubine, who remains anonymous.13 The Metamorphoses by Apuleius gives examples of fictional families who in their own way provide evidence of family customs of the time.14 The necropoleis which have been published, such as Sétif, Tipasa, and Pupput (which awaits final publication), provide more information on funerary rites than on family and family 10 CIL .  0001 CLE ; see Ladjimi Sebaï ; Khanoussi and Maurin : no. . 11 CIL . : mausoleum of M. Gargilius Fortunatus. 12 Khanoussi and Maurin : no. ; and , nos. –; –. 13 Shaw b: ; Fredouille . 14 Bradley b.
Family and Kinship in Roman Africa

structure.15 They are highly illuminating about the practice of funerary banquets at the family tomb, a pagan practice that was not renounced by Christians. In the eastern necropolis at Sétif a tomb dating between the end of the first and the mid-second century contained the inhumation of a lamb (along with a mirror and two long bone pins), its head facing east, legs south, in the same pose as deceased human beings, thereby indicating the practice of substitute sacrifice (the lamb offered in the place of a new-born) which is attested elsewhere, in particular in the second-century inscriptions at N’Gaous.16 The large necropolis at Pupput (mainly second century to the first half of the third century) presently being excavated by Aïcha Ben Abed and Marc Griesheimer was levelled to a height which leaves the cupulae intact in the funerary enclosures but all the identifying inscriptions have been destroyed. This site is particularly valuable for our knowledge of funerary banquets due to the presence of numerous masonry tables (mensae). It suggests the presence of family burials in which the bodies or ashes of close relatives, found under the massive structures that covered and individualized each of them, were grouped together. It is tempting to imagine a family name above the door of each enclosure or in another prominent position. Such collective enclosures are also characteristic of the western necropolis at Tipasa (third to fourth centuries), published by Mounir Bouchenaki, where a unique tessellated mensa with a Christian inscription demonstrates that the pagan custom of funerary banquets was retained by Christian families.17 At Dougga, which is surrounded by necropoleis, the oldest has only one extant mausoleum, known as ‘Libyco-punic’, but in the other necropoleis outside the town the proximity of certain tombs proves the existence of family burials. Beyond the suburban cemeteries, on the probable site of an estate, rose a grand tomb, the columbarium of the Remmii. Monumental cippi accommodated up to eight epigraphic fields, although not all were 15 Sétif: Février and Gaspary –; Février and Guéry ; see also Guéry . Tipasa: Lancel –; Bouchenaki, . Pupput: Ben Abed and Griesheimer  and . 16 Sétif: Février and Guéry : ; N’Gaous: Le Glay a: , . 17 In Chr(ist)o Deo pax et concordia sit convivio nostro: Bouchenaki  and ; Marrou  (see AE : ).

Mireille Corbier
inscribed.18 In contrast, the epigraphy of Cirta includes five examples of family burials (with four found out of context) that are identifiable by the family name alone.19 It seems that the term gens was used at Cirta to identify the family group. Considered from the point of view of the married couple, North African epitaphs show an unusual use of the term marita, and, more often than at Rome, sponsa and sponsus are used to refer to spouses and not only to the betrothed.20 It would be useful to establish a precise chronology for these epitaphs (in fact, from sponsus and sponsa come the names of husband and wife in several romance languages), but these terms seem to have been largely forgotten in studies concerning Roman marriage, which for the most part are based on the epigraphic documentation from Rome and Italy.21 The terms of kinship derived from remarriage are met infrequently, whereas remarriages can be attested by inscriptions, notably by the difference in the gentilicium of children in the case of the mother’s remarriage.22 One possibility is that these terms might not have been used in everyday language, another that family members other than stepchildren or stepparents might have taken care of funerary obligations. The Inscription as ‘Biography’ The moral portrait of a deceased man or woman drawn by several epitaphs cannot be read as ‘biography’. As a genre, funerary poetry 18 Khanoussi and Maurin : –, nos. –, and  – and –. 19 ILAlg. /. : oikos koimeseos (house of rest?) Fonteiorum; : memoriae Helviorum; : coemeteria memoriae gentis Lepidiorum; : memoriae gentis Scantianae; : memoria gentis Umbricior(um). 20 e.g. a commemorated sponsa died at the age of  (CIL . ). Note, however, that in CIL . , among the funerary inscriptions of Rome, is found a husband (coniunx), named Primus, who commemorates Quartilla, sponsa sua bene merens, who lived  years. 21 Roman marriage: Treggiari . 22 e.g. from a glance at the index of CIL : vitricus occurs twice, privignus once, and there is no example of noverca. Gascou : no. , shows a privignus (engraved priviginus) commemorated by his stepmother. For an example (at Timgad) of remarriage identified by the different gentilicia of the children, see Laronde  and Corbier a: .
Family and Kinship in Roman Africa

was much appreciated in Africa, where the local grammaticus or, if possible, a literate member of the family might compose a poem in a somewhat shaky metre, sometimes reminiscent of Virgilian verse. To judge from the epitaphs written by their husbands or sons, African women were endowed with the traditional qualities prized in Roman matrons, especially castitas and pudicitia. The two epithets juxtaposed as casta pudica are the beginning of a verse which is also found in Italy.23 The epitaph of Postumia Matronilla (‘with a predestined name’ as Leïla Ladjimi Sebaï put it) is a minor masterpiece that reveals precious little about her other than the fact that she had only one husband (whose name is not mentioned) and that she lived to see one or more of her grandchildren.24 The Roman ideal of the univira, a woman who had known only one man, seems to have been widespread,25 and the oft-cited doublet univira unicuba is particularly suggestive. The expression sola contenta maritu (for solo contenta marito, ‘happy with only one husband’) found at Cirta, in which we might be tempted to find a personal touch, is in fact the end of a hexameter, which is also present in Gaul, at Vesontio.26 Individualized eulogies extolling competence, culture, or charm are rare: paedagoga at Dougga,27 filosofa at Sousse,28 while at Haïdra Mevia Felicitas is praised for her beauty and her conversation.29 ‘Dedicated to her husband’, ‘a mother to her family’, these are the qualities worthy of praise in a woman. Julius Maximus of Mactar valued his wife’s sense of economy, ‘for her husband’s property as well as her own . . . she barely spent what was needed to cover her personal needs’, interesting evidence which hints at the division of property between spouses that was characteristic of ‘Roman’ marriages.30 Fertility is also praised: the twelve ILAlg. /. , at Cirta in Algeria; Galletier : . CIL .  0001 ILS  0001 ILTun. , epitaph on her mausoleum at Hr Zaatli in the region of Feriana, Tunisia. Ladjimi Sebaï . See Document no. . 25 ILAlg. /.  and : examples of Geminia Ingenua and Iuventia Rogata, at Cirta. 26 ILALg. /. , at Cirta, and CIL . , at Vesontio. 27 CIL .  0001 Khanoussi and Maurin : no. , and : no. . 28 AE : . 29 Fuit enim forma certior, moresque facundi (CIL, . , cf.  and p.  0001 CLE  0001 ILTun. ); see Ladjimi Sebaï . 30 CIL .  0001  0001 CLE ; see Picard : . 23 24

Mireille Corbier
children of Claudia Fortunata near Hippo and the five children and ten grandchildren of Sulpicia Victoria at Auzia must represent exceptional numbers worthy of commemoration.31 The recently published epitaph of Rubria Festa at Cherchel gives proof of the notion, common in the mid-second century (and reminiscent of Rousseau), that it was virtuous for a woman of high standing to breastfeed her own children. Fronto belongs to this trend in sentimentality, especially regarding young children.32 Whether in prose or verse, this funerary praise seems characteristic of the municipal elite, but it is hardly suitable for women of high social standing, for whom the rank of their father and/or husband, their noble (clarissimi) children, and their own acts of euergetism are sufficient finery. Much like their husbands, and often along with them, they are honoured in the public sphere or, more often, in the semi-public sphere created by the building activity of the family. In the Severan era, the wives of the Aradii of Bulla Regia are represented in the form of statues beside those of their husbands in the marketplace which they had constructed,33 while in the same period the baths offered by Julia Memmia, patrona [et alumna], to her native city (where she apparently did not reside) housed a statue of the benefactress whose pedestal records her generosity, exactly as would have been done for a man.34 At Cirta, Seia Gaetula, widow of a municipal magistrate and very proud to be the mother and the grandmother (from another daughter) of two women of senatorial rank, obtained from the town council public space on which to erect at least three statues whose bases still extol the prestigious marriage alliances made by her daughter and granddaughter with senators from the same town.35 As far as the relationship between husband and wife is concerned, the epitaphs seem marked by conformity. Indeed, the vast Hippo: CIL . ; Auzia: CIL . . Rubria Festa was the provincial flaminica (AE : ) See Document no. . On breastfeeding, see Bradley ; on Fronto, see Corbier b: – and on the sentimental trend, see Corbier a. 33 Corbier : – and –. 34 ILAfr. ; see Broise and Thébert : –, and Corbier a: . 35 CIL . – 0001 ILAlg. /. , , ; see Document no. ; also Corbier : –. 31 32
Family and Kinship in Roman Africa

majority of funerary inscriptions speak of legitimate unions and mutual love between spouses, as at Theveste, where a passionate widower adds to his eulogy of his wife digna coniunx te adoro.36 At Sétif, however, it is a concubine—Arria Dativa—who has a tumulus amoris, a ‘tomb of love’, dedicated to her by her anonymous lover.37 In Africa, as elsewhere, epitaphs are not a true reflection of conjugal life; thus, inequality, constraints, even violence, so astutely detected by Brent Shaw in the Confessions and other writings of Augustine, are not expressed in inscriptions.38 Tertullian even asserts that some pagans married to Christian wives would not hesitate to denounce them, making ‘the dowry the price of their silence’.39 For his part, Pliny the Elder brought back from Africa the image of women treated as beasts of burden: ‘There I saw the earth dug up after the rains by a plough to which were harnessed on one side a poor little donkey and on the other a woman’.40 The double standard applied to the sex lives of men and women which Augustine condemned he presents as normal in his era.41 A suggestion of it can be found in this long epitaph ‘prepared while living for the day of my death’ by Praecilius Fortunatus, a goldsmith from Cirta, who depicts himself as a bon vivant (‘I always loved to laugh with my dear friends and make the most of luxury’) and who evokes with emotion the memory of his marriage to Valeria; once a widower, he ‘could not find another wife as chaste as she’.42 The rituals of daily life are even less in evidence. It would not be enough, as L. Ladjimi Sebaï rightly notes, to juxtapose the representations of funerary banquets which are characteristic of the region of Bou Arada (on the stelae mentioned above), where the wife is seated at the foot of her husband’s bed (Fig. .), and the passage in Apuleius’ Metamorphoses, where Lucius, upon arriving at his host’s house recounts: ‘I entered just at the moment when, reclining on a little couch, he was about to take his
36 37 40 42
CIL .  0001 CLE , at Theveste; see Ladjimi Sebaï, . 38 Shaw b. 39 Tertullian, Ad uxor. . . . CIL . . 41 See Shaw b: . HN . . ILAlg. /. ; see Picard : –.

Mireille Corbier
. . Stela with banquet scene, region of Bou Arada
Family and Kinship in Roman Africa
. . Cippus with couple and daughter, provenance unknown


Mireille Corbier
supper. At his feet was seated his wife’,43 and then to conclude that African wives observed this ancient custom as an expression of deference to their husbands. The funerary images reproduce the iconographic motif of reclining at the banquet, which is well studied for the Near Eastern and Greek worlds,44 in its context, however, Apuleius’ description contains its share of derision. The Choice of Spouse The question of choice of spouse is one of the most important for the history of the family. I will discuss only briefly the somewhat modern notion of mixed marriages with regard to a recent publication which requires particular comment. For Tertullian too ‘mixed marriages’ were an important issue in African society at the end of the second and beginning of the third century . In his case, however, it was a question of marriage between Christian women and pagan men, for which the funerary inscriptions give no clue. Different kinds of statistical analyses have recently been conducted by David Cherry on several Roman sites in Algeria, the most important a comparison between Thubursicu Numidarum and Lambaesis.45 Focusing on the issue of indigenous acculturation, he has also tried to identify ‘mixed marriages’, which he defines as marriages between Roman or ‘Romanized’ and ‘un-Romanized’. The notion that the proportion of ‘mixed’ marriages can be taken as an indicator of the degree of ‘acculturation’ of ‘indigenous people’ in a ‘colonial’ society deserves to be discussed before being taken as self-evident. Marriage between people of different social status and cultures does not in itself imply the weakening of cultural difference, nor the acceptance of values and ways of thinking by members of one or the other cultures concerned, or by members of both. It implies only that the two families recognize that they share the same social rank, and that theirs is a relationship of equals. But even if this approach is accepted, it would still be necessary to rely on established categories. My two main reservations 43 45
Apuleius, Met. . ; Ladjimi Sebaï : –. Cherry  and .
44
Dentzer .
Family and Kinship in Roman Africa

are based on the insufficient attention given to the importance of juridical status in Roman society and an abusive use of onomastics. To distinguish between ‘non-Romanized’ and ‘Romanized’ using the adoption of a Latin single name by peregrini (a term Cherry himself rarely uses) is a somewhat unreliable method that leads the author to contradict himself. Why, for example, class a man named Gallus among the ‘Roman(ized) men’ because he has a Latin single name,46 and consider his wife Berecbal Secundi f(ilia)47 ‘un-Romanized’, when she is the daughter of Secundus who, according to the criteria used by the author, would be a ‘Roman(ized) man’? Within the same family Roman and African names can alternate without allowing us to measure, using this criterion, the degree of assimilation of its members. Jean-Marie Lassère noted the mix of African and Roman names in five families at Auzia and of Punic and Roman names in the Tapapii family at Lepcis Magna.48 Furthermore, names with a Latin sound such as Saturninus, -a, are common among peregrini. In married couples of peregrine status one (sometimes the husband, sometimes the wife) can bear a Latinized name and the other not. On the subject of matrimonial exchange, anthropologists have insisted on two opposed behaviours: in some societies marriage with close relatives is a common practice, in others it is expressly avoided or even prohibited.49 Roman Africa gives evidence of close-kin marriage (i.e. marriage between cousins), as well as examples of the remarriage of a widow to her husband’s brother; Apuleius’ own experience is informative on this second point.50 Born in Madauros, Apuleius spent several years of his life in Tripolitania following his marriage to Aemilia Pudentilla, a rich widow of Oea who was older than him. In  – Apuleius was accused by his wife’s family of having used magic to seduce her. He took on his own defence, and the speech he made on this occasion (Apologia) is a document rich in evidence for an anthropological study of the society of Tripolitania in the middle of the second century. Among the many facts that we can glean from it is the persistence of a form of remarriage not much practised by 47 ILAlg. . . Cherry : . Lassère : ; Auzia: CIL . , , , , ; Lepcis Magna: IRT , , –, , . 49 See, notably, Bonte . 50 On Apuleius’ personal history, see Bradley a. 46 48

Mireille Corbier
Romans, although it was not forbidden, in which the widow married her late husband’s brother.51 In Rome itself, there are at least two examples among the elite. One was the triumvir Marcus Licinius Crassus who, according to Plutarch, married his brother’s widow.52 Secondly, a passage in Tacitus can only be understood if we reconstruct behind the double kinship of two men, Mamercus Aemilius Scaurus and Lucius Cornelius Sulla, the remarriage of the mother of the latter to the half-brother of her first husband. A nephew should share the same gentilicium as his patruus, his paternal uncle, which is not the case here, since pater and patruus had the same mother but different fathers.53 Should we follow Philippe Moreau’s interpretation of certain passages of Ovid and Seneca, who seem to present relationships between brother-in-law and sister-in-law as ‘incestuous’?54 In both cases (Tereus and Philomela, sister of his wife Procne, and Thyestes, seducer of Aerope, wife of his brother Atreus), it is in fact a question of adulterous relationships and not remarriage. If they are presented by Ovid and Seneca as contrary to fas, divine law, it may be because of the particular seriousness of adultery in the bosom of the family itself. When her husband Sicinius Amicus died, Aemilia Pudentilla, who was not yet  and had two young sons, Sicinius Pontianus and Sicinius Pudens (who from then on passed into the paternal power of their grandfather), was under intense pressure from her father-in-law to marry her deceased husband’s brother, Sicinius Clarus. The father-in-law threatened to disinherit his two grandsons. Pudentilla had been able to resolve this delicate situation by accepting an indefinitely prolonged engagement. When she was about  years old and had attained more independence because her sons had meanwhile received their inheritance from their grandfather, Pudentilla met and married Apuleius. Several years later, a similar situation befell the next generation—Pontianus, Pudentilla’s elder son, had married the daughter of Herennius Rufinus, himself another respected citizen of Oea, despite the 51 This kind of remarriage with an affine (relative by marriage) of the same generation was prohibited by Constantius II in : Cod. Theod., . . . 52 Plutarch, Life of Crassus . . 53 So the two half-brothers did not belong to the same familia: Tacitus, Ann. . ; see Corbier : – with stemma. 54 Moreau : ; Ovid, Met. . , –; Seneca, Thy. –.
Family and Kinship in Roman Africa

caricature Apuleius draws of him. Pontianus died unexpectedly at a young age, leaving behind a widow who was also young and to whom he had bequeathed precious little. The young Herennia’s father endeavoured to arrange a remarriage for her with Pudens, the younger brother of her dead husband and the youngest son of Pudentilla. Although Apuleius recounts these arrangements in highly negative terms (in the context, it should be remembered, of his dispute with Pudentilla’s family), jurists and historians have recognized in these two marriage plans the practice of levirate, which was widespread in many societies in the Near East and which therefore could represent a Punic heritage among the families of Oea who were otherwise very ‘Romanized’.55 Levirate, an obligation imposed on the brother of a deceased man to marry his widow, was tied to the desire to perpetuate an existing marital alliance, and allowed eventually the preservation of the dead man’s rights over his descendants. To put it in more prosaic terms, in the two cases presented by Apuleius, the family sought to maintain an alliance which had been broken by the death of one of the interested parties. For the Sicinii, it was a question of keeping Pudentilla’s dowry and patrimony; then, for the Herennii, of maintaining an alliance formed in the next generation with the sons of her first husband. Are there other examples of this type of union beyond that of Pudentilla and her sons? Among the Aufidii, an elite family from Saldae with a son who attained equestrian rank, the remarriage of the mother to the brother of her first husband is certain. In an inscription on a statue base erected in memory of the son in  there is a reference to kinship ties which proves the existence of this type of union.56 The vocabulary used in the inscription allows us to reconstruct the family links: the double kinship relationship connecting the deceased young knight to his mother’s husband (patruus idemque vitricus) reveals that she married two brothers in succession. The relationship between onomastics and kinship vocabulary lets us hypothesize the existence of a kind of marriage which Norden ; Fantham ; Bradley a. See Document no. . CIL . ; on the family see also Mathieu :  with a prosopographic note in no.  and stemma no. . See Document no. : the Aufidii of Saldae. 55 56

Mireille Corbier
French anthropologists customarily describe as ‘remarquable’ as it implies the reduplication of matrimonial alliance: that of two brothers with two sisters. At Cirta, the homage offered by a freedman to his female patron suggests a union between two sisters, the Antoniae, and two brothers of a senatorial family, the Arrii Antonini. If the dedicant followed Roman kinship terms, Antonia, matertera of three Arrii of senatorial rank, is the sister of their mother; in other respects, the three clarissimi viri are the nephews of her husband, C(aius) Arrius Pacatus.57 But what about close-kin marriage, meaning marriage between cousins? In a work of fiction set in Greece, the same Apuleius describes the marriage between a ‘handsome young man, who was also my first cousin’ (speciosus adulescens . . . meus alioquin consobrinus), a marriage consisting of two stages, since it had been preceded by an engagement.58 From a remarkable epigraphic dossier at Chemtou (Simitthus) in Tunisia comes an example of a real marriage of this type.59 Although not exactly the ‘saga’ the editors of the three stelae deem it, since only two generations are concerned, this family history is nevertheless a beautiful piece of micro-history. The three funerary stelae, found in situ vertically aligned beside each other at Sidi Mohammed Lazrag, near Simitthus, are now exhibited in their original alignment in an attractive display in the museum at Chemtou (Figs. .–.). It would be valuable to know if we indeed have all of the occupants of the funerary enclosure or only some of them. The first stela (Fig. .) commemorates C(aius) Sulpicius Primus and his wife Laetoria Rufina. The second (Fig. .) tells of the marriage of Sulpicius Faustus and Sempronia Urbica, who came from another town (Masculitana, i.e. from Mascula), and who was much younger than her husband. Although she was left a widow with a young daughter named Faustina, Urbica did not return to Mascula or remarry, preferring instead to enjoy the good repute accorded to widows who remained ‘univira’, and in time she saw to the arrangement of a good marriage for her daughter. The third stela (Fig. .) sets 57 ILAlg. /. ; see Corbier : – with stemma. See Document no. : the Arrii and the Antoniae at Cirta. 58 Apuleius, Meta. . . ; see Treggiari : –, who uses it to illustrate Roman marriage. 59 Benzina Ben Abdallah and Khanoussi  (see AE : –). See Document no. : the Sulpicii.
Family and Kinship in Roman Africa

. . Stela of C. Sulpicius Primus and Laetoria Rufina, Sidi Mohammed Lazrag
. . Stela of Sulpicius Faustus and Sempronia Urbica, Sidi Mohammed Lazrag

Mireille Corbier
. . Stela of Sulpicius Primus and [Sulpicia] Faustina, Sidi Mohammed Lazrag
Family and Kinship in Roman Africa

out a married couple of the next generation, revealing that the husband chosen by destiny (and helped along by her mother) was in fact her first cousin, the son bearing the same name, Sulpicius Primus, as the deceased on the first stela. It is interesting to note that Sempronia Urbica is designated as Masculitana, ‘a native of Mascula’; Sulpicius Faustus did not marry a woman from his own town. But did he go as far as Mascula in Numidia to find her? She might instead have come from Masculula, a small ancient town near Simitthus, and the stone-carver might simply have left out a syllable.60 The detail which most sparks my interest, however, is the indication of the choice of a husband made by the mother from among the many suitors. Sempronia Urbica made an excellent choice for her daughter since, according to the inscription, she became destiny’s spokeswoman! As the editors put it, ‘it was therefore inevitable (fatum) that two first cousins, born of two very close brothers, should marry’. In the era of the Sulpicii, the closest blood relatives who were permitted to marry by Roman standards were indeed first cousins (as it is in contemporary French law). Of interest here are several fragments of jurisconsults integrated by the compilers of the Digest during Justinian’s reign and a rescript of Caracalla, dated  , which was included in the Code of Justinian.61 In the context of a comment on patria potestas, the jurist Paul specifies that the grandfather alone had the right to allow a marriage between his grandson and granddaughter, born of two different sons of his, who are under his paternal power.62 This remark confirms, if we still need confirmation, the legitimacy of the union between parallel patrilateral first cousins. The jurists Marcellus and Papinian envisaged the writing of wills in which a brother favoured the one of his two brothers who would marry their first cousin (consobrina).63 The rescript of Caracalla was issued in response to a request by a certain Cassia, who would not submit to her mother’s condition that she marry her matrilateral first cousin in order to become her heir and therefore saw her rights 60
A suggestion made to me by Azedine Beschaouch. In Justinian’s era marriage with a first cousin was still permitted: Justinian, Institutiones . . . 62 Paul, Digest . . . 63 Marcellus, Digest . . ; Papinian, Digest . . . 61

