Sir Bevis of Hampton in Literary Tradition 1843841738, 9781843841739

Citation preview

Sir Bevis of Hampton in Literary Tradition

Sir Bevis of Hampton is one of the most widespread and important Middle English romances. This book – the first ever full-length study to be devoted to it – considers it in its historical and literary contexts, and its Anglo-Norman, Welsh, Irish and Icelandic versions. It also offers detailed textual analyses, and discusses particular aspects of the story, its ‘afterlife’ and its influence during the early modern period.

Studies in Medieval Romance ISSN 1479–9308

Series Editors Corinne Saunders Roger Dalrymple This series aims to provide a forum for critical studies of the medieval romance, a genre which plays a crucial role in literary history, clearly reveals medieval secular concerns, and raises complex questions regarding social structures, human relationships, and the psyche. Its scope extends from the early middle ages into the Renaissance period, and although its main focus is on English literature, comparative studies are welcomed. Proposals or queries should be sent in the first instance to one of the addresses given below; all submissions will receive prompt and informed consideration. Dr Corinne Saunders, Department of English, University of Durham, Durham, DH1 3AY Boydell & Brewer Limited, PO Box 9, Woodbridge, Suffolk, IP12 3DF Volumes already published I: The Orient in Chaucer and Medieval Romance, Carol F. Heffernan, 2003 II: Cultural Encounters in the Romance of Medieval England, edited by Corinne Saunders, 2005 III: The Idea of Anglo-Saxon England in Middle English Romance, Robert Allen Rouse, 2005 IV: Guy of Warwick: Icon and Ancestor, edited by Alison Wiggins and Rosalind Field, 2007 V: The Sea and Medieval English Literature, Sebastian I. Sobecki, 2008 VI: Boundaries in Medieval Romance, edited by Neil Cartlidge, 2008 VII: Naming and Namelessness in Medieval Romance, Jane Bliss, 2008

Sir Bevis of Hampton in Literary Tradition

Edited by

Jennifer Fellows Ivana DjordjeviĆ

D. S. BREWER

©  Contributors 2008 All Rights Reserved. Except as permitted under current legislation no part of this work may be photocopied, stored in a retrieval system, published, performed in public, adapted, broadcast, transmitted, recorded or reproduced in any form or by any means, without the prior permission of the copyright owner First published 2008 D. S. Brewer, Cambridge ISBN  978–1-84384–173–9

D. S. Brewer is an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge, Suffolk IP12 3DF, UK and of Boydell & Brewer Inc. 668 Mount Hope Ave, Rochester, NY 14604, USA website: www.boydellandbrewer.com A CIP catalogue record for this book is available from the British Library This publication is printed on acid-free paper Printed in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire

Contents Illustrations

vi

Contributors

vii

Abbreviations and Sigla

ix

Boeve/Bevis: A Synopsis

xi

Introduction Ivana Djordjević and Jennifer Fellows

1

1 The Anglo-Norman Boeve de Haumtone as a chanson de geste Marianne Ailes

9

2 Mestre and Son: The Role of Sabaoth and Terri in Boeve de Haumtone Judith Weiss

25

3 Rewriting Bevis in Wales and Ireland Erich Poppe and Regine Reck

37

4 Bevers saga in the Context of Old Norse Historical Prose Christopher Sanders

51

5 From Boeve to Bevis: The Translator at Work Ivana Djordjević

67

6 The Middle English and Renaissance Bevis: A Textual Survey Jennifer Fellows

80

7 For King and Country? The Tension between National and Regional Identities in Sir Bevis of Hampton Robert Allen Rouse

114

8 Defining Christian Knighthood in a Saracen World: Changing Depictions of the Protagonist in Sir Bevis of Hampton Siobhain Bly Calkin

127

9 Ascopard’s Betrayal: A Narrative Problem Melissa Furrow

145

10 Gender, Virtue and Wisdom in Sir Bevis of Hampton Corinne Saunders

161

11 Sir Bevis of Hampton: Renaissance Influence and Reception Andrew King

176

Bibliography of Bevis Scholarship

193

Index

203

Illustrations Plate 1 Naples, Biblioteca Nazionale, MS XIII.B.29, p. 30. 86 (Reproduced by permission of the Ministero per i Beni e le Attività Culturali.) The tail-rhyme passage in column b, where the merchants attempt to sell Bevis, is shared only by the Egerton MS (S). 88 Plate 2 Naples, Biblioteca Nazionale, MS XIII.B.29, p. 76. (Reproduced by permission of the Ministero per i Beni e le Attività Culturali.) 92 Plate 3 Cambridge, Trinity College, MS O.2.13, fol. 152r. (Reproduced by permission of the Master and Fellows, Trinity College, Cambridge.) 95 Plate 4 Manchester, Chetham’s Library, MS 8009, fol. 165v. (Reproduced by permission of Chetham’s Library, Manchester.) The scribal correction in the ninth line suggests the use of more than one exemplar. 100 Plate 5 Thomas East’s edition (1582?), sigs [Eiv]v–F[i]r. (Reproduced by permission of the Provost and Fellows, King’s College, Cambridge.) This is the earliest surviving edition to set out each couplet as a single line. Plate 6 William Stansby’s edition (1630), sig. E3v, showing a change 102 in the style of illustration (cf. Plate 5 above). (Reproduced by permission of the Folger Shakespeare Library, Washington, DC.) Disclaimer: Some images in the printed version of this book are not available for inclusion in the eBook. To view these images please refer to the printed version of this book.

vi

Contributors Marianne Ailes is Lecturer in French at the University of Bristol. Her recent publications include The Song of Roland: On Absolutes and Relative Values (2002) and an edition and translation of Ambroise’s The History of the Holy War (with Malcolm Barber, 2003). Siobhain Bly Calkin is Associate Professor at Carleton University in Ottawa. She is the author of Saracens and the Making of English Identity: The ­Auchinleck Manuscript, as well as various articles. Her research interests include Saracens, romances in their manuscript contexts, and notions of closure in medieval texts. Ivana Djordjević is Assistant Professor in the Liberal Arts College at Concordia University, Montreal. She has published on Anglo-Norman and Middle English romance and on the poetics of rewriting, especially translation. Jennifer Fellows edits the Modern Humanities Research Association’s Annual Bibliography of English Language and Literature. Her own published work is mostly on Middle English romance and on the theory and practice of editing medieval popular texts. She has also published an edition of Richard ­Johnson’s late Elizabethan prose romance The Seven Champions of Christendom (Ashgate, 2003). Melissa Furrow is Professor of English at Dalhousie University in Halifax, Nova Scotia. She is currently working on a book on the reception of romances in medieval England. She will also be bringing out a new teaching edition of Ten Fifteenth-Century Comic Poems in the TEAMS series. Andrew King is Lecturer in Medieval and Renaissance English at University College Cork, Ireland. He is the author of ‘The Faerie Queene’ and Middle English Romance: The Matter of Just Memory, as well as articles on Spenser, Sidney, romance and Shakespeare. Erich Poppe is Professor of Celtic Studies and General Linguistics at the Philipps-Universität Marburg. One of his research interests is the interactions of the medieval Welsh and Irish literatures with the languages, literatures and cultures with which they came in contact and with the literary and cultural processes that shaped the resulting translations. His publications include articles on the medieval Welsh and Irish adaptations of the story of Bevis of Hampton. Regine Reck holds a Ph.D. in Celtic Studies from the Philipps-Universität, Marburg; her study (The Aesthetics of Combat in Medieval Welsh Literature) will be published by Harrassowitz in 2008. She has collaborated with Erich vii

Poppe on a detailed analysis of the medieval Welsh version of the Geste de Boeve de Haumtone (published in Zeitschrift für celtische Philologie) and on an edition of selected passages of Ystorya Bown o Hamtwn (forthcoming with the University of Wales Press). Robert Rouse is Associate Professor of Medieval Literature in the Department of English at the University of British Columbia. His research interests range across both Old and Middle English literary culture, addressing issues of historiography, nationalism and the medieval geographical imagination. He is the author of The Idea of Anglo-Saxon England in Middle English Romance (2005) and (with Cory J. Rushton) The Medieval Quest for Arthur (2005). Christopher Sanders is one of the editors of the Arnamagnæan Commission’s Dictionary of Old Norse Prose in Copenhagen. His edition of Bevers saga was published in 2001. Corinne Saunders is Professor in the Department of English Studies, ­University of Durham. She has published extensively on Middle English literature and is currently writing about magic and the supernatural in medieval romance and culture. She is an editor of Medium Ævum. Judith Weiss is Emeritus Fellow of Robinson College, Cambridge. Her interests lie mainly in the field of Anglo-Norman romance and historiography: her parallel text and translation of Wace’s Roman de Brut appeared in 1999, and her translation of Boeve de Haumtone and Gui de Warewic will be published in early 2008 for the French of England Translation Series.

viii

Abbreviations and Sigla Abbreviations BHF Jennifer L. Fellows, ‘Sir Beves of Hampton: Study and Edition’, 5 vols (unpub. Ph.D. diss., Univ. of Cambridge, 1980) BHK The Romance of Sir Beues of Hamtoun …, ed. Eugen Kölbing, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) Boeve Der anglonormannische Boeve de Haumtone, ed. Albert Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974) Bown Ystorya Bown de Hamtwn, ed. Morgan Watkin (Cardiff: University of Wales Press, 1958) Bs Bevers saga, ed. Christopher Sanders, Stofnun Árna Magnússonar á Íslandi 51 (Reykjavík: Stofnun Árna Magnússonar á Íslandi, 2001) CFMA Classiques français du moyen âge EEBO Early English Books Online EETS Early English Text Society Extra Series   es os Original Series   SATF Société des anciens textes français STC A Short-Title Catalogue of Books Printed in England, Scotland, & Ireland, and of English Books Printed Abroad, 1475–1640, ed. W. A. Jackson, F. S. Ferguson and Katharine F. Pantzer, rev. edn, 3 vols (London: Bibliographical Society, 1976–91) Short-Title Catalogue of Books Printed in England, Scotland, Wing Ireland, Wales, and British America and of English Books Printed in Other Countries 1641–1700, ed. Donald G. Wing, 3 vols (New York: Columbia University Press for the Index Society, 1948–51) Sigla A B C Cp E G K M

Edinburgh, National Library of Scotland, MS Advocates’ 19.2.1 (Auchinleck MS) Oxford, Bodleian Library, MS Eng. Poet. d.208 Cambridge University Library, MS Ff.2.38 Bevis edition (1560?) by William Copland (STC 1988.8) Cambridge, Gonville and Caius College, MS 175/96 Bevis edition (c. 1626) by G. W. for W. Lee (STC 1993) Bevis edition (1582?) by Thomas East (STC 1990) Manchester, Chetham’s Library, MS 8009 ix

N O Q S T V W

Naples, Biblioteca Nazionale, MS XIII.B.2 Bevis edition (c. 1503) by Richard Pynson (STC 1988) Bevis edition (1565?) by William Copland (STC 1989) London, British Library, MS Egerton 2862 Cambridge, Trinity College, MS O.2.13 Bevis edition (c. 1610) attributed to T. Snodham (STC 1992) Fragments of a Bevis edition (c. 1533) by Wynkyn de Worde (STC 1988.6)



Boeve/Bevis: A Synopsis The following plot summary is based on the Anglo-Norman narrative. Major differences between the Anglo-Norman (AN) and Middle English (ME) versions are indicated in parentheses. The ten-year-old Boeve (the seven-year-old Bevis in the ME) is sold into slavery through the machinations of his wicked mother, who arranges the death of her elderly husband (Gui, earl of Southampton) at the hands of her lover Doun (unnamed in most ME texts), the emperor of Germany, whom she then marries. Bevis finds favour at the court of the pagan King Hermine of Egypt (Armenia in the ME), whose daughter Josiane falls in love with him and presents him with the horse Arundel. After Boeve has led Hermine’s armies to victory against the invading King Brademond, Josiane declares her love; Boeve eventually agrees to love her provided that she embraces ­Christianity. At this point Boeve is falsely accused to the king of having seduced Josiane and is sent as a messenger to King Brademond in Damascus, bearing a letter which orders his own death. On the way he meets the son of his ‘master’/foster-father Sabaoth (his uncle Saber in the ME), who is protecting his interests in England. Bevis is incarcerated in Brademond’s dungeon, while Josiane is forced to marry King Yvori of Monbrant but manages to preserve her virginity. After seven years Boeve escapes and, disguised as a palmer, comes to Yvori’s castle. He rescues Josiane and Arundel and flees into the forest. Here he kills two lions and subdues the giant Escopart (Ascopard in the ME), who has been sent by Yvori to bring Josiane back, and who becomes Boeve’s ‘page’. The three travel to Cologne, where Boeve’s uncle is bishop; Josiane and (in the AN but not in the ME) Escopart are baptized. (In the ME, Bevis proceeds to deliver the city from a dragon.) Boeve returns to England to reclaim his lands, only to be recalled by Escopart to Cologne, where Josiane has been forced into marriage with Earl Miles. Having strangled him on their wedding-night, she is condemned to be burned at the stake but is rescued by Boeve and Escopart. They all go to England, where Boeve wrests his earldom from his stepfather. Doun comes to an ignominious end in a cauldron of boiling lead; his wife, the hero’s mother, flings herself from a tower. Boeve and Josiane are married. Boeve wins a race on Arundel at Whitsun and with the prize money builds Arundel Castle, but he is driven into exile when King Edgar’s son attempts to steal Arundel and is kicked to death. A pregnant Josiane accompanies her husband into exile. After giving birth to twin sons in a forest, she is carried off by the traitor Escopart, who has again been sent in pursuit of her by Yvori. xi

Learning of Josiane’s predicament through a dream, Sabaoth sets out to rescue her; he kills Escopart. For seven years, Sabaoth and Josiane seek Boeve, who meanwhile has married the duchess of Civile, though their marriage remains unconsummated. After the family is finally reunited, Boeve’s son Gui becomes king of Egypt upon his maternal grandfather’s death. Boeve kills Yvori in single combat and becomes king of Monbrant. Boeve returns to England again, to support Sabaoth’s son Robant against King Edgar. (In the ME, the king’s steward stirs up the citizens of London against Bevis; a street battle ensues, in which Bevis kills many thousands of men.) The king sues for peace with Boeve, offering his daughter in marriage to Boeve’s son Miles. Boeve and Josiane return to Monbrant, where they and Arundel all die on the same day.

xii

Introduction IVANA DJORDJEVIĆ and jennifer fellows Although it is relatively little known today, the story of Bevis of Hampton was among the most popular narratives of the medieval and early modern periods, its only serious rival in this respect being that of Guy of Warwick. There are many parallels between the textual and reception histories of the English versions of Bevis and of Guy, but also considerable differences. Both are translations of Anglo-Norman texts that are generally regarded as ‘ancestral romances’, associated with particular aristocratic families and specific localities; texts of both appear in two of the major medieval manuscript compilations of Middle English romances extant from the medieval period; both were translated into Irish in the fifteenth century; both were printed in the sixteenth century by Wynkyn de Worde and William Copland, the giants among Renaissance printers of medieval romance; the ‘vogue’ of both continued well into the seventeenth century, both being repeatedly singled out in Humanist and Puritan denunciations of popular secular literature; both Bevis and Guy have, to some extent, an extra-literary life as folk heroes;



 

 

 

Cf. Ronald S. Crane, ‘The vogue of “Guy of Warwick” from the close of the Middle Ages to the Romantic revival’, PMLA, 30 (1915), 125–94; Guy of Warwick: Icon and Ancestor, ed. by Alison Wiggins and Rosalind Field, Studies in Medieval Romance (Cambridge: Brewer, 2007). See, e.g., M. Dominica Legge, Anglo-Norman Literature and Its Background (Oxford: Clarendon Press, 1963), pp. 139–75. See The Auchinleck Manuscript: National Library of Scotland Advocates’ MS 19.2.1, introd. by Derek Pearsall and I. C. Cunningham (London: Scolar Press in association with the National Library of Scotland, 1977); Cambridge University Library MS Ff.2.38, introd. by Frances McSparran and P. R. Robinson (London: Scolar Press, 1979). See ‘The Irish lives of Guy of Warwick and Bevis of Hampton’, ed. and trans. by F. N. Robinson, Zeitschrift für celtische Philologie, 6 (1908), 9–180, 273–338. See Carol M. Meale, ‘Caxton, de Worde, and the publication of romance in late medieval England’, Library, 14 (1992), 283–98; A. S. G. Edwards, ‘William Copland and the identity of printed Middle English romance’, in The Matter of Identity in Medieval Romance, ed. Phillipa Hardman (Cambridge: Brewer, 2002), pp. 139–47. See Ronald S. Crane, The Vogue of Medieval Chivalric Romance during the English Renaissance (Menasha, WI: University of Wisconsin Press, 1919). See, e.g., David Griffith, ‘The visual history of Guy of Warwick’, in Guy of Warwick: Icon and Ancestor, ed. Wiggins and Field, pp. 110–32; Jennifer Fellows, ‘Sir Bevis of Hampton in popular tradition’, Proceedings of the Hampshire Field Club and Archaeological Society, 42 (1986), 139­–45.





Sir Bevis of Hampton in Literary Tradition

scenes from both romances are (or were) represented in a variety of visual media. On the other hand, while Guy gave rise to a generically more diverse group of texts than did Bevis, the latter is unique in having continued to be printed in its fifteenth-century metrical form from the beginning of the Tudor period until the early years of the eighteenth century.10 The respective lengths of the two romances are probably significant here: only about one-third the length of Guy, Bevis would lend itself more readily to cheap production for a mass market. By contrast, Guy tended to be adapted and excerpted, as well as accruing self-contained episodes (such as that of the Dun Cow) that formed no part of the Middle English romance.11 Curiously, given that the Guy of the romance that bears his name is not of noble birth, he was appropriated as an ancestor for the purposes of ‘baronial propaganda’ by the families associated with the earldom of Warwick in a way that the aristocratic Bevis seems not to have been by the earls of Arundel.12 This is perhaps because Guy was, from early on, taken seriously by chroniclers (including Knighton, Hardyng and Holinshed) as a part of English history,13 whereas the story of Bevis remained largely confined to the realms of fiction and folklore. Bevis belongs to a much larger and more complex literary tradition than does Guy, versions of his story spreading throughout Europe during the medieval and early modern periods. In the heyday of Continental philology, with its source-hunting obsession, theories about the origins of the Bevis story were numerous and often fiercely argued over. It was suggested, for example, that it was of German origin, Hamtone being a misidentified town not far from Mainz; that the home of the earliest version was in north-western France; that the roots of the story were to be found in a tenth-century Viking saga and that King Hermin’s country is not Armenia but Armorica, i.e. Brittany; that the story had to be of Celtic origin because adultery was a central motif in it; that this obviously Anglo-Saxon story was no more than an expanded and romanticized version of the tale of Horn; that the Persian–Armenian origin of the story was confirmed by its onomastics; that the story of Bevis was essentially the same as the story of Hamlet, ultimately derived from a conflation of the 



10 11 12 13

Jessica Brantley, ‘Images of the vernacular in the Taymouth Hours’, English Manuscript Studies, 1100–1700, 10 (2002), 83–113; Jennifer Fellows, ‘Romance among the choir-stalls: Middle English romance motifs on English misericords’, in Profane Images in Marginal Arts of the Middle Ages, ed. by Elaine C. Block and Malcolm Jones (Turnhout: Brepols, forthcoming). Cf. A. S. G. Edwards, ‘The Speculum Guy de Warwick and Lydgate’s Guy of Warwick: the nonromance Middle English tradition’, in Guy of Warwick: Icon and Ancestor, ed. Wiggins and Field, pp. 81–93. See pp. 109–13 below. Cf. Griffith, ‘The visual history of Guy of Warwick’, p. 116. See Emma Mason, ‘Legends of the Beauchamps’ ancestors: the use of baronial propaganda in medieval England’, Journal of Medieval History, 10:1 (1984), 25–40. See A Manual of the Writings in Middle English 1050–1500, ed. by J. Burke Severs, I: Romances (New Haven, CT: Connecticut Academy of Arts and Sciences, 1967), p. 30; Griffith, ‘The visual history of Guy of Warwick’, in Guy of Warwick: Icon and Ancestor, ed. Wiggins and Field, p. 118.





Introduction

stories of Bellerophon and of Brutus; and so forth.14 These views illustrate both the impossibility of establishing where the individual building-blocks of this not too original narrative came from and the difficulty, for these mostly Continental scholars, of separating the Insular versions of the narrative from the Continental ones with which they were also familiar – especially at a time when the dating of all these texts was even more uncertain and contentious than it is now.15 We may smile at the foibles of our scholarly ancestors, but there is a reason for their perplexity: the story’s pan-European success was so overwhelming that we should not be surprised to see its origins questioned. The Anglo-Norman version, whose Continental French reworkings inaugurated Bevis’s triumphant eastward march across Europe, reversing the customary direction of translatio studii, featured a hero whose Englishness is more tenuous than subsequent English recastings would indicate. In this respect further comparison with Guy of Warwick is illustrative. Mobilized for the promotion of a rather ostentatious kind of Englishness as early as its Anglo-Norman redactions, the story of Guy was enormously popular in England but did not spread very far beyond the area of English cultural influence. Thus both the fifteenth-century Irish Life of Sir Guy and the Romaunt de Guy de Warwik et de Herolt d’Ardenne, composed at roughly the same time in Continental French prose, appear to have been commiss­ioned by English patrons;16 and while some manuscripts of the French romance did make it to the Continent, they left no progeny. Guy’s emphatic Englishness did not travel well; Bevis, on the other hand, did not acquire the status of national hero until he was translated into English (see Robert Allen Rouse’s chapter below). The Anglo-Norman narrative was largely free of nationalist baggage, a feature that greatly facilitated its protean trans­formations in Europe. Only the Irish version, produced in the fifteenth century, was translated from a Middle English redaction; as suggested below by Erich Poppe and Regine Reck, it may have been produced for a family with strong ties to England.

14

For the views summarized above, see Pio Rajna, Ricerche intorno ai ‘Reali di Francia’ (Bologna: Romagnoli, 1872), p. 123 (Gaston Paris was of a similar opinion); Max Deutschbein, Studien zur Sagengeschichte Englands (Cöthen: Schulze, 1906), p. 205; Hermann Suchier, ‘Nachtrag’, in Der anglonormannische Boeve de Haumtone, ed. by Albert Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974), pp. cxcv–cxcvi; Richard Wülker, Geschichte der englischen Literatur von den älteste Zeiten bis zur Gegenwart (Leipzig; Vienna: Bibliographisches Institut, 1896), p. 98; Prentiss C. Hoyt, ‘The home of the Beves saga’, PMLA, 17:2 (1902), 237–46; Franz Settegast, Quellenstudien zur galloromanischen Epik (Leipzig: Harrassowitz, 1904), pp. 338–69; Rudolf Zenker, Boeve–Amlethus (Berlin: Felber, 1905). 15 A further problem is posed by the relationship between the Anglo-Norman Boeve de Haumtone and the Provençal chanson de geste of Daurel et Beton – a puzzle as yet without a satisfactory explanation: see A Critical Edition of the Old Provençal Epic ‘Daurel et Beton’, ed. by Arthur S. Kimmel, University of North Carolina Studies in the Romance Languages and Literatures 108 (Chapel Hill, NC: University of North Carolina Press, 1971), esp. pp. 43–6. 16 See Martha W. Driver, ‘ “In her owne persone semly and bewteus”: representing women in stories of Guy of Warwick’, in Guy of Warwick: Icon and Ancestor, ed. Wiggins and Field, pp. 133–53 (p. 134).





Sir Bevis of Hampton in Literary Tradition

All the other European versions go back to the Anglo-Norman text and its descendants, in a tradition that places little emphasis on the hero’s national origins. Boeve was the source of the Middle Welsh Ystorya Bown o Hamtwn, discussed here by Poppe and Reck, who study it alongside the Early Modern Irish Stair Bibuis; it was also the original of the Icelandic Bevers saga, examined in this volume by its editor, Christopher Sanders.17 The paucity of extant manuscripts of Boeve cannot be a true reflection of the story’s popularity in the century after its composition. A much better indicator of its attractiveness is the alacrity with which it was translated and otherwise appropriated. Within a few decades of its composition, most convincingly dated to the last decade of the twelfth century or the first few years of the thirteenth,18 Boeve was translated into Welsh (around the middle of the thirteenth century).19 The Icelandic translation is harder to date, but it too may have originated in the thirteenth century, as Sanders argues. There are two Faroese ballads of Boeve, Bevusar tættir and Bevusar ríma, based on material from the first few chapters of Bevers saga.20 Three Continental French verse redactions, all from the thirteenth century, expanded the briskly compact Anglo-Norman narrative to between ten and twenty thousand lines.21 The story was then recast in French prose in the early fifteenth century, in a redaction that subsequently went through several printings.22 From north-eastern France, the Bevis material eventually made its way to the Netherlands, where a verse redaction, of which little remains, preceded a much better-documented prose adaptation, printed texts of which begin to appear in the first decade of the sixteenth century.23 It was, however, the story’s southward move to Italy that was crucial to its subsequent dissemination as far as Russia and to its enduring appeal. The various stages in its journey were marked by narrative transformations, sometimes of a very radical nature. In Italy we encounter a multiplicity of versions composed in different dialects, shaped into different prosodic forms, and performing different functions in the broader context of the Italian narrative

17 18 19 20

21 22 23

Bevers saga, ed. by Christopher Sanders, Stofnun Árna Magnússonar á Íslandi 51 (Reykjavík: Stofnun Árna Magnússonar á Íslandi, 2001). See Judith Weiss, ‘The date of the Anglo-Norman Boeve de Haumtone’, Medium Ævum, 55 (1986), 237–41. See p. 37 n. 3 below. On the Faroese ballads, see ibid., pp. cxxxv–cxxxvi; ‘Bevusar tættir’, ed. by N. Djurhuus, in Fóroya kvæði, Corpus Carminum Færoensium 5 (Copenhagen: Akademisk Forlag, 1968), pp. 309–12; Michael Chesnutt, ‘Bevussrímur and Bevusar tættir: a case study of Icelandic influence on Faroese balladry’, in Opuscula, vol. XII, ed. by Britta Olrik Frederiksen, Bibliotheca Arnamagnæana 44 (Copenhagen: Reitzels, 2005), pp. 399–437. Der festländische Bueve de Hantone, ed. by Albert Stimming, 5 vols, Gesellschaft für romanische Literatur 25, 30, 34, 41, 42 (Dresden: Niemeyer, 1911–20). Marie-Madeleine Ival, Beufves de Hantonne: Version en prose (Aix-en-Provence: CUERMA; Marseilles: Lafitte, 1984). See A Dictionary of Medieval Heroes: Characters in Medieval Narrative Traditions and Their Afterlife in Literature, Theatre and the Visual Arts, ed. by Willem P. Gerritsen and Anthony G. van Melle, trans. by Tanis Guest (Woodbridge: Boydell Press, 1998), p. 63.





Introduction

tradition. Franco-Venetian cantari of the fourteenth century were followed by several Tuscan versions in ottava rima, as well as versions in prose. A 1497 printing of a verse tale of Buovo (or Bovo) d’Antona, as the hero is usually called in Italian, was the source of Elia Levita Bachur’s translation into Yiddish, Bovo-Buch, composed in 1507 and first published in Venice in 1541.24 Bovo-Buch was one of the most popular narratives in Yiddish secular literature for five hundred years; it was translated into Romanian as late as 1881. In Italy itself, an important fourteenth-century Franco-Italian redaction, which formed part of a compilation attempting to merge different stories from the Carolingian cycle into a unified whole, served as the basis of Andrea da Barberino’s lengthy prose romance I reali di Francia, written around the turn of the fifteenth century, first printed in 1491, and endlessly reprinted throughout the sixteenth century.25 Readers familiar with the Anglo-Norman and Middle English versions might not immediately recognize the muchamplified story or some of the characters. Sabot is here called Sinibaldo, though his son is still recognizable as Teris or Terigi, just as the hero’s father remains recognizable under his lightly Italianized name, Guido. The boy’s mother, who was nameless in the Insular texts, is here called Brandoria and is still an odd combination of the malmariée and the wicked stepmother, arranging to have her husband killed by her lover, Duodo di Maganza, and plotting the death of her son too. As in the Insular versions, merchants sell the young Buovo to King Erminione of Erminia. In due course the hero acquires a horse, Rondello, and attracts the attention of the king’s daughter, Drusiana. Adventures and characters proliferate, diverging more and more from earlier versions of the story and yet preserving its overall shape. Names change, sometimes beyond recognition, but Yvori is still identifiable in Marcabruno, and Bradmund in Lucafero. Escopart, no longer a giant, has become Pulicane, half-man, half-dog, product of the unnatural union of a woman and a mastiff. In spite of all these, and many other, substantial changes, the principal episodes of the original narrative remain in place until the latter part of the text, when Buovo engages in military campaigns in Central and SouthEastern Europe (rather than in the Middle East), before dying at the hands of his half-brother Gailone. It is in this form that the story attained its greatest popularity and, from the sixteenth century, spread eastwards by way of Venice. In 1549, twenty copies of the Italian Buovo d’Antona were shipped from Venice to Ragusa.26 With its sizeable population of educated Slavs fluent in Italian, this city-state on the Adriatic coast (present-day Dubrovnik, Croatia), was ideally placed to act as a conduit for Western cultural goods; thus when the story surfaced in

24 25

Elia Levita Bachur, Bovo-Buch, trans. by Jerry C. Smith (Tucson, AR: Fenestra, 2003). Andrea da Barberino, I reali di Francia, ed. by Giuseppe Vandelli and Giovanni Gambarin (Bari: Laterza, 1947). 26 Veselin Kostić, Kulturne veze izmedju jugoslovenskih zemalja i Engleske do 1700. godine (Belgrade: SANU, 1972), pp. 359–60.