Mireille Corbier
contested by the substitute heir. The emperor—that is, the jurists in his entourage—judged that the planned marriage was perfectly acceptable; as a result the substitution of heir could proceed.64 Was the ‘fated (fatum) marriage’ between Primus and his parallel patrilateral cousin Faustina a simple marriage between first cousins (consobrini)? Or was it the perpetuation of a ‘preference’ for a union with a cousin that could have been shared by some of the Berber societies before the arrival of the Romans? The two ideas are reconcilable: a local practice of preferential marriage could have been inserted within the framework of Roman law. In order, however, to verify the preferential character of this type of union, evidence for other value judgements, such as the one expressed here, is required (but is not extant), as well as objective data demonstrating the existence of this type of marriage. It is not possible to conclude that marriage between cousins was uncommon strictly on the statistical observation of the rather limited number of known spouses who carry the same gentilicium. This was the rather hasty conclusion drawn by R. P. Saller and B. Shaw from their onomastic study of the epitaphs of the western provinces.65 I would argue that it is necessary to work on each site with real families and try to reconstruct families over several generations wherever possible. At Thubursicu Numidarum a group of inscriptions allows us to identify the family relationships of the leading citizen (whose three sons attained equestrian rank) Q. Vetidius Iuvenalis (called Iuventius), and his wife, Gellia Honorata, who does not bear the same gentilicium as he does. She is the daughter of his first cousin Vetidia Mustacia, who was herself the daughter of his paternal uncle Vetidius Mustiolus.66 Cousins can therefore be hidden where we least expect to find them. The Flavii at Cillium were married to Flaviae, who could have been, but were not necessarily, their cousins, since several families of Cillium might have gained Roman citizenship under the Cod. Just. . . . Shaw and Saller . I will not here address this issue, which I have treated elsewhere; see Corbier : –; c: –; : . 66 ILAlg. . – with stemma; see Lassère : . See Document no. : the Vetidii at Thubursicu Numidarum. Here we see a direct delayed exchange between the Vetidii and the Gellii. 64 65
Family and Kinship in Roman Africa

Flavians.67 There is, however, no doubt concerning the Caecilii at Volubilis, who do not have an imperial gentilicium. Divided in two main branches, they clearly marry among themselves.68 Although it was a second marriage, one L. Caecilius even married a certain Caecilia Caeciliana, whose name could reveal that, in a previous generation, a Caecilius married a Caecilia.69 The recent publication in a single volume of hundreds of epitaphs from Dougga does not provide any contradictory evidence, although it is true that they rarely mention kin relationships.70 Whatever had been the frequency of close-kin marriages at the beginning of the third century, two centuries later Augustine, from Thagaste, then bishop of Hippo, gives a very different point of view in the City of God.71 He starts by stating that ‘in his time’ (nostris temporibus, although it is not certain that he means African society in particular) marriage between first cousins (consobrini) had become rare, implying that they had been frequent, because of changes in social customs (per mores), even though still permitted by law. Augustine justifies this fact by arguing that the degree of kinship is very close to that of brothers and sisters (germani). His reservations about this type of marriage are not based on divine law, which he himself points out imposes no ban, but rather on his desire to multiply alliances, by preventing only one person from accumulating two kinship relationships which could be distributed between two persons, and to enlarge the circle of kinship. He presents his argument less from the perspective of family interests—which are called upon to manage family capital as well as possible by diversifying and renewing investments—than from that of human society, in the belief that the caritas ensured by alliance and kinship guaranteed harmony and efficiently created the bases for ‘social life’.72 For the choice of spouse, he offers an ideal that is completely different from that of Sempronia Urbica, and which was shared in the same era by two other church
67 68 69 70 72
Les Flavii de Cillium, : . See Lassère : –; Gascou : , , ; Lefebvre . Gascou : no. ; see above, n. . 71 Augustine, City of God . . Khanoussi and Maurin . Corbier b: –.

Mireille Corbier
fathers, Ambrose of Milan and John Chrysostom,73 and which was beginning to be applied in imperial legislation. At the end of the fourth century Theodosius is credited by contemporary authors (Libanius in the east and Ambrose and Pseudo-Aurelius Victor in the west) with a law which is no longer extant forbidding marriage with first cousins (consobrinae), whether patrilateral (sorores patrueles) or not.74 Reaffirmed in  by Arcadius for the east, and in  by Honorius for the west, this interdiction was subsequently annulled in  by the same Arcadius for the eastern half of the empire, which had a long-standing tradition of close-kin marriages.75 Was this a new ‘Christian’ ideal in the sense that it was ‘inspired by Christianity’? Augustine, it must be said, refers to the transformation of social mores in society as if it were an objective fact, an evolution of social practice. In a recent work, Philippe Moreau underscores certain passages in pagan authors of the fourth century, specially Firmicus Maternus and the Emperor Julian, who, according to him, also condemn this type of union as incestuous, leading him to see ‘a general tendency to assimilate cousins and sisters’ (the former coming to be considered, as the latter, prohibited spouses).76 In fact, the astrological treatise which Firmicus Maternus wrote before his conversion to Christianity, upon which the author relies, presents sexual relationships between a woman and the sons of her paternal uncle as incestuous.77 The reproach made by the Emperor Julian to ‘marriages which are not marriages’ is very allusive: commentators have accepted that this general formulation intended to criticize the marriage of Constantius II, Constantine’s son, with his first patrilateral cousin Constantia; but Julian himself had contracted a similar type of marriage.78 73 Ambrose, in his correspondence with Paternus: Ep. ; John Chrysostom, Homilies on the first letter to the Corinthians . . See O’Roark : –, and Moreau : . 74 Libanius, Oration  (de angariis), ; Ambrose, Ep. ; Pseudo-Aurelius Victor, Epitome of the Caesars . . 75 Cod. Theod. . .  (); . .  (); Cod. Just. . . . For details on these measures, read the classic study by Roda  as well as Puliatti:  – and Moreau : –. 76 77 Moreau : –. Firmicus Maternus, Mathesis . . . 78 Julian, Against Heraklios the Cynic . c.
Family and Kinship in Roman Africa

There is one certainty, however: in less than two centuries marriage between cousins, from being legally and socially acceptable as it still was at the beginning of the third century, had become suspicious. The break is neither dated nor explained. A process of broadened social horizons tied to the Roman empire might be the origin of this change in perspective. Even if this new ‘ideal’ is not actually rooted in Christianity, it certainly became a Christian ideal and rule for centuries in the west, as medieval theologians borrowed Augustine’s justifications for the ban on marriage between first cousins. Thus, after the reception of Islam in the Maghreb, when some ancient traditions in the family sphere could have been reinforced, and then in Spain, with the Arab invasion in , different models of family structure emerged, creating different choices, some still relevant today: the system of descent, the status of the conjugal pair (e.g. polygyny versus monogamy), the matrimonial alliance (a major theme of this chapter), the social position of women, and the notion of honour.79 Documents . Epitaph of Postumia Matronilla CIL .  0001 ILS  0001 ILTun.  (Hr Zaatli, region of Feriana, Tunisia); photo in Ladjimi Sebaï . D(iis) M(anibus) s(acrum). Postumia Matronilla inconpa/rabilis coniux, mater bona, avia piissima, pudica religiosa laborio/sa frugi efficaxs vigilans sollicita uniuira unicuba [t]otius industriae et fidei matrona, vixit annis n(umero) LIII mensibus n(umero) V diebus tribus. To the gods of the Underworld. Postumia Matronilla, incomparable spouse, good mother, most devoted grandmother, modest, pious, industrious, well-behaved, efficient, attentive, zealous; married once, she shared one bed, she was a conscientious and faithful matrona. She lived  years,  months, and three days.
79 These were clearly shown by Pierre Guichard , for Muslim Spain and re-examined by Jack Goody . On the topic of kinship and alliance in the Roman world, see Corbier .

Mireille Corbier
. Funerary poem of Rubria Festa AE :  (Cherchel, Algeria) [Ha]nc struem perennis arae posuit his in sedibus [I]ulius Festae Secundus coniugi karissimae. Vixit annos sextriginta bisque viginti dies. Pondus uteri enisa decimum luce rapta est tertia. Nata claro Rubriorum genere de primoribus sancta mores pulchra visu praecluens prudentia exornata summo honore magno iudicio patrum aurea uitta et corona Mauricae provinciae haec et divum consecuta est summa pro meritis bona. Quinque natos lacte mater ipsa quos aluit suo sospites superstitesque liquit votorum potens. Julius Secundus had an altar made on this eternal spot for Festa, his dearest wife. She lived  years and  days. Bringing into the world the weight of her womb for the tenth time, she was carried off on the third day. Born of the race of the Rubrii, famous among the great families, chaste in her ways, fair of face, very well known for her wisdom, she received the gold fillet and crown of the province of Mauretania, the highest honour that could be accorded to her by order of the patres, and she attained the greatest good of the gods for her merits. She left safe and sound five children, whom their mother had fed with her own milk, fulfilling her vow.
. Relatives of Seia Gaetula After three statue bases erected to her household at Cirta Corbier : – CIL .  0001 ILAlg. /.  (Cirta, Algeria) Seiae M(arci) f(iliae) Gaetu lae, uxori Naeui Censiti, matri Nae viarum Marcia nae et Naevillae, c(larissimae) m(emoriae) f(eminae), nuptae Fulvio Faustino, praeto rio viro, aviae Sabi niae Celsinae, c(larissimae) f(eminae), nuptae Geminio Mo desto, praetorio vi ro, eadem Gaetula d(ecreto) d(ecurionum) s(ua) p(ecunia) p(osuit).
CIL .  0001 ILAlg. /.  M(arco) Naevio M(arci) f(ilio) Seiano, fratri Naevi arum Marcianae et Naevillae, c(larissimae) m(emoriae) f(eminae), nuptae Fulvio Faus tino, praetorio viro, avonculo Sabiniae Celsinae, c(larissimae) f(eminae), Seia Gae tula mater d(ecreto) d(ecurionum) s(ua) p(ecunia) p(osuit).
Family and Kinship in Roman Africa

CIL .  0001 ILAlg. /.  Naeviae Naevillae, c(larissimae) m(emoriae) f(eminae), Naevi Cen/siti fil(iae), nuptae Fulvio Fausti no, praetorio viro, Seia Gae tula, mater d(ecreto) d(ecurionum) s(ua) p(ecunia) p(osuit).
Stemma (M. Seius) M. Naevius Censitus
Seia M. f. Gaetula
Q. Fulvius Faustus
Naevia Fulvius Faustinus M. Naevius Naevia Naevilla M. f. Seianus Marciana (Sabinius) clarissimae memoriae praetorius vir femina Sabinia Celsina clarissima femina
Geminius Modestus praetorius vir
. The Sicinii of Oea (Libya) After Apuleius’ Apology Stemma (Sicinius)
Apuleius = () Aemilia Pudentilla () = Sicinius Amicus Sicinius Clarus Sicinius Aemilianus decrepitus senex Herennius Rufinus Herennia
Sicinius Pontianus splendidissimus eques
Sicinius Pudens

Mireille Corbier
. The Aufidii of Saldae CIL .  (Saldae, Algeria). Date:  January  . M. Aufidio M. fil(io) Arn(ensi tribu) Honorato eq(uo) pub(lico) ornato dec(urioni) col(oniae) C. Aufidius L. f(ilius) Arn(ensi tribu) Honoratus patruus idemque vitricus et Sellia Q. fil(ia) Satura ma ter filio piissim[o] secundum volunt tem [sic] eius statu am posuerunt dedicaverunt que. Dedicata VIIII kal(endas) Ianuarias T. Sextio Laterano C. C[u]s[pi]o Rufino co(n)s(ulibus) a(nno) p(rovinciae) CLVIII ob cuius dedicatio nem sportulae datae sunt l(oco) ab ord(ine) sanc(tissimo) accepto.
Stemma Q. Sellius =
L. Aufidius =
C. Aufidius L. f. Arn. Honoratus patruus idemque vitricus
M. Aufidius L. f. Arn. = Sellia Q. f. fil. Satura mater 

M. Aufidius M. fil. Arn. Honoratus equo publico ornatus, decurio coloniae
. The Arrii and the Antoniae of Cirta (Numidia) ILAlg. /.  (Cirta) Antoniae L(ucii) fil(iae) Saturninae, coniugi C(ai) Arri Pacati, materterae Arrio rum Antonini Maxi mi Pacati, clarissi morum virorum, L(ucius) Antonius Cassianus lib(ertus) patronae merenti l(ibens) a(nimo) s(ua) p(ecunia) p(osuit) d(ecreto) d(ecurionum).
Stemma C. Arrius Antoninus
(C. Arrius Antoninus)
Arrius Antoninus
(Antonia)
Arrius Maximus clarissimi viri
Antonia L. fil. Saturnina matertera Arriorum (. . . ) clarissimorum virorum
Arrius Pacatus
C. Arrius Pacatus
Family and Kinship in Roman Africa

. The Sulpicii of the region of Simitthus See Ben Abdallah and Khanoussi : – 0001 AE : – (Sidi Mohammed Lazrag, region of Chemtou, Tunisia) Stela I (Fig. .) Laetoria Rufina L(ucii) f(ilia) pi(a) u(ixit) a(nnis) XLVII. H(ic) s(ita). O(ssa) t(ibi) b(ene) q(uiescant). D(iis) M(anibus) s(acrum). C(aius) Sulpicius Primus pius vi(xit) an(nis) LXXVI. H(ic) s(itus). O(ssa) t(ibi) b(ene) q(uiescant). Laetoria Rufina, daughter of Lucius, attentive to duty, lived  years. Here she lies. May her bones rest in peace. Consecrated to the gods of the Underworld. Gaius Sulpicius Primus, attentive to duty, lived  years. He lies here. May his bones rest in peace.
Stela II (Fig. .) Sulpicius Faustus uxore ducta Semproniam Vrbicam Masculitana [s]uscepit filiam parvam Faustinam; optabat vita ut coniungeret illam nec licuit illi optinere vota: annis sup(p)letis moritur sexaginta. Mater aeducat cum magna molessa petitorum cuinam sit coniungenda: electo multis bono tradidit viro. Vixit cu<m> cura annis sexaginta. Sulpicius Faustus, having married a woman from Mascula [or Masculula, suggested by A. Beschaouch], Sempronia Urbica, had a daughter, Faustina. He wanted to live long enough to marry her, but he was not allowed to see this wish come true: he died with  years completed. Her mother raised the daughter, and had great difficulty knowing to which suitor she should marry her. She gave her to a good man, having chosen from several. She lived with the desire to do the right thing for  years.

Mireille Corbier
Stela III (Fig. .) [Sulpi]cii Primus et [F]aust[ina] [nas]cuntur duo fratribus u[nanimis?] [c]oniuncti bene sunt fato vole[nte], matre Vrbica fato praedican[te]; [l]uxerunt bona industria ma[trem]. Mortua uxore non altera ducta annis invitus vixit octoginta et coniux illi iuncta quinquaginta. Sulpicius Primus and [Sulpicia] Faustina were born of two brothers who were very close. They were happily married, as destiny decided, with Urbica, their mother, lending her voice to destiny; they mourned this devoted mother. After the death of his wife, he (Primus) did not remarry; he lived, against his will, for  years and he was united in marriage with his wife for  years.
Stemma [Sulpicius]
Stela I (nd cent.) Laetoria Rufina dead at  years
C. Sulpicius Primus dead at  years
Stela II (end nd–rd cent.)
Sulpicius Faustus dead at  years
Sulpicius Primus dead at  years
Sempronia Vrbica Masculitana dead at  years
Sulpicia Faustina dead c.  years Stela III (rd cent.)
Stemma (after Lassère : )
Family and Kinship in Roman Africa

. The Vetidii of Thubursicu Numidarum ILAlg. . – with complete stemma (Thubursicu Numidarum, Algeria)
Vetidius Mustus
Petronia Frontilla Vetidius Mustiolus Q. Vetidius Felix
Vetidius Vetidianus
Vetidia Mustacia
Q. Gellius Honoratus
Gellia Honorata
Q. Vetidius Iuvenalis
This page intentionally left blank
 Children and Parents on the Tombstones of Pannonia Mary T. Boatwright
A bias and the turbulent modern history of central Europe have contributed to the relative obscurity of Roman Pannonia in the English-speaking world. Yet this northern frontier province at the bend of the Danube was key to the empire’s history, and the artefacts from its Roman period are visually striking and informative. Pannonian stelae are especially remarkable. Combining image and Latin text, these often massive tombstones uniquely illuminate the individuals they memorialize as well as aspects of life on this frontier. This chapter focuses on one feature of the Romano-Pannonian tombstones, a recurrent emphasis on ‘family’ that is evidenced both visually and in the epitaphs. We can see this in a very pronounced form on a third-century stela commemorating Aelius Munatius and his family (Fig. .).1 Research on this chapter was aided by Duke University’s Arts and Sciences Committee on Faculty Research and by a National Endowment for the Humanities Summer Stipend (). Special thanks are due to Zs. Mvár, A. Facsády, M. Sasel Kos, and the many other Hungarian, Slovenian, and Austrian colleagues who opened their museum storerooms and expertise to me. I am grateful also to C. Brian Rose and others at the University of Cincinnati, where I gave an early version of this chapter. Finally, although any mistakes remain my own, the encouragement and suggestions of K. Bradley and M. George were instrumental for this piece. This chapter is dedicated to the memory of my mother 1 RIU  0001 AE .  0001 ILS ; limestone,  cm. H,  W,  D; letter height, .–: D M Ael(io) Munatio caps(ario) coh(ortis) (milliariae) Hem(e)s(enorum) stup(endiorum) XXVIII dom(o) Sam(osata) Aur. Cansauna con [or con(iux)] Ant(onio) Basso vex(illario) sec(undo) her(ede) sanctiss(imo) coniug[i] con se natib(us)q(ue) suis fecit m(onumentum) m(emoriae). Further bibliography found in Die römischen Inschriften Ungarns, ad loc. This corpus of inscriptions hereafter referred to as RIU.

Mary T. Boatwright
. . Stela of Aelius Munatius, Aurelia Cansauna, and family, Intercisa (Hungarian National Museum, Budapest, RIU )
Children on the Tombstones of Pannonia

The tombstone was raised at the cavalry camp Intercisa (in Pannonia Inferior) for Aelius Munatius, Aurelia Cansauna’s ‘most revered husband’ (sanctissimo coniugi), who served the cohort there as a specialized attendant and had come from Samosata; Aurelia and a second heir were responsible for the monument. In its portrait zone husband and wife are depicted with four children, one an infant whom Aurelia holds to her bared left breast. This startling image, like others from Pannonian tombstones memorializing families, appears all the more peculiar when contrasted with funerary portraiture from Rome and its environs. The analysis of the Pannonian tombstones sheds light on cultural ideals in this frontier province, where Roman and non-Roman alike apparently construed ‘family’ in terms of affection between husband and wife, parent and child, and siblings. Some background information on Pannonia and its Roman history is in order. Pannonia, much of which roughly coincides with western Hungary, was conquered by the Romans in the late Republic and the Augustan period. It remained part of the empire for over four centuries, though not a particularly secure part.2 The area was critical to communications in Europe. Through it ran the great Amber Route, linking Aquileia and the Baltic Sea. More significantly, the Sava, Drava, and Danube river valleys, ultimately connecting Europe to the Black Sea, provided Rome’s major links between its north-west and north-east territories. In –  Octavian subjugated the southern region of later Pannonia as part of Illyricum.3 The peoples the Romans confronted were Illyrians mixed with Celts, who had begun arriving from the west three centuries before. The Celtic presence is more visible in the north. Fiercer Illyrians dominated in the south.4 The Romans advanced the boundaries of Illyricum to the Danube in – , enslaving and deporting most of the adult 2 For this history, see Millar : –; Wilkes a; and Wilkes b, focusing on southern Pannonia Inferior. 3 This province was later divided into Dalmatia (along the Adriatic coast) and Pannonia: Wilkes a: –; b: –. 4 See the elder Pliny, HN . . Alföldy (: ) locates the tribes once called ‘Illyrii’ but now considered ‘Pannonii’ between the Sava and the Drava, and the Celtic tribes to the west, north, and extreme south-east; Mócsy : . None had a centralized power structure as strong as that of tribes in Noricum or Dacia.

Mary T. Boatwright
males south of the Drava.5 Such brutal treatment resulted in the great Pannonian–Dalmatian Revolt of  –, savagely fought in the south.6 The northern tribes, such as the Boii and Eravisci, were more acquiescent, apparently preferring the Romans to the Transdanubians on their east and north. In this northern region the Roman province Pannonia was established in  ,7 divided into Upper and Lower Pannonia in , and further subdivided under Diocletian.8 Pannonian life was not tranquil. The presence and movements of non-indigenous peoples were constants. Most noticeable were the Roman soldiers required by the exposed location of the province. By the end of the first century  four legions and numerous auxiliaries served here, and the numbers rose to some , men, probably  per cent of all Roman troops, by the early third century.9 Although many auxiliaries serving here were 5 Wilkes b: , and a: –: Tiberius led the troops from  to   and the Romans established four forts: Siscia and Sirmium on the Sava, and Poetovio and Mursa farther north along the Drava. Material considerations were apparently inconsequential for the Romans: according to Strabo (. . ), the only imports from the Danubian provinces were slaves, cattle, and animal skins. 6 Velleius Paterculus . . : the Pannonian Illyrians (Desidiates and Perustae) of modern central and eastern Bosnia had to be virtually exterminated before the area could be pacified. Wilkes (a:  n. ) notes that Tiberius’ victory is ‘the likely subject’ of the Gemma Augustea, and that the Perustae were among the ethne featured in the Julio-Claudian Sebasteion at Aphrodisias (see Smith : –, and his pl. ; the ethne’s names are spelled differently in the sources). For the campaign, see also Suetonius, Tiberius , Cassius Dio . . ; Velleius Paterculus . ; and Wilkes a: –. 7 The presumable Tiberian fort or settlement at Carnuntum, the first legionary camp in the north of the new province, has not yet been attested archaeologically. The Roman troops in Pannonia revolted in  , only five years after the province was established: Tacitus, Annales . –. 8 Alföldy : . The division of  separated the German and Sarmatian fronts into Superior and Inferior (respectively). During the Tetrarchy Pannonia Superior was subdivided into Pannonia Prima in the north (cap. Savaria), and Pannonia Ripariensis or Savia in the south (cap. Siscia); Pannonia Inferior became Pannonia Valeria in the north (chief sites, Aquincum and Sopianae), and Pannonia Secunda in the south (cap. Sirmium). 9 Visy : –: the four legionary forts were at Vindobona, Carnuntum, and Brigetio in Pannonia Superior, and at Aquincum in Pannonia Inferior. At least  cavalry alae and  cohortes are known from diplomata. See also Wilkes b: .
Children on the Tombstones of Pannonia

Pannonians themselves,10 other auxiliaries and legionaries came from elsewhere: the Rhine area, North Africa, and, especially from the s until the s, Syria.11 Many not forming part of Rome’s military forces also were brought into the province. Tiberius seems responsible for moving the Illyrian Azali from south in the province to between the more northern Boii and Eravisci.12 As early as   the Romans settled Germanic Transdanubians south of the Danube,13 and by the end of the century they had transplanted to Pannonia at least one Celtic group.14 The Romans continued such intrusions periodically through the fourth century.15 When in  the Constitutio Antoniniana granted Roman citizenship to all free inhabitants of Rome’s dominion, its boon may not have extended to dedicitii, non-Romans transplanted to Roman lands. If that is the case, the law’s effects may have been less profound in this province, which saw new groups of Germanic, Celtic, and other tribes from across the Danube sporadically settled as aliens.16 Civilian Romans were also integral to the province, and at least originally they represented an occupying group. By the mid-first century  Claudius established Savaria as the Romans’ first From the mid-st cent. on: Mócsy : , –. See e.g. for North Africans, Di Vita-Evrard ; Fitz , for Syrians; and see nn.  and  below. 12 Mócsy : . 13 In   Romans transplanted here the followers of Vannius, a ‘friendly king’ who with Rome’s help had ruled the Suebi above the Danube until ousted by rebellion (Tacitus, Ann. . –). This is analogous to the better known settlement on the lower Danube of ‘more than , Transdanubian peoples, along with wives and children, chiefs or kings’ by Tiberius Plautius Silvanus Aelianus during Nero’s reign: ILS . See Wilkes a: –. Mócsy (: –), with archaeological evidence, argues for other transfers of ‘barbarians’ into northern Pannonia in the Julio-Claudian period. 14 The Cotini, apparently Celtic (Tacitus, Germania ), were relocated south of Lake Balaton: Mócsy : . 15 During the temporary truce of , and upon demand from the ‘barbarians’ themselves, Marcus Aurelius settled unknown numbers of Transdanubians in Pannonia, Moesia, Dacia, Thrace, and even Italy: Cassius Dio . . –. Analogous settlements occurred in the rd and th cents.: e.g. Wolfram : . 16 Without comment, Wolfram (: –) excludes from the grant dedicitii and their descendants (laeti). But the matter is not so clear: see SherwinWhite : –. 10 11

Mary T. Boatwright
veteran colony in northern Pannonia, the area furnishing the tombstones I discuss.17 From the Flavian era into the third century other northern towns were settled as colonies, granted municipal status, or granted colonial status (beginning in the second century).18 But Pannonia was never as urbanized as Italy, for example, and its cities developed slowly in terms of political and physical infrastructures alike.19 Pannonian life was often disrupted by matters outside the province itself. Its troops were moved elsewhere when Rome needed: for example, it served as a springboard for some of Trajan’s Dacian campaigns,20 and in  Septimius Severus launched his bid for the imperial power with the three legions of Pannonia Superior that followed him as their governor.21 At other times hostile invasions occurred. From  to  Quadi, Marcomanni, and Iazyges swept into the province from the east and north almost yearly. On the Column of Marcus Aurelius in Rome the Transdanubians are typified as barbarians: men are shown as trousered, with unkempt beards and wild hair, and women are depicted as dishevelled, often with one breast exposed, and at the prey of the Roman soldiers. Despite the Roman superiority over the barbarian such iconography proclaims,22 in the late third, fourth, and early fifth centuries Transdanubians again invaded the province in thrusts towards 17 Savaria is actually Pannonia’s second colony, since the first, of veterans and Italian citizens, was established under Augustus (perhaps as early as the s ) at Emona, on the site which the camp legio XV Apollinaris vacated when moved north to Carnuntum. 18 Borhy (: ) illustrates this process with the civilian settlement at Brigetio. Vespasian established Siscia and Sirmium as colonies; Trajan, Poetovio; Hadrian, Mursa. The three Flavian municipia are Neviodunum, Andautonia, and Scarbantia. Settlement was more rapid along the Sava and Drava than farther north: Borhy : . 19 About % of Pannonia’s population lived in cities, making it about as urbanized as northern Gaul; cf. Italy’s %: Alföldy : ; Woolf : –; Millar : . Pannonian towns developed only in the late nd cent. and the rd: e.g. Mócsy : , . 20 e.g. Gabler . 21 He later liberally thanked the province for its support, encouraging great prosperity in the early rd cent.: Fitz . 22 See Pirson  and Zanker . Millar (: ) cites damage from this period attested archaeologically in Aquincum and Carnuntum.
Children on the Tombstones of Pannonia

Italy.23 By the mid-fifth century Romans gave up the area as Hunnish.24 The Pannonian tombstones are particularly valuable in light of the scarce, and biased, notice Pannonia receives in ancient literary sources. For Romans in Rome, Pannonia was peripheral, the land and its inhabitants barely civilized. Positioning the area outside civilization and humanitas, the elder Pliny characterizes Pannonia as ‘acorn-bearing’ (glandifera Pannoniae, HN . , implicitly contrasting the province to the more civilized grainbearing Mediterranean regions). Aurelius Victor dwells on its frightful forests (immanes silvae, Caesares . ).25 Almost no senators are from here, and Decius and other memorable emperors who hailed from this region in the third and fourth centuries are typed as uncultured men of action, great warriors but unpredictable primitives.26 Cassius Dio, who served in Pannonia in the s, describes life in the province as miserable and uncultured. But he ends, ‘For all that the Pannonians are considered the bravest of all men . . . for they are very high-spirited and bloodthirsty, as men who possess nothing that makes an honorable life worth while’ (. . ; Loeb translation by E. Cary, ).27 23 Commodus constructed army lookout posts at spots along the Danube ‘exposed to hidden crossings by bandits’ (CIL . ). The more concerted, later invasions were by Iuthungi, Vandals, Suebi, and Sarmatians: Millar : –. The invasion of  reached as far as Italy, and was stopped only by Aurelian. 24 By  , the time of Priscus’ famous visit to Attila’s Hunnish court north of the Danube, all of the area within five days’ travel south of the mighty river was part of Attila’s ‘frontier’: Whittaker (: ), citing Priscus, History, fr. , p.  Blockley. 25 Herodian (. . –) reports criticism of Pannonia’s cold and foggy climate, which made for poor harvests; Tacitus has the insurgent troops of Pannonia lament that at retirement there they receive not fields (nomen agrorum), but morasses and mountain slopes (uligines paludum vel inculta montium: Tacitus, Ann. . ). 26 See e.g. Aurelius Victor, Caesares . ; Historia Augusta, Aurelian . ; and Birley . 27 Earlier, Velleius Paterculus said that Pannonians were not to be trusted: they had adopted Roman ways too quickly and only superficially (. . ). See also Fronto, Principia Historiae . , and Herodian . . , who remarks that ‘the inhabitants of the district of Pannonia are tall men of fine physique, natural and fierce fighters, but intellectually dull and slow-witted when it comes to crafty words or subtle actions’ (Loeb trans. by C. R. Whittaker, ).

Mary T. Boatwright
Yet most Pannonian stelae apparently espouse what Cassius Dio would probably consider ‘honourable’ and Roman, the family. Pannonia is one of the areas of the Roman world in which the deceased were commemorated by tombstones carrying Latin inscriptions and portrait relief. Although other types of funerary monuments were also used in Pannonia, stelae are more common, particularly in the northern area from Carnuntum down to Sopianae (modern Pécs; the region corresponds roughly with western Hungary today).28 They are usually worked on only one side. Of limestone, sandstone, or (rarely) marble, the Pannonian tombstones tend to be massively conspicuous, over  m. tall, about – cm. wide, and – cm. deep.29 They were in use from the second half of the first century  into the fourth, a longer period than for comparable tombstones from elsewhere in the empire, and their floruit is the second and third centuries . When with funerary portrait relief, the common configuration during the second and third centuries, the portrait ‘zone’ is placed near the top of the stela, often as a slightly elongated rectangle (see Fig. .). When more than two individuals are depicted, they frequently overlap in the relatively restricted space, contributing to an impression of interaction and affection. In Pannonia the use of figured and inscribed stelae began under influence from northern Italy and the Rhineland. Before the Romans arrived, the inhabitants of Pannonia practised nothing similar and apparently had no written language; further, the Roman occupation of north Pannonia began with soldiers from north Italy and Germany. Some of the earliest tombstones here are to soldiers from those regions.30 In the appearance of this 28 Mócsy explains this geographical incidence as reflecting where Romans settled first, where legionary and auxiliary camps were clustered, and where trade contacts were most frequent and there was a pre-existing Celtic presence (: –). 29 Most are of local limestone, perhaps a fifth or less are of sandstone, and only a handful are of (imported) marble or some other stone. The limestone can be good—that of nearby Noricum is compact, crystalline, and fairly white—or fissured, dark, and with inclusions. The red or yellow sandstone is friable. It is often held that the incidence of commemoration varied locally according to the availability of suitable stone, as in Britain: see Mann . 30 See e.g. the tombstone of Castricius Victor from Como, who served and died in Aquincum in the late st cent. and is depicted standing in his military uniform (CIL .  0001 Schober, no. ), and see Mócsy (: –), who
Children on the Tombstones of Pannonia

Roman custom in the new province we may see a desire, on the part of the Roman military, ‘to assert and maintain its identity’ in its protectorate.31 The relationship between military and civilian was very close in Pannonia,32 however, and the use of figurative tombstones spread outside the camps almost immediately. (See the late first-century tombstone of Cassus and Strubilo from Scarbantia, discussed below.) Such tombstones were revitalized in Pannonia when, from the s until the s, Intercisa was manned by Syrian auxiliary troops, who maintained many of their original traditions in this ,-man cavalry camp overlooking the Danube.33 Even outside of Intercisa the use of stelae continued relatively frequently into the third century.34 Whatever their origins and influences, Pannonian tombstones are distinctive in many ways.35 One is their emphasis on what seems to be the ‘nuclear family’—husband and wife, parent and children, and siblings. This aspect emerges in even greater prominence in comparison to funerary depictions of families from Rome and its environs. The emphasis can be seen even on some tombstones for soldiers erected before , when Septimius Severus lifted the legal ban on soldiers’ marrying while in service.36 Grouped individuals, apparently nuclear families of varying configurations, are frequently shown on Pannonian tombstones. An elaborate marble stela (second-century) from Savaria, for notes as influenced by the Rhineland Aquincum’s st-cent. full-length figured military tombstones, and those with eagles with outstretched wings (e.g. CIL .  0001 Schober, no. ). 31 Cf. Hope b: , on the appearance of military tombstones in Roman Britain. 32 Mócsy : ; Borhy () examines Brigetio as a case study. 33 The cohort I Hemesenorum, an equestrian auxiliary cohort from Syrian Emesa, was transferred to Intercisa under Marcus or Commodus, and was reinforced with troops from Syria under Septimius Severus: Mócsy : , ; Fitz . My Figs. . and . are stones raised by or to soldiers in this cohort. 34 The decline of figured tombstones in Germany after the st cent., and their relative rarity in Britain (for both, see Hope b: –), emphasize the distinctive longevity and frequency of the custom in Pannonia. 35 One might infer this from the recurrent epigraphic specification found in these epitaphs, that the one erecting the monument made it while still alive: see n.  below. 36 See Debrunner-Hall , on inscriptions from Carnuntum; Phang : esp. ch. ; and nn.  and  below.

Mary T. Boatwright
example, depicts together a mother, father, and son (Fig. .).37 The inscription specifies that a ‘most wretched’ mother, Comminia Valagenta, erected the monument, while still alive, for her son, Tartonius Secundinus (a man of ‘incomparable piety’ (inconparabili pietate) who died, aged , while on a military expedition). Also commemorated by her tombstone are her husband, Tartonius Finitianus Daiber (a veteran who died aged ), and another man, Comminius Optatus. Comminius’ name indicates that he is her brother. Since his age at death is not expressed in the epitaph, he may have been alive when Comminia commissioned the tombstone, and he seems not to be depicted. In the portrait a woman (l.), undoubtedly Comminia, drapes her left arm naturally over the shoulder of the full-grown youth to her left. Her left hand appears on his left shoulder, and with the first two fingers of her right she points at him. The youth holds a military sword, probably a sign that he is Tartonius Secundinus. To his left and a bit behind stands an older, slightly larger and bearded man, who holds a scroll. This is probably the father, a Pannonian who received Roman citizenship (signified by the scroll) after serving in Rome’s auxiliary troops. Other stelae are more humble, but express familial links just as conspicuously. For example, Scarbantia, a town along the Amber Route that received municipal status in the Flavian period, provides a late first-century tombstone attesting Cassus, a slave (ser[vus]) of Musa, and his wife (uxor) Strubilo, the freedwoman of Scalleo.38 The rough portrait (not illustrated in this chapter) shows the heads of a woman (l.) and man. The woman wears a Norico-Pannonian turban, an element of native dress.39 The stone was commissioned by the children of this indigenous couple. Despite Cassus’ servile status and presumable inability to 37 RIU  0001 CIL .  0001 Schober, no. ;  H,  W,  D, letter height, .–.: D M Comminia Valagenta an vv v(iva) f(ecit) sibi et Tartonio Secundi[no mil(iti) leg(ionis) X]IIII. G(eminae), de[func]to in expediti[o]ne inconparabili pietate an(norum) XXV mater infelicissima et Tart(onio) Finitiano Daiberi vet(erano) LXX et Com(minio) Optato mil(iti) l(egionis) XIIII an(norum). 38 CIL .  0001 0001 CSIR Österreich ., Scarbantia, no. : Cassus Musa [sic] ser(vus) annor(um) C, Strubilo Scalleo(nis) lib(erta) uxor ann(orum) LX. H(ic) s(ita) e(st). Fili(i) posierun(t). Correct Latin would read: Cassus Musae servus . . . et Strubilo . . . H(ic) s(iti) s(unt). 39 For this type of turban and other indigenous dress, see Garbsch : – and passim, with illustrations.
Children on the Tombstones of Pannonia

. . Stela of Comminia Valagenta and family, Savaria (Hungarian National Museum, Budapest, RIU )

Mary T. Boatwright
contract a legal marriage or make a binding will, his children identify themselves as sons/children ( filii) and call their mother Cassus’ wife (uxor) in the short epitaph. In terms of understanding the spread of ‘Roman’ inscribed and figured tombstones in Pannonia, we should note that Cassus and Strubilo were not the dependants of Roman citizens. Further, the stated servile status of Cassus, and the portrait’s depiction of him and his wife alone (without their presumably more privileged children), caution against simplistically applying to Pannonian tombstones the purpose of memorializing social mobility.40 Trios or couples like the examples discussed above have some similarities with funerary relief from Rome and its environs. Less common in other locales are slightly more extended family units that seem equally loving. On a second-century stela, found in Intercisa but perhaps made in Aquincum, we see three individuals identified by the epitaph as Demiuncus, son of Coucus; his wife (con[iunx]) Anculata, daughter of Caupio; and a sister (soro[r]) Veneria. The stone was raised by two liberti (Fig. .).41 The strikingly indigenous dress of the two women, who are depicted at left and slightly larger than the man, includes winged fibulae and the Norico-Pannonian turban with veil. The bouffant hair of the man assimilates his portrait to the female ones so dominated by the turbans.42 Again the figures overlap, each one’s left side slightly hidden by the figure to its left. The somewhat descending heights of the three from left to right emphasizes the left hands of the two women, which are placed on the left shoulders of the person next to them. The gestures of the women seem to me to exclude the notion that Veneria is the sister of one of the liberti; perhaps Veneria is the sister of Anculata, and the order is Veneria, Anculata, and Demiuncus. 40 Cf. Woolf (: esp. –), who begins with noting that monuments may generally be seen as ‘responses to perceptions of insecurity’ (), but in his investigation of Roman inscriptions goes on to stress social considerations over more essential ones of life and death. 41 RIU  0001 Schober, no. ; I follow the text proposed by Visy (: –, no. ; RIU reads: Angulata Campionis f and Batalus). Limestone,  H,  W,  D, letter height .: Osa Demiunci lib(erta). D M Demiuncus Couci f(ilius) an(norum) C et Anculata Caupionis f(ilia) an(norum) C con(iunx) t(itulum) p(osuerunt) Bataulus et Louco lib(erti) f(aciendum) c(uraverunt) [et] Veneriae soro[ri vi]vae. 42 The small axe he holds may also indicate non-Roman status.
Children on the Tombstones of Pannonia

. . Stela of Demiuncus and family, Intercisa (Hungarian National Museum, Budapest, RIU )

Mary T. Boatwright
A remarkable number of Pannonia’s tombstones represent children, at all ages, together with other members of their family. A fragmentary stela from Intercisa unusually depicts outside the portrait zone what seems to be the younger of two deceased daughters. The image of the girl, who is called Otiouna, is partly below, partly protruding into, the squarish portrait area that shows the more regular grouping of mother, father, and almost adult child (Fig. .: RIU ; the older sister was  when she died).43 The -year-old girl’s smaller size and youth seem reflected in the parents’ choice of parvae rather than filiae to designate their two daughters. A damaged stela from Ulcisia Castra dedicated to members of the Flavii family retains most of its epitaph and portrait zone (Fig. .).44 We see three adults—again apparently mother, older sister, and father—in front of whom stand two smaller children. The incomplete epitaph identifies a boy who died at age , a sister who died aged , and another sister who died aged . The mother protectively clasps the little boy to herself with her right hand. Although the small girl is not touched by her relatives, she appears sheltered between the fruit basket her elder sister carries, and the scroll her father holds in his left hand. Many other Pannonian tombstones depict children and parents exhibiting apparently affectionate gestures and interaction. As in illustration ., young children tend to appear in the bosom of their family rather than alone.45 Indeed, as we see below with RIU  and  (Figs. . and .), some Pannonian stelae feature children in the portrait zone but do not name or mention them in the epitaph. Perhaps in this region, conventionally scorned as uncultured by writers in Rome, a greater significance was assigned to imagery than to text. The 43 Limestone;  H,  W,  D, letter height, : Otiouna vixit an(nos) XII et Regilia an(norum) IIII h(ic) s(itae) sunt. Ianuarius (et) Otio[u]na p[a]rvis (et) sibi [vivi] posuerunt. The unusual placement of the child’s depiction suggests that Regilia died after the stone was commissioned at the death of her sister Otiouna. 44 RIU  0001 CIL . 2; limestone, now  H,  W,  D, letter height .–.: D M T(ito) Fl(avio) Constantino ann(orum) V et Flaviae Exsuperatae ann(orum) XX et Flaviae Constantinae ann(orum) II fili(i)s T—tas et[. . . . 45 My still incomplete database, which now includes only stelae (and some sarcophagi) with largely extant inscriptions and portrait relief that are catalogued in RIU, has only one example of a child portrayed alone (RIU ), against  examples of children in group settings.
Children on the Tombstones of Pannonia

. . Stela of Otiouna, Intercisa (Hungarian National Museum, Budapest, RIU )

Mary T. Boatwright
. . Stela of the Flavii, Ulcisia Castra (Aquincum Museum, Budapest, RIU )
Children on the Tombstones of Pannonia

incongruity may also suggest prefabrication of stelae to the point of their epitaphs, which could then be filled in by customers who bought their tombstone ‘ready made’. Even if this latter explanation is valid, however, the presence of children on pre-carved funerary portraits indicates widespread desire in the province to be commemorated as a family of parents and child(ren). A first reaction to the Pannonian tombstones might be that such a stress on family is Roman.46 In  R. Saller and B. Shaw used the epitaphs of military Pannonian stelae, although not their images, as part of a wide-ranging article on Roman tombstones and family relations. According to their analysis, over  per cent of Pannonia’s stelae commemorate family units, being raised by the husband, wife, son, daughter, or parent of the deceased.47 The concurrence of the Pannonian evidence with epigraphic data from other western Roman provinces led Saller and Shaw to argue that the ‘nuclear family unit’ was a Roman phenomenon, not something from a more modern period. Their conclusions have been challenged on various grounds,48 but for Pannonia my analysis of the tombstones corroborates theirs. My incomplete database of more than  figured Pannonian tombstones includes  dedicated by a wife,  by a husband,  by a father (in  cases also identified as a husband), and  by a mother (identified also as wife in  cases). Thirty-six include images of children. Further, according to extant epitaphs,  tombstones of my database commemorate  family members; 46 Indeed, in the only remark about children’s presence on the Pannonian stones have seen, Schober (: p. ), comments on the stela commemorating Aelius Munatius and his family (my Fig. .) that the depiction of four children refers to the ius liberorum. This interpretation is less persuasive when we consider how many tombstones with children are for non-Romans. But see George, in this volume, on funerary reliefs from Roman Italy and the appropriation of ‘Roman’ self-representation by individuals and families originally excluded from Roman society. 47 Saller and Shaw : esp. . They restrict themselves to the military population in Pannonia, relying on Schober, Hofmann, and CIL . – (: ). 48 For instance, Roxan () holds that the proportion of commemorated marriages among military populations is less than Saller and Shaw maintain; she uses Pannonia as her counter-example and maintains the evidence is skewed by the high proportion of stones from Intercisa. In a more general analysis of Roman tombstones, Meyer () downplays affection to highlight legal issues, while arguing that the installation of tombstones was motivated by inheritance laws. The high number of non-Romans on my stones discredits her interpretation.