Sir Bevis of Hampton in Literary Tradition

a Byelorussian manuscript (now in Poznań, Poland) a few decades later, it was in a form that showed clearly both its derivation from an Italian version current in Venice and traces of its passage through Balkan lands.27 The quickly multiplying Russian versions can all be traced back to the Byelorussian text represented by the Poznań manuscript. It is in Russia that Bevis, now called Bova, reached the apex of his social ascent. Whereas in the West the story was almost certainly never read at royal courts, in 1693 an illustrated copy of a Russian redaction was among the books enjoyed by Peter the Great’s young son Alexis. The tsarevich must have made good use of the book, ‘many pages of which were torn out and spoiled’, as was recorded at the time.28 In Russia the story appealed to all social strata and was transmitted orally, in manuscript form, in chapbooks and broadsheets, well into the nineteenth and, in the case of oral folk narratives, the twentieth century. In Italy too there is evidence of continuing interest in the story beyond the sixteenth century. In the 1750s, the Italian composer Tommaso Traetta ­capitalized on the popularity of Buovo d’Antona and, armed with a libretto by Carlo Goldoni, turned a romantic episode from the story into a dramma giocoso or comic opera in three acts, first performed in Venice in 1758.29 The comedy must have derived part of its appeal from a radical reversal of the audience’s expectations. To the extent that they were familiar with Buovo’s life and adventures, listeners would have expected his courtship of the Erminian princess Drusiana to result in their marriage, as in the narrative tradition; instead, Drusiana is tricked into marrying Buovo’s rival ­Maccabruno, while the hero, after some aristocratic misgivings, finds happiness with ­Menichina, a miller’s daughter. In some parts of Italy, just as in Russia, Buovo’s exploits attracted and entertained popular audiences as recently as the twentieth century. This was the case in Sicily, where the story became part of the repertoire of storytellers (cuntastorie) and puppeteers (pupari), who helped preserve the name and fame of Buovo, his father and his twin sons. Sephardic ballads, ultimately derived from the Continental French Beuve, were sung on Rhodes in the eighteenth and nineteenth centuries.30 A seventeenth-century Russian copy of the story ends with the assurance that ‘Bova’s fame shall endure henceforth unto all generations’. The words were prophetic, at least for a century or two. The fame of Bova/Bevis/Boeve/ Buovo/Beuve/Bovo no longer endures, but few other medieval heroes can boast of such a long, distinguished, and socially and geographically wideranging career.

27

For an in-depth study of the Russian versions of the story, see V. D. Kuzmina, Rytsarskii roman na Rusi (Moscow: Nauka, 1964), pp. 17–132, 245–64. 28 History of Russian Literature, 11th–17th Centuries, ed. by Dimitry Likhachev, trans. by Kathleen Cook-Horujy (Moscow: Raduga, 1989), p. 472. 29 A recording made in 1994 and conducted by Alan Curtis, with the tenor Howard Crook in the title role, is available on the Opus 111 label. 30 A Dictionary of Medieval Heroes, ed. Gerritsen and van Melle, pp. 63–4.





Introduction

* The scope of this volume is limited to the Anglo-Norman Boeve de ­Haumtone, the Middle English Sir Bevis of Hampton, and their direct descendants. In common with most popular romances and their chanson de geste sources, Boeve/Bevis received very little serious scholarly or critical attention until twenty or thirty years ago. At the end of the nineteenth century and the beginning of the twentieth, attention was focused primarily on sources and analogues and on the interrelationships of the various European versions of the story; 31 the approach in such scholarship was literary-historical rather than critical. The boom in medieval translation studies in the past quartercentury or so has, however, generated increased interest in processes of linguistic and cultural translation in relation to ‘popular’ texts; along with this has come a belated recognition of the literary qualities of such works and a growing appreciation of their poetics. At the same time, Anglo-Norman litera­ ture has become more and more a subject of serious study – with particular emphasis on its social, political and historical contexts. With the burgeoning of codicological studies and the production of Scolar Press facsimiles of two important romance collections in the 1970s,32 the Middle English Bevis came to be considered in its manuscript context, in its thematic relation to other texts (particularly romances) that appear in the same anthologies. Foremost among collections that have been studied in this way is the Auchinleck MS, which has attracted the attention of, among others, postcolonialist and feminist critics – the primary focus being on issues of national, religious and gender identity as they are reflected in this manuscript.33 Book history has also been a growth area in recent years, as has the literary-historical study of the post-medieval reception and influence of Middle English texts.34 All these approaches are reflected in the present volume. Marianne Ailes discusses the genre and technique of the Anglo-Norman Boeve, relating its stylistic features to the rhetoric of the schools. Erich Poppe and Regine Reck, Christopher Sanders, and Ivana Djordjević all deal with issues of translation and cultural transfer – from Anglo-Norman to Welsh, Icelandic and Middle English, and from Middle English to Irish. Judith Weiss examines the charac­ter of the hero’s mestre, Sabaot, in Boeve in relation to social practices in Anglo31

Cf., e.g., Christian Boje, Über den altfranzösischen Roman von Beuve de Hamtone, Beihefte zur Zeitschrift für romanische Philologie 19 (Halle: Niemeyer, 1909); John E. Matzke, ‘The oldest form of the Beves legend’, Modern Philology, 10 (1912/13), 19–54. 32 See n. 3 above. 33 E.g. Thorlac Turville-Petre, England the Nation: Language, Literature, and National Identity, 1290–1340 (Oxford: Clarendon Press, 1996), passim; Siobhain Bly Calkin, Saracens and the Making of English Identity: The Auchinleck Manuscript, Studies in Medieval History and Culture (London; New York: Routledge, 2005). Felicity Riddy, in an unpublished paper presented at the third Romance in Medieval England conference (Bristol, 1992), discussed the Auchinleck MS as a ‘women’s manuscript’. 34 See, e.g., Andrew King, ‘The Faerie Queene’ and Middle English Romance: The Matter of Just Memory, Oxford English Monographs (Oxford: Clarendon Press, 2000); Helen Cooper, The English Romance in Time: Transforming Motifs from Geoffrey of Monmouth to the Death of Shakespeare (Oxford; New York: Oxford University Press, 2004).





Sir Bevis of Hampton in Literary Tradition

Norman England and shows how it contributes to the construction of English identity within the poem. Issues of identity – with respect to nationality, religion and gender – are also the subject of the chapters by Robert Allen Rouse, Siobhain Bly Calkin, Melissa Furrow and Corinne Saunders, each of these treating a particular theme (Bevis as a distinctively English ‘kniȝt of cristene lawe’) or characters (Ascopard, Josian) in the Middle English Bevis and demonstrating how they complement and help define the character of the protagonist. Jennifer Fellows addresses the complex textual history of the English Bevis in the Middle English and Renaissance periods; while Andrew King describes the nature of the English romance’s influence on the works of such post-medieval writers as Spenser, Richard Johnson, Drayton and Bunyan. The genesis of this volume was in a one-day colloquium organized by Dr Mishtooni Bose in Southampton in May 2004. We are very grateful to her for her interest in our hero and for bringing together, for the first time, Boeve/ Bevis scholars from diverse linguistic disciplines and thus sowing the seeds of fruitful interaction and collaboration; we hope that she approves of the fruits that these have borne. Bevis studies have undergone a modest resurgence in the past few years; perhaps the present collection of essays will stimulate further interest in this long neglected, but once hugely popular, romance.



1 The Anglo-Norman Boeve de Haumtone as a chanson de geste MARIANNE AILES

T

he enigma that is Boeve de Haumtone is summed up in M. Dominica Legge’s seminal study of Anglo-Norman literature, where she discusses the text under the chapter heading ‘Ancestral Romance’, and describes it as belonging to ‘the class labelled romance … cast in the form of a chanson de geste’. As form is a major generic marker, her ambivalence over the nature of the text invites examination. In this chapter, I shall examine the way the poet presents Boeve as a chanson de geste, and the use of both chanson de geste discourse, based on the laisse, and more scholarly rhetoric, such as chiasmus and annominatio, as a way of approaching this question of genre. Legge’s term ‘ancestral romance’ has not gone unquestioned. Susan Crane, in her examination of Insular literature, debunks the theory of ‘ancestral romance’ linked to individual patrons but continues to classify Boeve and the other texts that have English heroes as ‘romances’. The catalogue of AngloNorman texts and manuscripts compiled by Ruth J. Dean and Maureen B. M. Boulton does not endorse the concept of ancestral romance, though the texts usually listed under this category are there grouped together. The authors describe Boeve as a romance, but add that ‘there are also three continental versions usually considered chansons de geste’. François Suard describes Boeve as a ‘chanson d’errance’, but he is concerned more with the later, Continental versions of the tale. My concern is with the Anglo-Norman



M. Dominica Legge, Anglo-Norman Literature and Its Background (Oxford: Clarendon Press, 1963), p. 156. All references to the text of Boeve are to Der anglonormannische Boeve de ­Haumtone, ed. by Albert Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974) (hereafter Boeve).  Susan Crane, Insular Romance: Politics, Faith, and Culture in Anglo-Norman and Middle English Literature (Berkeley, CA: University of California Press, 1986), pp. 13, 16–18; also Susan Dannenbaum, ‘Anglo-Norman romances of English heroes: “ancestral romance”?’, Romance Philology, 35 (1981/2), 601–8.  Ruth J. Dean and Maureen B. M. Boulton, Anglo-Norman Literature: A Guide to Texts and Manuscripts, ANTS Occasional Publications (London: ANTS, 2000), p. 89.  ‘Le Beuves de Hantonne en prose: importance et expression du sentiment amoureux’, in François Suard, Chanson de geste et tradition épique en France au Moyen Âge (Caen: Paradigme, 1994), pp. 399–414 (p. 399).





Marianne Ailes

text, generally considered the oldest extant version; I am accepting Dean and Boulton’s description of this as thirteenth-century. We can certainly recognize generic similarities between the romances of Horn, Haveloc, Gui de Warewic, Fouke le Fitz Waryn, Waldef and Boeve – particularly in terms of narrative pattern, one Crane characterizes as ‘a pattern of dispossession and reinstatement, the hero regaining through his admixture of courage and legal knowledge a rightful inheritance wrongly seized from him’. A pattern of dispossession and reinstatement is not unique to the Insular texts, but in the Continental chanson de geste the regaining of a rightful inheritance would more probably be achieved by physical re­conquest. The difficulty in attempting to classify Boeve in relation to chanson de geste is brought out by Crane, who considers that all these stories of English heroes, dealing with external, political forces, ‘may seem close to epic’. The crucial difference for Crane is that each of these Insular heroes is essentially self-centred, not entirely ‘representative of his community, bent on winning its survival even at the expense of his own life’, as might be expected of an epic hero. A problem here is our imprecise use of terms such as ‘epic’ and ‘romance’. Instead of imposing modern generic classifications on Boeve, we need to consider how the text presents itself, what horizons of expectation are set up in the audience/readership and how we are invited to read the text. Let us begin at the beginning, for the prologue establishes the expectations of listeners/readers. We are immediately invited to enter the world of the chanson de geste: i Seingnurs barons, ore entendez a mei, si ws dirrai gestes, que jeo diverses sai, de Boefs de Haumtone, li chevaler curtays,



On the dating of Boeve, cf. p. 25 n. 1 below. The text was preserved in two manuscripts, which form the basis of Stimming’s edition. These are the fourteenth-century Paris, Bibliothèque Nationale, MS nouv. acq. fr. 4532 (B) and the thirteenth-century MS Firmin Didot (D), which perished in World War II (see Judith Weiss, ‘The Anglo-Norman Boeve de Haumtone: a fragment of a new manuscript’, Modern Language Review, 95 (2000), 305–10). Both were incomplete, but they were complementary: the former contained lines 1–1268 of Stimming’s edition, and the latter lines 913–3850. A further 62-line fragment, corresponding to lines 1003–65 of the printed edition, was recently discovered on a pastedown in the Hunterian Library of the University of Glasgow and has been transcribed by Weiss (ibid.). The overlapping portion of the poem shows that while the three manuscripts differed in the quality of the copyists’ work, linguistic features and relatively minor details, they all seem to have represented the same version of the narrative. Dean and Boulton, Anglo-Norman Literature, p. 89, mention some fragments of Boeve in binding fragments in London, Lambeth Palace, MS 1237, nos. 1, 2, dating them to the second half of the thirteenth century.  Crane, Insular Romance, p. 18.  As in Raoul de Cambrai, ed. by Sarah Kay (Oxford: Clarendon Press, 1991). Cf. Jean-Louis Picherit, ‘L’évolution de quelques thèmes épiques, la dépossession, l’exhérédation et la reconquête du fief’, Olifant, 11 (1986), 115–28.  Crane, Insular Romance, p. 14.

10



Boeve de Haumtone as a chanson de geste ke par coup de espeie conquist tant bons reys. Si vus volez oyer, jeo vus en dirrai; unkes ne oistes meyllur, si com jeo crai. ii Seingnurs, si de lui oyer desirez, jeo vus en dirrai, kar jeo sai asez; primes vus en dirrai de soun parentez. A Haumtone fu li quens plein de bontez, il out a noun Guioun, chevaler fu prisez; meilour de lui ne fust en son tens trovez. iii Seignurs, iceo quens Guioun dount vus chaunt estoit bon chavaler, pruz e combataunt; mes de une chose lui alout home blamaunt, k’ainz ne vout femme prendre en tot son vivaunt, dunt pus se repenti par le men ascient.  (1–17)

(My lords, now pay attention to me. I shall tell you tales – I know different ones – about Boeve of Hampton, the courteous knight, who defeated so many good kings by dint of his sword. If you wish to hear, I will tell you; I believe you never heard anything better. / My lords, if you wish to hear about him, I will tell you, for I know much. First I will tell you of his parentage. The good count, named Gui, lived at Hampton. He was a renowned knight; in his time there was no one better than him. / My lords, this count Gui of whom I speak was a good knight, valiant and brave; but there was one thing for which men blamed him, that he had refused ever to take a wife, which I believe he later regretted.)

We are invited here to listen, and specifically to listen to tales of deeds (gestes, line 2). Jean Rychner, in his magisterial analysis of the ‘art’ of the jongleur in the chanson de geste, points out that of the nine poems that form the core of his study, six ‘s’ouvrent par un prologue, qui nous montre le jongleur annonçant sa chanson, vantant sa marchandise devant son auditoire’.10 Tony Hunt, in his analysis of the prologue in Chrétien de Troyes’s Yvain, does not distinguish between different genres, but he does give examples of different ways of obtaining the benevolentia of the audience; the chanson de geste figures largely among those prologues that praise the protagonists. The genre



All translations of Boeve are quoted from ‘Boeve de Haumtone’ and ‘Gui de Warewic’: Two Anglo-Norman Romances, trans. by Judith Weiss, French of England Translation Series (Tempe, AZ: Medieval and Renaissance Texts and Studies, 2008). 10 Jean Rychner, La Chanson de geste: Essai sur l’art épique des jongleurs (Geneva: Droz, 1955), p. 10.

11



Marianne Ailes

is also well represented in the list of prologues that summarize the content of the text that is to follow.11 Jean-Pierre Martin divides the prologues of the chanson de geste into four sections, each corresponding to a different function: namely, the announcement of the chanson, a statement of the origins of the chanson, a summary of the narrative, and the beginning of the narrative. His first section, the announcement of the chanson (‘song’), is essentially the captatio benevolentiae.12 Our poet reinforces his opening address to the audience by announcing his song, using both forms identified by Martin: the imperative (line 1) and the interrogative (lines 6, 8). The origins of the song are not given in Boeve; nor does the poet go on to give a summary of the narrative; he does, however, prepare the narrative by introducing the protagonist and telling of his conquest of many kings ‘by the blows of his sword’, before moving on to Martin’s fourth element, the beginning of the narrative. Of course Martin’s schema is something of an abstraction, and many texts do not fit exactly – particularly when, as he says, parallelism across laisses is involved. In the case of Boeve there is also a considerable amount of what Martin calls ‘le jeu des laisses’, of reprise over the three laisses of the prologue.13 In all this our poet sets his text up following the established pattern of the chanson de geste. Rychner’s examples frequently include the word chanson in their prologues.14 Our poet is more ambivalent about the nature of his text; he will tell his tale (dirrai, lines 1, 5, 8) but a few lines later we hear that he will sing about Count Gui (‘quens Guioun dount vus chaunt’, line 13). His closing lines are equally ambivalent: ‘Nostre chançon finist, ne dure plus avant; / jeo ne vus dirrai plus en dist ne en chant’ (‘Our song is ended, it lasts no longer; I will tell you no more in words or in song’, lines 3845–6). He does contrive to include in both opening and closing sections of the poem the word geste (lines 1, 3847) and some reference to his tale as a song (lines 13, 3845–6).15 The poet is thus inviting us to listen to his text as a chanson de geste. While we are dealing with a poem that has undoubtedly been composed as a written text, it is also clearly marked for oral dissemination. Thus in the opening laisses we find the verbs entendre (line 1), oïr (lines 5, 6) and dire (lines 2, 5), with oïr and dire picked up again in laisse ii (lines 7, 8, 9), 11

12 13 14

15

Tony Hunt, ‘The rhetorical background to the Arthurian prologue’, Forum for Modern Language Studies, 6 (1970), 1–23 (p. 3). Other prologues are described as attracting the benevolentia of the audience by praising the poet, denigrating rivals or praising the narrative. Jean-Pierre Martin, ‘Sur les prologues des chansons de geste: structures rhétoriques et fonctions discursives’, Moyen Âge, 93 (1987), 185–201 (p. 186). Ibid., p. 197. Thus the Chanson de Guillaume has ‘oimas commence la chancun d’Willame’ (line 1); the Couronnement Louis is described as ‘une chancun bien faite et evenante’ (line 2) while the Charroi de Nîmes (line 3), the Prise d’Orange (line 3) and the Moniage Guillaume (line 1) are each a ‘bone chanson’ and Raoul de Cambrai is a ‘chancon de joie et de bauder’ (line 1): cf. Rychner, La Chanson de geste, esp. pp. 68–74. Cf. Legge, Anglo-Norman Literature, p. 156.

12



Boeve de Haumtone as a chanson de geste

and chanter added in laisse iii. The narrating persona of the jongleur, who addresses an assumed audience, is also clearly present, in what Suard, in his introductory study of the genre, describes as a dialogue between the public and the jongleur.16 While all this is typical of the chanson de geste genre, more unusual is the splitting of the prologue over several laisses, with the assumed audience apostrophized anaphorically at the beginnings of consecutive laisses (lines 1, 7, 13). The reprise of the end of one laisse at the beginning of the next is a characteristic of chanson de geste discourse; here the reference to Gui in line 11 is picked up again in line 13. Less characteristic of the genre is the chiastic structure of the reprise between laisses i and ii: Si vus volez oyer, jeo vus en dirrai … Seingnurs, si de lui oyer desirez, jeo vus en dirrai …  (5, 7–8) (If you wish to hear, I will tell you … / My lords, if you wish to hear about him, I will tell you …)

Here line 7 inverts the order of the first hemistich of line 5, and the whole of line 5 is then divided over two lines. The division of the prologue over several laisses thus not only sets the pattern of laisse structure and length, but also gives an opportunity to use the chiastic reprise that is characteristic of the text. The anticipation of disaster in line 17 recalls the Chanson de Roland, that most famous of chansons de geste, in which we are told that the Saracen Marsile will face disaster.17 In many ways, then, the prologue sets up the expectation that we are about to hear a chanson de geste, though the rhetorical structuring of this prologue prepares us for a chanson de geste with a rather different kind of discourse – that is, one that uses chanson de geste forms combined with rhetorical structures. When Suard asks ‘Qu’est-ce qu’une chanson de geste?’,18 he begins, like Rychner, with the poem as a song. He then goes on to say that the chanson de geste has distinctive formal characteristics: ‘D’un bout à l’autre du Moyen Age, une chanson de geste se reconnaît à des critères formels identiques: une succession de strophes de longueur inégale, bâties sur une seule assonance ou une seule rime.’19 16

François Suard, La Chanson de geste, 2nd edn, Que sais-je? (Paris: Presses Universitaires de France, 1993), p. 19; Martin, ‘Sur les prologues’, p. 185, writes of the tradition tolerating a voice in such extra-diegetic elements as the prologue, part of discours rather than narrative, the narrator’s voice being unobtrusive in the narrative itself. 17 La Chanson de Roland, ed. by Frederick Whitehead, rev. by T. D. Hemming, French Texts (London: Bristol Classical Press, 1993), laisse i; for a translation, see The Song of Roland, trans. by Glyn Burgess (Harmondsworth: Penguin, 1990). 18 Suard, La Chanson de geste, p. 7. 19 Ibid., p. 9; Edward A. Heinemann, L’Art métrique de la chanson de geste: Essai sur la musicalité du récit, Publications romanes et françaises 205 (Geneva: Droz, 1993), p. 139, writes of the laisse as a constant of the chanson de geste.

13



Marianne Ailes

The first generic marker for the chanson de geste is the use of the laisse form. Boeve uses as its basic form the traditional laisse: that is, as Suard defines it, strophes of varying length united by rhyme or assonance. The length of the laisse in Boeve has attracted some attention, mostly from Middle English scholars, as there appears to be a link between the laisse and the different forms of versification used in most versions of the Middle English Bevis.20 It has been noted that there is a marked change in laisse length after laisse 66 (line 415). Up to that point the laisses are very short, on average 6.28 lines; Middle English scholars have considered this to resemble the six-line tail-rhyme strophe.21 The variation in length is, however, consider­able, from five to eight lines, though it does suggest that the Insular poet is comfortable with short strophes and may be familiar with Middle English as well as French forms. The remainder of the poem has an average laisse length of 24.7 lines (3435 lines over 139 laisses), still relatively short. In the poem taken as a whole, the average laisse length is only 18.8 lines (3850 lines over 205 laisses). This is quite surprising, as many Continental chansons de geste of the thirteenth century have very long laisses, apparently continuing until the poet needs to change his rhyme. Such long laisses, described as ‘favouring the narrative at the expense of the lyric dimension’,22 in such texts as, for example, Maugis d’Aigremont, seem to have little relevance to the structure of the narrative. Other chansons de geste, and in particular early texts such as the Chanson de Roland and the Chanson de Guillaume, have much shorter laisses. Normally in these texts the laisse is used to express a single idea.23 Many of the laisses in Boeve work in this way, particularly the short laisses at the beginning of the text. Our poet, it could be argued, is emphasizing the importance of the laisse as a unit and a poetic form by using a shorter laisse length at the beginning of his text. He has thus firmly established our expectations about the text. Other laisses are structured differently, with a complete change of focus in the middle, a change highlighted by the use of a rhetorical transitio. The vast majority of authorial interventions in Boeve are, in fact, transitional phrases, used as a structuring device. Such examples of transitio, a not particularly subtle example of rhetoric, are almost always used mid-laisse. This is an 20

Albert C. Baugh, ‘Improvisation in the Middle English romance’, Proceedings of the American Philosophical Society, 103 (1959), 418–54 (esp. p. 431); Ivana Djordjević, ‘Versification and translation in Sir Bevis of Hampton’, Medium Ævum, 74 (2005), 41–59 (p. 42). 21 Baugh, ‘Improvisation’, p. 431; Djordjević, ‘Versification and translation’, pp. 53–4 n. 9. 22 Catherine Jones, ‘Dispersed parallelism in Hervis de Mes’, Olifant, 13 (1988), 29–40 (p. 29). Where a laisse can extend to 1400 lines, as in Hervis de Metz, it is difficult to see what its function is; Jones argues that a different kind of parallelism is being used in this text. 23 Rychner, La Chanson de geste, pp. 68–74, writes of the unity of the laisse, but this really only holds true for some chansons de geste; Rychner writes in particular about the unity of the laisse in the Chanson de Roland, an atypical text in the tightness of its laisse structure; Heinemann, L’Art métrique, p. 142, emphasizes the variety in laisse structure. Cf., e.g., the Couronnement Louis, analysed in Philip E. Bennett, ‘Poetic structures in the Couronnement de Louis’, in Litera et Sensus: Essays on Form and Meaning in Medieval French Literature Presented to John Fox, ed. by D. A. Trotter (Exeter: Exeter University Press, 1989), pp. 17–30.

14



Boeve de Haumtone as a chanson de geste

indicator of how flexibly the Boeve poet is using the laisse structure. Thus the poet begins, for example, laisse xxxiii with a transitional phrase: ‘Ore dirrum de le fiz au counte Guioun’ (‘Now we shall tell of count Gui’s son’, line 207). Yet elsewhere such phrases will come in the middle of what may be a long laisse. Laisse clxxxiii is divided with a transitio part-way through: ‘Ore lerrom de Hermin ester’ (‘Now we shall leave Hermin’, line 3120).24 It seems that the poet is adapting the laisse – a feature of the oral, performative tradition of the chanson de geste – to a written format by incorporating, as a structuring element, signposts, not uncommon in the chanson de geste, which would be equally helpful in a performing context. The varied placing of these devices mid-laisse or at the beginning of the laisse draws attention to a laisse structure that is very different from the structure of a Chanson de Roland, in that it may lead to a climax at the end of the laisse rather than in the middle. There are even several cases where a piece of direct discourse is broken up by a laisse change.25 This careful highlighting of the structure suggests a deliberate exploitation of one of the models offered by the tradition. The enchaînement between laisses is an important feature of chanson de geste discourse.26 In Boeve we find a particular development of the laisse structure: chiasmus, similar to that linking the opening laisses, is used to link laisses throughout the poem. Thus, the reprise of the end of laisse xxxvii at the beginning of laisse xxxviii uses this ‘a b b a’ structure; Sabot has been ordered to kill the child Boeve and prepares to deceive Boeve’s mother: si ensenglenta le dras a le enfaunt. xxxviii Quant Sabot out fet les dras ensenglenter …  (236–7) (and bloodied the child’s clothes. / When Sabot had bloodied the clothes …)

On occasion the chiasmus is spread over several lines and may be combined with a reprise, as at the end of laisse lxxxii and the beginning of laisse lxxxiii: ‘Ma bele damoisele,’ dist Boefs a vis fer, ‘iceste fol amour pur dieu lessez ester, kar li roi me freit honir e vergunder.

24 25 26

Cf., e.g., lines 2790, 3120, 3409, 3331. Cf., e.g., lines 750, 960. Cf. Rychner, La Chanson de geste, pp. 74–93; Edward A. Heinemann, ‘On the metric artistry of the chanson de geste’, Olifant, 16 (1991), 5–59, and L’Art métrique, pp. 205–20; Suard, La Chanson de geste, p. 29.

15



Marianne Ailes lxxxiii  Ma bele damoisele,’ ceo li dist Bovoun,

‘pur dieu lessez ester ceste grant folesoun …’  (677–81) (‘My fair lady’, said fierce-faced Boeve, ‘for God’s sake give up this foolish love, for the king will have me shamed and disgraced.’ / ‘My fair lady’, said Boeve, ‘for God’s sake give up this great folly …’)

Here line 680 picks up 677, while 681 reverses the hemistich order of 678, forming a chiastic reprise in a text clearly marked as intended for oral dissemination. Sometimes the chiasmus is incomplete, with perhaps the second hemistich of one line picked up in the first hemistich of the second: ignelement se sont armé. cxcvi Ore sont armé sur chivals de pris …  (3581–2) (Quickly they armed themselves. / Now, armed and on excellent horses …)

Chiasmus is also used elsewhere in the text, in ways that have no impact on the laisse structure but can suggest formulaic repetition, characteristic of the chanson de geste genre, which frequently requires alteration in wordorder to accommodate changes in rhyme or assonance. While in these cases the parallels are used primarily as a structuring device, linking the different laisses, elsewhere parallels seem to invite the reader/listener to make links that are not made explicitly in the text. Thus links are made between the belle Sarrasine Josiane (Boeve’s beloved), and Boeve’s horse, Arundel. Josiane has a magic girdle which will prevent any man from sleeping with her: ‘il n’i avereit homme en secle vivant / ki de cocher ove li avereit accun talent’ (‘there was not a man in the world who would have any desire to sleep with her’, lines 1003–4). Equally no man dares groom the horse: ‘il n’i out homme en secle ke le osast tocher’ (‘there was no person in the world who dared touch it’, line 1014). A later chiasmus is used to link laisses separated in the text by two other laisses: ‘Josian out mult dolent le qer … / Kant Josiane le veit, mult ot le qer dolent’ (‘Josiane’s heart was desolate … / When Josiane saw him, she was sick at heart’, lines 1423, 1495). In this particular case the parallelism itself is more significant than the chiasmus that reinforces it; I would describe this as chiastic, although the elements are separated, since it is part of a technique of parallelism. These laisses enclose a moving scene in which Boeve, having arrived at Josiane’s husband’s court disguised as a palmer, has been identified because the horse allows him to mount. Josiane, like other belles Sarrasines, does not hesitate to express her wishes. Boeve initially rejects her – in part 16



Boeve de Haumtone as a chanson de geste

because she has been married for seven years and he has been told he must marry only a virgin; Josiane then explains that she has remained a virgin. Her continued sorrow, now provoked by her husband Yvori’s imminent return from hunting, is emphasized by the parallelism between line 1423, at the beginning of the episode, and line 1495 at the end. Josiane is the subject of another effective chiasmus, one not connected to the laisse structure. She has been forced into marriage, and a whole series of rhetorical devices emphasizes that she had no choice: Ore vus dirrai de Miles l’adverser ke fist Josian mal gre le sun esposer. Mal gre le sun la mena a muster, mal gre le sun la fist la nuit cocher.  (2099–102) (Now I shall tell you about the fiendish Miles, who had married Josiane against her will. Against her will he led her to church, against her will he took her to bed at night.)