Mary T. Boatwright
 commemorate ; , ;  stones commemorate  family members;  stones, ; and  tombstones commemorate . Less than ten tombstones, however, mention liberti/libertae.49 The nuclear family predominates on the Pannonian stones. Saller and Shaw’s work has figured in other controversies, for instance about the intensity of parental affection in antiquity.50 Although I personally find the Pannonian images of parents and children quite moving, I do not assume these tombstones tell us the true feelings and emotions of the deceased or their survivors. On the other hand, the Pannonian evidence can reveal the cultural values and ideas of the society in which the stones were raised. But the ‘Roman’ society the Pannonian tombstones illustrate looks different from that of Rome itself. Many modern scholars concur with Aulus Gellius (Attic Nights . . , . . ), that (legitimate) child-bearing and child-rearing were the purpose of Roman marriage.51 The stupifyingly high rate of infant mortality in the Roman world—it is roughly estimated that half the children born in any year died before the age of — helps explain this aim.52 Albeit greatly desired by most, children were problematic, and they were not often depicted in Roman art. They appear on surviving Roman works only at the end of the Republic.53 Although infants were always under-represented, children became more regular subjects after Augustus promoted 49 Only three, RIU , , and , are solely from alumni or liberti to a patron. Merely  stones were raised by someone identified as an heir or heirs, and of this group three of the heirs are also identified as wife, one as libertus, four as sons, and one as daughter. See also n.  above, for images of children on the tombstones. 50 See now Saller : , referring (e.g.) to Veyne and Foucault. A related discussion concerns the effect on parents of the death of infants and small children: see (e.g.) Parkin : , with references. I do not discuss here the wider controversy pertaining to the Romans’ conception of childhood. 51 See Treggiari : esp. –; Saller : ; Rawson a: . 52 Bradley (b: –) points out that Roman sources focus more on ‘the older child, the one . . . who has successfully negotiated [infant mortality]’; Parkin : –. 53 See Rawson a; and the forthcoming monograph by J. D. Uzzi on children in official Roman art (Cambridge). Rawson’s earliest image of a child on a funerary monument is the Servilii relief, which she dates to – : see references in n.  below. George  discusses the standing relief of a mother and daughter from the Palazzo dei Conservatori collection of the Capitoline Museums, dating it to c. .
Children on the Tombstones of Pannonia

citizen marriage and procreation, and grandstanded his adoptive sons Gaius and Lucius.54 But my reading of the visual evidence, much of which comes from Rome and its environs, is that most images do not show parents interacting with their children.55 The early Augustan relief of the Servilii family may serve as an extreme example. On it the freeborn child P. Servilius Globulus appears physically detached from his parents, separated from their likeness by a pilaster.56 The few representations of parent–child interaction tend to show one-sided interest; as on the Ara Pacis, the child tugs at an often indifferent or preoccupied mother.57 More children appeared in private and public art in the second century .58 Since the earliest images of this group date to the Trajanic era, such as the children carried on their fathers’ shoulders and in the personified city’s arms on the Arch of Beneventum,59 some tie the apparent rise of children’s depictions 54 Augustus’ marriage legislation, known from Propertius . . , Justinian, Institutions . . , and Gaius . – (inter alia), is discussed by Treggiari (: –); Parkin (: –); and Rawson (a: –), with bibliography. Rawson (a: –), tentatively dates to the time of C. and L. Caesar’s prominence ( ) the appearance of private busts and statues of boys and girls, and notes the number of children on funerary reliefs greatly increases between   and  . Bradley (b: esp. ) discusses the underrepresentation of infants in epitaphs, literature, and art. 55 The (Tiberian–Claudian) funerary relief of the Sertorii family, Kockel (: pl.  A and B), does show some affectionate interaction between the parents and their children: George : –. At times interaction depicted on images from Rome and environs seems unnatural: on Vat.  from the Museo Chiaramonti, for example, the father’s arm draped over his child’s shoulder is unrealistically attenuated. 56 CIL .  0001 Musei Vaticani, Museo Gregoriano Profano, inv. , often dated to – : see Kockel : –, and pl.  B; Zanker : fig. ; Rawson a: , fig. .. Kleiner (: cat. no. ) dates the relief to  – . Bradley (b: esp. ) argues that among the elite at Rome the widespread practice of wet-nursing and reliance on dependants for primary tasks in child-rearing combined with the high incidence of child mortality, for a minimum of parental interaction with a child. 57 A similar image comes from the Villa Doria Pamphilj: see Rawson a:  and fig. . ; Kleiner : cat. no. ; George : – and pl. .; Kockel : – and pl.  B. Kockel argues that the motif of the tugging child predates the Ara Pacis; Kleiner, that the motif was influenced by it. 58 Though Rawson (a: ) notes her earliest examples of monuments to young girls on their own come from the Flavian period (e.g. Kleiner : no. : an -year-old freeborn daughter of a man and his freedwoman wife). 59 See now Currie .

Mary T. Boatwright
to social and political changes in that period.60 Yet many children still appear alone, as on kline monuments.61 The monument of a young woman now in the Getty Museum may exemplify this type: on the lid of a sarcophagus and depicted as if on a kline, the melancholy adolescent toys with a small dog, her only companion, and isolates herself by her averted gaze.62 Depictions of parents and their children, often including more than one child, become more frequent in the second half of the second century. This is when Faustina the Younger bore Marcus Aurelius perhaps as many as fifteen sons and daughters.63 Although only six survived (one of whom was Commodus), their births and lives as children were celebrated on coinage that was widely disseminated.64 This emphasis was carried a step further in coinage struck for Lucius Verus and Lucilla, Faustina’s daughter, which depicts Fecunditas (or Lucilla), apparently about to nurse her child.65 The unusual image of a nursing mother and child has parallels on a few children’s sarcophagi that show the life cycle of the deceased. (We return to the nursing image at the end of the chapter.) This type of sarcophagus, which begins in the second century , now has only about eighteen (largely) extant examples, and two spectacular ones clearly include the depiction of a mother nursing her child.66 The sarcophagus of M. Cornelius Statius in the Louvre, which seems to come from Ostia and to 60 Rawson a: ; , looking particularly at the alimenta that Nerva began and Trajan expanded in Italy. 61 Kleiner : , ; cf. Toynbee : . 62 Wrede : –, on Getty inv. no.  AA., dating this controversial piece to the Trajanic period. 63 Fittschen : . 64 See Fittschen . 65 e.g. sestertius,  –, BMC . , pl. . : Lucilla or Fecundity, seated with infant; child standing r. & l.; legend: F[] SC. 66 For the type, see Huskinson ; Amedick : –; further on these two sarcophagi, see George : –. The example I do not discuss (Huskinson no. .  0001 Amedick no. , late nd cent.; dimensions unrecorded; photo in Amedick, pl. . ), originally from Tivoli, is now in the Villa Doria Pamphilj in Rome. Most of the few nursing women depicted on such sarcophagi are identified as (servile) wet-nurses by kerchiefs on their heads: Amedick : –. Some presumable mothers do not nurse but simply carry their babies at their breasts, as the mother in the carriage on Amedick no.  (pp. , –; Rome, Museo Nazionale Romano; c.  ). The Terra Mater relief of the Ara Pacis offered an exemplar for nursing scenes, as did images of Isis Lactans (Tran ). See also Sˇasˇel Kos : –, with further references.
Children on the Tombstones of Pannonia

date to  –, displays milestones of the young boy’s life on its front. From left to right we see him at four different stages. As a newborn, he is nursed by his mother with his father looking on. Still an infant but no longer in swaddling clothes, he is held in his father’s arms; then, ‘portrayed as a small-scale adult’, he playfully drives a ram-drawn chariot. Finally, he is shown declaiming in front of his father.67 The boy progresses quickly to adulthood, at least in his facial features, in a motif so common that it has been dubbed ‘puer-senex’.68 Despite such second-century examples, monuments from Rome and its environs show relatively few children interacting affectionately with their parents; instead, children are depicted more frequently in images reflecting status relations within the household.69 In contrast, on Pannonian tombstones children appear frequently, and in number, with their parents. As on some stelae illustrated in this chapter, such apparently loving families often include more than one child. Such imagery occurs equally on tombstones commemorating Romans and on ones for individuals identifiable by name and/or costume as indigenous. The imagery is sometimes employed even on stelae whose epitaphs are silent about the children depicted. For example, a second-century limestone stela raised by Suriacus Secuindinus [sic] for his ‘incomparable wife’ Aelia Vitalina (coniugi inconparabili) shows three adults and three children, although only the husband and wife are mentioned in the inscription (Fig. .: RIU ).70 The woman on the left, presumably Aelia Vitalina, has her left arm over the shoulder of the man (or woman?) next to her; her right hand rests on the right shoulder of the child in front of her. The man on our right, probably Secuindinus, rests his right hand protectively on the shoulder of the girl who stands before him 67 CIL . ; Huskinson : no. ., pp. –; Amedick : no. , pp. –: Louvre Ma : Luna marble, . H,  W, .–. D. Quote from Huskinson, p. . Kampen (b: –) remarks on the substitution of private life for more traditional public virtues found on adults’ biographical sarcophagi. 68 Cf. Huskinson : –; cf. Bradley b: . 69 George . 70 Otherwise unpublished, from east of Brigetio;  H,  W,  D, letter height .–.–.: D M et perpetu(a)e securitati Aeliae Vitalin(a)e coniugi inconparabili qu(a)e vixit annos XXXIII menses V dies XVIIII. Suriacus Secuindinus [sic] dec(urio) m(unicipii) M(ogetianae? or -ursella?) quattorvirus [sic] maritus f(aciendum) c(uravit). The term ‘perpetuae securitati’ is found also on later Christian monuments.

Mary T. Boatwright
. . Stela of Suriacus Secuindinus and family, Brigetio (Hungarian National Museum, Budapest, RIU )
Children on the Tombstones of Pannonia

holding an apple. Although the inscription does not identify everyone portrayed, it does specify that Suriacus Secuindinus was a decurion of a municipium (perhaps Mogetiana) and a quattorvir, asserting his Roman status and values.71 He seems to wear the toga. His wife’s Roman status is signalled by her duo nomina.72 In comparison to the commemorated Mira, discussed immediately below, and to some other women illustrated in this chapter, Aelia Vitalina appears Roman in hairdo and dress. Our next example, from the first half of the second century, shows family imagery and language on a Pannonian stela raised to and/or by individuals who do not obviously identify themselves as Romans. The ‘well-deserving’ wife Mira (con(iugi) b(ene) m(erenti) ), the daughter of Crescens, is commemorated by a sandstone stela in Ulcisia Castra raised by her husband, M. Attius Rufus, a veteran of the Second Legion (Fig. .: RIU ).73 As a veteran Attius Rufus wears the toga, but his wife Mira is marked as non-Roman by her single name, filiation from the non-Roman Crescens, and indigenous dress. She wears a NoricoPannonian turban (here with a long veil behind), a heavy torque, and huge fibulae to hold up the shoulders of her dress.74 A boy’s likeness is central to the portrait zone, despite the epitaph’s silence about any child. The adults’ hands closely hold between them their presumable young son, also dressed in a toga. Their gesture, which must have linked their arms behind his back, pushes the boy into the foreground of the group portrait. Such family imagery seems deliberate. These affectionate family groups are found on tombstones whose diverse workshops are 71 Municipal positions are not commonly noted in Pannonian epitaphs. The index for RIU – has three pages of references to ‘Städtewesen’ (of which one page comprises references to ‘augustales et seviri’ and to ‘collegia’), but  pages for ‘Militärwesen’. 72 Phang (: ) rightly notes that such women may be Roman citizens or Junian Latins, freedpersons informally manumitted by Roman citizens and with lesser rights. 73  H,  W,  D, letter height .–.: Mira Crescentis f(ilia) an(norum) XXX t(itulum) p(osuit) M. At(t)ius Rufus vet(eranus) leg(ionis) II. Ad(iutricis) sibi et con(iugi) b(ene) m(erenti) pos(uit). Visy (: , no. , and pl. ) provides the date. 74 Further affirmations of ‘ethnicity’ are the motifs of the intermediate zone: the horse-drawn carriage, and the tripod and servant at the funeral feast. See Visy .

Mary T. Boatwright
. . Stela of M. Attius Rufus and family, Ulcisia Castra (Balassa Bálint Museum, Esztergom, RIU )
Children on the Tombstones of Pannonia

revealed by the stelae’s materials as well as by their execution: for example, sandstone and limestone would be carved by different sculptors. Further, patrons’ choice and agency are indicated by the formula vivus/viva sibi fecit (or posuit)—made (or installed) this for himself/herself while alive—which frequently appears.75 We see it, for example, on a late second- or early third-century limestone stela from Intercisa (Fig. .: RIU ).76 While still alive Germanius Valens, who served in the cavalry troop transferred here from Syria, raised the memorial to his wife (uxor) Aurelia Baracha (who died age ), his two daughters (successively named Aurelia Germanilla before their deaths at age  and ), and his mother, Immosta. The portrait zone represents, under an elaborately segmented half-vault, three adults with two children in front of them. The two flanking adults, an older woman (l.) and a bearded man (r.), place their innermost hands on the shoulders of the central woman, whose importance is further signalled by the older woman’s pointing to her with her right hand. The younger woman pulls to herself with both hands a little girl with a variation of the top-plait hairstyle. In front of the man and sheltered with his left hand stands a slightly smaller girl with a different hairstyle. Here the affection of the close-knit family conspicuously transcends death, since the two daughters are shown as coexistent, simultaneously embraced by their parents.77 Although they died very young, they did not simply lapse from memory, as regrettably yet inevitably lost before their time.
75 About % of the inscriptions in my incomplete database carry this formula, which appears on slightly less than half of the tombstones discussed in this article: RIU , ,  (Figs. ., ., .), and CIL .  (Fig. .). It is not used on either tombstone that portrays a child without specifying a child in the epitaph. 76 0001 AE :  0001 Schober, no. ; limestone:  H,  W,  D, letter height .–.: D M Aureliae Barachae vixit ann(os) XXXV et Aur<e>l(iae) Germanillae vixit ann(os) IIII et altera filia Aurelia Germanilla vi[xi]t ann(os) II et Immostae matri su(a)e vixit ann(os) XL. Germanius Valens mil(es) coh(ortis) (milliariae) Hemes(enorum) uxori et matri et filiis posuit et sibi {v}vi(v)us fecit. 77 The relative ages at death of the mother, , and her daughter-in-law, , also suggest that the stone commemorates individuals who died at very different dates: women’s age at marriage was probably –. The identification of the mother on the stone suggests Immosta came with her son to Pannonia.

Mary T. Boatwright
. . Stela of Germanius Valens and family, Intercisa (Ferenczy Museum, Szentendre, RIU )
Children on the Tombstones of Pannonia

We should return to the striking image with which I began, the memorial of Aelius Munatius, Aurelia Cansauna, and their children (Fig. ., RIU ). Although the children are not named in the epitaph, only referred to in the phrase con se natibusque suis (‘with herself and her children’), they are as prominent on this tombstone as children on other Pannonian stelae. We have now seen other examples of multiple children individualized by different heights and hairstyles (Figs. ., ., and .). The overlapping of figures and the placement of young children in front of their parents must now also seem familiar (cf. Figs. ., ., ., and .). The emphasis on touch, as Aurelia’s hand supporting her baby grazes the head of the little girl before her, is manifest on all the Pannonian stones I have discussed. What is more unusual, however, is that Aurelia is shown about to nurse, with her breast bared. I have found two Pannonian comparanda for the nursing scene. One commemorates Flavia Aiulo, who died aged  in Aquincum; the now headless portrait shows a tightly swaddled infant, held with its face upwards, lying angularly across the mother’s clothed chest (CIL . ).78 The other, my Fig. ., comes from second-century Scarbantia. This elaborate limestone stela, mostly complete, memorializes Claudia Julia and her parents Tiberius Claudius Surus and Ulpia Restituta (CIL .  and p.  0001 CSIR Österreich ., Scarbantia, no. ).79 The epitaph of Fig. ., CIL . , asserts that the parents commissioned the stone while alive, for themselves and for their daughter who had died at the age of . Incongruous with the text, the portrait shows only a young mother cradling an infant at her breast. The mother is fully clothed, including conspicuous Pannonian fibulae at her shoulders and a heavy necklace and bracelet. Her baby is depicted realistically, nuzzling its mother and stretching out its arm, with its little hand blissfully curling in anticipation of nursing. M.-L. Krüger, who published this stela in CSIR Österreich ., ventures that the young woman is Claudia 78 Also published, with photograph, in Szilágyi :  and pl. : Flavia Aiulo ann(orum) XX h(ic) s(ita) e(st). Gallio sorori pientis(simae) et Avitus consobrin[i] [sic] pientissim[---]. Flavia’s name suggests a terminus post quem of the early nd cent. 79 Limestone,  H,  W (no depth given), letter height –.: Ti(berius) Claudius Surus et Ulpia Restituta v(ivi) s(ibi) et Cl(audiae) Iuliae fil(iae) an(norum) XX h(ic) s(itae) fec(erunt).

Mary T. Boatwright
. . Stela of Claudia Julia, Scarbantia (City Museum, Wiener Neustadt, CIL . )
Children on the Tombstones of Pannonia

Julia, who died in childbirth.80 But the analogy of other Pannonian stones like RIU  (Fig. .), which display emblematic affectionate family scenes, suggests that the portrait here may be of Ulpia Restituta, with Claudia Julia portrayed as an infant. Despite the uncertain identification of those portrayed, the image on the Scarbantia stela is a striking one. Nursing mothers are extremely rare in Roman art. In part this relates to the negative connotations the naked breast usually carried in Graeco-Roman art.81 Further, at least among the elite in Italy and the Greek East, Roman mothers seem not to have nursed their own children regularly, instead handing them to servile or otherwise dependent wet-nurses.82 Other than the three Pannonian stelae, the coins honouring Lucilla, and the two children’s sarcophagi discussed above, I know of only one other depiction of a mother who nursed her child. An early firstcentury  limestone stela from Cologne depicts Bella Rema, the daughter of Vonucus, holding her infant tightly to her clothed chest and partially enveloped in her mantle; it was erected by her husband Longinus.83 80 CSIR Österreich ., Scarbantia, no. , p. . Some support for Krüger’s interpretation might be the ‘declamation’ scene right of the portrait zone, which she notes as unusual in Pannonia. Perhaps here we have two elements featured elsewhere on childhood sarcophagi, the deceased’s birth and education: cf. Kampen b: –. The tombstone of Flavia Aiulo, CIL .  in n.  above, also depicts a young mother without mentioning her husband, but the epitaph is fragmentary. 81 See Cohen ; Bonfante () advances social, religious, and magical reasons for the reluctance to depict naked breasts of actual women. But the breasts of ‘barbarian’ women, such as on the Column of Marcus Aurelius, were often bared (n.  above). 82 Bradley b; Bonfante : –. The practice occasioned much controversy: Plutarch (Consolatio ad uxorem –, De amore prolis –, and the spurious De libris educandis ), and Aulus Gellius (Attic Nights . , referring to Favorinus); see Bradley  and Tacitus at n.  below. 83 Galsterer and Galsterer : no. , pl. :  H,  W,  D, letter height .–.: Bellae Vonuci f Remae Longinus vir illaeius fecit pie. The more famous nursing woman from Cologne, Severina nutrix (wet-nurse), is depicted on the top part of an early-rd cent. limestone funerary altar for a (now) unnamed man that was probably commissioned by Severina (or perhaps vice versa): see Galsterer and Galsterer : no. , pl. ; cf. Fildes : pl. . , and George : . Severina is shown on one side nursing an infant, and on the other leaning over a swaddled baby in what seems to be a cradle. I do not address here non-human nursing mothers; see nn.  and .

Mary T. Boatwright
The provenance of Bella Rema’s tombstone, like that of the three nursing Pannonian mothers, is significant. In Germania . , written at the end of the first century , Tacitus notes that maternal nursing was prevalent among German tribes. Although he adduces this notice in oblique criticism of Roman mothers’ handing their babies over to wet-nurses, a practice censored also by other authors,84 German and Celtic traditions, especially those relating to women, were particularly tenacious in Pannonia at least into the second century.85 Through the middle of the third century Poetovio (in southern Pannonia Superior), for example, had a well-attested epichoric cult of the Nutrices Augustae. These deities, originally Celtic but later Romanized and distinguished by the epithet Augustae, were depicted nursing. They have similarities to the various Matres and Matronae, ‘divine Mothers’ that were worshipped in the Germanies.86 Further, slaves were relatively rare in Pannonia in our period,87 inhibiting the use of female slave wet-nurses and childminders, the nutrix so scorned by Tacitus and others. The representation of a mother nursing her infant, found on three Pannonian stelae and depicted in my Figs. . and ., may reflect actual practices in the region. But even simply as an image, it is simultaneously powerful and tender. The image allows us to address directly the question of Romanization in Pannonia, since all the nursing mothers here are marked as Roman by the use of duo nomina. The ‘native’ cognomen of Flavia Aiulo (from Aquincum; CIL .  discussed above) suggests she was only recently enfranchised as a Roman, receiving citizenship and her nomen Flavia either personally or through her parents.88 The duo nomina of Claudia Julia, as well as the names of her parents (CIL . ; Fig. .), are Latin names more common in this province and elsewhere. Yet the costume of the young mother from Scarbantia, whom I tentatively identify as Claudia Julia, is decidedly non-Roman, with its large See also Tacitus, Dialogus . –. , and references in n.  above. Harl () investigates some  Pannonian items (not all stelae, and excluding Roman women) for the information they provide for continuation of Celtic women’s traditional power despite living in Roman Pannonia. 86 ˇ 87 Sasˇel Kos : –, with further references. Mócsy . 88 Moreover, her right shoulder, still visible despite the stela’s break, shows a large Pannonian fibula. 84 85
Children on the Tombstones of Pannonia

Pannonian fibulae and other jewellery.89 The slightly later depiction of Aurelia Cansauna and her family (RIU ; Fig. .) is even more confounding and interesting. Its provenance at Intercisa, and her husband’s specified origin of Samosata, argue that the family had Syrian origins. Yet Aurelia’s pose with her infant, and the prominent, affectionate display of her various young children in front of her and her husband, have parallels with Pannonian funerary material not from this Syrian enclave. These similarities indicate that ethnic and ideological boundaries were not rigid in this province. These nursing mothers, although extreme within the general iconography of Roman funerary portraiture, are less peculiar in the wider context of the Pannonian tombstones’ distinct emphasis on family. The emphasis united non-Roman and Roman in this province. The Pannonian stones’ familial emphasis, constant through the remarkably long use of figured tombstones here, cuts across simplistic binary oppositions of ‘native’ and Roman, civilian and military. Affectionate family portraits are frequent among all presumable groups, and distinct from what was common in Rome itself.90 The family portraits here let us see how ‘the Roman family’ was constructed in this border province, with its GermanoCeltic traditions and the exigencies of its location and strategic Roman roles. The many images of (multiple) children and their parents, interacting affectionately with one another, reveal that the nuclear family itself was valued. Less evident in the epitaphs and portraits, and thus presumably less significant overall, are the various distinctions of social status used frequently in Rome, Italy, and elsewhere to denote and demarcate servile, freed, and free, Roman and non-Roman, soldier and civilian. The stelae disclose strong idealization of the nuclear family in Pannonia. This can be linked to two interconnected aspects of this frontier zone’s history. One is the uncertainty of day-to-day 89 She may also be wearing a Pannonian turban rather than have her hair wrapped around her head in a more Roman coiffure. Phang (: ) notes that such ‘colorless’ names ‘may mask a native ancestry’; see also n.  above. Surus may indicate a Syrian origin for Tiberius Claudius. 90 Thus the use of figured tombstones in Pannonia differs from that in Britain, where Hope argues that they ‘were of particular relevance to immigrants and outsiders who used the medium to assert their identity in a strange land’ (b: ).

Mary T. Boatwright
existence, which may have rendered family life especially precious. From the first to the fifth centuries , Pannonia was repeatedly occupied and overrun by alien groups, be they Romans, Romano-Syrian cavalry, Transdanubians, or others.91 Male children enrolled in the auxiliary forces and legions at the age of , if not before. Upon marriage at an equally early age, women had little assurance that their husbands, and later their sons, would not be killed in battle or while patrolling the borders. It must also have been painfully clear that a family’s integrity and safety could not be guaranteed during invasions and at other unsettled times, and throughout history women and children have been the most numerous victims of wars, especially in occupied lands. Rome’s traditional but vague exaltation of family, particularly pronounced during and after the desperate Marcomannic Wars and attendant Antonine plague,92 may have resonated especially strongly here. Further, the affectionate nuclear family enabled presumably distinct identities in Pannonia to be negotiated and accommodated. We may see this hybridization most clearly in RIU  (Fig. .), where native Mira, daughter of Crescens, and her Roman veteran husband, M. Attius Rufus, both resplendent with marks of status and distinction, display their son between them. But such family imagery spread widely in Pannonia. As we have seen, it was used for families of quite different status and origins, including servile and freed natives (Cassus and Strubilo, in CIL .  0001 ), free non-Roman citizens (RIU , Fig. .), enlisted soldiers from Syria (RIU , Fig. .), and Roman municipal magistrates and wives (RIU , Fig. .). Pannonia’s tombstones, themselves intriguing in their imagery and epitaphs, also give us a glimpse into the process of Roman self-definition in this frontier province. 91 As Whittaker () and others now convincingly argue, Roman frontiers were zones of exchange, not impenetrable barriers. 92 This period was one when official Roman art and ideology most promoted the family, and when the empress Faustina and at least one daughter visited Pannonia in the company of Marcus Aurelius (in ): Philostratus, Vita sophistarum . .  (. . ); for date, see Fittschen : . Cf. Historia Augusta, Marcus . .
REFERENCES
A, A. (), Die munizipale Mittelschicht im kaiserzeitlichen Italien, Europaische Hochschulschriften: ser.  Geschichte und ihre Hilfswissenschaften, vol.  (Berlin). A, C. (), Things Fall Apart (New York). A, J. N. (), The Vulgar Latin of the Letters of Claudius Terentianus (P. Mich. VIII, –) (Manchester). A F, M. L. (), Organizaciones suprafamiliares en la Hispania antigua, Studia archaeologica,  (Valladolid). —— (), ‘Organizaciones suprafamiliares en la Hispania antigua (II)’, Boletín del Seminario de Estudios de Arte y Arqueología, : –. —— (), ‘Onomastique personnelle indigène de la péninsule ibérique sous la domination romaine’, Aufstieg und Niedergang der römischen Welt, . . : –. A, S. () (ed.), The Early Roman Empire in the East (Oxford). A, M. (), Trials in the Late Roman Republic,  to  BC (Toronto). A, G. (), ‘Notes sur la relation entre le droit de cité et la nomenclature dans l’Empire romain’, Latomus, : –. —— (), ‘La Pannonia e l’Impero romano’, in Hajnóczi (), –. —— (), Städte, Eliten und Gesellschaft in der Gallia Cisalpina (Stuttgart). A, S. (), ‘Quelques aspects du mariage dans l’Égypte ancienne’, Journal of Egyptian Archaeology, : –. A, P. (), ‘The Relationship between Wall-Decoration and Room-Type in Pompeian Houses: A Case Study of the Casa della Caccia Antica’, Journal of Roman Archaeology, : –. A, D. F.  (), Egitânia (Lisbon). A, R. (), Soldier and Society in Roman Egypt: A Social History (London, New York). —— (), ‘Houses and Households in Roman Egypt’, in Laurence and Wallace-Hadrill (), –. —— (), ‘The Ties that Bind: Soldiers and Societies’, in A. Goldsworthy and I. Haynes (eds.), The Roman Army as a Community, Journal of Roman Archaeology suppl.  (Portsmouth, RI), –.

References
A, R. (), The City in Roman and Byzantine Egypt (London, New York). Á S  B, J. (), ‘Museo Arqueológico de Mérida’, Memorias de los Museos Arqueológicos Provinciales, : –. A, R. (), Die Sarkophage mit Darstellungen aus dem Menschenleben, part : Vita privata (Berlin). A, M. (), Il testamento romano attraverso la prassi documentale (Florence). A, D. W. (), ‘Romanticizing the Ancient Medical Profession: The Characterization of the Physician in the GraecoRoman Novel’, Bulletin of the History of Medicine : –. A, J., and B, H. () (eds.), Parenté et stratégies familiales dans l’antiquité romaine, Actes de la table ronde des – octobre , Coll. EFR  (Rome). A-R, V. (), La successione testamentaria secondo i papiri greco-egizi (Palermo). A, L. J. (), Her Price is Beyond Rubies: The Jewish Woman in Graeco-Roman Palestine (Sheffield). A, A. (), Women and Law in Late Antiquity (Oxford). —— (), ‘The Guardianship of Women in Roman Egypt’, Akten des . Internationalen Papyrologenkongresses (Stuttgart and Berlin), –. A, P. (), ‘Marital Disputes in Greco-Roman Egypt’, Journal of Juristic Papyrology, : –. A, J.-J. (), Business Managers in Ancient Rome: A Social and Economic Study of Institores,  B.C.–A.D.  (Leiden). A, A., G, J., and W, P. (), ‘Inscriptions latines découvertes à Lyon’, Revue des études anciennes, : –. A, N. (), Discovering Jerusalem: Recent Archaeological Excavations in the Upper City (Oxford). B, R. S., and F, B. W. (), The Demography of Roman Egypt (Cambridge). B, C., et al. (), La Maison urbaine d’époque romaine en Gaule narbonnaise et dans les provinces voisines, Actes du colloque d’Avignon – novembre  with Atlas des maisons de Gaule narbonnaise, Documents d’Archéologie Vauclusienne ,  vols. (Avignon). B, J. P. V. D. (), Life and Leisure in Ancient Rome (London). B, D. C. (), ‘The Place of Residence of the Divorced Wife in Roman Egypt’, Akten des  Internationalen Papyrologenkongresses (Stuttgart and Berlin), –. B, J. M. (), ‘Celibacy’, in Schiffman and VanderKam (), –. B, D., and C, J.-L. () (eds.), De la ferme indigène à la villa romaine: La Romanisation des campagnes de la Gaule, Actes du
References

IIième colloque de l’association AGER tenu à Amiens (Somme) du  au  Septembre , Revue Archéologique de Picardie, no. spécial  (Amiens). B, M. (), Roman Nature: The Thought of Pliny the Elder (Oxford). B, M. (), ‘Looking (Harder) for Roman Myth: Dumézil, Declamation and the Problems of Definition’, in F. Graf (ed.), Mythos in mythenloser Gesellschadft. Das Paradigma Roms (Stuttgart and Leipzig), –. B, H. I. (), ‘Diplomata Antinoitica’, Aegyptus, : –. —— (), ‘Antinoopolis: A Hadrianic Foundation in Egypt’, Journal of Roman Studies : –. B A, A., and G, M. (), ‘Fouilles de la nécropole romaine de Pupput (Tunisie)’, Comptes rendus de l’Académie des Inscriptions et Belles Lettres, –. —— —— () (eds.), La nécropole romaine de Pupput (Rome). B, P., M, J. T., and  V, R. (), Les Grottes de Murabba‘ât, Discoveries in the Judaean Desert, , (Oxford) (0001 DJD ). B B A, Z. (), Catalogue des inscriptions latines païennes du Musée du Bardo, Coll. EFR  (Rome). —— (), ‘Un nouvel affranchi impérial de l’époque des Sévères’, in M. Khanoussi, P. Ruggeri, and C. Vismara (eds.), L’Africa romana: Atti del XII convegno di studio, Olbia, – dicembre  (Sassari), –. —— and K, M. () ‘La Saga des Sulpicii’, in M. Khanoussi, P. Ruggeri, and C. Vismara (eds.), L’Africa romana: Atti del XII convegno di studio, Olbia, – dicembre  (Sassari), –. B, L. K. (), ‘The Stem Family and the Developmental Cycle of the Peasant Household: An Eighteenth Century Austrian Example’, American Historical Review, : –. B, D. H. (), Cicero pro P. Sulla Oratio (Cambridge). B, J. (), ‘Household Artefacts: Towards a Re-interpretation of Roman Domestic Space’, in Laurence and Wallace-Hadrill (), –. B, J. (), ‘Enfants malades et maladies des enfants dans le Corpus hippocratique’, in P. Potter, G. Maloney, J. Desautels (eds.), La Maladie et les maladies dans la Collection hippocratique: Actes du VIe Colloque International hippocratique (Québec), –. —— (), ‘La Médecine des enfants à l’époque impériale,’ Aufstieg und Niedergang der römischen Welt, . . : –. B, L. () Le stele funerarie della Dacia, Archaeologica,  (Rome). B, A. R. (), ‘Decius Reconsidered’, in E. Frézouls and H. Jouffroy (eds.), Les Empereurs illyriens: Actes du colloque de Strasbourg (– octobre ) (Strasbourg), –.

References
B, M. V. (), ‘L’archivio di Tryphon, tessitore di Oxyrhynchos’, Aegyptus, : –; –. B, T. F. C. and M, M. () (eds.), The Early Roman Empire in the West (Oxford). B M, J. M. (), ‘Religión y sociedad en las inscripciones de Salamanca’, in M. Mayer (ed.), Religio Deorum: Actas del coloquio internacional de Epigrafía de la A.I.E.G.L.: Culto y sociedad en Occidente (Sabadell), –. B, B. (), ‘Acts and the House Church’, in D. W. J. Gill and C. Gempf (eds.), The Book of Acts in its First Century Setting, : The Book of Acts in its Graeco-Roman Setting (Grand Rapids, Mich.), –. B, S. E., and J, R. L. F. () (eds.), Sequence and Space in Pompeii, Oxbow monographs,  (Oxford). B, V. (), ‘Les Maladies des enfants et leur traitement d’après le témoignage de Pline l’Ancien’, in C. Deroux (ed.), Maladie et maladies dans les textes latins antiques et médiévaux: Actes du Ve Colloque International, Collection Latomus,  (Brussels), –. B, L. (), ‘Nursing Mothers in Classical Art’, in KoloskiOstrow and Lyons (), –. —— and S, J. L. () (eds.), The World of Roman Costume (Madison). B, C. (), Studies in Magical Amulets Chiefly Greco-Egyptian (Ann Arbor). B, P. () (ed.), Épouser au plus proche. Inceste, prohibitions et stratégies matrimoniales autour de la Méditerranée, EHESS (Paris). B, B. (), ‘Das Gesicht der Aufsteiger: Römische Freigelassene und die Ideologie der Elite’, in Maximilian Braun, Andreas Haltenhoff, Fritz-Heiner Mutschler (eds.), Moribus antiquis res stat Romana. Römische Werte und römische Literatur im . und . Jh. v. Chr. (Munich and Leipzig), –. B, L. (), ‘Romani e Pannoni. Aspetti dell’acculturazione in una provincia di frontiera’, in Hajnóczi (), –. B, D. (), Antike Grabaltäre aus den Nekropolen Roms (Bern). B, M. (), ‘Nouvelle inscription à Tipasa (Maurétanie césarienne)’, Mitteilungen des deutschen archaeologischen Instituts, Röm. Abt. : –. —— () Fouilles de la nécropole occidentale de Tipasa (–) (Algiers). B, P. (), Esquisse d’une théorie de la pratique, précédé de trois études d’ethnologie kabyle (Paris). B, K. R. (), ‘Wet-Nursing at Rome: A Study in Social Relations’, in B. Rawson (a), –.
References

—— (), Slaves and Masters in the Roman Empire (New York and Oxford). —— (), Discovering the Roman Family (New York). —— (), ‘Writing the History of the Roman Family’, Classical Philology, : –. —— (a) Slavery and Society at Rome (Cambridge). —— (b), ‘The Nurse and the Child at Rome: Duty, Affect and Socialisation’, Thamyris, /: –. —— (), ‘Images of Childhood: The Evidence of Plutarch’, in S. B. Pomeroy (ed.), Plutarch’s Advice to the Bride and Groom and A Consolation to His Wife (New York and Oxford), –. —— (a), ‘Romanitas and the Roman Family: The Evidence of Apuleius’ Apology’, Canadian Journal of History, : –. —— (b), ‘Fictive Families: Family and Household in the Metamorphoses of Apuleius’, Phoenix : –. —— (), ‘Children and Dreams’, in Dixon (a), –. B, F. (), Les Stèles funéraires à personnages de Bordeaux, Ier–IIIe siècles (Paris). B, D., and G, C. () (eds.), Myth, History and Culture in Republican Rome: Studies in Honour of T. P. Wiseman (Exeter). B, R.  (), The Limits of Participation. Women and Civic Life in the Greek East in the Hellenistic and Roman Periods (Amsterdam). B, E. H. (), ‘A Weaver of Oxyrhynchus: Sketch of a Humble Life in Roman Egypt’, Transactions and Proceedings of the American Philological Association, : –. B, H., and T, Y. (), Recherches archéologiques francotunisiennes à Bulla Regia, ii. Les Architectures, : Les Thermes memmiens, Coll. EFR  (Rome). B, B. (), ‘Konnten Frauen im alten Judentum die Scheidung betreiben?’, Evangelische Theologie, : –. B, M. (), ‘Excavations in the House of Caiaphas, Mount Zion’, in Y. Yadin (ed.), Jerusalem Revealed (Jerusalem and New Haven), –. —— (), ‘The Archaeology of Palestine’, in W. Horbury, W. D. Davies, and J. Sturdy (eds.), The Cambridge History of Judaism, iii (Cambridge), –. B, G. J. (), ‘Rethinking the Couvade: Cross-Cultural Evidence’, American Anthropologist, : –. B, P. A. (), Italian Manpower (Oxford). B, Y. (), ‘Les Alliances matrimoniales des sénateurs et chevaliers gallo-romains’, in Andreau and Bruhns (), –.

References
C, J. C. (), ‘Necrópoles e ritos funerários no ocidente da Lusitânia romana’, in Vaquerizo (), –. C, R. (), ‘The Household Formation Pattern of a Vlach Mountain Community of Greece: Syrrako, –’, Journal of Family History, : –. C, C., C, C., and N, G. (), ‘A presença romana do concelho de Vila Franca de Xira: Investigar, divulgar e animar’, in M. da G. Filipe and J. M. C. Raposo (eds.), Ocupação romana dos estuários do Tejo e do Sado. Actas das primeiras jornadas sobre Romanização dos estuários do Tejo e do Sado, Seixal,  (Lisbon and Seixal), –. C, A. (), ‘The Historical Dimension in Mortuary Expressions of Status and Sentiment’, Current Anthropology, /: –. C, A. T. (), ‘Household Histories’, in R. McC. Netting, R. R. Wilk, and E. J. Arnold (eds.), Households: Comparative and Historical Studies of the Domestic Group (Berkeley, Los Angeles, London), –. C, E. (), Final Judgements: Duty and Emotion in Roman Wills  B.C.–A.D.  (Oxford). C, D. (), ‘Marriage and Acculturation in Roman Algeria’, Classical Philology, : –. —— (), Frontier and Society in Roman North Africa (Oxford). C, G. E. F. (), Cisalpine Gaul: Social and Economic History from  B.C. to the Death of Trajan (Oxford). C R, P. (), Il sepolcro del fornaio Marco Virgilio Eurisace a Porta Maggiore, Monumenti Romani,  (Rome). C, C. J. (), Recht, Rhetorik, Politik. Untersuchungen zu Ciceros rhetorischer Strategie (Darmstadt). —— (), Diritto, Retorica, Politica. La Strategia retorica di Cicerone (Bologna). C, F. (), ‘Il sepolcro degli Scipioni’, Dialoghi di archeologia, : –. —— and T, Y. (), ‘Architecture funéraire et pouvoir: réflexions sur l’hellénisme numide’, Mélanges de l’École française de Rome, Antiquité, : –. C, B. (), ‘Divesting the Female Breast of Clothes in Classical Sculpture’, in Koloski-Ostrow and Lyons (), –. C, N. (), ‘Jewish Names as Cultural Indicators in Antiquity’, Journal for the Study of Judaism : –. —— (), ‘The Names of the Translators in the Letter of Aristeas: A Study in the Dynamics of Cultural Transition’, Journal for the Study of Judaism, : –. C, S. J. D. () (ed.), The Jewish Family in Antiquity (Atlanta).
References

C, J. J. (), ‘Marriage, Divorce and Family in Second Temple Judaism’, in L. G. Perdue, J. Blenkinsopp, J. J. Collins, and C. Meyers (eds.), Families in Ancient Israel (Louisville, Ky.), –. C, D., and D, M. (), ‘Households and “Hidden” Kin in Early-Nineteenth-Century England: Four Case Studies in Suburban Exeter, –’, Continuity and Change, : –. C, P. (), The Roman Law of Marriage (Oxford). C, M. (), ‘Les familles clarissimes d’Afrique proconsulaire (Ier–IIIe siècles)’, in Atti del Colloquio Internazionale AIEGL su Epigrafia e ordine senatorio (Roma, – maggio ), , Edizioni di storia e letteratura (Tituli, ) (Rome), –. —— (), ‘Pour une pluralité des approches prosopographiques’, Mélanges de l’École française de Rome, Moyen Âge, Temps modernes, /: –. —— (a), ‘Usages publics du vocabulaire de la parenté: Patronus et alumnus de la cité dans l’Afrique romaine’, in A. Mastino (ed.), L’Africa romana. Atti del VII convegno di studio, Sassari, – dicembre  (Sassari), –. —— (b) ‘Construire sa parenté à Rome’, Revue historique,  (juillet–sept.), –. —— (a), ‘Divorce and Adoption as Roman Familial Strategies’, in B. Rawson (a), –. —— (b), ‘Family Behavior of the Roman Aristocracy’, in S. Pomeroy (ed.), Women’s History and Ancient History (London), –. —— (c), ‘Constructing Kinship in Rome: Marriage and Divorce. Filiation and Adoption’, in Kertzer and Saller (), –. —— (), ‘Épigraphie et parenté’, in Y. Le Bohec and Y. Roman (eds.), Épigraphie et histoire: acquis et problèmes. Actes du Congrès de la SoPHAU, Lyon-Chambéry, – mai  (Lyon), –. —— (a), ‘La petite enfance à Rome: loi, normes, pratiques individuelles et collectives’, Annales. Histoire, Sciences sociales, : –. —— (b) (ed.), Adoption et fosterage (Paris). —— (), ‘Parenté et alliance dans le monde romain (IIIe siècle avant J.-C.–VIe siècle après J.-C.)’, in Parenté, sexe et genre dans le monde grec, de l’Antiquité à l’Âge moderne (Volos, – juin ) (Bordeaux). C, S. (), The Empire of the Tetrarchs: Imperial Pronouncements and Government A.D. – (Oxford). C, J.-P. (), ‘La Technique du rescrit à la fin du Principat’, Studia et Documenta Historiae et Iuris, : –. C, H. (), Documentary Letters of Recommendation in Latin from the Roman Empire, Beiträge zur klassischen Phililogie,  (Königstein).

References
C, H. (), ‘The Guardianship of Jesus son of Babatha: Roman and Local Law in the Province of Arabia’, Journal of Roman Studies, : –. —— (), ‘The Rabbis and the Documents’, in M. Goodman (ed.), Jews in a Graeco-Roman World (Oxford), –. —— and Q, E. (), ‘XH . ev/Se. ar.  of  or  C.E.: A Wife’s Renunciation of Claims’, Journal of Jewish Studies, : –. —— and Y, A. (), Aramaic, Hebrew and Greek Documentary Texts from N . ahal H . ever and Other Sites with an Appendix Containing Alleged Qumran Texts (The Seiyâl Collection II), Discoveries in the Judaean Desert,  (Oxford) (0001 DJD ). C, J. A. (), Law and Life of Rome (London and Southampton). C, L. A. (a), ‘La Famille lusitano-romaine’, in Gorges and Nogales Basarrate (), –. —— (b), ‘Aetates mortuorum: Études quantitatives sur les âges des défunts en Lusitanie’, Conimbriga, : –. C, S. (), ‘The Empire of Adults: The Representation of Children on Trajan’s Arch at Beneventum’, in J. Elsner (ed.), Art and Text in Roman Culture (Cambridge), –. D, G. (), ‘The Significance of the Handshake Motif in Classical Funerary Art’, American Journal of Archaeology, : –. D H, M. (), ‘Eine reine Männerwelt? Frauen um das römische Heer’, in M. H. Dettenhofer (ed.), Reine Männersache? Frauen in Männerdomänen der antiken Welt (Cologne), –. D, D. (), Alexandrian Citizenship during the Roman Principate (Atlanta). D, M. (), Ellenismo e romanizzazione nella X regio: La scultura delle elites locali dall’età repubblicana ai guilio-claudi (Rome). D, J.-M. (), Le motif du banquet couché dans le Proche-Orient et le monde grec du VIIe au IVe siècle avant J.-C., Bibl. des Éc. franc. d’Athènes et de Rome,  (Rome). D, M. M. A (), ‘Da latinização onomástica à romanização onomástica no processo de aculturação dos Igaeditani’, Symbolae L. Mitxelena (Vitoria), –. D, M. W. (), Magic and Magicians in the Greco-Roman World (London and New York). D V-E, G. (), ‘Légionnaires africains en Pannonie au IIe s. ap. J.C.’, in Hajnóczi (), –. D, S. (), The Roman Mother (Norman, Okla.). —— () ‘The Sentimental Ideal of the Roman Family’, in Rawson (a), –. —— (), The Roman Family (Baltimore and London).
References

—— (), ‘Conflict in the Roman Family’, in Rawson and Weaver (), –. —— (a) (ed.), Childhood, Class and Kin in the Roman World (London and New York). —— (b), ‘Familia Veturia: Towards a Lower-Class Economic Prosopography’, in Dixon (a), –. D, F. L. (), ‘Coming of Age in Rome: The History and Social Significance of Assuming the Toga Virilis’, MA thesis (University of Victoria). Do, W. J. () (ed.), Roman Eloquence. Rhetoric in Society and Literature (London). D, W. A. (), ‘Sheep Ranchers and Sugar Growers: Property Transmission in the Basque Immigrant Family of the American West and Australia’, in R. McC. Netting, R. R. Wilk, E. J. Arnold (eds.), Households: Comparative and Historical Studies of the Domestic Group (Berkeley, Los Angeles, London), –. —— (), ‘Iberian Family History’, Journal of Family History, /: –. D, I. E. (), ‘Soranus and his System of Medicine,’ Bulletin of the History of Medicine, : –. D, J. F. (), ‘The Rise and Fall of the Gallic Julii’, Latomus, : –. —— (), ‘The Wool Textile Industry in Gallia Belgica and the Secundinii of Igel’, Textile History, /: –. —— (), ‘The Gallo-Roman Woollen Industry and the Great Debate: The Igel Column Revisited’, in D. Mattingly and J. Salmon (eds.), Economies beyond Agriculture in the Classical World (London), –.  B, J. (), ‘The Blood: Symbolic Relationships between Descent, Marriage, Incest Prohibitions and Spiritual Kinship in Greece’, Man, : –. D, A. M. (), Freedmen in the Early Roman Empire (Oxford). D, K. M. D., and D, M. W. (), ‘Invida rumpantur pectora: The Iconography of Phthonos/Invidia in Greco-Roman Art’, Jahrbuch für Antike und Christentum, : –. D-J, R. P. (), ‘Age-Rounding, Illiteracy and Social Differentiation in the Roman Empire’, Chiron, : –. —— (), Structure and Scale in the Roman Economy (Cambridge). —— (), ‘The Impact of the Antonine Plague’, Journal of Roman Archaeology, : –. D, R. (), ‘Recherches sur la répartition géographique et chronologique des termes sevir Augustalis, Augustalis et sevir dans l’Empire romain’, Epigraphische Studien, . –.