The address to the audience and the transitio of line 2099 are followed by semi-chiastic repetition, as a phrase from the second hemistich of line 2100 becomes the first hemistich of line 2101; this is then followed by the anaphora of lines 2101–2. Parallels and reprises are common in the chanson de geste, part of what we expect in chanson de geste discourse; but the Boeve poet uses chiastic structure in his repetitions, in a way that combines chanson de geste discourse with the rhetoric of the schools.27 Chiasmus is found occasionally in the chanson de geste and may be a result of nothing more than the need to change the rhyme, but nowhere have I seen it used more frequently than in Boeve. On another occasion, the kind of repetition and parallelism we expect in a chanson de geste is used in a more sophisticated way, to emphasize a change of perspective, as words spoken by the narrative voice, ‘L’endemain fu pentecoste en esté’ (‘The next day was Pentecost, in summer’, line 2461), are then echoed by a character: ‘hui est pentecoste en esté’ (‘today is Pentecost in summer’, line 2472). Other rhetorical devices used in the text include ones common in chanson de geste as well as ones that are more usual in romance texts. Among the latter is the occasional use of annominatio, as in the following examples: ‘Boefs corne un corn’ (line 559), ‘une ceinture / la ceinture / ceinte’ (lines 1000–2), ‘ben forcé / sa force’ (lines 1338–9), ‘e l’ewe fu freyde si li ad refreydez’ (line 1968), ‘le roi unt coroné, / Boves, li ad la corone fermé’ (lines 2465–6), ‘la offerant fet o bon volunté; / o li offerent ses princes alosé’ (lines

27

On the difference between the rhetoric of the schools and the more colloquial rhetoric of the chanson de geste, see Sarah Kay, ‘The nature of rhetoric in the chanson de geste’, Zeitschrift für romanische Philologie, 94 (1978), 305–20.

17



Marianne Ailes

2468–9), ‘sonja … / Le sunge conte …’ (lines 3439–41). The function of these devices seems to be nothing more than emphasis. In some instances it is unclear where the source, if any, of the use of a rhetorical device, lies. We find in Boeve a kind of anti-blazon, a description of a hideous veleyn, Escopart: par desuz un tertre vist un veleyn gesant, ke ben out nof pez de grant: en sa main tint un mace pesant, que dis homes a peine ne portassent, a son geron un bon branc trenchant, entre se deus oyls un pe out de grant, le front out large com croupe de olifant, plu neyr ou la char ke n’est arement, le nez out mesasis e cornus par devant, le jambes out longes e gros ensement, les pez larges e plaz, mult fu lede sergant, plu tost corust ke oysel n’est volant. Kant il parla, il baia si vilement, com ceo fust un vilen mastin abaiant. cxxxii Le veylen estoit mult grant e mult fers, le chivels out longes com come de destrer e les oyls granz com deus saucers e les dens longes com un sengler, la boche grant, mult fu lede bachelor.

cxxxiii E le vilen estoit grant e metailez, le brace out longes e enforcez, les ungles si dures, ben le sachez, ke il n’ad mure en cristientez, se il fust entur un jur, pur veritez, ke le mur n’ust tost acravantez.  (1744–68)

(… saw, reposing on top of a hill, a churl who was certainly nine feet tall. In his hand he held a heavy club, which ten men could hardly carry, and by his side a good sharp sword. The space between his eyes was a foot wide, his forehead was as large as an elephant’s buttocks, his skin was blacker than ink, his nose was misshapen and knobbly in front, his legs too were long and thick, and his feet long and flat. He was a hideous fellow, a faster runner than a bird on the wing. When he spoke, he barked as horribly as if he were a vile baying hound. / The churl was extremely large and very fierce; his hair was as long as a horse’s mane, his eyes as big as two saucers, his teeth as long as a boar’s, his mouth huge – he was a most ugly young man. / And the churl was large and misshapen; his arms were long and strong, his nails so long that I tell you, there’s no wall 18



Boeve de Haumtone as a chanson de geste in Christendom which, truthfully, if he were nearby one day, he would not quickly demolish.)

The first laisse recalls the descriptions in the chansons de geste of hideous giants, of which there are many; the second recalls the animal similes used by Calogrenant to describe the herdsman in Chrétien de Troyes’s Yvain.28 There are elements of accumulatio and anaphora in the description. The whole may have come from the poet’s general knowledge of rhetorical devices, or he may be deliberately combining chanson de geste and romance topoi to produce a description which in total becomes more extreme than either. Given the economy of the narrative, it may seem rather surprising to find such richness of rhetoric, but we must consider the function of that rhetoric. Here again, we find that our poet is working within the chanson de geste tradition: his rhetorical techniques are rarely about amplification and ‘mere’ decoration but, rather, are concerned with the ‘dramatic effect of the action’.29 The reprises and repetitions slow down the narrative momentum, while chiasmus and anaphora are used to heighten dramatic effect. Other forms of internal parallels (what Heinemann calls ‘disjunctive echoes’) invite the reader to make connections within the text.30 In this way the poet successfully combines the chanson de geste discourse with forms of discourse generally found in other narrative genres such as romance and verse chronicle. In Boeve a complex narrative, economically told, unfolds over a mere 3850 lines. The poem opens with the murder of Boeve’s father, Gui, at the instigation of Boeve’s mother, who marries his killer. Boeve himself is sold to a group of Saracen merchants. The Saracen princess Josiane falls in love with him. After various adventures the two lovers escape together and Josiane and the giant Escopart are baptized. Boeve’s stepfather is killed, Boeve’s mother kills herself, and the lovers marry. They are not, however, allowed to live happily ever after. The narrative is re-launched and the couple and their two sons face various adventures. In the end Josiane’s ex-husband, Yvori, is killed and Boeve takes his place as king of Monbrant. Certain narrative elements are commonplaces of the chanson de geste. The most typical is the love-affair between the Christian hero and the Saracen princess, against a background of Christian–Saracen conflict, with all the divided loyalties this entails. Conversion of the heroine is part of this package. Although Saracen giants are common, they are normally killed in the chanson de geste, but conversion itself is another significant element in many texts, and in Boeve we have the conversion not only of Escopart, but also of, among others, the emir de Cordes, linked through formulaic repeti28

Chrétien de Troyes, Yvain (Le Chevalier au lion), ed. by Wendelin Foerster, rev. by T. B. W. Reid, French Classics (Manchester: Manchester University Press, 1967), lines 295–307; for a trans­ lation, see D. D. R. Owen, Arthurian Romances, Everyman Classics (London: Dent, 1987), pp. 284–5. 29 Kay, ‘The nature of rhetoric’, p. 311. 30 Heinemann, L’Art métrique, pp. 237–81, and ‘On the metric artistry’, pp. 32–40.

19



Marianne Ailes

tion to that of the king of Damacle, both declaring that ‘lerrai Tervagant’ (‘I’ll abandon Tervagant’, lines 3631, 3659). In fact, so many want to be baptized that ‘la grant baptisement’ (‘the great baptism’) lasts for four months (lines 3679–80). Like many other chanson de geste Saracen leaders disappointed in their gods, King Brademound of Damascus beats the idol of Mahomet (line 1164), and towards the end of the poem an idol of Tervagant is destroyed (line 3665). While many medieval texts include prayers, the form of the prière du plus grand péril is peculiar to the chanson de geste.31 In Boeve we have some interesting developments. Boeve, in distress, caught between a fast-flowing river and his enemies, prays: ‘A deus!’, fet il, ‘beau rey de parays, ky de la virgine en Bedleem nasquis e en la beneyte croiz mort pur nus suffris e en le sepulcre fustes ensevelis e enfern brisas e’n outas tes amys e a la Madeleyne pardonas ses fous deliz e ore syés al destre tun pere le poestifs e vendras au dreyn jour jugger morz e vifs e solum sa decerte rendras chescun ses meryz, jeo te requer, ay merci, Jesu Crist; meuz eyme ester neyé e en ewe mausmys ke jeo ne seye isci de ceo paens pris.’  (1243–54)

(‘Oh God!’ he said, ‘fair king of Paradise, who was born of the Virgin in Bethlehem, and suffered death on the blessed cross for us, and was buried in the tomb, and harrowed hell, and rescued your friends, and pardoned the Magdalen her wanton pleasures, and now sits at the right hand of your powerful Father, and will come on the last day to judge the quick and the dead and, according to his deserts, will give each one his due, I beg you, Jesus Christ, for mercy: I would rather be in peril and drowned in the river than here seized by the heathen.’) In structure this is a typical prière du plus grand péril: God is addressed; there is a list of miracles attesting God’s power, followed by an appeal for help. However, in most prières du plus grand péril the miracles listed include some from the Old Testament; here the life of Christ, his miraculous birth, death and resurrection, are the focus, all emphasized by what might 31

On the prière du plus grand péril, see E. R. Labande, ‘À propos des prières dans les chansons de geste’, in Recueil de travaux offert à M. Clovis Brunel, 2 vols, Mémoires et documents 12 (Paris: Société de l’école des Chartes, 1955), II, 62–80; Jacques de Caluwé, ‘La prière épique dans les plus anciennes chansons de geste’, Olifant, 4 (1976), 4–20, and ‘L’originalité de quelques prières épiques’, Marche romane, 20:4 (1970), 59–74. The term was first used in Jean Frappier, Les Chansons de geste du cycle de Guillaume d’Orange, 2nd edn, vol. II (Paris: Société d’Édition d’Enseignement Supérieur, 1967), pp.132–40.

20



Boeve de Haumtone as a chanson de geste

be considered excessive anaphora. Other prayers in the text also echo the structure of the prière du plus grand péril. Earlier in the text we find Boeve praying in prison: Beau sire dieus, ke me deignastes a fourmer e en la beneite croiz de tun sanc achater, jeo te pri, beau duz sire, de fin quer e enter que tu ne me lessez si longement demurer, ke jeo ne sei fet pendre ou vif escorcher, ou tu me facez de ci toust eschaper.  (1040–5)

(Fair lord God, who deigned to create me and redeem me on the blessed cross with your blood, I beg you, fair sweet lord, sincerely and whole-heartedly, that you do not let me remain here long; let me rather be hanged or flayed alive, or help me soon to escape from here.) And later, more surprisingly, a greeting is based on this same structure: Deus vus salve, sire roi, e vostre baronez, ke pur pecheurs fu de la pucele nez, en la terre seynte trente ans penez, quarante jurs juna pur son poeple salver, pus li trai Judas pur sule trente deners, a Juys fu bailé pur son cors tormenter, en la crois pur nus deignoit morer e pus son cors en sepulcre poser, dedens les trois jors fu resuscité, a jur de jugement nus vendra tuz juger; cil salt le roi e tuz le baronez.  (2409–19)

(May God – who for sinners was born of the Virgin, endured thirty years in the Holy Land, fasted for forty days to save his people, then was betrayed by Judas for only thirty pieces, was given to the Jews to torture his body, deigned to die for us on the cross and have his body entombed, was resurrected within three days and will come to judge us all on Judgement Day – save the king and all the barons.) So here we have an adaptation of a topos typical of the chanson de geste, the prière du plus grand péril, its use extending beyond the danger-to-life context in which we would expect to find it. Suard’s description of the chanson de geste includes as a basic element the topos of the epic blow.32 In Boeve we find several such mighty blows, when Boeve seems to deal with his opponents in a very summary manner.33 In one 32 33

Suard, La Chanson de geste, p. 35. Cf., e.g., line 2847, ‘fert un quens, la teste ad coupez’ (‘he struck a count, and cut off his head’), and lines 3235–6.

21



Marianne Ailes

instance the horse is killed and the rider spared, though this is then followed by the mighty epic blow: Boves tret Morgelei, va ferir Yvori, grant coupe li doune sur le helme burni, les pers e les flors Boves ad trenché par mi, par devant l’arçon le chival attendi, jeskes a tere le chival parfendi. Kant le chival le sent, Yvori chai …  (3591–6)

(Boeve drew Morgeley and went to strike Yvori, giving him a great blow on his gleaming helmet; he cut the precious stones and ornaments down the middle, hit the horse in front of the saddle-bow and slit it in two. When this happened to the horse, Yvori fell …) Elsewhere Boeve has only to draw his sword to achieve the submission of his unhorsed opponent.34 Some narrative elements may recall other chansons de geste, though it is not always easy to know whether the allusion is to specific texts or simply to common narrative elements. The hero is sold to Saracen merchants like Vivien in Maugis d’Aigremont; Orable manages to keep herself pure while married to the pagan king in the Enfances Guillaume.35 Like Gui de ­ Bourgogne in Fierabras, or Horn in the Roman de Horn, Boeve initially refuses the Saracen princess’s overtures. As in La Chevalerie Ogier de Danemarche, the hero is blamed for the death of the king’s child.36 Boeve trapped before the flowing water, testing its depth with his lance, recalls Richard de Normendie in ­Fierabras; but, while Richard’s prayer is followed by direct divine intervention, Boeve’s is not.37 Such intertextual allusions strengthen our perception of the text as a chanson de geste. It seems that the values in Boeve are also those of the chanson de geste world. Marriage here seems to be concerned with dynastic alliances as much as with love. Admittedly Boeve himself marries his belle Sarrasine, but he puts regaining his inheritance as a higher priority, postponing his marriage. In the next generation Boeve’s daughter Beatrice is given to the son, Boeve, of 34 35

Cf., e.g., lines 2942–5, 3251–3. Maugis d’Aigremont: Chanson de geste, ed. by Philippe Vernay (Bern: Francke, 1980); on Orable, see Philip E. Bennett, ‘The storming of the Other World, the enamoured Muslim princess and the evolution of the legend of Guillaume d’Orange’, in Guillaume d’Orange and the Chanson de Geste: Essays Presented to Duncan McMillan …, ed. by Wolfgang van Emden and Philip E. Bennett (Reading: University of Reading, 1984), pp. 1–14 (p. 2). As Bennett points out (ibid.), the idea could have come from Chrétien de Troyes’s Cligès. 36 In Ogier, it is actually the hero’s son who causes the death of the king’s son. 37 Fierabras: Chanson de geste du xiie siècle, ed. by Marc Le Person, CFMA (Paris: Champion, 2003) lines 4512–15; Marianne Ailes, ‘Faith in Fierabras’, Charlemagne in the North: Proceedings of the Twelfth International Conference of the Société Rencesvals, Edinburgh, 4th to 11th August 1991, ed. by Philip E. Bennett, Anne Elizabeth Cobby and Graham A. Runnalls ­ (Edinburgh: Société Rencesvals, British Branch, 1993), pp. 125–33 (p. 130).

22



Boeve de Haumtone as a chanson de geste

Tierri (himself the son of Boeve’s mestre), as part of a male–male exchange. Boeve is speaking to his son: ‘Mandom Terri, qu’il nos soit aidant, / si dorrai vostre sour a Boun, son enfant’ (‘Let’s send for Terri to come and help us; I shall give your sister to his son Boeve’, lines 3519–20). Beatrice’s only function is to serve as a gift from one man to another.38 On the other hand, we find early on in the text the romance motif of the malmariée,39 with criticism of the person who gives a young wife to an old man, as in the marriage of Boeve’s parents: ‘Ore oiez, ke il fest graunt pecché / que doune jofne femme a viel homme barbé’ (‘Now listen to how great a crime it is, to give a young woman to a bearded old man!’, lines 120–1). This motif is in fact the trigger for the whole action, as Boeve’s mother plots the murder of her husband. The situation that precedes the marriage and opens the narrative, Gui’s long-term refusal to take a wife, is also one that is more to be expected in courtly texts. Refusal to take a wife or have a lover is generally considered with some suspicion, and sometimes even with hints of homosexuality.40 Thus, from the very beginning of the narrative, the text, shaped and presented to us as a chanson de geste, willingly borrows from the courtly tradition, as do many other, later chansons de geste. If we compare all this briefly with the Continental versions of Boeve, we find some interesting changes.41 It seems likely that the Continental texts are developed from the Anglo-Norman tradition. It is not surprising, then, that they retain many of the features of the Anglo-Norman. There are some cases of anaphora, but most of the chiastic structures are lost. The most obvious difference between the Continental and the Anglo-Norman versions is in the length; all three are much longer than the Anglo-Norman text, with longer laisses on average, as shown in the table below:42 Text Anglo-Norman BH Continental BH I Continental BH II Continental BH III

38 39

40 41

42

Number of lines Number of laisses   3,850 205 16,391 452 10,610 208 19,127 370

Average laisse length 19 36 51 52

On woman as gift in the chanson de geste, see Sarah Kay, The Chansons de Geste in the Age of Romance: Political Fictions (Oxford: Clarendon Press, 1995) pp. 37–48. This is a common topos in romance and other courlty genres such as Marie de France’s Lais; on its use in Boeve, see Ivana Djordjević, ‘Original and translation: Bevis’s mother in Anglo-Norman and Middle English’, in Cultural Encounters in the Romance of Medieval England, ed. by Corinne Saunders, Studies in Medieval Romance (Cambridge: Brewer, 2005), pp. 11–26. E.g., the knight in La Chatelaine de Vergi, or Aeneas in the Roman de Troie. There are three different Continental versions: see Der festländische Bueve de Hantone, ed. by Albert Stimming, 5 vols, Gesellschaft für romanische Literatur 25, 30, 34, 41, 42 (Dresden: Niemeyer, 1911–20). Details of the Continental texts are taken from A. Richard Hartman, ‘Laisse division in two later epics, Aiol and Parise la Duchesse’, Olifant, 12 (1987), 5–27 (pp. 26–7).

23



Marianne Ailes

When Hartman refers to ‘rambling works such as Beuve de Haumtone’, he is referring to the Continental versions,43 which would thus conform to the Continental pattern of progressively longer chansons de geste. In the Continental texts, we find a more conventional laisse structure, with longer laisses and quite frequent use of antistrophe and reprise. A few chiastic reprises are to be found, not always corresponding to chiastic structures in the AngloNorman. There are also a few cases of transitio or apostrophe in the middle of the laisse, but this is no longer the norm, which is now for changes in the narrative to occur at the beginning of a laisse. A more detailed study is needed to confirm this, but it would seem that in the process of reworking the narrative for a Continental audience some of the features we find in the Anglo-Norman text are lost, or at least become less prominent. According to Legge, ‘no trace of an Anglo-Norman chanson de geste exists’.44 I would argue that what we find in Boeve de Haumtone is an AngloNorman development of the genre – an appropriation of the chanson de geste for Insular culture. This conclusion can be reinforced by the fact that we find similar combinations of scholarly rhetoric and chanson de geste discourse in Horn and in the Anglo-Norman version of Fierabras.45 This otherwise rare combination of the rhetoric of the schools and chanson de geste discourse, of the matter of romance with the epic conflict of Christian and Saracen, of unjust lord and vassal, carrying echoes of the epic of revolt, results in a text that is undoubtedly different from early Continental chansons de geste. What is produced is a singular development of the chanson de geste rather than a true hybrid of epic and romance, for the framework and the narrative structure are those of the chanson de geste. Our poet sings the geste of Boeve.

43 44 45

Ibid., p. 8. Legge, Anglo-Norman Literature, p. 156. The Romance of Horn by Thomas, ed. by Mildred K. Pope, 2 vols, Anglo-Norman Texts 9–10, 12–13. Horn is translated in The Birth of Romance: An Anthology, trans. by Judith Weiss, ­Everyman’s Library (London: Dent; Rutland, VT: Tuttle, 1992), pp. 1–120; for an analysis of the rhetorical features of Horn, see ibid., pp.12–14. On the Anglo-Norman Fierabras, see Marianne Ailes, ‘Fierebras: an Anglo-Norman chanson de geste’, forthcoming in the Actes of the Société Rencesvals International Congress, Storrs, 2006.

24

2 Mestre and Son: The Role of Sabaoth and Terri in Boeve de Haumtone JUDITH WEISS

I

n the Anglo-Norman Boeve de Haumtone (late twelfth-century), the young hero has a mestre, Sabot or Sabaoth, who plays an important part in the narrative. He and later his son save Boeve from death and from his enemies. Mestre is a term which in the twelfth century replaces pedagogus, or nutricius, as the term for the person employed by noble households to educate their sons, and Insular texts before Boeve show knowledge of the word and the role. It is the purpose of this chapter to examine the way early AngloNorman romances use the historical figure of the mestre: briefly to look at its use in the Roman de Horn and Ipomedon before Boeve, and in Gui de Warewic after Boeve. In Boeve itself, I contend that the figure acquired a distinct flavour, peculiar to the poem, but one which was much attenuated in the later English version, Sir Bevis of Hampton. In Norman and Angevin England, as in Continental France, the custom in the twelfth century, and later too, was to allocate a magister or mestre to a youthful nobleman. Such a figure was usually a knight, generally with his own household. The young boy was either given the mestre within his parents’ household – this might happen if he were a prince – or sent out, at



See Judith Weiss, ‘The date of the Anglo-Norman Boeve de Haumtone’, Medium Ævum, 55 (1986), 237–41.  Sabot (in eight cases spelt Saboc) is the form of the name in B, the later of the two MSS of Boeve; Sabaoth is the form in D. On MSS B and D, see p. 10 n. 5 above. All references to the text of Boeve are by line number to Der anglonormannische Boeve de Haumtone, ed. by Albert ­Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974) (hereafter Boeve).  For a very different view of Sabaoth and his ‘Christic’ role, see Valérie Galent-Fasseur, ‘Un médiateur de la Providence: le personnage de Sabaoth dans la version anglo-normande et la version en prose de Beuve de Hantone’, Littérales, 20 (1998), 25–38.  Clerics were alternative, cheaper instructors: see Nicholas Orme, From Childhood to Chivalry: The Education of the English Kings and Aristocracy, 1066–1530 (London; New York: Methuen, 1984), p. 24; Marjorie Chibnall, Anglo-Norman England, 1066–1166 (Oxford: Blackwell, 1986), p. 212, on the education of Robert Curthose. For a good account of the education of princes, nobility and lesser ranks, see Frank Barlow, William Rufus (London: Methuen, 1983), pp. 14–25.

25



Judith Weiss

around seven or even younger, to the mestre’s household. There he may have found himself in a group of other young boys in a similar situation; they were all there to be educated. But the mestre was more than just an educator. The connotations of mestre and maistrie help us here. Mestre certainly could mean not just tutor but master, expert, manager, and was applied to doctors and lawyers, those with skills. But maistrie also carries meanings of authority, control and power. The mestre was appointed by a father to be a sort of mentor to his son: he protected and taught him, he accompanied him to the most profitable tourna­ ments, he restrained and advised him. He was not necessarily much older than the youth but he had greater experience. According to Georges Duby, it was the mestre who would accompany the restless youths of Northern France – the juvenes – after they took up arms, on lengthy expeditions, financed by their fathers. Duby takes as one of his examples the story (dated around 1225) of Henry the Young King and his mestre, William Marshal: Li reis ovoc sum fiz le mist; Granz biens a faire li pramist Por lui garder e esseigner … Tant li fist e tant l’avança Que par ce ke il out apris Monta li giembles reis en pris E en enor e en hautesce; A lui s’acompaigna proësce. Des ce ku’il out tantes bontez, Si fu il al plus beal contez De toz les princes terrïens Ne sarrazins ne crestïens … Les armes conut, e en sout Tant con vaislet saveir en pout; Molt li plout des armes li estres, E ce fu molt bel a sis mestres. (The king put him in the company of his son; he promised to do the marshal much good in return for his care and instruction … He did so much for him, and brought him on so, that, as a result of what he had learned, the young king’s reputation increased, along with his eminence and the honour paid to him; he also acquired the quality of valour. Now that he had so many qualities, he was reckoned to be the finest of all the princes on earth, be they pagan or Christian … He knew about the use of arms, as much as any young nobleman could be expected to



Georges Duby, The Chivalrous Society, trans. by Cynthia Postan (London: Arnold, 1977), p. 114.  The History of William Marshall, ed. by A. J. Holden, with translation by S. Gregory and histori­cal notes by D. Crouch, vol. I (London: ANTS from Birkbeck College, 2002), lines 1943–66.

26



Sabaoth and Terri in Boeve de Haumtone know. The life of combat pleased him well, which was very pleasing for his tutor.)

Such historical examples can show us how close the relationship often was between the mestre and his nurri (the one who was nurtured). In the case of Henry the Young King, both William Marshal and his sons became Henry’s good friends. The doomed son of Henry I, William, was at sixteen still in the care of his mestre, called Othuer, who clasped him in his arms and died with him as the White Ship sank. The person in charge of education and upbringing might actually be a member of the family: Robert of Gloucester, the maternal uncle of the future Henry II, supervised his instruction in ‘letters and manners’, between the ages of nine and fourteen, by one Master Matthew in Bristol. What kind of things were taught? Basic academic skills, perhaps not always well learnt; etiquette, learning to serve one’s betters at table; music – dancing, singing, playing instruments; athletics, riding, hunting, hawking and jousting. Above all, ‘the whole process of socialization’ was taught; this included learning how to survive as servants and among peers and being prepared for life as members or heads of households. From the twelfth century on, there emerge courtesy books such as the Facetus and Daniel of Beccles’s Liber urbanus, to help guide the guiders in the ways of moral and social behaviour.10 Boeve’s literary predecessors, whether dealing with fiction or ‘fact’, show us the mestre and his nurri. According to Wace’s Roman de Rou (started 1160), Boton of Bayeux is mestre to William Longsword, duke of Normandy, and tells his nurri off for considering a cowardly course of action; he is later asked to look after William’s young son, so continuing his part to the next generation. But the most extended treatment of a mestre comes in the early part of Horn (c. 1170), with the portrait of the seneschal Herland. Since it is clear to me that the poet of Boeve knew Horn and used it in part as a model, I shall sketch Herland’s function in the poem, as concisely as I can. Horn is a fatherless exile from ‘Suddene’, so it is not until he arrives, aged ten, at the Breton court that he is allocated the king’s seneschal as his mestre, though the word used constantly of their relationship is norrir (‘to bring up, nurture’). The education given Horn and his best friend, Haderof, is described in some detail and includes music, hunting, hawking and swordsmanship, and it is clear later that he also knows how to serve in hall, play chess, put the stone



Orme, From Childhood to Chivalry, p. 19; W. L. Warren, Henry II (London: Eyre Methuen, 1973), pp. 38–9.  According to Barlow, William Rufus, p. 21, Henry II was the first literate descendant of the Conqueror.  Kate Mertes, The English Noble Household 1250–1600: Good Governance and Politic Rule (Oxford: Blackwell, 1988), p. 175. 10 Daniel of Beccles was a member of Henry II’s household: see Orme, From Childhood to Chivalry, p. 136.

27



Judith Weiss

and, above all, perform superlatively on the harp. Horn’s social accomplishments are part of his courtliness: he has been educated to be an excellent member of a court. One of the mestre’s duties Herland does not have to perform is that of advising or restraining. The young Horn is already so extraordinarily selfcontrolled and wise that the seneschal does not have to tell him how to behave when the king’s daughter propositions him. Indeed, it is, rather, the seneschal who gets himself into difficulties in this episode, providing the poem with two superb scenes when the princess first bribes him to bring Horn to her and then flies into a rage because he has not done so. Once this part of the poem is over, Herland does not reappear in person. We find out later from his son Joceran that he suffers once his protégé Horn is in disgrace with the Breton king; Herland is dispossessed and exiled, so that his son has to assume the guise of a poor palmer when he seeks for Horn in Ireland to ask for help. Though Horn tries to conceal his identity, he is soon recognized, whereupon he promises to repay his mestre: Tut le bien ke joe sai e tut l’afetement M’aprist il, e jo.l sai de sun enseignement. S’il ad perdu par mei, coe peise mei forment. Ferai l’en, quant purrai, mut bon restorement.11 (All the good that I know, and all my attainments, he taught me. I know them through his teaching. If he is deprived because of me, it brings me much sorrow. When I can, I will amply pay him back.)

Though Horn acknowledges his debt to Herland, his best friend is not the seneschal’s son Joceran but Haderof, son of his father’s seneschal, and it is Haderof that he later rewards with an Irish heiress. His closest male bonds are thus inherited from his father, while his bond with his own mestre does not last into the next generation. I have argued elsewhere that Hue de Rotelande’s Ipomedon (c. 1180) had an influence on Boeve,12 but not in the relationship between mestre and nurri. The equivalent figure in Ipomedon is Tholomeu, in other words Ptolemy.13 At the outset Tholomeu seems very like Herland:

11

Thomas, The Romance of Horn, ed. by M. K. Pope, vol. I, Anglo-Norman Texts 9 (Oxford: Blackwell, 1955), lines 3758–61. 12 Judith Weiss, ‘ “The courteous warrior”: epic, romance and comedy in the making of Boeve de Haumtone’, in Boundaries in Medieval Romance, ed. by Neil Cartlidge, Studies in Medieval Romance (Cambridge: Brewer, 2008), pp. 149–60. 13 In his romances, set in southern Italian kingdoms, Hue de Rotelande uses names taken from ­classical sources and the romans antiques. Ptolemy was one of Alexander’s generals, so appears as one of his douze pairs in Alexander romances. But the later Claudius Ptolemaios (c. ad 100–170), a Greek astronomer, is the author of the Almageste, much cited as a source of wisdom in medieval texts. Hue may be conflating these two figures.

28



Sabaoth and Terri in Boeve de Haumtone Il ot un mestre si curteys, Ke al mund n’out si riche reys K’il nel soust mult bien servir E les custumes retenir.14 (He had such a courteous tutor that there was not so noble a king in the whole world that he would not have known how to serve him well and observe the customs.)

Tholomeu sees to it that Ipomedon is well educated in handling birds and dogs, and is well lettered. Himself ‘very wise’, and fluent in several languages, he accompanies his nurri to the court of Calabria, where ­Ipomedon’s courteous accomplishments (such as serving the cup in hall, like Horn) are praised. Ipomedon is evidently close to his mestre, whose advice he asks and in whom he confides. But for much of the poem Tholomeu seems to be no more than a superior kind of servant, arranging dinners, taking cloaks and abetting his pupil in schemes. There is no sign that he possesses much status or any lands of his own; it is only at the end that he is rewarded with Burgundy and a wife. Above all, he is not a fighter – and it is crucial he not be so, since much of Ipomedon is concerned with the hero’s apparent lack of proesce, that vital counterpart to curteisie. The somewhat colourless figure of the mestre has the role of supporting Ipomedon’s courteous appearance, while not suggesting it is backed up by any more substantial achievements. Sabaoth’s role in Boeve can be summarized as follows: Sabaoth first appears when Boeve’s mother is trying to have her son killed and asks his mestre to do the job. He pretends to do so, but the child’s refusal to follow his mestre’s advice sabotages this plan, and Boeve’s mother sells him to pagan merchants, who take him to Egypt. Here the careers of the two figures diverge. We discover later that Sabaoth has suffered for his love of his nurri by being expelled from his lands. An unnamed son of his, dressed as a palmer, goes looking for Boeve; unlike Horn, however, Boeve successfully deceives the palmer into thinking that he is dead. Sabaoth reappears when Boeve is trying to reclaim his inheritance: he uses his fighting skills to see his charge regain his lands – only to lose them again when he falls foul of the English King Edgar. Sabaoth thus takes over these lands while his nurri is banished to foreign parts. His son Terri becomes Boeve’s bosom friend and his companion in his adventures. Sabaoth meanwhile dreams of the theft of Boeve’s horse, which his wife correctly interprets as the abduction of Josiane, Boeve’s wife. Disguising himself as a pilgrim, he rescues Josiane from Saracens, helps her to travel about in disguise and is in turn cared for by her when he falls ill.

14

Ipomedon: Poème de Hue de Rotelande (fin du xiie siècle), ed. by A. J. Holden, Bibliothèque française et romane, B17 (Paris: Klincksieck, 1979), lines 199–202.

29



Judith Weiss The two of them find Boeve and Terri in the dukedom of Civile, whose duchess wants to marry Boeve. Terri replaces Boeve as husband and duke. Thereafter he continually arrives to help Boeve when the hero is in need. He calls his son, who will eventually marry Boeve’s daughter, Boves. Meanwhile Sabaoth has another dream about damage to Boeve and, again disguised as a pilgrim, rescues his nurri’s stolen horse from Yvori, king of Monbrant; he cheekily challenges king and city alike before escaping from hordes of pursuing Saracens. While present at the coronation of Boeve in Monbrant, he hears that the English king has dispossessed his son Robant, thus once again taking Boeve’s lands. Sabaoth, his son and his grandsons, and Boeve and his sons, all go to England with a huge army; Edgar capitulates, offering his daughter to Miles, one of Boeve’s sons. We hear no more about what happens to Robant, but Miles becomes king of England on the death of Edgar, and Sabaoth promises to Boeve he will look after Miles all his life; he has in effect become his mestre too.

The elements in this story that are already familiar to us from earlier narratives are the mestre advising his nurri, and suffering and being exiled for him, and the son of the mestre looking for the nurri. What is missing is, conspicuously, any courtly training of the hero by the mestre. The only remnant of what we saw in Horn or Ipomedon is Boeve acting as cup-bearer to King Hermine in Egypt, the country of his exile, and a later reference to his doing the same for King Edgar (laisse clxvii). Otherwise he apparently receives no teaching – in either skills or, notably, etiquette. Boeve in the first part of the poem must be amongst the rudest heroes in Insular romance, and this is only partly explained by his circumstances. He never asks for advice and, when he is given good counsel, either rejects it or takes some time to be persuaded into sensible courses. All this is at odds, possibly deliberately and ironically so, with one of his constant epithets – li sené (‘the wise’).15 A long passage on the value of bon conseil, which the hero asks from his mestre and friends in Hue de Rotelande’s contemporary romance ­Protheselaus (1190s) may be parodic; nevertheless, the difference from the world of Boeve is clear: Ki ne creit cunseil, sil requert, Fols est enfin e tuz jors ert. Tant sunt fol e musard plusor Ne creent cunseil fors le lor; Co teng a grant sorquiderie E sin avent sovent folie. Qui par cunseil ovre sovent N’est blamé se il mesprent; Qui mesprent de son eindegré E n’avra cunseil demandé Vus savez que l’em dirra: 15

See Weiss, ‘ “The courteous warrior” ’, pp. 151–3.

30



Sabaoth and Terri in Boeve de Haumtone Fols est, dehait qui le pleindra … Que qu’en avenge, mal u ben, Mult valt cunseil sor tute ren.16 (He who does not believe counsel if he asks for it is in the end a fool and always will be. Many are foolish and stupid who believe no advice except their own. This smacks of great arrogance, and folly often comes of it. Who often acts by advice is not blamed if he goes wrong. Who goes wrong of his own free will and has not asked for advice, you know what will be said about him: he’s an idiot, bad luck to whoever pities him … Whatever happens, bad or good, counsel is supremely worthwhile.)

If, then, the headstrong hero finds it hard to follow advice, his mestre Sabaoth, by comparison, emerges as distinctly wiser, and this wisdom is demonstrated through the knowledge imparted to him through his dreams. At the same time, he is a most doughty warrior and an achiever who can do more than just fight: at times he appears more effective than Boeve himself, as when he rescues his nurri’s wife and his horse. At the end, he has survived Boeve’s death and is still keeping an eye on Boeve’s son. The second part of the poem shows us close links being made between Boeve and his mestre’s family. The readiness with which Terri can take his friend’s place in the marriage-bed of the duchess of Civile carries perhaps a fleeting reminder of the interchangeability of Horn and Haderof; it certainly suggests that Terri is similar to Boeve and equally acceptable. The two friends are on intimate terms: when they meet again they exchange family news – ‘How’s your wife?’ ‘Fine, thanks: we have three sons and I’ve called the eldest after you’ (lines 3195–7) – and marriage is planned between their children. The assiduity with which each family helps the other suggests the importance the poet attaches to families and to their preservation and support, though the ending unsurprisingly demonstrates that Boeve’s family is the more important: the sudden introduction of Sabaoth’s other, hitherto unmentioned, son Robant and his dispossession is no more than a transparent and not very satisfactory plot device to ensure that each son of Boeve will become a king. The way Sabaoth will continue to look after Boeve’s child is reminiscent of Boton and the son of William Longsword in the Roman de Rou: the bond between mestre and nurri is extended to the next generation too. But there is no precedent for the bonds between the children: Boeve and Terri, and later their offspring.

16

Hue de Rotelande, Protheselaus, ed. by A. J. Holden, 2 vols (London: ANTS, 1991), lines 3393– 3408.

31



Judith Weiss

What influence did this story-pattern have on the next generation of Insular romances? As far as I can see, Waldef (1200 × 1210) is unaffected by it.17 Gui de Warewic (c. 1210 × 1215),18 on the other hand, shows clear signs of influence. In the way it uses the role of the mestre it tells us, I believe, much about some of its concerns, and comparing it with Boeve again makes us realize the particular flavour of the earlier work. Gui is not the son of a king, like Horn or Boeve, but he nevertheless is given a mestre, Heralt of Arderne. The education of Gui reminds us at once of Horn and Ipomedon: Heralt, himself curteis and enseigniez, teaches his charge about dogs, birds, hawks and falcons. As if illustrating Duby’s ideas about the juvenes, he is asked by Gui’s father to take his nurri on an educative overseas trip; they thus participate in tournaments, so that Gui has a chance to show his mettle and make himself worthier of the love of Felice. Thereafter the career of Heralt intersects with Gui’s in a series of separations and reunions. When the mestre, like Gui, gets involved in Continental feuds, he is apparently killed but marvellously revived by a skilful monk, thus providing the first occasion for a loving and tearful reunion with his nurri. Here Heralt has suffered for Gui, and meets him again dressed as a poor man (recalling Herland’s impoverishment), but there is no dysfunctional family in the background, as in Boeve. Despite the mestre’s near-death, the story seems tamer, less dramatic – one suspects that his brush with mortality was only narrated in order to provide a touching meeting. Heralt is again in difficulties later when captured by an enemy; he is rescued by Gui. This is significant: whereas in Boeve the hero was often less active or capable than his mestre, here it is the opposite – Gui does not need Heralt as Boeve needed Sabaoth, because Gui is himself active and capable and takes advice, even inviting Heralt to give it. In short, the relationship is less interesting and dynamic. We can still see traces of the Boeve model: in the second part of Gui de Warewic, Heralt looks after Gui’s son, Reinbrun – or, rather, tries to (most of the time he is looking for the abducted child, who grows up largely on his own) – and he survives the death of his nurri. But he does not see Gui again after Gui has, in short order, got married, conceived a son and left his wife in order to serve God. The emotional bond between mestre and nurri has been replaced by another: by that of friendship with Terri. The choice of name is significant: it at once reminds us of the good friend of Boeve. But Terri there was the son of the mestre, Sabaoth. Here, he is unrelated to Heralt and not even English: he is Terri of Worms, a Continental nobleman in need of as much, if not more, rescuing as Heralt.19 His separations from Gui, their 17

The hero’s cousin Florenz acts, like Sabaoth, as a kind of nutricius, by saving Waldef from a plot to kill him, but he is his contemporary rather than an older figure. 18 For this dating, see Judith Weiss, ‘Gui de Warewic at home and abroad: a hero for Europe’, in Guy of Warwick: Icon and Ancestor, ed. by Alison Wiggins and Rosalind Field, Studies in Medieval Romance (Cambridge: Brewer, 2007), pp. 1–11 (esp. pp. 4, 7–8). 19 Heralt does have a son, Aselac, but he appears very late in the day, to furnish an adventure for Reinbrun on his way back to Warwick.

32



Sabaoth and Terri in Boeve de Haumtone

swearing of friendship, their love, their reunions after periods in disguise are all positively operatic, milked for sentiment. Gui’s relationship with his wife, Felice, is conventional and bland by comparison, and it is Terri who after Gui’s death triumphs over the wife’s claim by removing his friend’s body to an abbey in Lorraine, against her and Gui’s wishes.20 This provokes two questions. Why did the poet of Gui cut the link between the mestre and Terri? And why is friend Terri not English? To answer the first: I think this poet is less interested in the bonds between families over several generations, and much more interested in male peer relationships. To answer the second: I suspect that this has to do with the way in which England is seen in the romance, at least until after the death of Athelstan, as a safe, orderly, untroubled place to return to, as opposed to turbulent Northern Europe, specifically the territories of the Holy Roman Emperor. Gui needs a Continental friend as a means of highlighting European feuds, jealousies and mismanagement. The poet of the English version of Boeve, Sir Bevis of Hampton, made several small though telling changes to the figure of the mestre and his relationship with the hero. Saber is still called meister, 21 but he is the hero’s uncle, presumably on the father’s side, since he is not addressed as ‘brother’ by Bevis’s mother.22 He thus has a vested interest in protecting his nephew and the heir of the family. His own estates lie in the Isle of Wight, part of Hampshire, of which he ends up earl (p. 216); Saber is thus closely associated with a particular region, which Sabaoth never was. He dreams of misfortunes befalling Bevis, which he then rectifies, but he does not make through those dreams the intimate association between Josian and the horse Arundel that Sabaoth and his wife did.23 The English narrative clarifies and rationalizes passages such as Saber’s rescue of Josian from Yvor’s Saracens but, because inevitably it does not reproduce the style of a chanson de geste, it omits the repeated epithets that allude to Sabaoth’s age24 and remind us that this old man frequently behaves like a young one – never more so than when he emulates Wace’s Gawain in provoking thousands of pagans to pursue him.25 The most important change of emphasis, however, comes in the relationship between Saber’s family and Bevis’s. The detail that Terri has named one of his children after Boeve, the future marriage between this godson and

20 21 22

23 24 25

See Weiss, ‘Gui de Warewic at home and abroad’, p. 10. The Romance of Sir Beues of Hamtoun …, ed. by Eugen Kölbing, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) (hereafter BHK), p. 14. The version represented by the printed texts of Bevis and by parts of MS M tries to make this relationship very clear from the start: Saber is ‘sir Gyes broþure’ (p. 15). Oddly, the Auchinleck MS uses the same name for Bevis’s Cologne uncle, calling him ‘Saber Florentin’ (p. 121). Compare Boeve 2735–41 / BHK, pp. 179–80; Boeve 3437–43 / BHK, pp. 191–2. E.g. li veil barbé (‘the old bearded man’), li blanc (‘the white-haired’), le flori (‘the hoary’): Boeve 3458, 3475, 3617, etc. Boeve 3468–79; cf. Roman de Brut de Wace, ed. by Ivor Arnold, vol. II (Paris: SATF, 1940), lines 11,741–880.

33



Judith Weiss

Boeve’s daughter, the intimate and friendly conversation between Boeve and Terri that I mentioned earlier, are all omitted. Presumably the English poet had no interest in the continuation of links between the two families, and this accords with the fact that Saber is made Bevis’s uncle: kin matters, more so than friendships extending across two generations. Returning to Boeve, and with the benefit of hindsight, we can surmise that its author was deliberately doing several things against the grain of narratives that had preceded it. The second part of the poem, in particular, is interested in the hero as part of a close-knit group, not an individual. His individuality and especially his courtliness are less important in such a context. Hence the poet’s virtual disregard for any training or education supplied by the mestre. He may even be pointedly omitting the courtly manners that were becoming a necessary characteristic of romance heroes. On the other hand, he attaches great value to the close bonds initiated between pupil and father-figure in childhood. In this respect, a detail may carry some weight. When Boeve’s evil mother berates his mestre for saving the child’s life, she addresses him in English.26 This suggests that she may be thought of as English – the fact that she is the daughter of the king of Scotland does not invalidate this, since the brother of the Scottish king in the 1170s was called ‘English’.27 It also suggests that Sabaoth understands English, may indeed be English, and thus he would be bound to his nurri by a shared language, the language of the nursery, from Boeve’s earliest years.28 Who, then, was Sabaoth? Or, rather, from whom was his name taken? It is a very rare name, meaning ‘armies, hosts’.29 MS B offers the alternative form of Sabot (and possibly Saboc, but this may be the result of scribal corruption) and – just once – Saber (line 3225), as in Bevis. This may offer us a precious clue. A king of the East Saxons in the early seventh century was called Sabert or Sæberht. He was a nephew of Æthelberht of Kent, through whom he was converted to Christianity, and together they founded St Paul’s Cathedral in London, Sabert’s capital city. Bede tells us that he was the first king of Essex to embrace Christianity, though his three sons reverted to paganism when he

26 27

Boeve 331. Hugh M. Thomas, The English and the Normans: Ethnic Hostility, Assimilation and Identity 1066–c.1270, 3rd edn (Oxford: Oxford University Press, 2003), p. 79; he also (p. 77) makes us aware of how complicated the question of ethnicity is, since earls descended from Norman immigrants could be described as English. 28 On the aristocracy’s knowledge of English in the mid-twelfth century, see Cecily Clark, ‘Women’s names in post-Conquest England: observations and speculations’, Speculum, 53 (1978), 223–51 (p. 223). M. D. Legge, ‘Anglo-Norman as a spoken language’, Proceedings of the Battle Conference on Anglo-Norman Studies, 2 (1979), 108–17 (pp. 110–11), discusses the same passage in Boeve, calling it ‘a rare reference to a Norman household with English servants’ and positing that the mother ‘knows enough English to address an inferior’. 29 According to the Oxford Dictionary of the Christian Church, ed. by F. L. Cross and E. A. ­Livingstone, 3rd edn (Oxford: Oxford University Press, 1997), p. 1432.

34



Sabaoth and Terri in Boeve de Haumtone

died in 616/617. He also tells us that Saba was the shorter, more familiar, form of his name.30 Because Bede was by far the most important Insular source for those writing chronicles about the English between 1066 and the 1130s, later historians did not forget this man, for several important reasons: his closeness to Æthelberht, his conversion as the first of his dynasty, and the connection with St Paul’s. He is mentioned again in John of Worcester’s Chronicle, in William of Malmesbury’s Gesta regum Anglorum and, along with his genealogy, in Henry of Huntingdon’s Historia Anglorum; Gaimar also refers to him.31 The poet of Boeve, then, has perhaps carefully selected, via these historians, a name from Anglo-Saxon history for his hero’s tutor, who may have educated his pupil not in Anglo-Norman, the language of acculturation, what Ian Short calls ‘the high-status, class-exclusive sociolect of a cultural and political elite’, but in English, the ‘class-inclusive vernacular’.32 If we remember that Sabaoth, besides understanding English, has a wife whose name (Eneborc) is obviously Anglo-Saxon, it seems that the poet is making a point about his hero, if not in the clearest fashion throughout. In the first part of the poem, Boeve is brought up by English speakers and often refers to England as the land of his birth, as well as to his determination to return there. In the second part, he is (inconsistently) described as ‘French’.33 Nevertheless, it seems clear that the original poet’s intention was to connect the d’Albinis, earls of Arundel, the likely patrons of the poem,34 with distant Anglo-Saxon history through Boeve – and through his mestre. The d’Albinis were francophone but, as Hugh Thomas and Ian Short have reminded us, like other members 30

31

32

33 34

Bede’s Ecclesiastical History of the English People, ed. by Bertram Colgrave and R. A. B. Mynors (Oxford: Clarendon Press, 1969), pp. 142, 152, 153 n. 2 (where Saba is described as ‘the familiar shortening of Sæberht’), 562; Barbara Yorke, Kings and Kingdoms of Early Anglo-Saxon England (London: Seaby, 1990), pp. 28, 46. M. Parker Pearson, R. van de Noort and A. Woolf, ‘Three men and a boat: Sutton Hoo and the East Saxon kingdom’, Anglo-Saxon England, 22 (1993), 27–50 (p. 45), have suggested that Sæberht (the name means ‘sea-bright’) may be the incumbent in Mound I at Sutton Hoo. The Chronicle of John of Worcester, ed. by R. R. Darlington and P. McGurk, trans. by Jennifer Bray and P. McGurk, 3 vols (Oxford, 1995), II, 76, 82; William of Malmesbury, Gesta regum Anglorum, ed. and trans. by R. A. B. Mynors, completed by R. M. Thomson and M. Winterbottom, 3 vols, Oxford Medieval Texts (Oxford: Clarendon Press, 1998), I, 144; Henry of Huntingdon, Historia Anglorum, ed. and trans. by Diana E. Greenway, Oxford Medieval Texts (Oxford: Clarendon Press, 1996), pp. 100, 116, 166; Geffrei Gaimar, Estoire des Engleis, ed. Alexander Bell, 3 vols, Anglo-Norman Texts 14–16 (Oxford: Blackwell for the ANTS, 1960), lines 1069–74. Ian Short, ‘Tam Angli quam Franci: self-definition in Anglo-Norman England’, Anglo-Norman Studies, 18 (1996), 153–75 (p. 156). Hugh Thomas, The English and the Normans, p. 177, points out that English nursemaids and servants probably played a part in promoting bilingualism. Clark, ‘Women’s names’, pp. 230–1, remarks that nurses and waiting-women in general were expected to be English-speaking. Jean-Pierre Martin, ‘Beuve de Hantone entre roman et chanson de geste’, Littérales, 31 (2003), 97–112, maintains (p. 100) that this means just ‘Christian’, as in Crusading narratives. On this patronage, see M. Dominica Legge, Anglo-Norman Literature and Its Background (Oxford: Clarendon Press, 1963), p. 159; Weiss, ‘The date of the Anglo-Norman Boeve de ­ Haumtone’; Ivana Djordjević, ‘Mapping Medieval Translation: Methodological Problems and a Case Study’ (unpub. Ph.D. diss., McGill Univ., 2003), pp. 260–74.

35



Judith Weiss

of the Anglo-Norman ruling elite from the 1150s on, they considered themselves English.35 The romance offers them a foundational history of their family which seeks to persuade them that really they are English. In this process of persuasion, Sabaoth plays a vital part.

35

Hugh Thomas, The English and the Normans, pp. 79, 319, 333; Short, ‘Tam Angli quam Franci’, pp. 156–7, 165, 173–4.

36

3 Rewriting Bevis in Wales and Ireland Erich Poppe and Regine Reck

T

he medieval vernacular literatures of Wales and Ireland have been often neglected in comparative literary studies, Arthurian Studies probably being the one major exception. However, these literatures participated in many pan-European trends and fashions through the translation and adaptation of foreign secular and religious literary sources. Cases in point are the story about Troy attributed to Dares Phrygius, narratives about Charlemagne, and the story of Mary of Egypt, all of which were translated into both Welsh and Irish. The Anglo-Norman Boeve de Haumtone (hereafter Boeve) also proved to be very successful in the island (or Insular) literatures of the British Isles and Iceland, and it is the immediate source for the Middle Welsh Ystorya Bown o Hamtwn and the indirect source for the (incomplete) Early Modern Irish version, which is based on a Middle English version of the romance. The Middle Welsh Bown was probably translated around the middle of the thirteenth century from a text close, but not identical, to one of the known texts of Boeve. It thus belongs to the earliest phase of such translations into Welsh and is more or less contemporaneous with the translation of the core texts of the cycle about Charlemagne, namely of pseudo-Turpin’s Historia and of the Chanson de Roland and the Pelèrinage de Charlemagne. The only 

Witness, e.g., Karl der Große in den europäischen Literaturen des Mittelalters: Produktion eines Mythos, ed. by Bernd Bastert (Tübingen: Niemeyer, 2004), in which the Welsh and Irish materials are not even mentioned.  For these concepts, see H. Munro Chadwick and N. Kershaw Chadwick, The Growth of Literature, vol. I: The Ancient Literatures of Europe (Cambridge: University Press, 1932), pp. 1–5; Hildegard L. C. Tristram, ‘Introduction’, in The Legend of Mary of Egypt in Medieval Insular Hagiography, ed. by Erich Poppe and Bianca Ross, Medieval Studies (Blackrock, Co. Dublin: Four Courts Press, 1996), pp. 1–17.  Dating suggested in Ystorya Bown de Hamtwn, ed. by Morgan Watkin (Cardiff: University of Wales Press, 1958) (hereafter Bown), pp. lv–lix, and not challenged since. All references to the text of Bown are by line number to Watkin’s edition. The discussion here is based on our analysis in Erich Poppe and Regine Reck, ‘A French romance in Wales, Ystorya Bown o Hamtwn: ­processes of medieval translation’, Zeitschrift für celtische Philologie, 55 (2006), 122–80 (second part forthcoming ibid., 56), which see for further details and references. Cf. also Erich Poppe, ‘Adaptation und Akkulturation: narrative Techniken in der mittelkymrischen Ystorya Bown de Hamtwn’, in Übersetzung, Adaptation und Akkulturation im insularen Mittelalter, ed. by Erich Poppe and Hildegard L. C. Tristram, Studien und Texte zur Keltologie 4 (Münster: Nodus, 1999), pp. 305–17.

37



Erich Poppe and Regine Reck

two medieval manuscripts in which Bown is transmitted are Llyfr Gwyn Rhydderch (the White Book of Rhydderch), written c. 1350, and Llyfr Coch Hergest (the Red Book of Hergest), written c. 1400. These are two large collections which contain, among other things, what was perceived to be the corpus of medieval Welsh secular narratives of both native and foreign origin – the inclusion of Bown is therefore an indication of its literary status. The interest in adaptations of popular foreign literature shared by both Rhydderch ab Ieuan Llwyd and Hopcyn ap Tomas Rhydderch, for whom the White Book and the Red Book respectively were produced, is probably not surprising in the political and cultural contexts of fourteenth-century Wales. The two manuscripts share a significant number of prose texts, and the relationship between them is problematic. The drift of modern scholarship in this matter inclines to the view that it is generally more likely that they derive from common sources, rather than that the texts in the Red Book are copied from the White Book. The differences between the two texts of Bown are not dramatic, and it would be difficult to argue on that basis that one manuscript preserves a significantly superior text. There is some small-scale stylistic variation between them, with regard to, for example, individual words or phrases, and the use of direct and indirect speech or the narrative present, but overall the text of Bown was fairly fixed in its medieval transmission. The version in the White Book, however, exhibits one enigmatic feature, namely an alternative beginning: this first beginning of forty-nine lines starts on fol. 118 and breaks up in the middle of its second column, the rest of which is left blank. The narrative then starts afresh on fol. 119. The first twenty-one lines of the two beginnings agree very closely, but the remaining lines of the first deviate significantly from the second: its narrative is much more condensed and, by contrast to the Anglo-Norman text and the second beginning, does not employ direct speech – a characteristic feature of native Welsh tales. Some details present in the second beginning as well as in the Anglo-Norman text, but not in the first beginning, indicate that the second beginning represents



Aberystwyth, National Library of Wales, MS Peniarth 4–5; Bown is written by this manuscript’s scribe C and is his only contribution.  Oxford, Jesus College, MS 111; Bown is written by this manuscript’s main scribe, Hywel Vychan. Bown is also found in three post-medieval manuscripts which are derived from the medieval ones: Aberystwyth, National Library of Wales, MSS 3043B (Mostyn 135), transcribed between c. 1580 and 1600 from the White Book; Peniarth 118 (Llyfr Sion Dafydd Rhys), a sixteenth-century compilation in which the text of Bown is an incomplete copy of the Red Book; and Gwenogvryn Evans 1A, written c. 1608, where the incomplete text of Bown would seem to belong to the same tradition as the White Book.  Cf. Daniel Huws, Medieval Welsh Manuscripts (Cardiff: University of Wales Press; National Library of Wales, 2000), pp. 246, 255. See also, with specific reference to the text of Peredur, Peter Wynn Thomas, ‘Cydberthynas y pedair fersiwn ganoloesol’, in Canhwyll Marchogyon: Cyd-destunoli ‘Peredur’, ed. by Sioned Davies and Peter Wynn Thomas (Cardiff: University of Wales Press, 2000), pp. 10–49, esp. his stemma (p. 41). There is also some evidence that one of the scribes of the Red Book had access to the White Book.  E.g. the references to the first day of May and to presents which the messenger receives from the German emperor: cf. Bown 78, 98–100; Der anglonormannische Boeve de Haumtone, ed.

38



Rewriting Bevis in Wales and Ireland

a new attempt at a translation, rather than a creative and expanded retelling of the first. The structural and stylistic differences between the two versions illustrate impressively the freedom with which medieval Welsh translators (or scribes) treated their sources. The story-line of Boeve and its details are faithfully preserved in Bown. The interest in the plot of the Anglo-Norman author and of the Welsh redactor would appear to have been very similar, with regard to both the elements of a successful story and their outlook on the world. Both present within a framework of martial Christian values a series of exciting events whose hero is characterized as ‘the best man/knight in Christendom’. The transferral of such social and literary ideals from the Anglo-Norman environment into a Welsh one did not create any problems. If there are any changes in Bown on the level of content, these are subtle and minor and do not change the text’s basic concerns but, rather, focus it even more securely within a martial Christian social context through a slightly stronger insistence on piety and the overt expression of Christian affiliations. It is on the levels of narrative structure and of style that the Welsh redactor significantly rewrote his source. At this point, a brief digression on terminological and conceptual matters is appropriate. ‘Rewriting’ is a concept we have adopted from Michael Cronin, who said about medieval Irish translations: What we have in effect is translation as a form of … ‘rewriting’. These target-oriented, prospective translations were source texts rewritten for an Irish audience ... The central concern of literary translators in the period was to provide translations that would be wholly acceptable to the native reader and which would function within the native literature.10

Cronin’s observation that medieval translations were produced with a view to acceptability for their new audience has, of course, already been widely echoed for other Insular cultures without employment of the term ‘rewriting’.11

 

10 11

by Albert Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974) (hereafter Boeve), lines 87, 102–3. Cf. Boeve 2640–2, 3097; Bown 3075–6, 3549–50. Cf. also Regine Reck, ‘Heiligere Streiter und keuschere Jungfrauen: religiöse Elemente in der kymrischen Adaptation des anglo-normannischen Boeve de Haumtone’, in Übersetzung, Adaptation und Akkulturation im insularen Mittelalter, ed. Poppe and Tristram, pp. 289–304. Michael Cronin, Translating Ireland: Translation, Languages, Cultures (Cork: Cork University Press, 1996), pp. 21–2. Cf., e.g., Ivana Djordjević, ‘Mapping medieval translation’, in Medieval Insular Romance: Translation and Innovation, ed. by Judith Weiss, Jennifer Fellows and Morgan Dickson (Cambridge: Brewer, 2000), pp. 7–23; J. D. Burnley, ‘Late medieval English translation: types and reflections’, in The Medieval Translator: The Theory and Practice of Translation in the Middle Ages …, ed. by Roger Ellis et al. (Cambridge: Brewer, 1989), pp. 37–53; Andreas Kraß, ‘Spielräume mittelalterlichen Übersetzens: zu Bearbeitungen der Mariensequenz “Stabat Mater Dolorosa” ’, Wolfram-Studien, 14 (1996), 87–108; Stefanie Würth, ‘Alexanders saga: literarische und kulturelle Adaptation einer lateinischen Vorlage’, in Hansische Literaturbeziehungen: das Beispiel der ‘Þiðreks saga’ und verwandter Literatur, ed. by Susanne Kramarz-Bein (Berlin; New York: de Gruyter, 1996), pp. 290–315; Gabriele Röder, ‘Die chansons de geste in der altnordischen

39



Erich Poppe and Regine Reck

With regard to the conventions of native literature according to which foreign sources were rewritten in order to be acceptable, we see these as the conventions considered appropriate by medieval authors/redactors originally for the presentation of stories which form part of their own cultural and historical heritage – in Wales the yardstick for such conventions would be the narrative techniques employed in the so-called Mabinogion corpus of tales, which is also preserved in the White Book and in the Red Book.12 The term ‘native’ is not meant to imply that these conventions are necessarily of venerable age, based on oral tradition, or uninfluenced by foreign models, stylistic or otherwise. Notwithstanding possibly diverse origins of individual elements, they serve specific textual and narrative purposes, and these include the rewriting of foreign narratives. In accordance with such native conventions, the Welsh redactor of Boeve changed his source’s macro-form from Anglo-Norman verse to Welsh prose. Characteristics of native narrative syntax employed in the Mabinogion corpus as well as in Bown are unmarked narration in sequences of clauses mostly coordinated by a(c) (‘and’) or other sentence connectives; a more marked variant with preposed temporal clauses that resume the verbal event of the preceding sentence; the use of cataphoric expressions to direct and guide the attention of the audience;13 and employment of the narrative present in a Welsh fashion (that is, mostly in subordinate clauses, but very rarely in main clauses)14 and of narrative modes with a non-finite form of the verb either on its own or combined in a periphrastic construction with a finite form of the semantically empty verb gwneuthur (‘to do’). The cataphoric expressions and the narrative modes with verbal nouns have no formal parallels in Anglo-Norman. A number of idiomatic Middle Welsh expressions for oaths and similar exclamations occur in the Mabinogion corpus, and the majority of these also appear in Bown: for example, yr Duw (‘by God’), ym kyffes (‘by my confession’), and myn vyg cret (‘by my faith’).15 The highly idiomatic phrase y rof a Duw (‘between me and God’) is frequent in the native corpus

12

13 14

15

Karlamagnús saga: Übersetzungen oder Adaptationen?’, in The Medieval Translator / Traduire au Moyen Âge, vol. 6, ed. by Roger Ellis, René Tixier and Bernd Weitmeier (Turnhout: Brepols, 1998), pp. 134–58. These are Pedeir Keinc y Mabinogi, Culhwch ac Olwen, Breudwyt Ronabwy, Breudwyt Maxen and Kyfranc Llud a Lleuelys, as well as Owein (or Chwedyl Iarlles y Ffynnawn), Gereint, and Peredur, which are in complex ways derived from Chrétien de Troyes’s Yvain, Erec and Perceval. For a comprehensive discussion of narrative techniques employed in this corpus, see Sioned Davies, Crefft y Cyfarwydd: Astudiaeth o Dechnegau Naratif yn y Mabinogion (Cardiff: University of Wales Press, 1995). These are mainly nachaf and llyma, both often approximately translated as ‘behold’, but also sef ; on the skewed distribution of the latter in the text, see below. Cf. also Erich Poppe, ‘Notes on the narrative present in Middle Welsh’, in Hispano-Gallo­Brittonica: Essays in Honour of Professor D. Ellis Evans on the Occasion of his Sixty-Fifth Birthday, ed. by J. E. Eska, R. Geraint Gruffydd and Nicolas Jacobs (Cardiff: University of Wales Press 1995), pp. 138–50. See Davies, Crefft y Cyfarwydd, p. 128, and cf. Bown 2292, 1808, 1285, 3230–1, 4006; Boeve 2023, 1639, 1220, 2797, 3531.