References
D, N., and P, H.-G. () (ed.), L’Onomastique latine. Actes du colloque international (Paris). E, M. (), Capuanische Grabsteine: Untersuchungen zu den Grabsteinen römischer Freigelassener aus Capua, BAR  (Oxford). E, E. J., and E, L. (), Asclepius: Collection and Interpretation of the Testimonies (Baltimore; st pub. ). E, J. C. (), ‘Epigraphy and History of Roman Hispania: The New Edition of CIL II’, Journal of Roman Archaeology, : –. —— (), ‘Conmemoración funeraria y relaciones familiares en Augusta Emerita’, in Gorges and Nogales Basarrate (), –. ——(), ‘Enquête de la famille romaine à la Civitas lgaeditanorum (Idanha-a-Velha) en Lusitánie au Haut-Empire’, in C. Auliard and L. Bodiou (eds.), Au jardin des Hespérides. Histoire, société et épigraphie des mondes anciens: Mélanges A. Tranoy (Rennes), ‒. —— (forthcoming), ‘Inmigración y sociedad local en Augusta Emerita, a.C–d.C’, in J.-G. Gorges, T. Nogales Basarrate, and E. Cerrillo Martín de Cáceres (eds.), Las comunicaciones en la Lusitania romana. Actas de la V Mesa Redonda Internacional sobre Lusitania, Cáceres, – noviembre de  (Madrid). —— N B, T., and T, W. (), Imagen y memoria. Monumentos funerarios con retratos en la colonia Augusta Emerita, Monografías Emeritenses, ; Bibliotheca Archaeologica Hispana,  (Madrid). E, C. (), The Politics of Immorality in Ancient Rome (Cambridge). E, J.  (). ‘Inscrições romanas do conventus Pacensis. Aditamento’, Trabalhos de Arqueologia do Sul, : –. —— (), ‘Roma e as primeiras culturas epigráficas de Lusitânia ocidental’, in F. Beltrán Lloris (ed.), Roma y el nacimiento de la cultura epigráfica en Occidente. Actas del coloquio ‘Roma y las primeras culturas epigráficas del Occidente mediterráneo, Siglos II a.E.–I d.E.’ (Zaragoza, – de noviembre de ) (Zaragoza), –. —— (), ‘Morrer aos  anos na Lusitânia romana’, in Gorges and Nogales Basarrate (), –. —— and  S, J. C (), ‘Catálogo da epigrafia romana de Abrantes’, Abrantes, : –. É, R. (), ‘L’Horloge de la Civitas Igaeditanorum et la création de la province de Lusitanie’, Revue des études anciennes, : –. —— F, G., L R, P., and T, A. (), ‘Les Dimensions sociales de la romanisation dans la péninsule ibérique des origines à la fin de l’Empire’, in D. M. Pippidi (ed.), Assimilation et résistance à la culture gréco-romaine dans le monde ancien (Bucharest and Paris), –.
References

E G, J. (), Law and Family in Late Antiquity: The Emperor Constantine’s Marriage Legislation (Oxford). —— (), Women and the Law in the Roman Empire: A Sourcebook on Marriage, Divorce, and Widowhood (London and New York). F, G. (), Libertus: Recherches sur les rapports patron–affranchi à la fin de la république romaine, Coll. EFR  (Rome). F, R. E. (), ‘Ciceronian Conciliare and Aristotelian Ethos’, Phoenix, : –. —— (), ‘Aemilia Pudentilla: Or the Wealthy Widow’s Choice’, in R. Hawley and B. Levick (eds.), Women in Antiquity: New Assessments (London), –. —— (), ‘The Contexts and Occasions of Roman Public Rhetoric’, in Dominik (), –. —— (), Ovid Fasti Book IV (Cambridge). F, W. (), Die Grabstelen des . und . Jahrhunderts im Rheingebiet (Cologne). F, L. H., and R, M. () (eds.), Jewish Life and Thought among Greeks and Romans (Minneapolis and Edinburgh). F, E., and A, S. () (eds.), Romanization and the City: Creations, Transformations, and Failures, Journal of Roman Archaeology suppl.  (Ann Arbor). F-G, F. (), ‘Grabinschriften im archäologischen Kontext. Komplementarität von Schrift und Bild?’, in Heinzelmann et al. (), –. F, N. (), ‘Grandes stèles à décor architectural de la région de Bou Arada (Aradi) en Tunisie’, Mitteilungen des deutschen archaeologischen Instituts, Röm. Abt. : –. —— (), ‘Le Mausolée de Q. Apuleus Maxsimus à El Amrouni’, Papers of the British School at Rome, : –. —— (), ‘Histoire antique et architecture dans la haute steppe en Afrique proconsulaire’, in F. Bejaoui (ed.), Histoire des hautes steppes. Antiquité–Moyen Age. Actes du colloque de Sbeitla, Sessions  et  (Tunis), –. F, L.  S (– []), ‘A presença da mulher na epigrafia do Conventus Scallabitanus’, Portugalia,  –: –. F, G. B. (), ‘Roman Lay Attitudes towards Medical Experimentation’, Bulletin of the History of Medicine, : –. F, P.-A., and G, A. (–), ‘La nécropole orientale de Sétif: Rapport préliminaire sur les fouilles effectuées de  à ’, Bulletin d’archéologie algérienne, : –. —— and G, R. (), ‘Rites funéraires de la nécropole orientale de Sétif’, Antiquités Africaines, : –.

References
Les Flavii de Cillium: Étude architecturale, épigraphique, historique et littéraire du Mausoleé de Kasserine (CIL VIII, –) (), Coll. EFR  (Rome). F, V. (), Breasts, Bottles and Babies: A History of Infant Feeding (Edinburgh). F, S. (), ‘A Note on Ossuary Burial and the Resurrection of the Dead in First-Century Jerusalem’, Journal of Jewish Studies, : –. F, K. (), Die Bildnistypen der Faustina minor und die Fecunditas Augustae, Abh. Akad. Göttingen (Göttingen). —— (), Prinzenbildnisse antoninischer Zeit (Mainz). F, J. (), Les Syriens à Intercisa (Brussels). —— (), The Great Age of Pannonia (AD –), trans. I. Varga (Budapest). F, R. (), Medicine and the Making of Roman Women (Oxford). F, M. B. (), ‘Family in familia: Kinship and Community in Slavery’, American Journal of Ancient History, : –. F, H. (), Ancestor Masks and Aristocratic Power in Roman Culture (Oxford). F, G. (), ‘Hellenistic and Roman Trends in the Herodian Architecture of Masada’, in K. Fittschen and G. Foerster (eds.), Judaea in the Greco-Roman World in the Time of Herod in the Light of Archaeological Evidence (Göttingen), –. F, D. (), L’archivio di Kronion (Milan). F, A. (), ‘The Survival of the Family Name and Pharaonic Order’, in A. Burgière, C. Klapisch-Zuber, M. Segalen, and F. Zonabend (eds.), A History of the Family, i: Distant Worlds, Ancient Worlds (Cambridge and Oxford), –. F, R. (), Kinship and Marriage: An Anthropological Perspective, rev. edn. (Cambridge). F, J.-C. (), ‘Les confessions d’Augustin, autobiographie au présent’, in M.-F. Baslez, P. Hoffmann, and L. Pernot (eds.), L’Invention de l’autobiographie: d’Hésiode à Augustin (Paris), –. F, R. (), Ancient Natural History (London and New York). F, H. G. (), Römische Grabreliefs in Mittel- und Süditalien (Rome). F, B. (), The Rise of the Roman Jurists: Studies in Cicero’s pro Caecina (Princeton). —— (), ‘Demography’, in A. K. Bowman, P. Garnsey, D. Rathbone (eds.), The Cambridge Ancient History, nd edn., xi (Cambridge), –. G, H. (), ‘Römische Kinder in Toga Praetexta’, Jahrbuch des Deutschen Archäologischen Instituts, : –.
References

G, D. (), ‘Traiano e la Pannonia’, in G. A. Popescu (ed.), Traiano: Ai confini dell’Impero (Milan), –. G, J. (), Apollon romain (Paris). G, T., K, L., and MN, B. E. (), ‘A FirstCentury Archive from Oxyrhynchus or Oxyrhynchite Loan Contracts and Egyptian Marriage’, in J. H. Johnson (ed.), Life in a Multicultural Society: Egypt from Cambysses to Constantine and Beyond, Studies in Ancient Oriental Civilization,  (Chicago), –. G, E. (), Étude sur la poésie funéraire romaine d’après les inscriptions (Paris). G, P., and W, P. (), ‘Familial Structures in Roman Italy: A Regional Approach’, in Rawson and Weaver (), –. G, B., and G, H. (), Die römischen Steininschriften aus Köln. Cologne: Wissenschaftliche Kataloge des Römisch-Germanischen Museums Köln,  (Cologne). G, G. (), Formen römischer Altäre auf der hispanischen Halbinsel, Madrider Forschungen,  (Mainz). G, J. (), Die norisch-pannonische Frauentracht im . und . Jahrhundert (Munich). G, J. M. (), ‘Epigrafia e romanização de Castelo Branco’, Conimbriga, : –. G  B, A. (), Esculturas romanas de España y Portugal (Madrid). —— (), ‘El elemento forastero en Hispania romana’, Boletín de la Real Academia de la Historia, : –. G, J. F. (), Women in Roman Law and Society (London and New York). —— (), Being a Roman Citizen (London and New York). —— (), Family and Familia in Roman Law and Life (Oxford). —— (), ‘Making Citizens. The Operation of the Lex Irnitana’, in Administration, Prosopography and Appointment Policies in the Roman Empire: Proceedings of the first workshop of the international network, Impact of Empire (Roman Empire,  B.C.–A.D. ), Leiden,  June– July , edited Lukas de Blois (Amsterdam), –. G, P. D. A. (), ‘Child Rearing in Ancient Italy’, in Kertzer and Saller (), –. —— (), Ideas of Slavery from Aristotle to Augustine (Cambridge). —— (), Food and Society in Classical Antiquity (Cambridge). G, J. (), Inscriptions antiques du Maroc, : Inscriptions latines (Paris). G, M. (), ‘Servus and domus: The Slave in the Roman House’, in Laurence and Wallace-Hadrill (), –.

References
G, M. (), ‘Family and Familia on Roman Biographical Sarcophagi’, Mitteilungen des deutschen archaeologischen Instituts, Röm. Abt. : –. —— (), ‘A Roman Funerary Monument with a Mother and Daughter’, in Dixon (a), –. —— (), ‘Slave Disguise at Ancient Rome’, Slavery and Abolition, /: –. G, E. H. (), ‘The Archives of the Temple of Soknobraisis at Bacchias’, Yale Classical Studies, : –. G, J. F. (), ‘The Plague under Marcus Aurelius’, American Journal of Philology, : –. G, M. W. (), Making Men: Sophists and Self-Presentation in Ancient Rome (Princeton). G-C P, M. P. (–), ‘Inscripciones funerarias colectivas de época romana en el distrito de Castelo Branco (Portugal)’, Lucentum, –: –. G R, M. C. (), Las unidades indígenas del área indoeuropea de Hispania, Veleia, Anejo  (Vitoria/Gasteiz). —— (), ‘Reflexiones sobre las unidades organizativas indígenas del área indoeuropea’, in M. C. González Rodríguez and J. Santos (eds.), Las estructuras sociales indígenas del norte de la península ibérica, Revisiones de Historia Antigua,  (Vitoria and Gasteiz), –. G, D. (), ‘Dating Documents in Provincia Iudaea: A Note on Papyri Murabba‘ât  and ’, Israel Exploration Journal, : –. G, M. (), ‘Babatha’s Story’, Journal of Roman Studies, : –. —— (), The Roman World  BC–AD  (London and New York). G, J. (), The Development of the Family and Marriage in Europe (Cambridge). G, R. (), ‘The Healing Event in Graeco-Roman FolkMedicine’, Clio Medica : –. G, J.-G., and N B, T. () (eds.), Sociedad y cultura en Lusitania romana (Mérida). G, G. (), ‘Childhood in Eastern Patristic Thought: Some Problems of Theology and Theological Anthropology’, Studies in Church History, : –. G, Z. (), ‘The Caiaphas Tomb in North Talpiyot’, in H. Geva (ed.), Ancient Jerusalem Revealed (Jerusalem), –. G, M. D. (), Diseases in the Ancient Greek World (Baltimore and London). G, E. S. (), Heritage and Hellenism: The Reinvention of Jewish Tradition (Berkeley, Los Angeles, and London).
References

G M (), Atlas antroponímico de la Lusitania romana (Bordeaux and Mérida). G, R. (), La nécropole orientale de Sitifis (Sétif, Algérie): Fouilles de – (Paris). G, P. (), Structures sociales ‘orientales’ et ‘occidentales’ dans l’Espagne musulmane (Paris). G, S. (), ‘The Family in First-Century Galilee’, in H. Moxnes (ed.), Constructing Early Christian Families: Family as Social Reality and Metaphor (London and New York), –. H, T. (), Schools of Gaul: A Study of Pagan and Christian Education in the Last Century of the Western Empire (Oxford). H, R. (),’The Goliath Family in Jericho: Funerary Inscriptions from a First Century A.D. Jewish Monumental Tomb’, Bulletin of the American Schools of Oriental Research, : –. —— and K, A. (), ‘Jewish Funerary Customs during the Second Temple Period, in the Light of the Excavations at the Jericho Necropolis’, Palestine Exploration Quarterly, : –. H, G. () (ed.), La Pannonia e l’Impero romano (Milan). H, E. W. (), Migration and Economy in Roman Imperial Spain, Aurea Saecula,  (Barcelona). H, J. S. (), ‘Scribonius Largus on the Medical Profession’, Bulletin of the History of Medicine, : –. H, E. A. (), ‘Household Structure in Fourteenth Century Macedonia’, Journal of Family History, : –. —— (), ‘On the *** of Studying Household Form and Function’, in R. McC. Netting, R. R. Wilk, and E. J. Arnold (eds.), Households: Comparative and Historical Studies of the Domestic Group (Berkeley, Los Angeles, London), –. H, A. E. (), ‘Topographical Arrangement of Tax Documents in the Philadelphia Tax Archive’, in A. Bülow-Jacobsen (ed.), Proceedings of the XXth International Congress of Papyrologists, Copenhagen, – August,  (Copenhagen), –. H, K. C. (–), ‘The Herodians and Mediterranean Kinship’, Biblical Theology Bulletin, /: –; /: –; /: –. H, O. (), ‘Die Stellung der Frau bei den einheimischen Stämmen Nordpannoniens. Eine sozial- und kunstgeschichtliche Studie’, Budapest Régiségie, : –. H, B. (), ‘Some Wandering Potters,’ in J. Dore and K. Greene (eds.), Roman Pottery Studies in Britain and beyond, BAR  (Oxford), –. H, R. (), Blood & Fire: William and Catherine Booth and their Salvation Army (London).

References
H, M. (a), ‘Grabarchitektur, Bestattungsbrauch und Sozialstruktur: Zur Rolle der familia’, in Heinzelmann et al. (), –. —— O, J., F, P., and W, M. () (eds.), Römischer Bestattungsbrauch und Beigabensitten in Rom, Norditalien und der Nordwestprovinzen von der später Republik bis in die Kaiserzeit / Culti dei morti e costumi funerari romani: Roma, Italia settentrionale e province nord-occidentali della tarda Reppublica all’età imperiale, Palilia,  (Wiesbaden). H, D., and K-Z, C. (), Tuscans and their Families (New Haven and London). H, H.  (), Römische Grabbauten (Darmstadt). H, L. R. (), ‘Your Mother-in-Law is Poison’, Man, : –. H, R. (), Rural Settlement in Roman Britain (London). H, Y. (), The Palestinian Dwelling in the RomanByzantine Period (Jerusalem). H, D. W. (), ‘The Role of Women in the Economic Life of Roman Egypt: A Case Study from First Century Tebtunis’, Échos du monde classique/Classical views, : –. —— (), ‘House and Household in Roman Egypt’, Yale Classical Studies, : –. —— (), ‘Naming Practices in Roman Egypt’, Bulletin of the American Society of Papyrologists, : –. H, H. (), Herod Antipas (Cambridge). H, H. (), Römische Militärgrabsteine der Donauländer (Vienna). H, N. (), ‘Le Vétéran Lucius Bellienus Gemellus’, Études de Papyrologie, : –. H, T. (), Emperors and Lawyers, nd edn. (Oxford). H, V. M. (a), ‘A Roof over the Dead: Communal Tombs and Family Structure’, in Laurence and Wallace-Hadrill (), –. —— (b), ‘Words and Pictures: The Interpretation of RomanoBritish Tombstones’, Britannia, : –. —— (), Constructing Identity: The Roman Funerary Monuments of Aquileia, Mainz and Nîmes, BAR Int. ser.  (Oxford). H, K. (), ‘The Age of Roman Girls at Marriage’, Population Studies, : –. —— (), ‘On the Probable Age Structure of the Roman Population’, Population Studies, : –. —— (), ‘Economic Growth and Towns in Classical Antiquity’, in P. Abrams and E. A. Wrigley (eds.), Towns in Societies: Essays in Economic History and Historical Sociology (Cambridge), –.
References

—— (), ‘Brother–Sister Marriage in Roman Egypt’, Comparative Studies in Society and History, : –. —— (), Death and Renewal (Cambridge). —— (), ‘Graveyards for Historians’, in F. Hinard (ed.), La Mort, les morts et l’au-delà dans le monde romain (Caen), –. —— (), ‘Seven Missing Papers’, in Andreau and Bruhns (), –. H, P., and P, N. (), The Corrupting Sea: A Study in Mediterranean History (Oxford). H, C. (), Das Kind und seine Krankheiten in der griechischen Medizin (Frankfurt). H, J. (), Roman Children’s Sarcophagi: Their Decoration and its Social Significance (Oxford and New York). —— (), ‘Iconography: Another Perspective’, in Rawson and Weaver (), –. —— (), ‘Élite Culture and the Identity of Empire’, in J. Huskinson (ed.), Experiencing Rome: Culture, Identity and Power in the Roman Empire (London), –. H, G. O. (), Cicero’s Correspondence. A Literary Study (Oxford). I, T. (), ‘Notes on the Distribution of Jewish Women’s Names in Palestine in the Second Temple and Mishnaic Periods’, Journal of Jewish Studies, : –. —— (), Jewish Women in Greco-Roman Palestine (Tübingen). —— (), ‘Notes and Observations on a Newly Published Divorce Bill from the Judaean Desert’, Harvard Theological Review, : –. I, A. Z. (), Galeni De optimo medico cognoscendo (Berlin). J, R. (), Doctors and Diseases in the Roman Empire (London). J, S. (), The Atlantic Celts. Ancient People or Modern Invention? (London). J, C. (), Sex or Symbol? Erotic Images of Greece and Rome (New York). J, S. (), Work, Identity, and Legal Status at Rome (Norman and London). J, J. (), Hippocrates (Baltimore and London). K, L. (), ‘The Soter Tomb in Thebes’, in S. P. Vleeming (ed.), Hundred-Gated Thebes: Acts of a Colloquium on Thebes and the Theban Area in the Graeco-Roman Period, Pap. Lugd. Bat.  (Leiden, Cologne, New York), –. K, N. (a), Image and Status: Roman Working Women in Ostia (Berlin).

References
K, N. (b), ‘Biographical Narration and Roman Funerary Art’, American Journal of Archaeology, /: –. —— (), ‘Between Public and Private: Women as Historical Subjects in Roman Art’, in S. B. Pomeroy (ed.), Women’s History and Ancient History (Chapel Hill), –. K, R. (), ‘Polygamy in P. Yadin?’, Zeitschrift für Papyrologie und Epigraphik, : –. K, S. J. (), ‘Romanization and the Hispaniae’, in Keay and Terrenato (), –. —— and T, N. () (eds.), Italy and the West: Comparative Issues in Romanization (Oxford). K, J. G. (), ‘The Will of Gaius Longinus Castor’, Bulletin of the American Society of Papyrologists, : –. K, D., and B, M. (), Family Life in Early Modern Times, – (New Haven). —— and S, R. P. () (eds.), The Family in Italy from Antiquity to the Present (New Haven). —— H, D. P. and K, N. (), ‘Kinship beyond the Household in a Nineteenth-Century Italian Town’, Continuity and Change, : –. K, M., and M, L. () (eds.), Dougga, fragments d’histoire. Choix d’inscriptions latines éditées, traduites et commentées (Ier–IVe siècles), Ausonius-Publications, Mémoires  (Bordeaux). —— —— (), Mourir à Dougga: Recueil des inscriptions funéraires, Ausonius-Publications, Mémoires  (Bordeaux). K, D. E. E. (), Roman Group Portraiture (New York). —— (), ‘The Great Frieze of the Ara Pacis Augustae’, Mélanges de l’École française de Rome, /: –. —— (), Roman Imperial Funerary Altars with Portraits (Rome). —— (), Roman Sculpture (New Haven). K, V. (), Porträtsreliefs stadtrömischer Grabbauten. Ein Beitrag zur Geschichte und zum Verständnis des spätrepublikanischfrühkaiserzeitlichen Privatporträts (Mainz). K, N. (), The Herodian Dynasty: Origins, Role in Society and Eclipse (Sheffield). —— (forthcoming), ‘The Royal Court of the Herods’, in N. Kokkinos (ed.), The World of the Herods. K-O, A. O., and L, C. L. () (eds.), Naked Truths: Women, Sexuality, and Gender in Classical Art and Archaeology (London and New York). K, M. (). ‘In commemorationem mortuorum: Text and Image along the “streets of tombs” ’, in J. Elsner (ed.), Art and Text in Roman Culture (Cambridge), –.
References

K, R. (), ‘Incantations and Prayers for Salvation on Inscribed Greek Amulets’ in C. A. Faraone and D. Obbink (eds.), Magika Hiera: Ancient Greek Magic and Religion (New York and Oxford). K, J.-U. (), ‘Familien- und Haushaltsstrukturen in spätantiken Gallien’, Klio, : –. K, H. (), Erbreachtliche Untersuchungen auf Grund der graecoaegyptischen Papyrusurkunden (Leipzig). L S, L. (), ‘L’amour en Afrique romaine: à propos d’une inscription métrique des environs de Dougga-Tunisie (CIL, VIII, ; CLE, )’, Antiquités africaines, : –. —— (), ‘La femme en Afrique à l’époque romaine (étude menée à partir de la documentation épigraphique)’, in S. Sotgiu (ed.), Attività di ricerca e di tutela del patrimonio archeologico e storicoartistico della Tunisia. Seminario di studi. Cagliari – Aprile  (Cagliari), –. —— (), ‘Saintes matrones ou dangereuses dévergondées: deux images des femmes du Maghreb à l’époque romaine’, Clio, : –. —— (), ‘L’inscription dédiée à Postumia Matronilla du Mausolée de Hr Ez-Zaatli (région de Fériana-Thelepte) CIL, VIII,  0001 ILS,  et ILT, ’, in F. Béjaoui (ed.), Histoire des hautes steppes. Antiquité–Moyen Age. Actes du colloque de Sbeitla. Sessions  et  (Tunis), –. L F, L. (), ‘The Costume of the Roman Bride’, in Bonfante and Sebesta (), –. L, G. (), Rhetoric Rampant: The Family under Siege in the Early Western Tradition (London, Ont.). L, S. (–), ‘Tipasitana I: Fouilles dans la nécropole occidentale de Tipasa’, Bulletin d’archéologie algérienne, : –. L, K. (), ‘The Soter Family: Geneology and Onomastics’, in S. P. Vleeming (ed.), Hundred-Gated Thebes: Acts of a Colloquium on Thebes and the Theban Area in the Graeco-Roman Period, Pap. Lugd. Bat.  (Leiden, Cologne, New York), –. L F, R. (), ‘Inheritance in the Greek World’, in Crux: Essays Presented to G. E. M. de Ste Croix (Exeter), –. L, A. (), ‘De Cyrène à Timgad: P. Flavius Pudens Pomponianus et sa famille’, Annali della Facoltà di Lettere e Filosofia dell’Università di Macerata, : –. L, P. (), The World we have Lost (London). —— (), Family Life and Illicit Love in Earlier Generations: Essays in Historical Sociology (Cambridge). —— (), The World we have Lost Further Explored (London) (revised and expanded rd edn. of Laslett ).