40



Rewriting Bevis in Wales and Ireland

and is attested six times in Bown, where it corresponds to functionally similar exclamations in Boeve in three instances.16 This use of a characteristic Welsh formula contrasts with an influence from the Anglo-Norman system of oaths: the phrase myn vymphen (‘by my head’) is not part of the native system, but it is relatively frequent in Bown, with eight attestations, and, more significantly, always renders Anglo-Norman par mun chef (‘by my head’).17 The system of exclamations in Bown provides in a nutshell an example of the complex tendencies reflected in the process of textual transfer: adaptation to native conventions as in y rof a Duw (‘between me and God’) side by side with imitation of the foreign source as in myn vymphen (‘by my head’). A thorough process of adaptation affects the accounts of single combat in Bown. As a rule, these scenes have been made to conform to the pattern of combats in the Mabinogion corpus. This consists of several more or less formulaic phrases which are combined to describe the attack of the two opponents, the parts of the body or armour that are attacked, the result of the blows, and the falling of the defeated combatant to the ground. The realiza­ tion of this pattern varies, from a skeleton description of the attack and the falling to the ground only, to very elaborate and detailed compositions. Syntactically, the attack and the blow delivered are typically given in main clauses, while the results of the blow are described in one or more postposed subordinate clauses introduced by the conjunction yny (‘so that, until’). There are some fourteen accounts of single combat in Bown, as well as several violent encounters, which as a rule all follow the same syntactic pattern. It is significant that the Welsh redactor prefers a fairly fixed and formulaic wording in the clause of result to describe the defeated combatant’s fall to the ground – ‘yny dygwydawd ynteu y’r llawr / yny dygwyd (ynteu) yn varw y’r llawr’ (‘so that he fell/falls (dead) to the ground’) – whereas the wording in Boeve varies considerably. In some instances the use of a present tense in these postposed clauses may have been conditioned by the Anglo-Norman source, but on the whole it agrees well with what has been described as the system typical for the native narrative corpus.18 In the Welsh narrative tradition, the narrator tends to remain inaudible, and it is therefore no surprise that the Welsh narrator of Bown is much less intrusive than his very audible Anglo-Norman counterpart. The latter’s prologue is replaced by the straightforward introduction of one of the leading characters, Bown’s father: ‘Yn Hamtwn yd oed iarll a elwit Giwn’ (‘In Hamtwn there was an earl called Giwn’).19 More interesting, however, are some instances where he does intrude, and these would appear to be the translator’s ‘lapses’, in that he allows himself to be influenced by his source and translates quite

16 17 18

See Reck, ‘Heiligere Streiter’, pp. 292–3; and cf. Bown 175, 1271, 705; Boeve 186, 1210, 692. See Bown, p. 228. Cf. also Regine Reck, The Aesthetics of Combat in Medieval Welsh Literature: A Literary Analysis (Wiesbaden: Harrassowitz, 2008). 19 Bown 50–1.

41



Erich Poppe and Regine Reck

faithfully a few narrator’s intrusions, comments and exclamations, a rhet­ orical question, and a reference to a written source,20 which are all wholly uncharacteristic of his own narrative background. There is even a small number of similar narrator’s intrusions which have no obvious parallels in the extant Anglo-Norman text.21 It must remain an open question, however, whether these phrases were contained in the Welsh redactor’s source, but not in the extant text of Boeve, or whether they are his own creative innovations, modelled on other instances of narrator’s intrusions in Boeve. A central device used by the narrator to structure and organize his text is paragraphing, the marking of narrative units. In Boeve, frequently used express­ions to mark a change of scene contain phrases in which the narrator intervenes, such as Ore dirrum de … (‘we will now talk of …’). Such formulae are not part of the Middle Welsh narrative repertoire; hence it is not surprising that they are often not translated in Bown, the narrative transition remaining unmarked. The first two instances of explicitly marked paragraphing in the Welsh text are somewhat odd in that a second person plural imperative is used, as if the narrator were addressed: Ymhoelwch in at … (‘Turn for us to …’)22 – the Anglo-Norman text here has the expected authorial first person plural, returnerum and dirrum respectively.23 In (approximately) the second half of the text, most of the Anglo-Norman phrases indicating such transitions are faithfully and idiomatically translated into Welsh. The Welsh redactor even appears to have settled for a favourite expression, bellach/weithon y dywedwn am … (‘now/further we will say about …’), which is employed three times for different phrases in the extant Anglo-Norman version.24 Medieval Welsh narrators sometimes switch quite effortlessly from in­direct quotation to direct quotation, apparently to emphasize the significant words that bear most emotional stress. There are some examples of this in Bown, most of them following the typical pattern identified in the native corpus. More importantly, they have no counterpart in the extant text of Boeve, and the majority appear to have been brought in artfully by the Welsh redactor to enhance the vividness of his narrative.25 An intriguing feature of Bown is the existence in it of two ‘regions’, ­characterized by a skewed distribution of linguistic and stylistic features. These include different preferences for morphological variants of verbal forms and prepositions, and, for region 2 in particular, a few isolated remains of the use of abbreviations (perhaps only for the name of the main character, Bown); the use of the loanword syr and of the religious interjection diolwch y 20 21 22

Cf., e.g., Bown 165–7, 662–4, 2873–5, 3108–10, 3783–5, 4301–4, 629–30, 3489–93. Cf., e.g., ibid. 664–6, 1422–6, 636–40, 3143–5, 371–4. Ibid. 1204–5, 2333–4. The Welsh syntax appears stilted; cf. also ibid., p. 105. A rare example of the Welsh narrator addressing his audience (ibid. 4182–3) reproduces the source (Boeve 3685). 23 Boeve 1146, 2051–2. 24 Cf. Bown 3493–4, 3869–71, 4031–2. 25 Cf. also Erich Poppe, ‘Slipping in some medieval Welsh texts: a preliminary survey’, Cahiers de l’ILSL, 18 (2005), 119–51.

42



Rewriting Bevis in Wales and Ireland

Duw (‘thanks to God’); the near-absence of compound adverbs; a preference for markers of direct speech unusual by the standards of the native narrative corpus; the rarity of cataphoric sef constructions; markers of transition closely modelled on the Anglo-Norman source; and a greater similarity in the scenes of combat to the Anglo-Norman text. Taken individually, these differences in distribution are probably not significant, but the collective evidence indicates a pattern that makes the differences between regions 1 and 2 appear neither fortuitous nor haphazard. The transition between the two regions is not sudden but gradual, and all features considered here characteristic of region 2 cluster, approximately, in the last third of the text. Apart from the peculiarities mentioned above, many of which are difficult to classify, the only consistent features of region 2 appear to be its greater closeness to the Anglo-Norman source and, arguably, a preference for a few narrative devices unusual by the standards of the native corpus.26 It is tempting to speculate that the differences between the two regions may reflect a decreasingly thorough revision of an earlier translation during its transmission.27 Our admittedly arbitrary dividing-line does not separate narrative units but, interestingly, it nearly coincides with the transition from the first part of Boeve to its second part at laisse clxvi, corresponding to Boeve’s decision to go to London in order to visit the king. Judith Weiss has suggested: The halves of Boeve de Haumtone (laisses i – clxv, and clxvi to end) are very different in character. There are numerous discrepancies between them in story, characters and style, and they are clumsily linked by a two-line prediction tacked on to the end of laisse clxv. People and episodes in the second half are referred to as if they had already been introduced earlier, when in fact they have not. In the first half, Boeve is an English hero; in the second, apparently French. Characters abruptly change allegiances between the two parts. Laisse style is handled mostly competently, occasionally very well, in the first half; in the second, it is merely a means of arbitrarily breaking up large hunks of narrative.28

It is, however, unlikely that the differences between regions 1 and 2 of Bown that we have identified, and those between the two halves of Boeve, are in 26

The two parts of Boeve were also treated differently by its Old Norse redactor, but he followed a different strategy: namely, greater distance from the source in the second part: cf. Christopher Sanders, ‘Bevers saga et la chanson anglo-normande Boeve d’Haumtone’, Revue des langues romanes, 102 (1998), 25–44 (p. 28): ‘Dans la première partie de la saga norroise, le nombre de déviations importantes par rapport à Boeve est remarquablement faible, mais dans la partie II, qu’on suppose être une addition secondaire, la situation est toute différente … Le récit n’en conserve pas moins ses traits essentiels, mais il semble que, de même que dans le cas du poème anglais Sir Bevis of Hampton, des inconsistances internes combinées avec le caractère souvent répétitif des événements ont amené le traducteur norrois à adopter un comportement plus indépendant que dans la partie I.’ 27 Alternatively one could think of the welding of two Welsh texts, but in this case the fairly extended transitional passage needs explaining. 28 Judith Weiss, ‘The date of the Anglo-Norman Boeve de Haumtone’, Medium Ævum, 55 (1986), 237–41 (p. 240).

43



Erich Poppe and Regine Reck

any way directly related, since the regions of Bown do not constitute different narrative units, as do the two halves of Boeve, but gradually fade into each other so that no exact dividing-line can be defined. The lexicon of Bown contains a number of Anglo-Norman loanwords, and nearly forty are thought to be first attested here. About thirty of these occur at the same place in the Anglo-Norman text. It is hence very likely that the Welsh redactor borrowed these directly from his source, which must have been quite close to the extant text(s) of Boeve. There are also a couple of emphatic negations in Bown which arose as imitations of Anglo-Norman idioms. A good example is the phrase werth vn uanec (literally, ‘value of a glove’), which is a direct translation of Boeve’s le vailant de un gant.29 Such emphatic negations have no counterpart in Welsh native texts and are best treated as loan phrases. A similar case is the quasi-formulaic expression ny chelaf / nyt ymgelaf ragot (‘I will not hide (it) from you’), which signals a speaker’s polite willingness to provide information. This is employed for three semantically matching Anglo-Norman expressions. Since to the best of our knowledge it does not occur in the native texts in the same formulaic way, it is another instance of a loan locution. There is, however, one significant difference between the Anglo-Norman model and its Middle Welsh realization, and this is the form of the address: whereas in Boeve the form of address in these phrases is consistently the second person plural, in Bown it is the second person singular.30 Although the Welsh redactor borrowed the basic phrase from Anglo-Norman, he changed the address to the form more natural to him. This is another illuminating example of the tensions between following the source and rewriting it that we often encounter in Bown. With regard to the relative distance between Boeve and Bown, it can be said in conclusion that the latter’s redactor followed the general principles formulated by his (near-)contemporary Gruffudd Bola ( fl. 1265–82) with reference to the translation of a religious core text, the Creed: ‘y troes i weitheu y geir yn y gilyd, a gveith ereill y dodeis synnvyr yn lle y synnvyr heruyd mod a phriodolder yn ieith ni’ (‘sometimes I turned the word into the corresponding one and at other times I gave the sense for the sense, according to the nature and diction of our language’).31 Bown is resolutely committed to the sense of the foreign source with regard to its plot and narrative intention, and at the same time intimately informed and moulded with regard to narrative

29 30

Bown 2006–9; Boeve 1796–7. Cf. Bown 680–2, 876–7, 1019, 2156, 3097; Boeve 670, 837, 976, 1922, 2661. There appears to be only one certain example in Bown of a second person plural as the form of address: here the first part of the Welsh sentence has a fairly close parallel, including the second person plural, in the extant text of Boeve, whereas the unparalleled second part reverts to the second person singular: cf. Bown 2107–11; Boeve 1887–8. The second person plural as form of address therefore seems to be modelled on Anglo-Norman usage. 31 Henry Lewis, ‘Credo Athanasius Sant’, Bulletin of the Board of Celtic Studies, 5 (1929–31), 193–203 (p. 196).

44



Rewriting Bevis in Wales and Ireland

techniques and strategies by the diction of its redactor’s language and literary tradition. The Early Modern Irish version of the story of Bevis of Hampton belongs to a second phase of Irish vernacular adaptations of foreign narratives with an interest in Continental matters during the fourteenth and fifteenth centuries – Charlemagne and English romances (William of Palerne, the Grail, Guy of Warwick, and Caxton’s version of the story of Hercules). The first phase of Irish adaptations, beginning probably as early as the tenth century, concentrated on texts about events in classical Antiquity, such as Dares’ Troy tale and Virgil’s Aeneid.32 The Irish version of Bevis is transmitted incomplete – it covers the events up to the killing of Earl Miles (Iarla Milis) by Josian (Sisian) and her threatened execution – and in only one manuscript, the second section of Trinity College Dublin, MS 1298 (formerly H.2.7), which has been dated to the second half of the fifteenth century. This section also contains Irish versions of the story of Hercules, based on Caxton’s Recuyell of the Historyes of Troye, and of Guy of Warwick, as well as a native tale, Stair Nuadat Find Femin.33 It was written by a well-known and prolific scribe, Uilliam Mac an Leagha. The number of formal and stylistic features shared by the four texts makes it likely that he was not the scribe only, but also the translator and the author. The same features have been noticed in five religious texts based on foreignlanguage sources in London, British Library, MS Add. 30512, also written by Mac an Leagha, and he is therefore thought to have been their translator as well.34 In the case of the nine texts, we may thus have the autographs of their redactor, Mac an Leagha. Since autographs of Irish texts are rare, this situation would provide a unique opportunity to analyse a process of adaptation without interference by later scribes and their re-interpretation of the original versions. But even if Mac an Leagha only copied, and perhaps revised, these texts, they nonetheless remain important evidence for the period’s literary mentalité. 32

For a useful brief survey, see Nessa Ní Shéaghdha, ‘Translations and adaptations into Irish’, Celtica, 16 (1984), 107–24. Our discussion here is based on Erich Poppe, ‘The Early Modern Irish version of Beves of Hamtoun’, Cambridge Medieval Celtic Studies, 23 (1992), 77–98, ‘Codes of conduct and honour in Stair Bibuis’, in Ogma: Essays in Celtic Studies in Honour of Próinséas Ní Chatháin, ed. by M. Richter and J.-M. Picard (Dublin: Four Courts Press, 2002), pp. 200–10, and ‘Narrative structure of medieval Irish adaptations: the case of Guy and Beues’, in Medieval Celtic Literature and Society, ed. by Helen Fulton (Dublin: Four Courts Press, 2005), pp. 205–29. 33 For these texts, see F. N. Robinson, ‘The Irish lives of Guy of Warwick and Bevis of Hampton’, Zeitschrift für celtische Philologie, 6 (1908), 9–180, 273–338, 556; Stair Ercuil ocus a Bás: The Life and Death of Hercules, ed. by Gordon Quin, Irish Texts Society 38 (Dublin: Irish Texts Society, 1939); Käthe Müller-Lisowski, ‘Stair Nuadat Find Femin’, Zeitschrift für celtische Philologie, 13 (1921), 195–250. An electronic version of Robinson’s texts and translations is supplied by the CELT project of the University of Cork at www.ucc.ie/celt/. 34 Cf. Stair Ercuil, ed. Quin, pp. xxxviii–xl; Diarmuid Ó Laoghaire, ‘Beathaí naomh iasachta i ndeireadh na meánaoise’, Léachtaí Cholm Cille, 15 (1985), 79–97.

45



Erich Poppe and Regine Reck

Mac an Leagha may have produced some of his manuscripts for a member of the Gaelicized Old English, or Hiberno-Norman, family of the Butlers, and the content of the manuscripts, as well as the decoration of at least one of them, seems to reflect the multi-cultural background, interests and tastes of this class.35 No name for the Irish version of Bevis can be deduced from the manuscript, and we therefore suggest Stair Bibuis, on the model of Stair Ercuil ocus a Bás introduced by Gordon Quin for the Irish adaptation of Caxton’s Recuyell in the same manuscript. The exact source of Stair Bibuis has not been identified, but Jennifer Fellows has argued that it was a Middle English version with close affinities to the texts of Bevis in Cambridge University Library, MS Ff.2.38, and Manchester, Chetham’s Library, MS 8009.36 Our comparison between source and adaptation will again be conducted on the levels of content, style and narrative structure. As in the case of Bown, the storyline of the source is quite faithfully preserved in Stair Bibuis, which allowed Fellows to establish the correspondence between the latter and two Middle English texts. There are, however, some minor changes on the level of content that reflect the redactor’s attempts to accommodate the expectations and experiences of the new audience, as well as his specific interests or his patron’s. Irish perceptions of the position of women and of the importance of hospitality may have conditioned some subtle changes in Stair Bibuis. Its emphasis on codes of conduct and honour, which the extant English versions lack, may be related to its redactor’s attitudes and preoccu­ pations, and perhaps also to those of his patron.37 When Brámon, the king of Damascus, demands the Egyptian king’s daughter, Sisian, in marriage, her father asks her in the Irish text if she will consent to be given to him, and she says no. In the Middle English version the king’s daughter is not given any choice;38 it is never even considered whether she might want to marry 35

For a discussion and further references, see Poppe, ‘Codes’, pp. 208–10, and, with some modifications, his ‘Stair Ercuil ocus a bás – rewriting Hercules in Ireland’, in Translations from Classical Literature: ‘Imtheachta Æniasa’ and ‘Stair Ercuil ocus a Bás’, ed. by Kevin Murray, Irish Texts Society, Subsidiary Series 17 (London: Irish Texts Society, 2006), pp. 37–68 (pp. 38–9). 36 See Jennifer Fellows, ‘Sir Beves of Hampton: Study and Edition’, 5 vols (unpub. Ph.D. diss., Univ. of Cambridge, 1980), I, 51. There are at least two passages where the Irish text appears closer to the English version in the Auchinleck MS (A) than to the version in the Chetham’s MS (M). First, during the fight with his rival, Bibus’s father asks for his life and the lives of his wife and son (see Robinson, ‘The Irish lives’, p. 275), the closest parallel to this in the extant English version being in A: see The Romance of Sir Beues of Hamtoun …, ed. by Eugen Kölbing, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) (hereafter BHK), p. 12. Secondly, in the first conversation between Bevis and Ermin, the statement that Ermin knew Bevis’s father and his reputation occurs only in M, but not in A or in the Irish version: cf. BHK, p. 26; Robinson, ‘The Irish lives’, p. 277, lines 31–3); see also Poppe, ‘Codes’, p. 206. 37 See Poppe, ‘Codes’, for a fuller discussion. All other manuscripts produced by Mac an Leagha are devotional in content. 38 The question of the king’s daughter’s choice does not arise in the Anglo-Norman, the Welsh or the Old Norse version: cf. Boeve 515–25; Bown 513–26; Bevers saga, ed. by Christopher Sanders, Stofnun Árna Magnússonar á Íslandi 51 (Reykjavík: Stofnun Árna Magnússonar á Íslandi, 2001) (hereafter Bs), p. 61.

46



Rewriting Bevis in Wales and Ireland

­ rademond or not. That the Irish redactor thought it conceivable that the B king’s daughter might have a view on, or even a choice in, the matter indicates his perception of the social and legal rights of an Irish noblewoman.39 Another episode of the English source seems to have been changed because of the importance of hospitality in Irish society. Having escaped from the dungeons of the king of Damascus, Bibus comes to a castle, asks for hospitality and food, and, of course, receives it. As it turns out, the owner of the castle is the brother of Grainnder, a knight Bibus had killed on his flight, and in the now ensuing fight between Bibus and the owner of the castle the latter is killed. In the English version, however, Bevis is not offered hospitality; instead he is advised to leave the castle, because its owner is a giant who hates Christians and will kill them immediately. The motivation for the fight between the castle’s owner and Bevis is different, then: in the English version it is only by fighting and killing him that Bevis can get the necessary food – that Bevis had earlier killed his brother enrages the giant even more, but this is not the primary motivation for their fight, as it is in the Irish version.40 In the same scene the giant accuses Bevis/Bibus of having stolen his brother’s horse. In the English version, Bevis ignores the accusation and tells the gruesome story of how he killed Grander by breaking open his skull and thus ‘made him prest’.41 Note that this joking remark, implying that as a result Grander now wears a tonsure, is not taken over by the Irish redactor. In the Irish version, Bibus explicitly rejects the accusation and denies having stolen Grainnder’s horse, but admits at the same time that he killed Grainnder and took his horse afterwards, thus presumably insisting on a subtle difference between theft and honorable acquisition through fighting.42 This is only one of a number of instances in Stair Bibuis in which concepts of conduct and honour and of moral standards appear to play a significantly more important and explicit role than in the known English texts. One or two such scenes would not suffice to argue that these are the redactor’s additions, but their accumulation might. It is clear, however, that such minor modifications do not impinge on the general progress and character of the storyline. Changes on the level of style are much more obvious. The highly ornamented, florid style of Stair Bibuis, with its alliterating phrases and strings of (near-)synonyms, reflects the stylistic norms of the receiving textual culture and is employed in Irish texts from the eleventh century onwards. It gives Stair Bibuis – as well as Mac an Leagha’s other adaptations – a distinct Irish flavour.43 It is found mainly in extended rhetorical set pieces such as descrip39 40 41 42 43

For further discussion, see Poppe, ‘The Irish version’, pp. 92, 95. For further discussion, see ibid., pp. 93–4. BHK, p. 94. For further discussion, see Poppe, ‘Codes’, p. 207. Cf. Poppe, ‘The Irish version’, pp. 82–7; Robinson, ‘The Irish lives’, p. 19: ‘the manner of the narrative is thoroughly Irish, and [the Irish texts] read in general like the native stories in the somewhat ornate prose of the period. The accumulation of adjectives and adverbs, often in alliterating groups of three, is characteristic of late Middle Irish, and the “Guy” and “Bevis” are by no

47



Erich Poppe and Regine Reck

tions of persons and battles and in other passages in which narrative speed is slowed down, but also in a variety of shorter passages and even single sentences. With regard to the Irish redactor’s treatment of the narrative structure of his Middle English sources, three different strategies emerge: namely, retention, suppression and accommodation. All three strategies must be seen as arising from a creative tension between source and native narrative conventions. An instance of retention of a structural feature of the source is the use of direct speech and dialogue in the Irish version. The redactor’s native narrative models, as well as his source, favour their employment, and the adaptation therefore replicates the source with insignificant differences.44 On the other hand, the presence of the narrator’s persona and his frequent explicit intrusion in the Middle English romances are not paralleled in native medieval Irish narratives, and this characteristic feature of the Middle English texts is accordingly suppressed by the redactor. He does not address the audience; nor does he comment on the plot-line or express emotions. He may be omniscient, but even his omniscience appears somewhat less explicit than in the Middle English texts. In many scenes in which the Middle English narrator shows access to his characters’ thoughts, the Irish protagonists’ minds are not presented from within; instead, the protagonists act or speak.45 The Middle English narrator’s prologue is replaced by the straightforward introduction of one of the leading characters, Bibus’s father – as in Bown and the Old Norse Bevers saga: ‘Bui iarla saidhbir, socarthanach a Saxanaibh doshinnrudh diarba comainm Sir Gyi o Hamtuir’ (‘There was a rich and charitable earl in England whose name was Sir Guy of Hamtoun’).46 It is to the borderline between the levels of content and of narrative structure that we would assign the change from Bevis’s monologue in the English version, about his appropriate course of action during the wedding-feast of his mother and her second husband, to an exhortation to act by his fellowswineherds in the Irish version.47 Psychological motivations are often externa­ lized in Irish narratives and take the form of a reproach for some socially unacceptable behaviour, such as cowardice. The differences between the two versions would seem to be conditioned by this narrative convention.

44

45 46

47

means extreme examples of the practice.’ Cf. also Quin on Stair Ercuil: ‘A tale which except for its central theme is thoroughly Irish’; ‘Although these [descriptions] are based on corresponding descriptions in Caxton they are purely Irish in diction and style’ (Stair Ercuil, ed. Quin, pp. xxvi, xxx). At least one instance of slipping from indirect quotation into direct quotation has been identified in Stair Bibuis: cf. Uáitéar Mac Gearailt, ‘Slipping as a narrative device in Irish’, Cahiers de l’ILSL, 18 (2005), 153–82 (p. 175). For further discussion with examples, see Poppe, ‘Narrative structure’, pp. 211–17. Robinson, ‘The Irish lives’, p. 273; cf. Bs, p. 3: ‘Gvion het einn ríkr jarl áá Einglandi hann sat þar sem Hamtun heiter’ (‘Gvion was the name of a mighty earl in England, he lived in a place called Hamtun’). See Poppe, ‘The Irish version’, p. 94; this monologue is retained in Bown 261–6, but not in Bevers saga, where it is replaced by immediate action: see Bs, p. 27.

48



Rewriting Bevis in Wales and Ireland

A simple but decisive example of structural accommodation to native conventions is the change in the macro-form from English verse to Irish prose, in line with the favoured native format for narrative. A more complex instance is the demarcation of episodes and sub-episodes, which is a narrative necessity in longer texts. A variety of devices is employed in the Middle English romances for this purpose, but the Irish redactor prefers, although not consistently, a format he developed on the basis of formal options available to him in native Irish prose: dala (‘concerning’) plus personal name to mark the beginning of an episode, and conidh (‘so that it/this is’) plus a one-term or one-sentence summary to mark its end, as in the following example: ‘Dala na luinigi … Conidh amlaidh-sin dosaeradh Bibus ar marbadh na ridiri’ (‘As for the boat … Thus Bibus was freed after killing the knights’).48 The Irish redactor of Stair Bibuis used the English romance of Bevis as his source but rewrote it in order to accommodate it to native literary expectations. His treatment of his source’s narrative structure in particular indicates that Irish narrative models were his yardstick. The motivation for the strategies of retention, suppression and accommodation is related to the relative proximity, or distance, of structural features of the source to structural features of native narrative. Even where he seems to have exercised some creative freedom, his inspiration is derived from native conventions rather than from the foreign source. He committed no obvious lapses as did the redactor of Bown. In a comparative perspective, a number of similarities between Bown and Stair Bibuis emerge, despite their different sources (Anglo-Norman and Middle English respectively) and their different dates. These are due, first, to their redactors’ shared approach to the adaptation of foreign narratives – namely, their intention to rewrite their sources in order to accommodate them to their, and their audiences’, literary expectations – and, secondly, to similarities in the narrative conventions of the receiving textual cultures: for example, prose as the medium for narrative, and the inaudibility of the narrator.49 In both cases, the relative distance between source and adaptation is small with regard to the plot and its message, but wide with regard to narrative presentation. In other words, the story of Bevis remains foreign and preserves its exotic appeal for its new Irish and Welsh audiences, while at the same time allowing identification with the Christian martial ethos; the narrative strategies and techniques employed become distinctly native, Irish and Welsh respectively, by a process of thorough rewriting. Finally, we should like to suggest that at least two factors contributed to the approach to literary transfer, or ‘translation’ in a wide sense, from one linguistic and cultural context to another, as it is reflected in Bown and Stair Bibuis: their 48

Robinson, ‘The Irish lives’, pp. 277, 279. See also Poppe, ‘The Irish version’, pp. 81–2, and ‘Narrative structure’, pp. 217–21, for further discussion and examples. 49 Further comparison with the Old Norse Bevers saga would be instructive.