References
L, P. and W, R. () (eds.), Household and Family in Past Time (Cambridge). L, J.-M. (), Ubique populus: Peuplement et mouvements de population dans l’Afrique romaine de la chute de Carthage à la fin de la dynastie des Sévères ( a. C.– p. C.) (Paris). L, J.-M. (), ‘Onomastique et acculturation’, in S. Gely (ed.), Sens et pouvoirs de la nomination dans les cultures hellénique et romaine: Actes du colloque de Montpellier, ,  mai  (Montpellier), . –. L, R. (), ‘Space and Text’, in Laurence and WallaceHadrill (), –. —— and B, J. () (eds.), Cultural Identity in the Roman Empire (London and New York). —— and W-H, A. () (eds.), Domestic Space in the Roman World: Pompeii and beyond, Journal of Roman Archaeology monograph,  (Portsmouth, RI). L, A. D. (), ‘Close-Kin Marriage in Late Antique Mesopotamia’, Greek, Roman and Byzantine Studies, : –. L, S. (), ‘Hommages publics et histoire sociale: Les Caecilii Caeciliani et la vie municipale de Volubilis (Maurétanie Tingitane)’, Mélanges de la Casa de Velázquez, : –. L G, M. (a), Saturne africain: Histoire, Bibl. des Éc. franc. d’Athènes et de Rome,  (Paris). —— (b), Saturne africain: Monuments,  vols. (Paris). L, A. (), ‘Burial Box of James the Brother of Jesus’, Biblical Archaeology Review, /: – and . L P, F. (), L’organisation de la famille (Paris). L R, P. (), L’armée romaine et l’organisation des provinces ibériques d’Auguste à l’invasion de  (Paris). —— (), ‘Droit latin et municipalisation en Lusitanie sous l’Empire’, in E. Ortiz de Urbina and J. Santos (eds.), Teoría y práctica del ordenamiento municipal en Hispania (Vitoria and Gasteiz), –. L, G. (), ‘Family and Kin—A Few Thoughts’, Journal of Family History, : –. Li-S, C. (), The Elementary Structures of Kinship (London). L, N. () (ed.), The Documents from the Bar-Kokhba Period in the Cave of Letters: Greek Papyri ( Jerusalem). —— (),‘Judah’s Bigamy’, Zeitschrift für Papyrologie und Epigraphik, : . —— K, R., and G, J. C. (), ‘Papyrus Yadin ’, Israel Exploration Journal, : –. L, B. (), ‘Appealing to Children’, Journal of Early Christian Studies, : –.
References

L, R. J., and L, M. L. (), ‘Galen and the Antonine Plague’, American Journal of Philology, : –. L, S. (), ‘The Extent of Jewish Polygamy in Talmudic Times’, Journal of Jewish Studies, : –. MD, W. R. (), ‘On the Expectation of Life in the Provinces of Hispania and Lusitania and Africa’, Biometrika, : –. MF, A., H, S., and J, C. (), Reconstructing Historical Communities (Oxford). MG, T. (), Prostitution, Sexuality, and the Law in Ancient Rome (Oxford). MM, R. (), Romanization in the Time of Augustus (New Haven). MW, J. (), ‘Children Among the Dead: The Influence of Urban Life on the Commemoration of Children on Tombstone Inscriptions’, in Dixon (a), –. M  M, J. (), Carta arqueológica de España: Salamanca (Salamanca). M, J. C. (), ‘Epigraphic Consciousness’, Journal of Roman Studies, : –. M, V. G. (), ‘As cidades marítimas da Lusitânia’, in J.-G. Gorges (ed.), Les Villes de Lusitanie romaine (Paris), –. M, L. T (), Funerary Monuments in Dacia Superior and Dacia Porolissensis, BAR  (Oxford). M, H.-I. (), Histoire de l’éducation dans l’antiquité (Paris). —— (), ‘Une inscription chrétienne de Tipasa et le refrigerium’, Antiquités africaines : –. M, D. B. (), ‘The Construction of the Ancient Family: Methodological Considerations’, Journal of Roman Studies, : –. M-K, S. (), ‘Mors immatura in the Roman World: A Mirror of Society and Tradition’, in J. Pierce, M. Millett, and M. Struck (eds.), Burial, Society and Context in the Roman World (Oxford), –. M V, R. (), ‘Las necrópolis del castro de Yecla de Yeltes: Datos arqueológicos y epigráficos para su estudio’, Zephyrus, –: –. M, N. (), Histoire d’un nom. Les Aufidii dans la vie politique, économique et sociale du monde romain. IIe siècle avant Jésus-Christ–IIIe siècle après Jésus-Christ (Rennes). M, G. (), ‘La puissance paternelle et le mariage des fils et filles de famille en droit romain’, in Studi in onore di Edoardo Volterra, vol. v (Milan), –.

References
M, S. P. (), ‘Physicians and the Roman Imperial Aristocracy: The Patronage of Therapeutics’, Bulletin of the History of Medicine, : –. M, M. (), The Gift: Forms and Functions of Exchange in Archaic Societies (London). M, J. P. (), Trials of Character. The Eloquence of Ciceronian Ethos (Chapel Hill). M’C, A. (), Aspects de l’évolution démographique et sociale à Mactaris aux IIe et IIIe siècles après J.-C. (Tunis). —— (), ‘Sculptures antiques de Hajeb el-Aioun: de la tradition punico-numide à la romanisation’, in A. Mrabet (ed.), Mobilité des hommes et des idées en Méditerranée (Sousse), –. M, L., and P, G. (), Stele romane in Piemonte, Monumenti Antichi serie misc. ; serie generale,  (Rome). M, J., M, M., R, N., and S, J. () (eds.), Integration in the Early Roman West: The Role of Culture and Ideology, Dossiers d’Archéologie du Musée National d’Histoire et d’Art,  (Luxembourg). M, R. (), ‘Land, Kinship and Consanguineous Marriage in Italy from the Seventeenth to the Nineteenth Centuries’, Journal of Family History, : –. M, E. A. (), ‘Explaining the Epigraphic Habit in the Roman Empire: The Evidence of Epitaphs’, Journal of Roman Studies, : –. M, F. (), ‘Eléments d’urbanisme à Bibracte: Les Maisons du Parc aux Chevaux’, in V. Guichard, S. Sievers, and O. H. Urban (eds.), Les Processus d’urbanisation à l’Age du Fer (Glux-en-Glenne), –. M Z, L. (), I Testamenti romani nei papiri e nelle tavolette d’Egitto: Silloge di documenti dal I al IV secolo DC (Turin). M, F. () (ed.), The Roman Empire and its Neighbours, nd edn. (New York). —— (), The Roman Near East  BC–AD  (Cambridge, Mass.). M, D. (), Material Culture and Mass Consumption (Oxford). M, M. (), The Romanization of Britain: An Archaeological Essay (Cambridge). M, A. (), ‘Die Entwicklung der Sklavenwirtschaft in Pannonien’, Acta Antiqua Academiae Scientiarum Hungaricae, : –. —— (), Pannonia and Upper Moesia: A History of the Middle Danube Provinces of the Roman Empire, trans. ed. Sheppard Frere (London and Boston). M, J. M. () ‘A propos de la tutelle dative des femmes dans l’Égypte romaine’, Akten des XIII. Internationalen
References

Papyrologenkongresses (Munich), – repr. in J. M. Modrzejewski (), Droit impérial et traditions locales dans l’Égypte romaine (Aldershot). M, O. (), ‘Richerche di sociologia nei documenti dell’Egitto greco-romano, . I: Testamenti’, Aegyptus, : –. M, D. (), Sex and Society in Greco-Roman Egypt (London and New York). M, P. () (ed.), Galien de Pergame: Souvenirs d’un médecin (Paris). M, P. (), Incestus et prohibitae nuptiae: L’inceste à Rome (Paris). M, T. (), Literate Education in the Hellenistic and Roman Worlds (Cambridge). M, I. (), Death Ritual and Social Structure in Classical Antiquity (Cambridge). M, J.-L. (), ‘The So-Called Letter of Domitian at the End of the Lex Irnitana’, Journal of Roman Studies, : –. M, H. (), Constructing Early Christian Families (London). N, D. (), ‘Imperial Expansion under the Roman Republic’, in M. Rowlands, M. Larsen, K. Kristiansen (eds.), Centre and Periphery in the Ancient World (Cambridge), –. N, G. (), The Family in Late Antiquity: The Rise of Christianity and the Endurance of Tradition (London and New York). N, J. M.  (), ‘Caracteres externos de las antiguas inscripciones salmantinas: Los epitafios de la zona occidental: su trascendencia epigráfica e histórica’, Boletín de la Real Academia de la Historia, : –. N, J. (), ‘The Ossuary Inscriptions from Giv‘at ha-Mivtar’, Israel Exploration Journal, : –. N, C. A. (), Status Declarations in Roman Egypt, American Studies in Papyrology,  (Amsterdam). N, R. McC., W, R. R., A, E. J. (), ‘Introduction’, in R. McC. Netting, R. R. Wilk, and E. J. Arnold (eds.), Households: Comparative and Historical Studies of the Domestic Group (Berkeley, Los Angeles, London), pp. xiii–xxxviii. N, E. (), ‘The Palaces Built by Herod: A Research Update’, in K. Fittschen and G. Foerster (eds.), Judaea in the Greco-Roman World in the Time of Herod in the Light of Archaeological Evidence (Göttingen), –. N, H. S. ( []), ‘The Physical Context of Roman Epitaphs and the Structure of “the Roman Family” ’, Analecta Romana Instituti Danici, : –. N, I. (; nd edn., ), Hellenistic Palaces: Tradition and Renewal (Aarhus).

References
N B, T., and M P, J. (), ‘Espacios y tipos funerarios en Augusta Emerita’, in Vaquerizo (), –. N, F. (), Apulejus von Madaura und das römische Privatrecht (Leipzig). N, D. (), ‘Wicked Stepmothers in Roman Society and Imagination’, Journal of Family History, : –. N, V. (), ‘Galen in the Eyes of his Contemporaries’, Bulletin of the History of Medicine, : –. —— (), ‘Murders and Miracles: Lay Attitudes towards Medicine in Classical Antiquity’, in R. Porter (ed.), Patients and Practitioners: Lay Perceptions of Medicine in Pre-Industrial Society (Cambridge), –. —— (), ‘The Perils of Patriotism: Pliny and Roman Medicine’, in R. French and F. Greenaway (eds.), Science in the Early Roman Empire: Pliny the Elder, his Sources and Influence (London and Sydney). —— (), ‘The Patient’s Choice: A New Treatise by Galen’, Classical Quarterly, : –. —— (), ‘The Medical Meeting Place’, Clio Medica, : –. O, N. (), Medieval Children (New Haven and London). O’R, D. (), ‘Close-Kin Marriage in Late Antiquity: The Evidence of Chrysostom’, Greek, Roman and Byzantine Studies, , –. O, C., and B, D. L. () (eds.), Families in the New Testament World: Households and House Churches (Louisville, Ky.). O, W. (), Priester und Tempel im hellenistischen Ägypten (Leipzig and Berlin). P, R. E. A. (), ‘Bullae insignia ingenuitatis’, American Journal of Ancient History, , –. P L, M. (), La onomástica personal pre-latina de la antigua Lusitania: Estudio lingüístico, Theses et studia philologica Salmanticensia,  (Salamanca). P, H. () ‘Loyal Slaves and Loyal Wives: The Crisis of the Outsider-within and Roman exemplum Literature’, in S. Joshel and S. Murnaghan (eds.), Women and Slaves in Greco-Roman Culture: Differential Equations (Routledge). P, T. G. (), Demography and Roman Society (Baltimore). —— (), Old Age in the Roman World (Baltimore). P, E. (), Pauvreté économique et pauvreté sociale à Byzance e–e siècles (Paris and The Hague). P, J. R. (), ‘Patronage, collegia and Burial in Imperial Rome’, in S. Bassett (ed.), Death in Towns: Urban Responses to the Dying and the Dead, – (Leicester), –.
References

—— (), ‘The collegia and the Transformation of Towns of Italy in the Second Century A.D.’, in L’Italie d’Auguste à Dioclétien, Coll. EFR  (Rome), –. P, O. () Slavery and Social Death: A Comparative Study (Cambridge, Mass.). P, G. (), ‘Urbs capta: Sketch of an Ancient Literary Motif’, Phoenix, : –. P, J. (), The Roman Villa: An Historical Introduction (London). P, M. A. H. (), ‘O dolium cinerário, com skyphos vidrado a verde da necrópole de Paredes (Alenquer)’, Conimbriga, : –. P, H. (), Römische Porträtstelen in Oberitalien (Mainz). P, S. E. (), The Marriage of Roman Soldiers ( B.C.–A.D. ): Law and Family in the Imperial Army (Leiden). P, G. C. (), La civilisation de l’Afrique romaine, nd edn. (Paris). —— L B, H., M, J. (), ‘Le cippe de Beccut’, Antiquités africaines, : –. P, G. B. (), Lettere Latine d’un soldato di Traiano, P. Mich. – (Bologna). —— (), De ludis saecularibus populi Romani Quiritium libri sex, nd edn. (Amsterdam). P, F. (), ‘Style and Message on the Column of Marcus Aurelius’, Papers of the British School at Rome, : –. P, A. (), Kinship in the Past (Oxford). P, S. B., and A, T. (), A Topographical Dictionary of Rome (Oxford). P, R., D V, A., D V-E, G., B, L. (), Libya: The Lost Cities of the Roman Empire (Cologne). P, Y., Y, E., and K, A. (), ‘Archaeological Remains at Jatt’, Atiqot, : – (Heb.) and *–* (English summary). P, S. R. F. (), Religions of the Ancient Greeks (Cambridge). P, S. (), Incesti crimina: regime giuridico da Augusto a Giustiniano (Milan). P, N. (forthcoming), ‘Romans in the Roman World’, in K. Galinsky (ed.), The Cambridge Companion to Augustus (Cambridge and New York). R. [sic], M. C.  (), ‘Estudio sobre la edad de la mortalidad en la Lusitania romana’, Caesaraugusta, : –. R, L. Y. (a), A Catalogue of Jewish Ossuaries in the Collections of the State of Israel (Jerusalem). —— (b), ‘Ossuaries and Ossilegium (Bone-Gathering) in the Late Second Temple Period’, in H. Geva (ed.), Ancient Jerusalem Revealed (Jerusalem), –.

References
R, T. (), Josephus (London). R S, J. L., and G G, E. (), ‘Las inscripciones de la necrópolis del Albarregas (Mérida) y su contexto arqueológico’, Veleia, : –. R, B. (), ‘Family Life among the Lower Classes at Rome in the First Two Centuries of the Empire’, Classical Philology, : –. —— (), ‘Roman Concubinage and Other De Facto Marriages’, Transactions of the American Philological Association, : –. —— (a) (ed.), The Family in Ancient Rome: New Perspectives (London, Sydney, Ithaca). —— (b), ‘Children in the Roman Familia’, in Rawson (a) –. —— (a), Marriage, Divorce and Children in Ancient Rome (Oxford and Canberra). —— (b), ‘Adult–Child Relationships in Roman Society’, in Rawson (a), –. —— (a), ‘The Iconography of Roman Childhood’, in Rawson and Weaver (), –. —— (b), ‘ “The Family” in the Ancient Mediterranean: Past, Present, Future’, Zeitschrift für Papyrologie und Epigraphik, : –. —— (), ‘Children as Cultural Symbols: Imperial Ideology in the Second Century’, in Dixon (a), –. —— (), Children and Childhood in Roman Italy (Oxford). —— and W, P. R. C. () (eds.), The Roman Family: Status, Sentiment, Space (Oxford). R, E. (), Intellectual Life in the Late Roman Republic (Baltimore). R, B. (), ‘Kinship and the Neighbourhood in NineteenthCentury Rural England: The Myth of the Autonomous Nuclear Family’, Journal of Family History, : –. R, R. (n.d.), My Roman Britain (Cirencester). R, A. (), ‘Philo on Infanticide’, Studia Philonica Annual, : –. R, C. (), ‘Concordia: Die Darstellung von Hochzeit und ehelicher Eintracht in der Spatantke’, in Spatantike und frühes Christentum, Austellungkatalog (Frankfurt), –. R, J. C (), ‘Felicitas Iulia Olisipo: Algumas considerações em torno do catálogo Lisboa Subterrânea’, Almadan, nd ser. : –. R, J. S. (), The Romans in Spain (Oxford). R, L., Jr. (), A New Topographical Dictionary of Ancient Rome (Baltimore and London). R, J. M. (), Dioscorides on Pharmacy and Medicine (Austin).
References

R, A. M. (), Crime and Community in Ciceronian Rome (Austin). R, L. (), ‘Androgynous Parents and Guest Children: The Huaorani Couvade’, Journal of the Royal Anthropological Institute, : –. R, J. B. () (ed. and tr.), Tacitus, Germania (Oxford). R, P. G. (), ‘The Couvade: A Problem Reborn’, Man,  : –. R, C., and M, K. (), The Archaeology of Disease, nd edn. (Ithaca). R, O. (), The Sources of Roman Law (London and New York). R, S. () ‘Il matrimonio fra cugini germani nella legislazione tardo imperiale’, Studia et documenta historiae et iuris, , –. R, D. (), The Building Program of Herod the Great (Berkeley). R, M. M. (), ‘Women on the Frontiers’, in V. A. Maxfield and M. J. Dobson (eds.), Roman Frontier Studies  (Exeter), –. R, N. (), ‘Romanisation, Cultural Identity and the Ethnic Discussion. The Integration of Lower Rhine Populations in the Roman Empire’, in Metzler, Millett, Roymans, and Slofstra (eds.), –. R, W. G. (), ‘Accelerating Social Mobility: The Case of Anglo-Saxon England’, Past and Present, : –. S, S. (a), ‘Home and Family’, in S. Safrai and M. Stern (eds.), The Jewish People in the First Century, vol. ii (Assen), –. —— (b), ‘Education and the Study of the Torah’, in S. Safrai and M. Stern (eds.), The Jewish People in the First Century, vol. ii (Assen), –. S, Z. (), The Economy of Roman Palestine (London). S  F, M. (), ‘Onomástica y sociedad en la epigrafía antigua de las provincias de Salamanca y Ávila’, Zephyrus, : –. S, R. (), The Ecology of the Ancient Greek World (London). —— (), Malaria and Rome (Oxford). S, R. P. (), ‘Familia, domus, and the Roman Conception of the Family’, Phoenix, : –. —— (a), ‘Men’s Age at Marriage and its Consequences in the Roman Family’, Classical Philology, : –. —— (b), ‘Slavery and the Roman Family’, Slavery & Abolition, : –. —— (), ‘European Family History and Roman Law’, Continuity and Change, : –. —— (), Patriarchy, Property and Death in the Roman Family (Cambridge).

References
S, R. P. (), ‘The Family and Society’, in J. Bodel (ed.), Epigraphic Evidence: Ancient History from Inscriptions (London and New York), –. —— and S, B. D. (), ‘Tombstones and Roman Family Relations in the Principate: Civilians, Soldiers and Slaves’, Journal of Roman Studies, , –. S, O. (), Die römischen Vornamen. Studien zur römischen Namengebung, Societas Scientiarum Fennica, Commentationes Humanarum Litterarum,  (Helsinki). S, B. (), ‘What’s in a Name? A Survey of Roman Onomastic Practice from c.  B.C. to A.D. ’, Journal of Roman Studies, : –. S M, E. (), ‘A propósito de las gentilitates: Los grupos familiares del área vetona y su adecuación para la interpretación de la organización social prerromana’, Veleia, : –. S, A., and V, A. (), I ceti medi in Cisalpina: Atti del colloquio internazionale – settembre  (Milan). Sˇˇ K, M. (), Pre-Roman Divinities of the Eastern Alps and Adriatic (Ljubljana). S, P. (), Judeophobia: Attitudes toward the Jews in the Ancient World (Cambridge, Mass.). S, W. (), ‘Libitina’s Bitter Gains: Seasonal Mortality and Endemic Disease in the Ancient City of Rome’, Ancient Society, : –. —— (), ‘Incest Revisited: Three Notes on the Demography of Sibling Marriage in Roman Egypt’, Bulletin of the American Society of Papyrologists, : –. —— (), Measuring Sex, Age and Death in the Roman Empire, Journal of Roman Archaeology suppl.  (Ann Arbor). —— (a) (ed.), Debating Roman Demography (Leiden). —— (b), Death on the Nile: Disease and the Demography of Roman Egypt (Leiden). S L. H., and VK, J. C. () (eds.), Encyclopedia of the Dead Sea Scrolls,  vols. (Oxford). S, A. (), Die römischen Grabsteine von Noricum und Pannonien, Sonderschriften des Österreichischen archäologischen Institutes in Wien,  (Vienna). S, B. I. (), Untersuchungen zur Tracht der römischen Matrona (Cologne). S, E. (–), The History of the Jewish People in the Age of Jesus Christ, revised G. Vermes, F. Millar, M. Black, and M. Goodman,  vols. (Edinburgh).
References

S, A. (), ‘Slums, Sanitation and Mortality in the Roman World’, Klio, : –. S, J. L. (), ‘Symbolism in the Costume of the Roman Woman’, in Bonfante and Sebesta (), –. S, M. (), ‘Nuclear is not Independent: Organisation of the Household in the Pays Bïgauden Sud in the Nineteenth and Twentieth Centuries’, in R. McC. Netting, R. R. Wilk, E. J. Arnold (eds.), Households: Comparative and Historical Studies of the Domestic Group (Berkeley, Los Angeles, London), –. S, L. (/), ‘Ornatus e status sociale delle donne romane’, Annali della Facoltà di Lettere e Filosofia di Perugia, : –. S, S. (), Childhood in the Middle Ages (London and New York). S, B. D. (), ‘Latin Funerary Epigraphy and Family Relationships in the Later Empire’, Historia, : –. —— (a), ‘The Age of Roman Girls at Marriage: Some Reconsiderations’, Journal of Roman Studies, : –. —— (b), ‘The Family in Late Antiquity’, Past and Present, : –. —— (), ‘The Cultural Meaning of Death: Age and Gender in the Roman Family’, in Kertzer and Saller (), –. —— (), ‘Explaining Incest: Brother–Sister Marriage in GraecoRoman Egypt’, Man, : –. —— () ‘Seasons of Death: Aspects of Mortality in Imperial Rome’, Journal of Roman Studies, : –. —— and S, R. P. (), ‘Close-Kin Marriage in Roman Society?’, Man, : –. S, M. (), ‘Choice of Marriage Partner in the Middle Ages: Development and Mode of Application of a Theory of Marriage’, Studies in Medieval and Renaissance History, : –, repr. in M. Sheehan (), Marriage, Family, and Law in Medieval Europe: Collected Studies (Toronto and Buffalo), –. S–W, A. N. (), The Roman Citizenship, nd edn. (Oxford). S, R. E. (), Galen’s System of Physiology and Medicine (Basel). S, A. V  (), Epigrafia de Olisipo: Subsídios para a história da Lisboa romana (Lisbon). S, F. (), Stadtrömische Marmorurnen (Mainz am Rhein). —— and F, K. (), Katalog der Skulpturen, Museo gregoriano profano ex lateranense, ii: Die Ausstattung des Hateriergrabes (Mainz). S, J. () ‘Changing Settlement Systems in the MeuseDemer-Scheldt Area during the Early Roman Period’, in N. Roymans, and F. Theuws (eds.), Images of the Past: Studies on

References
Ancient Societies in Northwestern Europe, Studies in prae- en protohistorie,  (Amsterdam), –. S, E. M. (), The Jews under Roman Rule from Pompey to Diocletian (Leiden). S, J. T. (), Roman Villas. A Study in Social Structure (London). S, R. R. R. (), ‘ “Simulacra Gentium”: The Ethne from the Sebasteion at Aphrodisias’, Journal of Roman Studies, : –. S, G. F. (), Ante Pacem: Archaeological Evidence of Church Life before Constantine (Macon, Ga.). S, F. H. Jr. (), ‘Roman Provincial Influence: An Onomastic Study of the Alentejo Region of Lusitania’, Beiträge zur Namenforschung, : –. S, M. (–), Greek and Latin Authors on Jews and Judaism,  vols. (Jerusalem). S, C. E. (), ‘Roman Gaul’, in J. M. Wallace-Hadrill and J. McManners (eds.), France, Government and Society, nd edn. (London), –. S, L. (), Family and Fortune: Studies in Aristocratic Finance in the Sixteenth and Seventeenth Centuries (Oxford). —— (), The Family, Sex and Marriage in England, – (New York). S, S. (), ‘The Toga: From National to Ceremonial Costume’, in Bonfante and Sebesta (), –. S, A. U. (), ‘Von Emil Hübner zur Neuauflage von CIL II’, Mitteilungen des deutschen archaeologischen Instituts. Abteilung Madrider, : –. S, J. (), Aquincum (Budapest). —— (), ‘Die Sterblichkeit in den Städten Mittel- und Süd-Italiens sowie in Hispanien in der römischen Kaiserzeit’, Acta archaeologica Academiae Scientiarum Hungaricae, : –. T, F. (), ‘Sévirat et promotion sociale en Italie nordorientale’, in M. Cébeillac-Gervasoni (ed.), Les élites municipales de l’Italie péninsulaire de la mort de César à la mort de Domitien entre continuité et rupture, Coll. EFR  (Rome), –. T, R. (), The Law of Greco-Roman Egypt in the Light of the Papyri  BC– AD (Warsaw). T, L. R. (), ‘New Light on the History of the Secular Games’, American Journal of Philology, : –. —— (), ‘Freedmen and Freeborn in the Epitaphs of Imperial Rome’, American Journal of Philology, : –. T, V. A., F, A., and S, M. (–), Corpus Papyrorum Judaicarum,  vols. (Cambridge, Mass.). T, G. (), Le harem et les cousins (Paris).
References

T, J. M. C. (), Death and Burial in the Roman World (London). T, V. T T (), Isis Lactans: Corpus des monuments grécoromains d’Isis allaitant Harpocrate (Leiden). T, A. (), La Galice romaine: recherches sur le nord-ouest de la péninsule ibérique dans l’antiquité (Paris). T, S. M. (), Roman Freedmen during the Late Republic (Oxford). —— (), ‘Jobs in the Household of Livia’, Papers of the British School at Rome, : –. —— (), ‘Contubernales in CIL VI.’, Phoenix : –. —— (), ‘Consent to Roman Marriage: Some Aspects of Law and Reality’, Échos du monde classique/Classical Views, ,  : –. —— (), Roman Marriage: Iusti Coniuges from the Time of Cicero to the Time of Ulpian (Oxford). —— (), ‘Leges sine Moribus’, Ancient History Bulletin, . . –. —— (), ‘Ancestral Virtues and Vices: Cicero on Nature, Nurture and Presentation’, in Braund and Gill (), –. —— (forthcoming a), ‘Women of the Time of Augustus’, in K. Galinsky (ed.), Cambridge Companion to the Age of Augustus (Cambridge). —— (forthcoming b), ‘Marriage and Family’, in S. Harrison (ed.), A Companion to Latin Literature (Oxford). T, A. (), The Letters of Anthony Trollope, ed. John N. Hall (Stanford). U, J. (), Elementos de un atlas antroponímico de la Hispania antigua, Bibliotheca Praehistorica Hispana,  (Madrid). V, D. () (ed.), Espacios y usos funerarios en el Occidente romano (Córdoba). V, F. (–), ‘La bilingue latina e neopunica di El Amrouni’, Helicon, –: –. V, J. L. I (), A Civitas de Viseu: Espaço e sociedade, História Regional e Local,  (Coimbra). V, G. (), The Complete Dead Sea Scrolls in English (Harmondsworth). V, P. (), ‘The Roman Empire’, in P. Veyne (ed.), A History of Private Life, i: From Pagan Rome to Byzantium (Cambridge, Mass., and London) –. V, Z. (), ‘Some notes on the Defence System of Pannonia in the nd and rd Centuries A.D’, in Hajnóczi (), –. —— (), Die Wagendarstellungen der Pannonischen Grabsteine (Pécs). V, F. (/), ‘Relievi funerari urbani con busti-ritratto’, Annali della Facoltà di Lettere e Filosofia di Perugia, ,  : –.

References
W, K., H, E. A., L, P. (), Statistical Studies of Historical Social Structure (New York). W, R., R, J., L, P. (), Family Forms in Historic Europe (Cambridge). W-H, A. (), ‘Elites and Trade in the Roman Town’, in J. Rich and A. Wallace-Hadrill (eds.), City and Country in the Roman World (London), –. —— (), Houses and Society in Pompeii and Herculaneum (Princeton). W, J. P. (–; repr. ), Étude historique sur les corporations professionnelles chez les Romains (New York). W-P, J. B. (), ‘From Republic to Empire: Reflections on the Early Imperial Provincial Architecture of the Roman West’, Journal of Roman Studies, : –. W, E. H. (), Remains of Old Latin,  (Cambridge, Mass.). —— (), Remains of Old Latin,  (Cambridge, Mass). W, A. (), ‘A Marriage Contract from the Province of Arabia Nova: Notes on Papyrus Yadin ’, Jewish Quarterly Review, : – (– discussion of marriage contract). W, P. A. (), Ancient Stepmothers. Myth, Misogyny and Reality, Mnemosyne suppl.  (Leiden). W, P. R. C. (), Familia Caesaris: A Social Study of the Emperor’s Freedmen and Slaves (Cambridge). W, J. (), ‘Creolizing the Roman Provinces’, American Journal of Archaeology, : –. W, J. E. G. (), ‘Tryphon’s Second Marriage (P. Oxy. II )’, Atti del XVII Congresso internazionale di Papirologia (Naples), iii: –. W, C. R. (), ‘Rural Labour in Three Roman Provinces’, in R. Garnsey (ed.), Non-Slave Labour in the Greco-Roman World, PCPhS suppl.  (Cambridge), –. —— (), Frontiers of the Roman Empire: A Social and Economic Study (Baltimore). W, T. (), Adults and Children in the Roman Empire (London). W, L. (), Die regionale Mobilität in Gallien nach den Inschriften des . bis . Jahrhunderts n.Chr.: Quantitative Studien zur Sozial- und Wirtschaftsgeschichte der westlichen Provinzen des römischen Reiches, Historia Einzelschriften,  (Stuttgart). W, E. M. () Roman Trier and the Treveri (London). —— (), ‘Rural Settlement in Roman Gaul’, Aufstieg und Niedergang der römischen Welt, . : –.
References

—— (), ‘Peasants and Potentates: An Investigation of Social Structures and Land Tenure in Roman Gaul’, American Journal of Ancient History, : –. —— (), Gallia Belgica (London). W, R. R., and N, R. McC. (), ‘Households: Changing Forms and Functions’, in R. McC. Netting, R. R. Wilk, and E. J. Arnold (eds.), Households: Comparative and Historical Studies of the Domestic Group (Berkeley, Los Angeles, London) –. W, J. J. (a), ‘The Danubian and Balkan Provinces’, in A. K. Bowman, E. Champlin, and A. Lintott (eds.), Cambridge Ancient History, x: The Augustan Empire,  B.C.–A.D. , nd edn. (Cambridge). —— (b), ‘Pannonia’s Identity and Entry into the Empire’, review of Hajnóczi (), Journal of Roman Archaeology, : –. W, M. H. (), ‘Palestinian Jewish Personal Names in Acts’, in R. Bauckham (ed.), The Book of Acts in its Palestinian Setting, The Book of Acts in its First Century Setting,  (Grand Rapids, Mich.), –. W, D. (), ‘Definition and Violation: Incest and Incest Taboos’, Man, : –. W, T. P. (a), Cinna the Poet and other Roman Essays (Leicester). —— (b), ‘The Last of the Metelli’, in Wiseman (a), –. W, R. (), Sexual Attraction and Childhood Association (Stanford). —— (), ‘Modelling Chinese Marriage Regimes’, Journal of Family History, : –. W, H. (), The Roman Empire and its Germanic Peoples, trans. by T. Dunlap (Berkeley and London). W, G. D. (), ‘World Systems Analysis and the Roman Empire’, Journal of Roman Archaeology, : –. —— (), ‘European Social Development and Roman Imperialism, Frontières d’Empire’, in P. Brun, S. van der Leeuw, and C. R. Whittaker (eds.), Nature et signification des frontières romaines (Nemours), –. —— (), ‘Monumental Writing and the Expansion of Roman Society in the Early Empire’, Journal of Roman Studies, : –. —— (), Becoming Roman. The Origins of Provincial Civilization in Gaul (Cambridge). —— (), ‘Urbanization and its Discontents in Early Roman Gaul’, in E. Fentress and S. Alcock (eds.), Romanization and the City: Creations, Transformations, and Failures, Journal of Roman Archaeology suppl.  (Providence, R. I.), –. W, H. (), Consecratio in Formam Deorum: Vergöttlichte Privatpersonen in der römischen Kaiserzeit (Mainz).

References
W, H. (), ‘Der Sarkophagdeckel eines Mädchens in Malibu und die frühen Klinensarkophage Roms, Athens und Kleinasiens’, in Roman Funerary Monuments in the J. Paul Getty Museum, Occasional Papers on Antiquities,  (Malibu, Calif.), –. Y, Y., G, J. C., and Y, A. (), ‘Babatha’s Ketubba’, Israel Exploration Journal, : –. Y, S. J. (), ‘Explicating Residence: A Cultural Analysis of Changing Households among Japanese-Americans’, in R. McC. Netting, R. R. Wilk, and E. J. Arnold (eds.), Households: Comparative and Historical Studies of the Domestic Group (Berkeley, Los Angeles, London), –. Y, A., and G, J. C. (), ‘A Receipt for a Ketubba’, in Y. Gafni, A. Oppenheimer, and D. Schwartz (eds.), The Jews in the Hellenistic-Roman World: Studies in Memory of Menahem Stern (Jerusalem), – (Heb.). Y, H. C. (), ‘: Law vs custom in Roman Egypt’, in Le Monde Grec: Hommages à Cl. Préaux (Brussels), –, repr. in H. C. Youtie (), Scriptiunculae Posteriores (Bonn), i. –. —— (a), ‘P. Mich. inv. : Arrangement Regarding Inheritance’, Zeitschrift für Papyrologie und Epigraphik, : –. —— (b) ‘P. Mich inv.  verso: The Rule of Precedent’, Zeitschrift für Papyrologie und Epigraphik, : –. Z, M. (), ‘Antinoopolis in Ägypten: Die hadrianische Gründung und ihre Privilegien in der neueren Forschung’, Aufstieg und Niedergang der römischen Welt, . . : –. Z, P. (), ‘Grabreliefs römischer Freigelassener’, Jahrbuch des Deutschen Archäologischen Instituts, : –. —— (), ‘Zur Rezeption des hellenistischen Individualporträts in Rom and in den italischen Städten’, in P. Zanker (ed.), Hellenismus in Mittelitalien (Göttingen), –. —— (), ‘Zur Bildnisrepräsentation führender Männer in mittelitalischen und campanischen Städten zur Zeit der späten Republik und der julischclaudischen kaiser’, in M. Cébeillac-Gervasoni (ed.), Les ‘Bourgeoisies’ municipales italiennes aux IIième et Iier siècles av. J.-C. (Paris), –. —— (), The Power of Images in the Age of Augustus (Ann Arbor). —— (), ‘The Hellenistic Grave Stelai from Smyrna: Identity and Self-Image in the Polis’, in A. Bulloch et al. (eds.), Images and Ideologies: Self Definition in the Hellenistic World (Berkeley), –. —— (), ‘Die Frauen und Kinder der Barbaren auf der Markussäule’, in J. Scheid and V. Huet (eds.), Autour de la colonne Aurélienne: Geste et image sur la colonne de Marc Aurèle à Rome (Turnhout), –.
References

—— and  H, H. (), Römische Gräbstraßen: Selbstdarstellung—Status—Standard (Munich). Z, P. (), Die antiken Gemmen (Munich). Z, G. (), Römische Berufsdarstellungen (Berlin). Z, M. (), Apuleius Madaurensis Metamorphoses Book X: Text, Introduction and Commentary (Groningen). Z, B. (), ‘Odd Tomb Out: Has Jerusalem’s Essene Cemetery been Found?’, Biblical Archaeology Review, /: – and . Z, F. () ‘An Anthropological Perspective on Kinship and the Family’, in A. Burgière, C. Klapisch-Zuber, M. Segalen, and F. Zonabend (eds.), A History of the Family, i: Distant Worlds, Ancient Worlds (Cambridge and Oxford), –.
This page intentionally left blank
INDEX
Abbius Oppianicus, Statius – adolescence –, –, – Aelius Aurelius Antoninus, M. (Marcus Aurelius) , , , , ,  Column of  Aemilia Pudentilla – affine  Alexander Severus (M. Aurelius Severus Alexander) , ,  Ambrose, St. of Milan  Annius Milo, T. (Milo) ,  Antonius, M. (Praetor Pro Consule  ) –, – Antonius, M. (Mark Antony) – Antonius, L.  Apuleius , , , , , –,  Artemidorus  Asia Minor  Atticus (Q. Caecilius Pomponius Atticus) ,  Augusta Emerita (modern Mérida) , , , ,  Augustales ,  Augustine, St, of Hippo , , , , –,  Augustus (C. Iulius Caesar Octavianus) , , , , , , , , , ,  Aurelius Commodus Antoninus, M. ,  Aurelius Severus Antoninus, M. (Caracalla) , , , , , – Babatha –, ,  Bologna ,  bonus vir ,  bride-price (mohar)  bulla , , –,  Caelius Rufus, M. , ,  Capernaum  Caracalla, see Aurelius Severus Antoninus, M.
Cassius Dio – Cato the Elder (M. Porcius Cato) –, , , ,  Celsus –, –, ,  census returns , , ,  Cherry, D. – children – amulets , –,  cholera  deafness  diarrhoea , , ,  dysentery ,  epilepsy , , , , , ,  malaria  malnutrition  mortality , , , – surgery –,  tuberculosis  typhoid  Christianity , , , , , – Cicero, see Tullius Cicero, M. circumcision, see Jews Cisalpine Gaul – citizenship – civitas Igaeditanorum, see Lusitania Clodius Pulcher, P. ,  Cluentius (Aulus Cluentius Habitus) –,  Code of Justinian (Codex Justinianus) –,  collegia ,  colliberti ,  commemoration analysed by gender – by children , –, –, –, , , , , , –, – by mothers , – by siblings – by spouse –, , –, , –,  joint – Commodus, see Aurelius Commodus Antoninus, M. concordia , , ,  conservi , , 

Index
Constantine, see Flavius Valerius Constantinus, C. contubernia ,  conubium  cousins , , , –, , , –, , , , , , , , – curator minorum  death: of children –, –, , – of infants , , ,  of parents ,  dedicitii  dextrarum iunctio ,  Digest , , , , , , ,  Diocletian (Valerius Diocletianus) , , , , , , , , ,  Dioscorides  divorce , , , –, , , , , – doctors , , , , , , , , , , ,  dowry , , , , –, , , ,  education –, – Egypt: households –, – letters – metropolite status –, – elderly , , , –, – endogamy, see marriage exogamy, see marriage extended family , , , , –, , –, , –,  familia , , , , – family affection –, , , –, , , , , , , , –, , – family law – family structure –, –, –, – Fasti  fathers , , , , , , ,  consent over marriage – rights over children – see also paterfamilias and patria potestas filiation  filiusfamilias – Flavius Valerius Constantinus, C. (Constantine) 
freedmen and freedwomen –, , , –, , , , , ,  Fronto, M. Cornelius , ,  Galen –, –, ,  Gaul: domestic architecture – education – family structure –, – Mont Beuvray ,  grandchildren –, , , ,  grandparents , –, , , , , , , , ,  guardianship , , – Hecataeus of Abdera –, –, ,  Herod (C. Julius Herodes) – Drusilla (Iulia Drusilla)  palaces –, – Salome (Iulia Salome) – Hippocrates  Hispania Tarraconensis  imperial cult  see also Augustales incest –, ,  infants –, , , , –, , , , , –,  ingenui , , , , ,  inheritance –, –, –, , –, – inofficiosum testamentum ,  Iulius Caesar, C. , –, ,  Iulius Caesar, Ti. , ,  iustum matrimonium , ,  Jericho , ,  Jerusalem , , , , ,  Jews: burial practices – circumcision , –, , ,  divorce (get) , , , – domestic architecture – dowry , –,  education – Gentiles , ,  Josephus , , , , , –, ,  marriage –, – onomastics –
Index ossuaries , –, , –,  polygamy –,  John Chrysostom  Kellis – kinship – bilateral ,  matrilineal , , , ,  patrilineal  Laslett, P. , – levirate ,  libertination ,  libertus, see freedmen Lusitania: commemorative patterns – onomastics , , – manumission , , –, ,  Marcus Aurelius, see Aelius Aurelius Antoninus, M. marriage , , –, –, – ‘Arab marriage’  between close-kin –, , – endogamy –, –,  exogamy , ,  monogamy – paternal consent –,  polygamy –,  remarriage ,  sibling marriage , , ,  sim-pua marriage  source of family conflict – Masada ,  matertera  matrona , , –,  matronymics  Maximian , , , , –,  metropolite status, see Egypt Milo, see Annius Milo, T. Minicia Marcella  miqvaot , – Modestinus (Herennius Modestinus)  monogamy, see marriage mos maiorum  mothers: as commemorators , – nursing , –, – municipia , , ,  nieces  North Africa: burial customs –

onomastics , –,  nuclear family –, , –, –, –, , –, , –, –, –, –, – nurses –,  obsequium ,  oikiai ,  Olisipo (modern Lisbon). see Lusitania onomastics –, –, , –,  indigenous  Oppianicus, see Abbius Oppianicus, Statius orphans  Ostia  palla , , ,  Pannonia: cult of Nutrices Augustae  Intercisa , , , , ,  Savaria –,  Papinian (Aemilius Papinianus) , ,  papyri , , , , –, , –,  P. Fam. Tebt – P. Yadin –, ,  parents –, – abused by children – parricide – pater , ,  paterfamilias , , , ,  paternal power, see patria potestas patria potestas , ,  patrons , –, , ,  patruus ,  Paulus (Iulius Paulus) , , –, ,  peregrine , –, , –, –,  petitions – Petronius  pietas , , , –, , , , , ,  plague –,  Plancius, Cn. – Pliny the Elder (C. Plinius Secundus) , –, , –, ,  Pliny the Younger (C. Plinius L. f. Secundus)  Plutarch ,  polygamy, see marriage pudicitia , 

Index
Quintilian (M. Fabius Quintilianus)  Quintus, brother of Cicero, see Tullius Quintus, nephew of Cicero, see Tullius Ravenna ,  rescripts – repudium  rhetoric , –,  rickets –,  Rimini  Romanization , –, , , , , –, , , –, – Saller, R. –, –, –, –, , , –, , – Sassia –, ,  Scarbantia , , , ,  Scribonius Largus  senate , , –, , – Septimius Severus, L. –, , , , ,  Sextus Roscius –, ,  Shaw, B. –, –, –, –, , , , –, , , , – sim-pua marriage see marriage Slaves –, ,  commemoration , , , , ,  family –, – manumission –, , ,  origins ,  soldiers , , , , –, –, , , , 
Soranus –, –, , ,  sponsus, sponsa  stola – subscriptiones – Terentius Varro, M.  Tertullian , ,  Tiberias ,  Tiberius, see Iulius Caesar, Ti. toga , , ,  toga praetexta , ,  toga virilis ,  Trajan, see Ulpius Nerva Traianus, M. Tullius Cicero, M. –, –,  Tullius Cicero, M. junior ,  Tullius Cicero, Q. ,  Tullius Cicero, Q. junior  tutores, see guardianship Ulpian (Domitius Ulpianus) , ,  Ulpius Nerva Traianus, M.  uncles , , , , ,  undutiful wills ,  univira ,  uxor , –,  Valerius Maximus , ,  Valesius –, ,  Varro, see Terentius Varro, M. verna  Verres, C. –,  vitta –, ,  wet-nursing –, ,  wills , , , –, –