49



Erich Poppe and Regine Reck

redactors’ strong confidence in the forms and conventions of their own narrative traditions, and the medieval attitude to textual authority. Recent discussions of the status of medieval texts, informed by the insights of the ‘New Philology’, stress that the manuscript exemplar did not automatically enjoy canonical status which guaranteed its unchanged and fixed transmission, and we believe that the same would apply to a foreign-language source in the process of its translinguistic and transcultural transfer, translation or adaptation – whichever modern term we adopt.50

50

We wish to thank Ceridwen Lloyd-Morgan and Oliver Padel for their generous help with earlier versions of this paper.

50

4 Bevers saga in the Context of Old Norse Historical Prose CHRISTOPHER SANDERS

I

n a forthcoming paper, I venture the following description of the difference between the Anglo-Norman Boeve de Haumtone (Boeve) and the Old Norse Bevers saga (Bevers): Boeve, a form of dynastic entertainment in which some of the exoticism of the East was quite well-combined with the material and status interests of the local British aristocracy – quite gracefully mixed with satire and fun inherent in an oral poetic tradition – becomes in Bevers an essentially more Christian chronology of one man’s life, in which the hero’s gradual regaining of material rights and enhancement of his personal status is more a feature of his own individual prowess and is more reliant on God’s support and approval; these features lend the tale a somewhat ‘exemplum-like’ tone, highlighting potentially exemplary personal development in a more emphatically outlined Christian frame. Elements that were clear potentials in Boeve are developed – as a biography of one man’s life, perhaps as history – but a great deal is lost in the process.

Here I shall consider this description and ask how such a change between the two text forms may have come about. Bevers, a riddarasaga (‘saga of a knight’), survives in manuscripts and fragments, all of them Icelandic, rather than Norwegian, ranging from vellum manuscripts from c. 1350 (a fragment), to paper manuscripts, some eighteen in all, from as late as 1900. There are three medieval versions, one of which (the Ormsbók text) is aberrant and will not be discussed here. The two principal manuscripts of the main tradition are from c. 1400 and c. 1470 respec-



‘Old Norse’ is deliberately used here, as is usual in British English, of the common languages and cultures of medieval Iceland and Norway; this usage is particularly appropriate in cases such as Bevers saga, where the issue of provenance should justifiably remain open.  Christopher Sanders, ‘A typology of the primary texts of Bevers saga’, in Rittersagas – Übersetzung, Überlieferung, Transmission, ed. by Jürg Glauser and Susanne Kramarz-Bein, Beiträge zur nordischen Philologie 43 (Tübingen; Basle: Francke, forthcoming).  Ibid.

51



Christopher Sanders

tively,  and the latter is slightly more rhetorical in style than the former. The saga is, of course, in prose and shows the features that are typical of the Old Norse renderings of French verse romances and of works based on chanson de geste, and, to some extent, regular features of certain Old Norse translations of historical works in Latin (see below). A dating for Bevers will be considered at the end of the chapter. Among such typical features, most of them witnessed in the Bevers texts, are the following: pronouns are substituted for names or nouns, and viceversa; the sequence of events is made more logical – gaps and sudden leaps in the narrative are filled; direct speech is substituted for indirect speech, and vice-versa; there is a tendency to reduce exaggerations; there is a reordering of the sequence of the lines. Typical omissions are: laisses similaires; long descriptions of clothes or landscape; anticipations and repetitions; formulaic epithets, as in Josiane la bele; interjections from the narrator. Typical additions are: explanations of features that were implicit in a context but not made clear; narrative links; the insertion of ‘missing’ inquits, making clear who said what; formalized beginnings and ends of episodes. One notes in this outline a common denominator: an insistence on clarity and sequentiality – a policy of emphasizing the storyline, often at the expense of other qualities. In my own approach to Bevers, I owe much to the influence of Judith Weiss’s 1968 doctoral thesis, which made me see the importance of recognizing the bipartite structure of the poem before going on to an analysis of subsequent translation into another language. It seems that the Norse translator reacted rather as the Middle English redactor did, in such a way that the more sprawling and disorganized nature of the last one-third of Boeve apparently inspired more change and adaptation than is evident in the preceding sections. The translation into Old Norse is often very close and it is competent – there are very few misunderstandings. A number of passages are cut back significantly. Typically, they are those that culturally were not easily transferable to the non-feudal North: Bevers’ instatement as a liege lord, for example, on one of his returns to England, and the subsequent celebratory tournament



These texts are presented, in parallel with Boeve, in Bevers saga, ed. by Christopher Sanders, Stofnun Árna Magnússonar á Íslandi 51 (Reykjavík: Stofnun Árna Magnússonar á Íslandi, 2001) (hereafter Bs). All references to Bevers will be to this edition, but with normalized spelling; o caudate is rendered as ö. On the suitability of the surviving Anglo-Norman text for comparison with the Old Norse, see Christopher Sanders, ‘Bevers saga et la chanson anglo-norman de Boeve d’Haumtone’, Revue des langues romanes, 102:1 (1998), 25–44.  Lists of these categories (‘changes’, ‘omissions’ and ‘additions’) are classic in discussions of Old Norse translations from Old French; the tradition can be traced back to Rudolf Meissner, Die Strengleikar: Ein Beitrag zur Geschichte der altnordischen Prosaliteratur (Halle: Niemeyer, 1902).  Judith E. Martin, ‘Studies in Some Early Middle English Romances’ (unpub. Ph.D. diss., Univ. of Cambridge, 1968). I gratefully acknowledge Dr Weiss’s generosity in allowing me to photocopy parts of her thesis.

52



Bevers saga and Old Norse Historical Prose

in which his horse Arundela runs so unfortunately well, inspiring the inevitable envy. Apart from minor and middle-sized adaptations, there are three major changes in the saga in relation to the Anglo-Norman, all of them in the latter part of the narrative. The recapture of Arundela is expanded and given more prominence, as is the presence of the second horse, Arundela’s foal, and this gives rise to the second major change – the description, not found in the surviving Boeve, of preparation for an extra battle-episode in which Yvori (Ivorius in the Norse) seeks revenge for the loss of these precious steeds. Thirdly, in the duel which is intended to resolve the long-standing rivalry and conflict between Ivorius and Bevers it is, surprisingly, the son of Bevers, Gvion, who intervenes, sensing that his father will not manage to vanquish his opponent. It is he and not Bevers himself who finally downs the heathen, and this understandably gives rise to great annoyance on the part of his father. More generally, there are patterns of changes that are recurrent enough to be called thematic. First, there is a tendency to be more courtly than the Anglo-Norman source. This operates, however, on a verbal rather than a narrative level: there are references to people’s knightliness and to their courtliness, mainly by means of the two adjectives hǿverskr and heyskr (both meaning ‘courtly’) and associated words, yet this is varnish – nothing that makes an inherent difference to the development of the story. Thus when Josivena first serves food for Bevers: ‘hon þjónaði honum með öllum alhuga ok hǿversku, þvíat hon var bǽði fögr ok listug’ (‘she served him with great care and courtesy for she was both beautiful and accomplished’). Secondly, and perhaps surprisingly, given that the Anglo-Norman ­original is closer to a chanson de geste than to a roman courtois, but probably in connection with the thematic tendency towards some sort of courtliness, there is a slight strengthening of the theme of the love between the leading characters, Bevers and Josivena, especially Josivena’s love for Bevers. So, for example, there is the following addition after Boeve 667, which continues the passage just quoted: ‘ok gaf honum at drekka með leyniligum ­þvinganarekka’ (‘and she served him drink with secret sighs of heaviness’).10 Yet this particu­lar emphasis may purely be a preparation for the reprobation Josivena later receives for her forward conduct. Probably to be seen in tandem with this indirect promotion of Bevers’ role is a tendency to emphasize the virtues and underplay the shortcomings of the hero, thereby missing or omitting some of the humour – a policy, as in other instances, that is not consistently followed, but which has been noted in other translations to Old Norse from Old French. Thus when Bevers has 

All three of these changes, which relate to Boeve 3468–515, 3515ff., and 3582–624, are discussed in some detail in Bs, pp. 376–8.  For further details of these largely thematic changes, see Sanders, ‘A typology of the primary texts of Bevers saga’, which is the basis for the introductory sections of the present article.  Bs C7.98–100 (p. 73); there is no equivalent to this in Boeve. 10 Bs C7.100–1 (p. 73).

53



Christopher Sanders

undertaken command of Erminríkr’s troops,11 we find after Boeve 515: ‘en þar fannsk engi sá í konungshirð at á hendr vildi takask at vera forstjóri fyrir liðinu, því at þar váru allir hrǿddir ok huglausir’ (‘there was no one at the king’s court who was prepared to be the commander of the troops because everyone there was frightened and uncourageous’).12 Other characters’ fear of Bevers is frequently enlarged on, and passages that entail some potential criticism of Bevers’ rather boorish and insensitive behaviour are certainly not reproduced. At the same time as Bevers is being boosted, there is a playing-down of Josivena’s active role as the wooer, and all her activities as a musician, her rote-playing, are ignored – they are simply omitted, for her intended role in the piece was apparently not sympathetically received.13 Generally, the humour of the Anglo-Norman original is missing; notably absent here is the comedy that surrounds the giant page Escopart, his appearance and behaviour, together with the small jokes that are made at the expense of the clergy, and the humour associated with Josiane’s behaviour in wooing Boeve. There is also a certain intensifying of the Christian tone of the epic – and therefore, to some extent, of an anti-Islamic, or anti-pagan, undercurrent, beyond any to be found in the Anglo-Norman original.14 Perhaps connected to this, there is an inextinguishable delight in battle scenes – so much so that in the latter part of the narrative there is a description of preparation for an extra battle-episode (mentioned above), calqued on earlier battle sequences and unique to the Norse Bevers.15

11 12 13

Erminríkr is the Hermine of the Anglo-Norman, king of Egypt and father of Josivena. Bs B7.12–14 (p. 61). Josivena’s behaviour in wooing Bevers is twice given extra condemnation in Bevers: apparently as a replacement for Boeve 687, there is the comment: ‘ok þér gerið þat bernsliga at þér beiðisk þvílíkra hluta’ (‘and you behave like a child in asking for such things’, Bs C8.12 (p. 75)), which may be compared to the way in which ‘que ele out mesfet dunt ele out ledengé’(Boeve 723) is rendered by ‘ok grét nú mjök ok iðraðisk mjök sinna misgerða’ (‘and cried now heavily and greatly repented her misdeeds’, Bs C8.37 (p. 79)), which intensifies the self-recrimination of the original. None of the references to Josiane’s rote-playing (Boeve 2784–5, 3029–30, 3100– 1) is reproduced in Bevers. It is, of course, conceivable that these are later additions to the Anglo-Norman tradition, as Stimming, the editor of the Anglo-Norman, himself implies of lines 3029–30: see Der anglonormannische Boeve de Haumtone, ed. by Albert Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899), p. clxviii). Yet they seem to be well integrated in the Anglo-Norman epic’s portrayal of the heroine as a person of initiative (cf. Judith Weiss, ‘The wooing woman in Anglo-Norman romance’, in Romance in Medieval England, ed. by Maldwyn Mills, Jennifer Fellows and Carol M. Meale (Cambridge: Brewer, 1991), pp. 149–62), and this feature of the narrative may well have been culturally difficult (for a Nordic cleric?). 14 Apart from phraseological changes and additions, there are two considerable examples of stronger religious influence on Bevers than on Boeve. The first instance concerns the desecration of the heathen god, Terugant (cf. Bs, p. 378), and the other is the very end of the saga, where the Christian setting of the deaths of Bevers and Josivena is emphasized. Stimming mentions a third, the Pope’s coronation of Bevers (Boeve, p. cxii), but the main addition he mentions in this connection, after Boeve 3697, is in fact an anticipation of lines 3712–14. 15 Bs B33.73–86 (pp. 329, 331); cf. ibid., pp. 376–9.

54



Bevers saga and Old Norse Historical Prose

Long speeches and the occasional monologue are made major opportunities for summarizing the text of the original (although much direct speech is actually retained and reproduced as such, rather than being represented by oratio obliqua). Both these features are perhaps so typical of translations from Old French to Old Norse that they should be described as genre shifts rather than thematic tendencies. While tracing these tendencies, it is necessary to note that they are not altogether consistent and do not seem to pull together to contribute to a clear interpretative view of the work as a whole: while, for example, there is an apparent interest in strengthening the theme of Christian combat against the infidel, passages of Christian content, such as invocations of the Virgin Mary and long prayers, can also be omitted – perhaps because they were regarded as repetitive. What seems the most extreme example of lack of consistency is, indeed, the duel scene in the latter part of the poem, where it is Bevers’ elder son Gvion, rather than Bevers himself, who secures the decisive duel against the significant rival, Ivorius – a rewriting which is completely out of keeping with the tendency, described above as being a feature elsewhere in the text, of enhancing rather than belittling Bevers’ status and standing. It was mentioned earlier that there are many passages which are carefully translated and close to the surviving Anglo-Norman original. Thus a revised passage, such as the representation of the hero’s escape from the Damascus dungeon, can immediately be followed by a passage of close and uncomplicated rendering.16 The closely reproduced passages, in terms of their basic respect for the original text, are symptomatic of clerical training, and the more adaptive passages may also be related to the same cultural background. The relative insistence in the saga on explanation and logicality on the microlevel (i.e. the linking of one sentence to another, and the managing of the transfer from one episode to another, ensuring always that the audience is not uncertain about what is happening and why) can be seen as belonging to a mindset imbued with a deep and fundamental respect for the text itself and with a concern to communicate and explicate the contents – a mindset formed by biblical exegesis of the type that clerics would have been trained in during their preparation for ordination, or that monks would have experienced during their Bible studies.17 16 17

See the commentary in Bs, p. 370. In the appendix to Sanders, ‘A typology of the primary texts of Bevers saga’, the translation of a laisse of Boeve is presented alongside a reconstruction of an archetype of Bevers, and the textual differences are discussed in detail. One of the conclusions drawn from this comparison is that the manner in which the whole laisse was apparently read through in detail and fully comprehended before it was translated could well be related to habits developed through the reading of marginal glosses of Scripture. These observations can be related to much that is brought to the forefront in Rita Copeland, Rhetoric, Hermeneutics, and Translation in the Middle Ages: Academic Traditions and Vernacular Texts (Cambridge: Brewer, 1991). Copeland writes in great detail about the way in which exegesis and rhetoric work together at various stages in the Middle Ages, but, as she states in her introduction (p. 6), she is concerned with ‘commentaries from the arts curriculum, not scriptural exegesis’.

55



Christopher Sanders

The mindset being described here in terms of tactics of communication and explanation fits conveniently with what might be called the ‘narrative grammar’ of native Icelandic saga-writing (scene-openers, linking passages between episodes, careful presentation of new characters on their first appearance in the narrative). It has long been assumed that it was clerics, or monks, who translated from Old French to Old Norse, as only they would have had the educational background to do so and were the most likely social group to be commissioned for manuscript work of this type. We know, however, remarkably little about the education of Nordic priests; there is only limited evidence of them travelling to France or England, but no reason to assume that they did not do so. Any foreign student aspiring to a theological education at a cathedral school in France (or Anglo-Norman England) would have had to study biblical exegesis, probably in the form of the Glossa ordinaria.18 In Iceland, schools for priests were slow to develop, and a good deal of clerical education may have taken the form of private instruction; the import­ ance of the Benedictine and the Augustine monasteries as centres of learning should not be underestimated.19 Biblical exegesis could traditionally be fourfold – literal, allegorical, tropological and anagogical – but in the Glossa ordinaria the inventory was frequently reduced to three categories – historical, allegorical, and moral; there would, regardless, have been opportunity to acquaint oneself with many types of exposition. Given our uncertainty as to the extent of learning in the medieval North, it is perhaps best not to imagine acquaintanceship with particular theories of translation, such as Cicero’s non verbum pro verbo on the one hand, and a stricter literalism as advised by Jerome in connection with translation of Scripture on the other.20 Instead, it might be safer to concentrate on habits of mind and thinking, such as those induced by frequent exposure to biblical exegesis. To return to Bevers: the witnesses of the main textual tradition show on the one hand, in the largely thematic tendencies mentioned earlier, something we can describe as a deliberate, but not consistently pursued, literary endeavour, dependent on some sort of genre or ‘literary-type’ awareness – 18

For references concerning the Glossa ordinaria, its usage, development and dating, see Sanders, ‘A typology of the primary texts of Bevers saga’. 19 On the education of clergy in medieval Iceland, including education abroad, see Hjalti Hugason, Frumkristni og upphaf kirkju, Kristni á Íslandi 1 (Reykjavík: Alþingi 2000, 2000), pp. 225–32; on the monasteries, see Gunnar F. Guðmundsson, Íslenskt samfélag og Rómakirkja, Kristni á Íslandi 2 (Reykjavík: Alþingi 2000, 2000), pp. 212–25. Concerning potential study visits abroad from either Norway or Iceland, we know most about the twelfth century – the period before translations from Old French are to be expected. There is regrettably no mention of a monk from a Scandinavian cloister in Thomas Sullivan, OSB, Benedictine Monks at the University of Paris a.d. 1229–1500: A Bibliographical Register, Education and Society in the Middle Ages and Renaissance 4 (Leiden; New York; Cologne: Brill, 1995) – a work which does not, however, claim to be exhaustive. 20 See Rita Copeland, ‘The fortunes of “non verbum pro verbo”: or, Why Jerome is not a Ciceronian’, in The Medieval Translator: The Theory and Practice of Translation in the Middle Ages: Papers Read at a Conference Held 20–23 August 1987 at the University of Wales Conference Centre, Gregynog Hall, ed. by Roger Ellis et al. (Cambridge: Brewer, 1989), pp. 15–35.

56



Bevers saga and Old Norse Historical Prose

whoever produced the surviving texts of the main tradition was presumably acquainted with a good deal of chivalric literature that already existed in Old Norse, and must have had a clear idea of the vocabulary and stereotypes with which it operates; yet this is something rather superficial, for, on the other hand, if one were to attempt more fundamentally to characterize the Old Norse text, the adjectives that readily come to mind are ‘plodding’ and ‘pedestrian’. This rather unflattering element in the textual profile could also be viewed in connection with another feature that will only be hinted at here: that the movement from the Anglo-Norman to the Old Norse could be described as a development from a Bakhtinian dialogic to a monologic text form, in the sense that there is a movement from a text that is capable of maintaining a distance from itself to a text that lacks such a self-­awareness.21 Recent scholarship on the riddarasögur as a group does not have a great deal to offer here, as it has tended to concentrate on the Arthurian and more romance-like material. This group consists of a number of lais largely attributed to Marie de France (Strengleikar), three translations of works by ­Chrétien de Troyes, Erec et Enide, Yvain and Perceval (Erex saga, Ívens saga and Parcevals saga), a rendering of the Thomas Tristan (Tristrams saga), a version of the fabliau Le lai du cort mantel (Möttuls saga), and a form of the Floire et Blancheflor tale (Flóress saga ok Blankiflúr); there is a translation of a version of Partonopeus de Blois (Partalopa saga), presumably based on a non-extant French text. For the type of French original closer to chanson de geste, there are versions of an Elie de Saint-Gille (Elíss saga ok Rósamundar); a Flóvents saga, almost certainly based on a French text now lost (probably a version of Flovant neveu de Constantin); and renderings of texts, including the Chanson de Roland, which make up the cyclically constructed Karlamagnúss saga.22 Þiðreks saga af Bern, which relates to German rather than French material, is also relevant; it too deals in a compilatory manner with the past in ways that are distinctly similar to ­ Karlamagnúss saga, but its background and development are too complex to be gone into here.23 In terms of literary comment, two schools of thought, focused mainly on the Arthurian material, developed in the 1970s and 1980s, and they have not been eager to compromise. Geraldine Barnes stressed the educational, didactic

21

This comment should probably be placed in perspective by reference to Sarah Kay’s listing of a tendency shown by medieval scholars to lend the works they study some addditional value by assigning them ‘dialogic status’: cf. Sarah Kay, The ‘Chansons de Geste’ in the Age of Romance: Political Fictions (Oxford: Clarendon Press, 1995), p. 4. 22 For a general introduction to the translated riddarasögur, see Geraldine Barnes, ‘Riddarasögur’, in Medieval Scandinavia: An Encyclopedia, ed. by Phillip Pulsiano et al., Garland Encyclopedias of the Middle Ages 1; Garland Reference Library of the Humanities 934 (New York; London: Garland, 1993), pp. 531–3; Jürg Glauser, ‘Romance (translated riddarasögur)’, in A Companion to Old Norse–Icelandic Literature and Culture, ed. by Rory McTurk, Blackwell Companions to Literature and Culture 31 (Oxford; Malden, MA: Blackwell, 2004), pp. 372–87. 23 For a recent major study, see Susanne Kramarz-Bein, Die ‘Þiðreks saga’ im Kontext der alt­norwegischen Literatur, Beiträge zur nordischen Philologie 33 (Tübingen; Basle: Francke, 2002).

57



Christopher Sanders

and, to some extent, hagiographic leanings of the Nordic translations; while Marianne Kalinke emphasized the entertainment function of these texts.24 The most successful attempt to date to reconcile the two positions is an article by Susanne Kramarz-Bein on Parcevals saga, in which she finds a balance between an ideological and a humorous intent – an element which in some respects was already present in Chrétien de Troyes’s Perceval.25 Crudely, in the case of Bevers one could say that, although some literary awareness is demonstrated in the process, what is principally lost in translation is what we might today regard as its inherent literary characteristics.26 From a work that can be regarded as being principally a piece of fictive entertainment (with many social implications), the text has become more an annal of events, a type of history or biography. The transformation here is, then, partly a move from mimesis to chronology. The clerics or monks who were involved in the translation (and/or editorial?) process probably saw their task primarily as being directly faithful to the single text before them, yet additionally a sense of responsibility towards some sort of truth that lay beyond and behind the text might be the background for the major changes listed above in the last one-third of the saga.27 Bevers as it appears now seems, in other words, when we peel off its courtly varnish, to be under two main influences: a commitment to strict faithfulness to the original text on the one hand, and an openness to explanation and interpretation, or amplification (such as the new battle-episode) on the other. What might the Old Norse background be for this? The relationship between a concept of truth, a concept of history, and enjoyment of something more ‘fictive’ or ‘story-like’ is complex in the medieval period. I would maintain that the Old Norse situation in this regard, though subject to common European influence, is at the same time a special one. As the title of this chapter suggests, I am particularly concerned to place Bevers in the context of the large volume of Old Norse works that can be regarded as historio24

For a rehearsal of various stages in the debate betwen these two positions, see Geraldine Barnes, ‘Some current issues in riddarasögur research’, Arkiv för nordisk filologi, 104 (1989), 73–88. 25 Susanne Kramarz-Bein, ‘Höfische Unterhaltung und ideologisches Ziel: Das Beispiel der alt­norwegischen Parcevals saga’, in Die Aktualität der Saga: Festschrift für Hans Schottmann, ed. by Stig Toftgaard Andersen, Ergänzungsbände zum Reallexikon der germanischen Altertumskunde 21 (Berlin; New York: de Gruyter, 1999), pp. 63–84. 26 Here there is a link to recent theorizing, since this line of thinking is similar to Gerd Wolfgang Weber’s approach. His focus is largely on the romances rather than the chansons de geste, but when he writes that there was a stripping-down to ‘traditional story-patterns which were auto­ chthonous: the fornaldarsaga, the fairy tale and the mythic story patterns’, he is concerned with an analogous ‘down-grading’: cf. Gerd Wolfgang Weber, ‘The decadence of feudal myth: towards a theory of ridarasaga and romance’, in Structure and Meaning in Old Norse Literature: New Approaches to Textual Analysis and Literary Criticism, ed. by John Lindow, Lars Lönnroth and Gerd Wolfgang Weber, Viking Collection 3 (Odense: Odense University Press, 1986), pp. 415–54 (p. 425). For some objections to his view, see Barnes, ‘Some current issues in riddarasögur research’, pp. 78, 82–3. 27 The issue of the extent to which the Bevers texts may have been revised in the course of their transmission will not be discussed here; for a brief outline of the possibilities, see Sanders, ‘A typology of the primary texts of Bevers saga’.

58



Bevers saga and Old Norse Historical Prose

graphic, and in this connection it is not inappropriate to look at the Íslendingasögur (‘sagas of Icelanders’). Preben Meulengracht Sørensen, in his Fortælling og ære (‘Story-telling and Honour’),28 sees the sagas of Icelanders, Njáls saga, Gísla saga, Grettis saga, Eyrbyggja saga, etc., as being balanced on a razor-edge between what he readily regards as two poles, history and literature; only when the narrative’s function and indebtedness to a historical purpose have been understood can the modern reader begin to treat and appreciate them as literature – that is to say, as fiction. This approach naturally gave rise to a certain amount of illfeeling and debate amongst scholars who felt that an anthropological school of thought was stealing its texts, but it also seems to have prompted others to proceed along similar lines. In the conclusion to his study of Eyrbyggja saga, Klaus Böldl explains the lack of clear composition in the work as the result of direct conflict between its political purpose (political legitimation of the Sturlunga family) and elements of tradition (oral material as accepted history) that it must incorporate.29 It can be argued, in other words, that the central, or at least best-known, body of Old Norse literature is itself poised, apparently somewhat precariously, on a historical pivot. I mentioned above that there are apparently two main tendencies to be looked at in the case of Bevers: respect and precision on the one hand, some interpretative licence on the other. It is in terms of these tendencies that I wish now to relate Bevers to historiographical writing in Old Norse. I shall therefore look briefly at the latter’s background and development. Some of the earliest Old Norse historical writings, Norwegian in origin and penned in Latin (such as Theodoricus monachus’ The Ancient History of the Norwegian Kings, and the anonymous Historia Norwegie), had no difficulty in combining an ostensible historical function with entertainment and light instruction;30 nor did the sometimes rather informal collection of kings’ sagas known from the description of its manuscript as Morkinskinna;31 nor did Snorri Sturluson and what were perhaps assistant or subsequent

28

Preben Meulengracht Sørensen, Fortælling og ære: Studier i islændingesagaerne (Århus: Århus Universitetsforlag, 1993). 29 Klaus Böldl, Eigi einhamr : Beiträge zum Weltbild der ‘Eyrbyggja’ und anderer Isländersagas, ed. by Heinrich Beck, Dieter Geuenich and Heiko Steuer, Ergänzungsbände zum Reallexikon der germanischen Altertumskunde 48 (Berlin; New York: de Gruyter, 2005), p. 255. 30 See, e.g., Theodoricus monachus, Historia de antiquitate regum Norwagiensium / An Account of the Ancient History of the Norwegian Kings, trans. and annotated by David and Ian McDougall (London: Viking Society for Northern Research, University College London, 1998), p. 2: ‘Moreover, in the manner of the ancient chroniclers, I have added digressions in appropriate places which, in my opinion, are not without value in serving to delight the mind of the reader.’ In the admittedly more restrained Historia Norwegie, passages such as section viii, the description of Iceland, (De Glaciali Insula) demonstrate considerable narrative delight: see Historia Norwegie, ed. by Inger Ekrem and Lars Boje Mortensen, trans. by Peter Fisher (Copenhagen: Museums Tusculanum Forlag, 2003), pp. 68–75. 31 For a more literary or ‘intentionalist’ interpretation of this ‘informality’, see Ármann ­Jakobsson, Staður í nýjum heimi: konungasagan Morkinskinna (Reykjavík: Háskólaútgáfan, 2002), pp. 78– 108.

59



Christopher Sanders

compilers in the construction of the major history of the Norwegian kings, the composition of which began in the 1230s.32 It seems that a contrasting of two functions, prodesse (‘to educate or inform’) and delectare (‘to delight/ entertain’) – a distinctive pair used by commentators when discussing medieval Latin literature and its relationship to vernacular literature – should be seen not as a potential bifurcation, but as the basis for an amalgam.33 Historywriting of the Old Norse type referred to here corresponds to what R. W. Southern described as the ‘historical ideal of the West in the Carolingian and post-­Carolingian period … derived from Sallust and Suetonius, from Virgil and Lucan, from Boethius and from the writers on rhetoric whose works were studied in the schools’.34 As Southern explains, ‘Early medieval scholars, who were educated in the rhetorical tradition of the ancient world, understood instinctively the liberties the ancient historians had taken, and they followed their example with enthusiasm.’35 To some extent, other central historiographic works in Old Norse can be seen as proponents of this mixed and constituently ‘artistic’ practice of history: while Ari the Learned’s ­Íslendingabók (‘Book of the Icelanders’), reads very much like an annal, albeit an interpretative one, the same cannot be said of Landnámabók (‘The Book of Settlements’), which exists in many versions. Styrmisbók, the first of them that was reasonably well expanded, dates from the first half of the thirteenth century. Generally, they tend to contain a blend of genealogical information with entertaining tales about ancestors; these short narratives may well have first entered the compilation in the version put together by Sturla Þorðarson (c. 1275–80).36 Somewhat similar features are to be found in large parts of Stjórn, the extensive Old Norse Bible translation, which may well to a considerable extent be Norwegian in origin, but which survives almost exclusively in Icelandic manuscripts. There are three sections, and of these especially Stjórn III37 shows a significant interest in maintaining

32

33

34

35 36

37

‘In the broader features of classical saga-style there is again [in Heimskringla] a kind of equilibrium – between scholarship and art, information and entertainment, fact and fiction’ (Diana Whaley, Heimskringla: An Introduction, Viking Society for Northern Research, Text Series 8 (London: Viking Society for Northern Research, University College London, 1991), p. 83). The concept, as quoted in Barnes, ‘Some current issues in riddarasögur research’, p. 82, is to be found in Horace, Ars poetica, lines 333–4: ‘Poets aim either to benefit, or to amuse, or to utter words at once both pleasing and helpful to life ... He has won every vote who has blended profit and pleasure, at once delighting and instructing the reader’ (Horace: Satires, Epistles and Ars Poetica, ed. and trans. by H. Rushton Fairclough, Loeb Classical Library 194 (Cambridge, Mass.: Harvard University Press, 1929), p. 479). R. W. Southern, ‘Presidential address: aspects of the European tradition of historical writing, 1: the classical tradition from Einhard to Geoffrey of Monmouth’, Transactions of the Royal Historical Society, 5th ser. 20 (1970), 173–96 (p. 177). Ibid., p. 178. Cf. Fortællinger fra Landnámabók, ed. by Jón Helgason, Nordisk filologi: tekster og lærebøger til universitetsbrug A3 (Copenhagen: Munksgaard; Oslo: Dreyers Forlag; Stockholm: Svenska Bokförlaget Norstedts, 1963), p. xii. This edition conveniently collects the shorter narratives in a single volume. On esp. Stjórn I and Stjórn III, see Svanhildur Óskarsdóttir, ‘Universal History in Fourteenth-

60



Bevers saga and Old Norse Historical Prose

dramatic excitement while reproducing the biblical narrative.38 This readiness to combine informative and entertainment functions in what are principally historiographical works is also a feature of many of the bishops’ sagas – some of which are highly successful as literature while still informing their audience about the past and, of course, promoting exemplary behaviour. The significant corpus of translated hagiographic literature, much of which would have been known to a translating cleric, could also be mentioned here (the increased pious tone of Bevers probably owes something to it), but the relationship between Old Norse hagiography and the riddarasögur deserves separate attention. The thesis here is that Old Norse prose works abound in examples of (historical) information and entertainment being completely intermingled, so that it is not unnatural that a tale such as Boeve should be treated principally as history and only superficially as fiction or tale in the process of translation. History is primary, and yet a ‘literary’, artistic or entertaining aspect nonetheless has a part to play; underlying this there is a suggestion that the mode in which the resulting ‘historical’ text is presented in Old Norse is much influenced by exegetical habits of mind. In all, this is little more than a reinforcement for the purposes of Icelandic literary history of the general European commonplace that for much of the Middle Ages there was little distinction between fiction and history. One of the main strands of historical writing in Old Norse can be traced through the Latin histories written in Norway and their influence on the work known as Ágrip, a short account of some of the kings of Norway, which is one of the sources for the subsequent flowering of kings’ sagas in Iceland. Ágrip itself is a good example of a mixed text: overtly historical but rhetorically decorated, somewhat anecdotal and, as such, entertaining.39 Towards the end of the twelfth century a veraldar saga (‘world history’), covering the six ages of the world from the the Creation to the reign of Frederick Barbarossa, was compiled. It also probably awakened an appetite for lengthier versions of the material on which it had touched only briefly. There are, however, two groups of texts of a principally historical nature that seem to be particularly relevant as comparative and probably influential material in relation to Bevers: there are what have been rather disparagingly Century Iceland: Studies in AM 764 4to’ (unpub. Ph.D. diss., University College, London, 2000). 38 Cf. Ian J. Kirby, Bible Translation in Old Norse, Université de Lausanne, Publications de la Faculté des Lettres 27 (Geneva: Droz, 1986), p. 60: ‘Stjórn III derives, essentially, from the Vulgate; but it is a considerably reworked version, with some omission and summarising, and a good deal of explanation and elaboration.’ See also Christine Elizabeth Fell, ‘The Old Norse of the Book of Joshua’, in Proceedings of the First International Saga Conference, University of Edinburgh 1971, ed. by Peter Foote, Hermann Pálsson and Desmond Slay (London: Viking Society for Northern Research, 1973), pp. 114–42 (p. 121) on a reworking of the Book of Joshua in Stjórn III: ‘The translator’s changes are towards greater drama, greater variety of presentation, and greater realism … The whole book of Joshua becomes centred on the activities of its hero.’ 39 On the work’s anecdotal nature, see Ágrip af Nóregskonungasögum: A Twelfth-Century Synoptic History of the Kings of Norway, ed. and trans. by M. J. Driscoll, Viking Society for Northern Research, Text Series 10 (London: Viking Society for Northern Research, 1995), p. xvii.

61



Christopher Sanders

described as the ‘pseudo-historical’ sagas,40 and there is the group of chansons de geste and associated texts that constitute the historical celebration of Charlemagne in Karlamagnúss saga. The pseudo-historical sagas are all translated from identifiable Latin texts.41 These works are the subject of a collective study by Stefanie Würth, who finds various common traits in the group as a whole. By comparison with the existing Latin parallels, these texts concentrate on the historical content and suppress moralizing and generalizing statements. Their focus is on conciseness of narrative (brevitas) and explanation, while in the process the foreign models are integrated into the native literary tradition.42 Here there is a clear parallel to the exegetical and explanatory nature of the Bevers texts, with emphasis on a chronological narrative, as brought out above. Würth is, however, anxious to differentiate between the riddarasögur and the pseudohistorical sagas; while accepting that there is an apparent similarity between the two groups (riddarasögur and the pseudo-historical works), she insists that stylistic differences point to a difference in taste between Norway and Iceland: the Norwegian court preferred rhetoric, the Icelandic public facts. Works such as Partalopa saga, Bevers and the whole of Karlamagnúss saga, which all exist in versions with remarkably little rhetorical embellishment, may, however, have been translated in Iceland.43 In a case such as that of the B-version of Karlamagnúss saga – which, unlike the A-version, is markedly rhetorical – it is the result of influence from the late Icelandic florissant (‘florid’) style;44 this stylistic tendency is particularly associated with the 40 41

42

43

44

For a general introduction, see Stefanie Würth, ‘Historiography and pseudo-history,’ in A Companion to Old Norse–Icelandic Literature and Culture, ed. McTurk, pp. 155–72. The Books of Maccabees and Peter Comestor’s Historia scholastica make up the backbone of Gyðinga saga, a history of the Jews. Sallust’s Bellum Iugurthinum and Coniuratio Catalinae and Lucan’s Pharsalia are the principal components of Rómverja saga, a history of the Romans. The work attributed to Dares Phrygius, De excidio Troiae historia, and Ilias latina and other minor sources make up an important link backwards in time in the form of Trójumanna saga ‘History of the people of Troy’. There is a version of Valtère de Châtillon’s Alexandreis, which becomes Alexanders saga, and a link between Troy and Northern Europe is provided by Breta sögur, which is based on Geoffrey of Monmouth’s Historia regum Britannie. Stefanie Würth, Der ‘Antikenroman’ in der isländischen Literatur des Mittelalters: Eine Unter­ suchung zur Übersetzung und Rezeption lateinischer Literatur im Norden, Beiträge zur nordischen Philologie 26 (Basle; Frankfurt: Helbing & Lichtenhahn, 1998), pp. 248–9. Cf., e.g., Partalopa saga, ed. by Lise Præstgaard Andersen, Editiones Arnamagnæanæ B 28 (Copenhagen: Reitzel, 1983), p. xx, on the origin of Partalopa saga: ‘The presupposition is, of course, that the only way in which an Icelandic author of the 14th century could have any knowledge of continental literature was through the medium of Norwegian translations. Although this belief is shared by most scholars in the field, it is ill-founded, and in fact quite improbable.’ Since Stefán Karlsson’s identification of the hand of the oldest fragment containing a Karlamagnús text as Icelandic, there is no clear remaining evidence that any of the A-texts (generally the more conservative) was Norwegian. Af Fru Olif og Landres syni hennar may be an exception, but it is represented in the B-version only, and this was clearly compiled in Iceland: see Stefán Karlsson, ‘Hverrar þjóðar er Karlamagnússaga? Orðfræðileg athugun’, in Festskrift til Finn Hødnebø 29. desember 1989 (Oslo: Novus, 1989), pp. 164–9, and ‘Elsta brot Karlamagnús sögu og rekaþáttr Þingeyrabókar’, in Eyvindarbók: Festskrift til Eyvind Fjeld Halvorsen 4. Mai 1992 (Oslo: Institutt for nordistikk og litteraturvitenskap, Universitetet i Oslo, 1992), pp. 302–18. For a description of this style, see Ole Widding, ‘Den florissante stil i norrøn prosa (isl.

62



Bevers saga and Old Norse Historical Prose

school of Bergr Sokkason, which produced many texts at the beginning of the fourteenth century: that is to say, comparatively late in terms of much textual development.45 The slightly more rhetorical style of the ‘C’ manuscript of Bevers from c. 1470, by comparison with the ‘B’ manuscript from c. 1400, may also be due to the same influence. Considerations of stylistic history provide one of the reasons why riddarasögur cannot be treated as a single homogeneous group. The pseudo-historical sagas collectively do, admittedly, show rather less in the way of direct revision of content than many riddara­ sögur, including Bevers, but this may reflect exactly the conciseness that is part of the influential textual background for a work such as Bevers. We should not ignore what seems on certain levels to be the similarity in intent, and I think it is important to recognize that the pseudo-historical texts probably had a considerable influence on the manner in which the Bevers texts now appear.46 An interesting element is the manner in which the manuscript AM 573 4to in its original state contained, apart from Trójumanna saga, both Breta sögur and Valvens þáttr (the continuation of Parcevals saga that contains Gawain’s adventures), and this was hardly on account of stylistic similarity, but at least partly because of the connection through King Arthur to the past of Britain – riddarasaga material becomes in this case, in terms of its reception, ‘history’.47 Ormsbók, the mid-fourteenth-century manuscript that perished in a fire in Stockholm in 1697, contained mainly riddarasögur but also two pseudo-historical works, Trójumanna saga and Breta sögur. Here, then, it is riddarasögur that ‘host’ the pseudo-histories, so the compiler of this codex presumably had some sense of an association between the works of these two types that was more than purely stylistic. The other group of texts that can be related to Bevers is Karlamagnúss saga, the Old Norse Charlemagne cycle. The textual history is daunting. There are two versions of many of its constituent parts or branches, and considerable uncertainty as to when the separate branches were combined into single entities.48 There is as yet only one edition, from the nineteenth century, that contains the entire cycle.49

45

46

47 48

49

skrúðstíllinn) specielt i forhold til den lærde stil’, Selskab for nordisk filologi: Årsberetning for 1977–1978 (1979), 7–11. One scholar is of the opinion that Bergr Sokkason was himself responsible for two branches of the B version of Karlamagnúss saga: see Peter Hallberg, Stilsignalement och författerskap i norrön sagaliteratur: Synspunkt och eksempel (Göteborg: Elanders Bogtryckeri Akktiebolag, 1968), p. 164. Further indication of the way in which reception of pseudo-historical sagas and riddarasögur is sometimes linked is treated in a forthcoming study by Stefanie Würth, ‘Die Transmission der Breta sögur als Beispiel für verschiedene Formen der translatio innerhalb der mittelalterlichen isländischen Literatur’, in Rittersagas, ed. Glauser and Kramarz-Bein. See also Würth, Der ‘Antikenroman’, pp. 57–9, on the manuscript transmission of Breta sögur. For a good survey of recent scholarship, see Hélène Tétrel, ‘La Chanson des Saxons’ et sa réception norroise: Avatars de la matière épique, Medievalia 53 (Orleans: Paradigme, 2006), pp. 73– 108. Karlamagnus saga ok kappa hans: Fortællinger om Keiser Karl Magnus og hans jævninger i norsk bearbeidelse fra det trettende aarhundrede, ed. by C. R. Unger (Christiania: Jensen,

63



Christopher Sanders

One of the striking features of its make-up is its disparity. The high seriousness of Agulandus þáttr (þáttr = ‘strand’, being the Icelandic term usually used of the separate branches), with its narrative of the constant struggle against the infidel, mixed with a recognition of the Saracen’s heroism and dignity, and Runzivals þáttr (= the Chanson de Roland), with its theme of honour pitched against practicality, is incorporated in the same scheme as the amusing (and moralizing) Old Norse version of the Pelèrinage.50 The texts of both Agulandus and Runzivals correspond to features noted in the Bevers text: passages of close translation can be punctuated by sections of what looks like editorial emendation. The common enthusiasm for battle scenes is unmistakable. What seems to have inspired this compilation of various texts that can be related to Charlemagne is a fundamental search for knowledge, and a wish to celebrate the past. That there was an ideological interest in promoting the image of a king is also highly likely.51 I have been postulating that a way of understanding the reception of Bevers in the North is to recognize what was understood as its historical appeal; it is only fair in this connection to point out that the two clearly encyclopaedic collections (that is to say, manuscript compilations aiming at some form of universal knowledge) do not contain Bevers in particular or any material that is usually regarded as belonging to the group recognized as riddarasögur. The two compilations in question are Hauksbók and AM 764 4to. They have completely different purposes and are therefore quite differently constructed.52 One thing they do, however, demonstrate is that 1860). The build-up of this edition, with a short description of the contents, is as follows: Branch I has no clear French source, although some have argued for the existence of such a text (see Tétrel, ‘La Chanson des Saxons’, p. 74); Branch II, Af Frú Olif og Landres syni hennar (Olif is Charlemagne’s sister), probably derives from an English version of Doon de la Roche (cf. H. M. Smyser, ‘The Middle English and Old Norse story of Olive’, PMLA, 56:1 (1941), 69–84); Branch III, Af Oddgeiri danska, is close to the first part of the Chevalerie Ogier de Danemarch; Branch IV, Af Agulando konungi, is principally based on the Chanson d’Aspremont, but also incorporates, amongst other material, chapters of the Latin Pseudo-Turpin Chronicle; Branch V, Af Guitalin saxa, is based on a French text which is different from Jean Bodel’s Chanson des Saisnes (see Tétrel, ‘La Chanson des Saxons’, esp. pp. 216–43); Branch VI, Af Otvel, is based on Chanson d’Otinel; Branch VII, Af Jorsalaferð, derives from the French text Pelèrinage de Charlemagne à Jérusalem; Branch VIII, Af Runzivals bardaga, is a translation of a version of Chanson de Roland; Branches IX & X, Af Vilhjalmi korneis and Um kraptaverk ok jartegnir, may derive from a lost Vie romancée de Charlemagne (see Povl Skårup, ‘Contenu, sources, rédactions’, in Ogier le Danois III: Karlamagnússaga, Branches. I, III, VII et IX, ed. by Agnete Loth et al. (Copenhagen: Société pour l’Étude de la Langue et de la Littérature Danoises, 1980), pp. 346–7). There is no single study of the entire Karlamagnúss collection, but there are good introductions to each branch in Karlamagnús saga: The Saga of Charlemagne and His Heroes, trans. by Constance B. Hieatt, 3 vols, Medieval Sources in Translation, 13, 17, 25 (Toronto: Pontifical Institute of Mediaeval Studies, 1975–80). 50 See Alfrún Gunnlaugsdóttir, ‘Jorsalaferð: le voyage de Charlemagne en Orient’, Gripla, 7 (1990), 203–25. 51 Cf. Kramarz-Bein, Die ‘Þiðreks saga’, pp. 164–5. 52 For AM 764 4to, see Svanhildur Óskarsdóttir, ‘Universal History in Fourteenth-Century Iceland’, alongside Würth, ‘Die Transmission der Breta sögur’. A convenient encyclopaedia entry on Hauksbók is that by Gunnar Harðarson and Stefán Karlsson in Medieval Scandinavia, ed. Pulsiano, pp. 271–2.

64



Bevers saga and Old Norse Historical Prose

the ancient Icelanders’ concern with the past has by no means diminished by the beginning of the fourteenth century. Hauksbók is motivated partly by the desire of the instigator, Haukr Erlendsson, to place his own dynasty in a universal historical context, whereas AM 764 4to is probably intended as a book of instruction for the nuns at a cloister at Reynistaðr in North Iceland. Riddarasögur did not, in other words, earn acceptance in the encyclopaedic works of medieval Icelandic instruction, but this does not fundamentally undermine the ‘historical’ potential of certain works within what is conventionally regarded as the riddarasaga genre. Finally, to illustrate my main point, a stencilling of a transfer from mimesis to instructive chronology, let us look at an issue raised in Erich Poppe and Regine Reck’s paper in the present volume on the Middle Welsh translation of Boeve53 – the function of the narrator. In Bevers the narratorial inter­ ventions in which the narrator expresses his own view on a situation are universally suppressed.54 The narratorial guides that assist the progression from one passage to another, of the ‘let us now speak of something else’ type are both reproduced and slightly increased in number, as these match a corresponding pattern in Old Norse narrative technique.55 The remaining narrative comments that are added in the Norse are of an explanatory nature: what is taken to be a lack of clarity in the original apparently needs to be resolved. Thus, after Erminríkr’s two retainers have lyingly told him that his daughter Josivena has been bedded by Bevers, there is the excusatory explanation on the part of the narrator that Bevers could not have done this as he did not wish to have carnal pleasure with her until she was baptized.56 This illustrates the movement from a text principally intended to entertain to one that was to be explained. History, information about the past, moves in the direction of edification, which in the nature of the contemporary culture is Christian and moral. We must accept in the process that quite a lot of fun gets lost – but since both the Anglo-Norman and the saga fortunately survived we can enjoy them side by side and draw from them whatever they can tell us about the literary cultures from which they derive. What are the implications of the above? This paper has attempted to portray some of the general background in terms of Nordic prose culture for the reception of a work such as the Anglo-Norman Boeve. As I am suggesting that the saga partly belongs to a well-established tradition of historio­graphical writing, what might be the potential datings for the saga in relation to this tradition? The versions of Landnámabók and Stjórn referred to above belong to various periods in the thirteenth century, mainly to the second half. In the

53 54 55

Cf. pp. 37–50 above. Cf., e.g., Boeve 120, 175, 374, 497, 773, etc. E.g. at Boeve 207, 972, 2730, 3121; examples of extra narratorial guides of this type are at the beginnings of Bevers chs 3, 24, 29 (MS C only): see Bs, pp. 13, 231, 241, 281. 56 ‘því at hann vildi eigi eiga líkamsmunúð við hana fyrr en hon var skírð’ (Bs C8.77 (p. 83)).

65



Christopher Sanders

group of pseudo-historical sagas, Rómverja saga and Veraldar saga have often been associated with the period around 1180, whereas there is some scholarly disagreement as to a dating of Trójumanna saga.57 All of this is quite a lot older than any likely dating for Bevers. As for Karlamagnúss saga, the oldest fragment can be dated to c. 1250,58 and the first compilation in the form of a cycle may date from the same period – around the middle of the thirteenth century.59 Just as it is difficult to determine whether the Bevers translation was originally Norwegian or Icelandic, there is, pending studies of vocabulary and formulaic phrases, very little to go on when it comes to dating the work. Bevers deviates completely from the one other work of a chanson de geste type for which a clear French original survives and which has not been ­incorporated in a cycle: Elíss saga ok Rósamundar, written in a highly alliterative style and preserved in a Norwegian manuscript from c. 1270. The oldest surviving manuscript of Bevers is a fragment from c. 1350.60 If we associate some of the character of the surviving Bevers texts with the A version of ­Karlamagnúss saga and with the pseudo-historical sagas in general, then a period after 1250 is appropriate, leaving 1250–1350 as the likely timespan within which the texts of Bevers emerged in their present form. As is suggested in the introduction to my recent edition,61 Bevers does not immediately fall into any pattern that would immediately associate it with the Norwegian court or with any literary policy that might be associated with King Hákon Hákonarson of Norway (1217–63). The potential influences on the Bevers texts that have been discussed above derive in many ways from Iceland; ethnicity is not a vital issue, yet it is of course interesting in terms of literary history – and as yet the burden of proof lies on the shoulders of whoever might wish to argue that the work is Norwegian in origin.

57 58

59

60 61

Würth, Der ‘Antikenroman’, p. 56, prefers the view adopted by Jakob Benediktsson that both Trójumanna saga and Breta saga belong to the beginning of the thirteenth century. Stefán Karlsson, ‘Elsta brot Karlamagnús sögu og rekaþáttr Þingeyrabókar’, in Eyvindarbók: Festskrift til Eyvind Fjeld Halvorsen, ed. by Finn Hødnebø et al. (Oslo: Institutt for Nordistikk og Litteraturvitenskap, Universitetet i Oslo, 1992), pp. 302–18 (p. 317). Cf. Peter G. Foote, The Pseudo-Turpin Chronicle in Iceland: A Contribution to the Study of the ‘Karlamagnús Saga’, London Mediæval Studies Monographs 4 (London: London Mediæval Studies, University College, 1959), p. 47. Ibid., p. lxxiv. Ibid., pp. cliv–clv.

66

5 From Boeve to Bevis: The Translator at Work ivana djordjeviĆ

T

o call the composer of the Middle English Sir Bevis of Hampton a ‘translator’ is not necessarily to imply that the text itself is a ‘translation’. Indeed, as Jennifer Fellows has pointed out, ‘Only to a limited extent can Beves accurately be described as a translation of Boeve.’ So what does it mean to describe the Middle English Bevis as a translation of the AngloNorman Boeve de Haumtone? The author (or authors) of Bevis used a much broader range of translational procedures than we tend to recognize. Lengthy passages are interpolated, while others are substantially cut or even omitted; changes are discernible in the portrayal of some characters, in the relative emphasis given to individual episodes, and in the attention accorded to the hero’s children. And yet, without necessarily agreeing that the whole of Bevis should be described as a translation of Boeve, critics have recognized that for at least some sections of the poem, such as the hero’s enfances, the term is accurate. That is to say, while ‘translation’ may not be the right word for the text as a whole, it is an appropriate designation for the process that produced large sections of it. Translation as a process, not a product, will therefore be my subject here: not what a translation is or should be, but what a translator does. I shall leave aside substantial omissions, lengthy interpolations and bold rewritings, all of which are obviously motivated by conscious agendas, whether aesthetic or ideological or both, but shall look instead at closely rendered passages, which shed light on the more mundane, though no less interesting, aspects of 

Jennifer L. Fellows, ‘Sir Beves of Hampton: Study and Edition’, 5 vols (unpub. Ph.D. diss., Univ. of Cambridge, 1980) (hereafter BHF), I, 52.  Judith E. Martin, ‘Studies in Some Early Middle English Romances’ (unpub. Ph.D. diss., Univ. of Cambridge, 1968), p. 114, remarks that in the first 555 lines of Bevis its author ‘borrows more words from Boeve than in any other section of his poem, and diverges very little from events or even details in the Anglo-Norman’. Comparing the early part of Bevis with Boeve, Albert C. Baugh, ‘Convention and individuality in the Middle English romance’, in Medieval Literature and Folklore Studies: Essays in Honor of Francis Lee Utley, ed. by Jerome Mandel and Bruce A. Rosenberg (New Brunswick, NJ: Rutgers University Press, 1970), pp. 123–46 (p. 126), found that ‘the incidents and the sequence of incidents are, almost without exception, the same throughout, the English reproducing many small and quite casual details’. The English poet, Baugh stressed, maintained ‘his fidelity to his source in incident and idea’ (Albert C. Baugh, ‘The making of Beves of Hampton’, Library Chronicle, 40 (1974), 15–37 (p. 18)).

67



Ivana Djordjević

translation as a straightforward attempt to convey to an imagined audience what one considers to be most important or most interesting about a text. The process involves a careful, and often not consciously articulated, balancing of fidelity to the textual source and appeal to the anticipated ­audience. Changes that occur at this level are subtler and often made automatically. Nevertheless, they can have a significant cumulative effect not only on individual translated texts but also on an entire genre, especially a popular one such as the Middle English romance, as I hope to show. Translators invariably have to deal with major translational constraints, including the two languages’ different structures and usages and, in the case of verse translation, prosody. But there is a great deal that they do without quite knowing that they are doing it, simply because, as hypothesized by some translation theorists, certain modes of thinking are inherent to translation. Such regularities in translators’ behaviour have been described as ‘laws of translation’ or ‘translation universals’. They include explicitation (the most common), implicitation (its opposite), and repertorization, which I examine in greater detail below. Before I do so, however, I should note that all textual rewriters, not just translators, tend to be subject to the same impulses. The study of medieval translation in particular is therefore complicated by the difficulty (or, at times, impossibility) of distinguishing between translational and scribal intervention – a problem especially acute in the study of texts with convoluted manuscript traditions, such as the Boeve–Bevis complex of narratives. It is impossible to tell which of the two incomplete Anglo-Norman texts the translator might have used (probably neither); as Jennifer Fellows shows in her chapter, it is just as hard to establish which of the Middle English texts known to us best preserves the original form of the resulting translation. The Auchinleck text (A) was long privileged because it is the oldest, and at some points significantly closer to the Anglo-Norman, but its authority has been challenged in recent years. The distribution of direct verbal echoes of the extant Anglo-Norman texts among the Middle English manuscripts indicates a process of textual transmission so intricate that no manuscript can justifiably be singled out. While it is a relatively safe generalization to say that the redactions preserved in earlier manuscripts tend to show a more direct 

For studies of the Bevis translator’s handling of such constraints elsewhere, see Ivana Djordjević, ‘Versification and translation in Sir Beves of Hampton’, Medium Ævum, 74 (2005), 41–60, which deals with the translator’s efforts to find a verse form to match the short laisses of his original, and ‘Translating courtesy in a Middle English romance’, Studia Neophilologica, 76 (2004), 140–51, a study of the way in which he negotiates differences between French and English pragmatics.  For concise accounts of the problem of translation universals, see, e.g., Gideon Toury, Descriptive Translation Studies and Beyond (Amsterdam: Benjamins, 1995), pp. 258–79; Theo Hermans, Translation in Systems: Descriptive and System-Oriented Approaches Explained (Manchester: St Jerome, 1999), pp. 91–4; Andrew Chesterman, Memes of Translation: The Spread of Ideas in Translation Theory (Amsterdam: Benjamins, 1997), pp. 70–3. On translation universals with special reference to Bevis, see Ivana Djordjević, ‘Mapping Medieval Translation: Methodological Problems and a Case Study’ (unpub. Ph.D. diss., McGill Univ., 2002), pp. 155–8, 174–6.

68



Boeve to Bevis: The Translator at Work

translational relationship to the Anglo-Norman text, a textual detail from the Anglo-Norman narrative, omitted in all the early Middle English redactions, will occasionally resurface only in a very late manuscript, such as C or M, or even in an early printing. According to Gideon Toury, ‘In translation, textual relations obtaining in the original are often modified, sometimes to the point of being totally ignored, in favour of [more] habitual options offered by a target repertoire.’ Toury goes on to explain that this tendency is a result of the pervasive codification of all sorts of phenomena, both linguistic and non-linguistic, which have semiotic value for the members of a community: Sets of codified items form repertoires, i.e., aggregates governed by systemic relations, which determine the relative availability of items pertaining to such an aggregate for any particular use within the community’s culture. In other words, a repertoire amounts to the range of choices which makes cultural functions realizable through real products and practices.

Of all translation laws formulated so far, this one is the most tentative. In many ways, however, it is the most illuminating for the study of popular medieval texts, precisely because they depend to a large extent on extensive and complex repertoires of narrative elements at all textual levels: anything from common tags, frequently occurring rhyme pairs and well-established ways of expressing particular attitudes, to habitual comparisons, conventional episodes, or standard character types. Brief narrative formulae (‘tags’) are an especially intriguing aspect of the Middle English metrical romance, a genre famous – or notorious – for its formulaic density. This feature of the corpus has caused much debate as critics have struggled to situate the romances on either side of the (increasingly problematic) divide between orality and literacy. It also bears some of the blame for the long tradition of critical dismissal of the romances, a legacy of New Critical aesthetic canons but also (going back centuries) of Chaucer’s ‘Sir Thopas’. Although the popular metrical romance has now been fully rehabilitated in academic criticism, and Chaucer’s satirical intentions in ‘Sir Thopas’ have been repeatedly questioned, the poem can still serve as a

   

Cf. p. 83 below. Toury, Descriptive Translation Studies, p. 268 (square brackets in the original). Ibid. Some forms of repertorization in Bevis at these higher textual levels are studied in Ivana Djordjević, ‘Original and translation: Bevis’s mother in Anglo-Norman and Middle English’, in Cultural Encounters in the Romance of Medieval England, ed. by Corinne Saunders (Cambridge: Brewer, 2005), pp. 11–26; the article concludes that in the most closely translated section of Bevis the portrayal of the hero’s parents changes under the influence of current conventions, both literary and more broadly cultural.

69



Ivana Djordjević

useful starting-point for a study of stereotypical expressions in Bevis, a text on which Chaucer seems to have drawn in his own tale. Bevis is one of the romance heroes mentioned by name in ‘Sir Thopas’ (v.899); the second fitt of Chaucer’s tale begins with what appears to be a direct borrowing from the opening lines of the A redaction of Bevis (vii.833– 5);10 and Chaucer’s basic prosodic form in this tale is a rare six-line tailrhyme stanza also used in the stanzaic section of Bevis. Beyond these specific points of contact, ‘Sir Thopas’ shares with Bevis and other Middle English metrical romances a number of very short stereotypical expressions which tend to occur in the rhyming position and are likely to be affected primarily by metrical concerns. Typically a half-line long or even shorter, such narrative formulae sometimes extend to a full line or even to an entire rhyming couplet. Some of them are part of a system of descriptive shorthand peculiar to Middle English narrative. Thus in ‘Sir Thopas’ the hero is ‘fair and gent’ (vii.715), his father is ‘a man ful free’ (vii.721), and the maidens who long for him are ‘bright in bour’ (vii.742). Other stereotypes include asseverating tags (‘verrayment’, vii.713; ‘in good certayn’, vii.728; ‘for sothe’, vii.749)11 and all-purpose adverbials especially applicable to various knightly activities (e.g. ‘as he were wood’, vii.774). This last expression, which occurs with some frequency in Middle English popular romance,12 proves quite serviceable in Bevis: it is used twice to modify the verb ‘to cry’ and six times to describe different aspects of fighting: 1 Brademond cride 2 [Þai] slowe Sarsins 3 [Brademond] cride 4 [Beues] leide on



10 11 12

13 14

ase he wer wod (1012 / p. 43)13 as hii wer wod (1125 / p. 47) alse he hadde be wod (1817 / p. 71) ase he wer wod (3612 / p. 130)14

Chaucer’s tale was seen by earlier critics as the first indictment of the genre on aesthetic grounds, but there is no critical consensus as to what exactly Chaucer was parodying in ‘Sir Thopas’; many of the stock phrases at which he seems to poke fun in the tale occur frequently in his ‘serious’ work. For a concise overview of the issue, see Nancy Mason Bradbury, Writing Aloud: Storytelling in Late Medieval England (Urbana and Chicago, IL: University of Illinois Press, 1998), pp. 176–80. Cf. Ivana Djordjević, ‘Le grant translateur between “Sir Thopas” and “The Tale of Melibee” ’, in The Medieval Translator 10 (Paris 2004), ed. by Jacqueline Jenkins and Olivier Bertrand (Turnhout: Brepols, 2007), pp. 255–69. All references to ‘Sir Thopas’ are to The Riverside Chaucer, ed. by Larry D. Benson, 3rd edn (Boston, MA: Houghton Mifflin, 1987), pp. 213–16. On the use of similar authority tags in Boeve, see Djordjević, ‘Versification and translation’, p. 56 n. 30. Besides Bevis, the phrase occurs in Richard Coer de Lyon (a line virtually identical with example 6 below and paralleled in The King of Tars), Havelok the Dane, Sir Eglamour (a line identical, except for the character’s name, with example 1 below), Guy of Warwick, Ferumbras, Libeaus Desconus, Sir Tristrem, Ipomadon A and Athelston. See Albert C. Baugh, ‘Improvisation in the Middle English romance’, Proceedings of the American Philosophical Society, 103 (1959), 418– 54 (p. 446), and notes in The Romance of Sir Beues of Hamtoun, ed. by Eugen Kölbing, EETS, es 46, 48, 65 (London, 1885–94) (hereafter BHK), passim. References to the text of Bevis are by line number to BHF, II, and by page number to BHK. The line occurs in an 8-line passage exclusive to A.

70



Boeve to Bevis: The Translator at Work 5 [Þe dragoun] fleȝ awei 6 Þai leide on

ase he wer wod (3649 / p. 131)15 ase hii were wode (4792 / p. 178)16

Only two instances of this stock phrase are found outside the rhyming position: 7 As he were wod he gan to fyȝt (SN 3134 / p. 117)17 8 [Yuor] fauȝt ase he were wod þer-fore (5387 / p. 198)18 In the rhyming position, the phrase is a convenient complement for ‘mod’, ‘god[e]’, and especially ‘stod’, with which it is paired four times in a rhyming couplet. The resulting couplets are themselves put to good use as higherlevel formulaic expressions. The description of Brademond shouting ‘ase he wer wod / To King Ermin, þar a stod’ (1012–13 / p. 43) is adapted in a later episode, when Brademond cries out ‘alse he hadde be wod / To hem alle aboute him stod’ (1817–18 / p. 71). The second time the couplet occurs, with minimal variation, it is no longer just a tag but becomes part of the characterization of Brademond: he is, we realize, the kind of person who is in the habit of yelling ‘ase he wer wod’. In A’s version of Bevis’s fight with the dragon, similarly paired couplets are used to particularly good effect. At a crucial point in the fight, after Bevis has implored God for assistance in a few lines of direct address, the narrator continues: ‘Þat seide Beues, þar a stod, / And leide on ase he wer wod’ (3611–12 / p. 130). This is followed by an inconclusive exchange of blows, in the midst of which the hero appeals to God (and ‘Marie, His Moder dere’) yet again, this time in a longer speech. ‘Þis herde þe dragoun, þer a stod, / And fleȝ awei ase he wer wod’ (3648–9 / p. 131). The tide has turned, and within the next few lines Bevis will slice the dragon’s head in half. The matching couplets serve as a frame for Bevis’s two prayers, focusing the audience’s attention on God’s intervention on the hero’s behalf. At the same time, the poet suggests that the two adversaries are equal in strength by having them occupy the same slot in two virtually identical couplets. Bevis is indeed a match for the dragon – and more than a match, for while he fights ‘ase he wer wod’, the dragon eventually flees with the same abandon. The dragon fight is an interpolation in the Middle English narrative; no such episode is to be found in the extant manuscripts of Boeve. Similarly, line 5387 belongs to the longest of all treatments of Bevis’s fight with Yvor, in which the Anglo-Norman narrative is much amplified. The line occurs, however, at a moment of heightened narrative tension, for which a parallel

15 16 17 18

The line occurs in a 6-line passage exclusive to A. The line occurs in a 22-line passage exclusive to A. There is a lacuna in A here. The line occurs in A’s extended account of Bevis’s last encounter with Yvor. (For just over 20 lines in the other manuscripts, A has more than 90.)

71



Ivana Djordjević

with Boeve may be (very) tentatively identified in the description of how Yvori, attacking Boeve one last time, grant coup li feri (‘struck him a hard blow’, 3600). In the remaining instances we may be able to pinpoint the expressions that triggered the English stock phrase with greater certainty. In line 1125 / p. 47, Bevis’s men, encouraged by their leader, ‘slowe Sarsins as hii wer wod’. In Boeve they i vunt a baundoun (‘go to it violently/vigorously’, 587). ‘As hii wer wod’ is not an absolutely accurate rendering of a baundoun, but it conveys a similar sense of gleeful carnage and, as a ready-made phrase of great currency, was likely to have come to the translator’s mind without any prompting. In line 1012 / p. 43, where Brademond threatens Ermin, Bevis’s ‘cride ase he wer wod’ corresponds to a simple dist (‘said’) in Boeve 503. But the Anglo-Norman villain’s direct address is preceded by what amounts to its summary, in the course of which we learn that ­Brademound and his cent mil paens (‘hundred thousand pagans’) have fortement manasé (‘harshly threatened’) Josiane’s father (lines 498–9). Boeve’s insistence on the violence and seriousness of the threat may have inspired the translator to emphasize them in his turn, in the most economical way he knew: namely, by supplying a ready-made phrase, for which, furthermore, a rhyming word is easy to provide. The next time Brademound is seen shouting at someone, in Boeve 937–8 (‘Brandon s’est crié: / “Ke fetes, seynurs? Pur quey ne le pernez?” ’),19 the translator will thriftily transplant the same couplet to a new context, changing only what must be changed. The closest Anglo-Norman match for the Middle English phrase is found in three other episodes, the first of which relates Boeve’s fight with the two lions. The episode is significantly longer in Bevis,20 but the added length is mostly in details, while the general outline remains the same. At a crucial point in the fight one of the lions attacks the hero: it opens its mouth com il fu devé (‘as though it were enraged/crazed’, line 1729). Because of the differences between original and translation here, no line in Bevis corresponds exactly to this statement, but the closest equivalent of the expression com il fu devé is used at roughly the same point in the narrative to refer to Bevis: ‘As he were wood he gan to fyȝt’. It is not inconceivable that the presence of the French phrase in Boeve reminded the translator of its literal English rendering, which he then used at the first opportunity.21 Something rather similar happens in another episode that, like the lion fight, has been extensively reworked in the translation. The battle for Civile, which in Boeve frames the hero’s chaste marriage to the local heiress, has

19 20

‘Bradmund cried: “Knights, what are you about? Why don’t you seize him?” ’ Less than 20 lines long in Boeve, the description of the encounter is expanded to 75 lines in NS. (A and E are both damaged at this point, but it is clear that the length of the episode was roughly the same in A.) Only in C is the entire fight dealt with in a mere four lines. 21 The initial position of the phrase in the English line suggests that there were no significant metrical reasons for its use here.

72



Boeve to Bevis: The Translator at Work

been recast in Bevis as a tournament.22 Nevertheless, much of the action described is still of the same kind: violent encounters between opposing groups of armed men. In Boeve, the second attack on Civile begins with a description of the assailants, who ‘poynent ensemble lur chevals com desvé’ (‘all spur their horses like mad’, line 2912). In Bevis, as the tournament starts the participants ‘leide on ase hii were wode’. Although the treatment of the episode is different, here too the English expression may well have been ­t riggered by the presence of the French one. Towards the end of Boeve, when Yvori leads his men to their final defeat in the attack on the city of Abreford, they ‘chivachent ensemble com gent devé, / jeskes a Abreford ne sunt aresté’ (‘ride together like madmen, not stopping until Abreford’, lines 3555–6). At this point the translator has not yet started to embellish and follows his original quite closely. In translating these two lines, he tones down the fury of the king’s followers and says merely: ‘To Ermonie [Yuore] gan hem lede’ (5267 / p. 195).23 As I noted above, however, the French expression is picked up later in the same episode and used to describe the desperate intensity of Yvor’s last stand. It cannot be a coincidence that a fixed expression that occurs four times in Boeve is echoed three times in the corresponding passages of Bevis, even if it is not used in quite the same way.24 But why is it not used in the same way? Is the difference to be explained solely by the great liberty that the translator takes with his original in the episodes under scrutiny? The very loose nature of textual matching in those passages is no doubt a factor, but another important reason may lie in the translator’s awareness of the different connotations of the phrase in French and in English.25 Close scrutiny reveals that many of the ‘line-fillers’ in Boeve are used with a degree of sophistication that may have escaped post­medieval critics but that was obvious to the Middle English translator. Thus even in dealing with forms of address, a seemingly insignificant aspect of the conventional repertoire of Middle English romance, the translator of Boeve paid careful attention to the subtleties of his original in spite of the heavy linguistic and prosodic constraints under which he was operating.26 Similarly, the translator must have noticed that in French texts, both Anglo-Norman 22

23 24

25

26

This is itself an interesting example of what is most likely a repertorizing change, but at the level of plot. By the time Boeve came to be translated, winning a lady in a tournament (as opposed to winning her in battle) had become such an established convention in narratives of entertainment, in English as well as Anglo-Norman, that the pull of the repertoire was hard to resist. ‘To Ambyford þey went ryȝt’ in NSC. We first encounter the phrase in Boeve when the famished hero, having escaped from prison and bullied his way to dinner (by killing the giant who had refused him his hospitality and threatening his wife), ‘mange, com il fust devé’ (‘ate like a madman’, line 1336). Critics have been slow to recognize the connotative and other subtleties of the use in medieval narrative of stock expressions, which they have tended to see as little more than line-fillers. Only recently did some scholars begin to study the use of narrative stereotypes ‘to bring nuance to narrative context, to influence tone and to foreground theme’, as explained in Roger Dalrymple, Language and Piety in Middle English Romance (Cambridge: Brewer, 2000), p. ix. See Djordjević, ‘Translating courtesy’.

73



Ivana Djordjević

and Continental, the phrase com il fu devé is never complimentary to the character whose actions it helps describe. It may be applied to the hero, but only rarely and in special circumstances – never to underline his prowess and fighting spirit.27 It may be used to highlight the mindless temerity of the hero’s adversaries, as in the other examples in Boeve; more often, though not in Boeve, it is used of women – as an indication of their weakness and emotional instability.28 Even if the English formula ‘as … wod’ first emerged as a translation of come … devé, a possibility that should not be discounted although it cannot be proved, it obviously acquired different connotations. In Bevis and the other Middle English texts in which it occurs, it has no a priori positive or negative value that would restrict its application to either the hero or his enemies. The connotative subtleties of its use depend on the verbs with which it collocates. In Bevis, the connotation is negative when the phrase is coupled with the verbs ‘to cry’ (used only of Brademond) and ‘to flee’ (used of the dragon), but almost invariably positive when paired with verbs of violent action (‘to fight’, ‘to slay’, ‘to lay on’). The sole exception is example 8 above, where the subject is Bevis’s arch-enemy Yvor.29 Elsewhere in Middle English romance, the phrase likewise collocates favourably with verbs that designate aspects of fighting: ‘to fight’ in Havelok and the Cotton MS of Lybeaus Desconus, ‘to lay on’ in The Erle of Toulous, and ‘to rush forward’ (‘lepe’) in Havelok.30 With ‘to cry’, on the other hand, the connotations are negative, and in Sir Eglamour of Artois it is a ‘fowle gyaunt’ who ‘cryed as he were wode’.31 In Middle English metrical romance, as a rule, to fight ‘like crazy’ is commendable, but to shout ‘like crazy’ is the mark of a villain. In French romance, on the other hand, no respectable character does anything ‘like crazy’, except in extreme circumstances. The translator is aware of these nuances, and never uses ‘as … wod’ as a direct translation of come … devé, but he does react to the presence of the French phrase in his original. Every time he comes across the French expression he seems to be reminded of the usefulness of the English equivalent, which he includes in his rendering of the same episode 27

28 29

30

31

Hue de Rotelande uses it in both Ipomedon and Protheselaus. The example from Boeve quoted in n. 23 above plays up Boeve’s desperate hunger, perhaps in order to condone his killing of the giant who had initially denied him food. Such uses can be found in Hue’s Protheselaus and in Chrétien’s Yvain, as well as in the first part of the Roman de la rose. It could be argued that in his last duel with the hero, Yvor rises to the status of an almost respected adversary, as suggested by Bevis’s offer, soon afterwards, to spare the Saracen king’s life if he is willing to convert to Christianity. (Yvor refuses and is quickly dispatched.) Havelok, ed. by G. V. Smithers (Oxford: Clarendon Press, 1987), line 2662; Lybeaus Desconus, ed. by M. Mills, EETS, os 261 (London: Oxford University Press for the EETS, 1969), line 1161; The Erle of Tolous, in Of Love and Chivalry: An Anthology of Middle English Romance, ed. by Jennifer Fellows, Everyman’s Library (London: Dent; Rutland, VT: Tuttle, 1993), line 88; Havelok, line 1897. On one occasion, also in Havelok (line 2610), it refers to the hero’s enemies, who ‘stirte forth so he were wode’. Sir Eglamour of Artois, ed. by Frances E. Richardson, EETS, os 256 (London: Oxford University Press for the EETS, 1965), lines 533–4.

74



Boeve to Bevis: The Translator at Work

– a good example of the translator’s sophisticated repertorization, a constant productive tension between the urge to remain close to the source, and the need to adapt the text to the repertoire of the genre in the target language. ‘As [he were] wod’ is more frequent in Bevis than its French counterpart is in Boeve; this would seem to confirm the impression that Middle English metrical romances are more formulaic than their French originals. But if we look at some other kinds of formulaic expression we find the opposite to be the case. True, the maidens ‘bright in bour’ and knights ‘worly under wede’ of ‘Sir Thopas’ do not appear in Bevis, whose hero, unlike Chaucer’s, is not ‘fair and gent’, but there are other fixed epithets, ornamental ­ adjective or noun phrases that collocate with particular nouns or proper names: thirty or so appositional phrases are appended to the names of the principal characters (including the horse Arundel) throughout the poem. Thus Bevis is ‘þe ­Cristene kniȝt’ (1126 / p. 48) or ‘þe kniȝt of Cristene lawe’ (5381 / p. 198), ‘þe fer’ (1294 / p. 55), ‘þat hendi kniȝt’ (1311 / p. 55; 2585 / p. 97), ‘þat gentil kniȝt’ (2067 / p. 81), and ‘corteis & hende’ (5454 / p. 201). Josian is ‘þis faire mai’ or ‘maide’ (621 / p. 26; 1213 / p. 51; 2771 / p. 103), ‘þat maide briȝt’ (1010 / p. 43), ‘briȝt of hiwe’ (1224 / p. 52), ‘briȝt and schene’ (5440 / p. 200), ‘þe fre’ (5991 / p. 217), or simply ‘þat maide’ (856 / p. 37; 1943 / p. 75). Sabot is ‘a wel strong kniȝt’ (3803 / p. 138; ‘an hardy knyȝt’ in NSC) or, like Bevis, ‘þe hende kniȝt’ (5445 / p. 200). On one occasion he is also ‘wis of dede’ (4952 / p. 182), but he is more often ‘þe hore’ (4150 / p. 154; 4165 / p. 155) or, more specifically, ‘wiþ is berde hore’ (4247 / p. 158). ‘Þe hore’ is a phrase also used of Ermin (812 / p. 35, 5042 / p. 186), who is referred to on one occasion as ‘þe riche king’ (805 / p. 35), a phrase applied also to Yvor (4638 / p. 173). Fixed epithets such as these are much more numerous and varied in the Anglo-Norman text. A rough count yields more than 130 examples, a number that should not surprise us, for the profusion of fixed epithets is one of the most distinctive stylistic traits of the chanson de geste, and it was inevitable that in adopting the chanson’s prosodic form the author of Boeve would also inherit the fixed epithet, often a prosodically significant textual feature. Boeve is li vaillaunt (‘the valiant’ × 3) or le chevaler vailant (‘the valiant knight’), le sené (‘the wise’ / ‘the sensible’ × 13), li hardis (‘the bold’ × 2), a vis fer (‘stern-faced’ × 2) or od lo corage fer (‘fierce of heart’ or ‘fierce of mettle’ × 3) or simply le fer (‘the fierce’ / ‘the proud’ / ‘the bold’ × 2), li ber (‘the brave fighter’ / ‘the hero’), li vassal alosé (‘the warrior of high repute’), le prisé (‘the renowned’ × 2) or ke taunt fet a priser (‘who is so praiseworthy’), le pussaunt (‘the mighty’), od hardi qer (‘the bold-hearted’), le honuré (‘the honoured’), le chevaler menbré (‘the powerfully built (or strong) knight’), le pruz (‘the worthy’), li sire (‘the lord’), li curteis guerer (‘the fine warrior’), li alosez (‘the esteemed’ × 4, all concentrated towards the end of the narrative), le combatant (‘the brave’), le gent (‘the noble’ or ‘the handsome’), or any combination of the above qualities, such as ke mult e fort e fer, li gentil e li fer, li pruz e li sené, li pruz e li alosez, li hardi combatant. A similar 75



Ivana Djordjević

wealth of postmodifying phrases and (occasionally) clauses is applied to the heroine, who is la bele (‘the fair’ × 7), a cler vis (‘the bright-faced’ × 6) or o le vis cler (× 2), also expanded to a clause as ke avoit cleer le vis (‘who had a bright face’), a cors gent (‘with a shapely body’ × 4), od le cors avenaunt (‘with an attractive body’ × 3), o le cors honuré (‘with the honoured body’ × 3), la loé (‘the praised’ / ‘the renowned’), la pucele (‘the maiden’ × 4) or la gentil pucele (‘the noble maiden’), and ke tant est colorie (‘who is so freshcomplexioned’). Since many of these phrases occupy the final position in a line, their rhyming (or assonantal) potential would have been of particular importance for the author, but this was not his only concern. Although the characters’ names in Boeve are sometimes postmodified by randomly assigned epithets, in many cases there are good reasons for using one phrase rather than another – reasons that go beyond the needs of prosody. Some expressions are spread throughout the text, while others are clustered in particular episodes. Although the clustering may be partly due to authorial (or scribal) inertia, deliberate characterization and narrative pointing are often also brought into play. Josiane is thus a stereotypical romance lady in that her defining quality is her beauty: she is most often la bele or a cler vis. But other, more specific, epithets are designed to fit a particular context. Thus the first reference to her cors gent (lines 911, 972) appears in the two knights’ accusations against Boeve, where the integrity of her virginal body is indeed at stake. Phrases focused on her body are especially frequent in this section, which begins with the denunciation of Boeve and ends with Josiane’s marriage to Yvori, who will prove unable to penetrate her cors avenaunt (997). As Josiane’s body is of great importance throughout the narrative, large stretches of action are motivated by what happens to it. It is therefore not surprising to find a concentration of similar phrases in the episode of Josiane’s forced marriage to Earl Miles, in which the heroine’s efforts to preserve her virginity provoke a major crisis. As the circumstances of the wedding are described, Josiane is referred to simply by name. But after she has strangled her undesired bridegroom on their wedding-night, she is la pucele or la gentil pucele throughout the ensuing narrative of how she is sentenced to burning at the stake, only to be rescued by Boeve at the last moment (lines 2115–85). Analysis of the distribution of epithets in the poem also makes it clear that while the heroine is beautiful and charming by nature, this does not automatically entitle her to honour and praise. Only after her baptism is she Josian la loé (line 2051), and only as a wife is she Josian o le cors honuré (line 2629).32 32

Yvori refers to Josiane as his mulier … o le cors honoré (line 3176). This is a face-saving rhetorical strategy, for at the point when he is trying to enlist the support of his vassals for an allout attack against Hermine, he could hardly admit to unfamiliarity with his wife’s body. Earlier, before her rescue from Yvori’s court, Josiane was referred to as la pucele o le cors honuré (line 1538), a maiden wife in other words, in what must be deliberate irony at the expense of a character for whom the narrator reserves his strongest animosity: although the good qualities of Escopart, Brademound, and even the hero’s mother are grudgingly acknowledged, Yvori is invariably either li

76



Boeve to Bevis: The Translator at Work

The nuanced use of appositional epithets in the Anglo-Norman poem is not matched by the translator, who is content with just a sprinkling of phrases, all of which emphasize the heroine’s beauty, and that in the most conventional way. But of the eight appositional phrases used to describe Josian in Bevis, only two have no parallel in Boeve. The remaining six all seem to have been triggered by the presence of explanatory or descriptive phrases in the ­original. The translator first responds to the use of such phrases in his original in line 621 / p. 26, where the description of Josian as ‘þis faire mai’ is a distant echo of ke tant est colorie in Boeve 401.33 Thereafter, however, he is guided chiefly by the target-language repertoire. As he chooses his formulae from it, he makes no attempt to mirror either the content of individual descriptive phrases or their often quite appropriate use in Boeve in general. The treatment of the hero is similar, though the translator here seems more interested in trying to imitate – albeit in a very limited way – the nuances of the original. Three of the seven appositional phrases applied to the hero in Bevis are not based on anything in the Anglo-Norman original, while the others can be interpreted as approximations, sometimes very loose, of phrases from Boeve. ‘Sire Beues, þe Cristene kniȝt’ (1126 / p. 48) occupies the same slot as ‘Boefs, li hardis (line 616) in Boeve, but draws attention to a different aspect of the hero’s characterization. ‘Sire Beues, corteis & hende’ (5454 / p. 201) corresponds to the Anglo-Norman Boves le gent (line 3724) in both position and content. ‘Beues, þat hendi kniȝt’ (2585 / p. 97) is a good match for Boefs le prisé (line 746), since both expressions are highly conventional yet well suited to this particular context. In spite of these limited efforts, however, the paucity of epithets applied to the hero inevitably reduces the impact of the ones that do occur in the Middle English translation. The loss of the dozen references to the protagonist as le sené cannot but affect the portrayal of the hero. If, reading Boeve, we see the hero as generally shrewd, at times even wise, this is not only because of what he says and does, but also because of the subliminal effect of the narrator’s formulaic insistence on this aspect of his character. Should the hero of Bevis then strike us as more impetuous and perhaps less provident, the absence of any reference to his judiciousness is at least partly to blame.34 Nevertheless, in at least one situation the translator is able to use an epithet to excellent effect. When Josiane, rejected by Boeve, dispatches to him a servant with a conciliatory message, Boefs, li ber (line 730), courteous but stern, sends him back empty-handed. In Middle English he becomes ‘Beues þe adverser (‘the enemy’, 3121) or li faus esprové (‘the proven deceiver’, 3551). The former expression, also used of the Devil in other texts, indicates the force of the narrator’s disapproval. 33 The parallel is more obvious in C, where the formulation, ‘þat is so feyre’, matches the syntax of the original. 34 If we extend the analysis to all adjectives used to describe the main characters, regardless of syntactic position, we find that their distribution is very much the same as in the more restricted sample provided by the appositional phrases I am studying here. Bevis is nearly always ‘hende’ or ‘gentil’, and Josian ‘bright’and ‘fre’.

77



Ivana Djordjević

fer’ (1294 / p. 55), which makes for a very effective contrast with the ‘gentil’, ‘hende’ and ‘corteis’ knight that we are accustomed to. Moreover, as the adjective is elsewhere applied either to villains (Bevis’s mother, or ­ Brademond’s vassal Grander) or to groups of armed men ready to fight, it spotlights the comedy inherent both in the hero’s temporary assumption of the role of villain and in his reaction to Josian’s wooing as if it were a military campaign. The borrowing of the French adjective (less frequent in Middle English) here was almost certainly prompted by its frequency in the Boeve passages that immediately precede this episode, where it is used of Brademound (lines 597, 657), Hermine (line 643), and Boeve himself (line 677). The translator allows himself to be guided by his original, from which he borrows a frequently used expression. He uses it, however, in a way that both acknowledges the rules of the target-language repertoire and successfully plays with them.35 It is in the formulaic portrayal of a subsidiary character, Sabot, that the translator remains closest to his original, while adapting the technique of the Anglo-Norman author to the poetics of Middle English narrative. As in the case of Bevis and Josian, the number of appositional epithets is significantly reduced, from more than twenty in Boeve to just five in Bevis. Of those five, however, four match the Anglo-Norman text fairly closely.36 Furthermore, those four phrases all highlight the aspect of Sabot’s character that the author of the original had also chosen to emphasize – his age. It is true that the author of Boeve describes Sabot as le combataunt, li senez, le guerrer, l’alosez, le ferant (‘the hard-hitting’), le chevaler vailant, and le franc (‘the distinguished’ or ‘the noble’ or ‘the brave’), but he most often refers to him as le flori (‘the hoary’ × 6) or o la barbe florie (‘the white-bearded’), le barbé (‘the bearded’ × 4), le blanc (‘the white-haired’ × 3) or le vels (‘the old’). The translator faithfully renders Sabaoth le franc (3699) as ‘þe hende kniȝt, Saber’ (5445 / p. 200), but the other references that he preserves are to Saber ‘þe hore’ (4150 / p. 154; 4165 / p. 155) and ‘Saber wiþ is berde hore’ (4247 / p. 158). They correspond, respectively, to Sabaoth le vels (2243), Sabaoth le flori (2275), and Sabaoth … o la barbe florie (2297). Despite the critical stereotype according to which Middle English romances are more conventional than their French originals, the above analysis of one

35

A puzzling feature of Bevis is the translator’s apparent reluctance to resort to the simple adoption, from the source text, of a word that has already been domesticated in Middle English. This kind of borrowing is surprisingly infrequent in Bevis, even in lexical sets of a rather technical nature, for which such a procedure would be perfectly natural. To take just one example, in rendering the many terms used to designate horses the translator shows both precision and consistency, but while he will occasionally use the naturalized ‘deistrer’, ‘mule’ or ‘palfrei’ to translate the Anglo-Norman destrer, mulete and palefrei, he does so much less often than he could have. For a list of direct borrowings in the Middle English romance, see Martin, ‘Studies in Some Early Middle English Romances’, pp. 109–13. See also p. 82 below on the A redactor’s avoidance of direct transfer in a passage packed with nouns of French derivation. 36 The one exception is the context-specific ‘þat was wis of dede’ (4952 / p. 182), which occurs only in A, in an episode that differs from the corresponding passages both in Boeve and in other Middle English redactions.

78



Boeve to Bevis: The Translator at Work

particular kind of convention in Boeve and Bevis, the appositional fixed epithet, suggests that the Middle English text is in this respect less conventional than the Anglo-Norman. It also shows that the comparatively few fixed epithets retained in Bevis were often at least indirectly triggered by the presence of equivalent phrases in Boeve. At times, indeed, the Middle English epithets are directly translated from the Anglo-Norman, and sporadic attempts are even made to mimic the contextual appropriateness of a number of such expressions in Boeve. The significantly reduced number of formulaic phrases of this kind in Bevis provides telling evidence of the influence of the target-language repertoire, yet the effort to match Middle English phrase to Anglo-Norman original indicates that translation from French had an important role to play in the formation of such a repertoire. Although oral transmission is usually associated with a profusion of stereotypical expressions, translation too, with its inherently standardizing drift, encourages formulaic density. Translators always welcome the convenience of having ready-made answers to recurrent translational problems. Composed of a relatively limited range of stock situations narrated in similar terms, and peopled by a relatively limited range of stock characters described in similar terms, romances invite the translator to be thrifty and recycle. The translator of Bevis is a skilled recycler, always happy to reuse a line, a couplet or even a longer passage. In some cases he reuses lines he has borrowed from the common generic repertoire, lines with which other translators have rendered similar material. In others, he reuses his own lines, which will then be reused by others and become productive members of the same repertoire, by virtue of their recurrence in an undoubtedly widely distributed text.37 The repertoire of the genre is thus simultaneously drawn from and added to, in a mechanism that has yet to be properly studied.38

37

Much work has been done on textual borrowings among the Auchinleck romances, especially in connection with Laura Hibbard Loomis’s ‘bookshop’ theory, long influential though now superseded by more recent scholarship: see Laura Hibbard Loomis, ‘Chaucer and the Auchinleck Manuscript: Thopas and Guy of Warwick’, in Essays and Studies in Honor of Carleton Brown (New York: New York University Press, 1940), pp. 111–28, and ‘The Auchinleck Manuscript and a possible London bookshop of 1330–1340’, PMLA, 57 (1942), 595–627. For current thinking on the way the celebrated manuscript was put together, see Alison Wiggins, ‘Imagining the compiler: Guy of Warwick and the compilation of the Auchinleck Manuscript’, in Imagining the Book, ed. by Stephen Kelly and John J. Thompson, Medieval Texts and Cultures of Northern Europe 7 (Turnhout: Brepols, 2005), pp. 61–73. As far as I know, however, only Maldwyn Mills has studied the correlation between translation and the likelihood of borrowing. He proposed an account of the creation of the Middle English Guy of Warwick in which borrowings from Amis and Amiloun feature prominently, and made a very convincing case for the theory that the translator of Guy treated Amis as a ‘repertoire of extended rhyme-sequences’ which he used for a number of translational purposes: see Maldwyn Mills, ‘Techniques of translation in the Middle English versions of Guy of Warwick’, in The Medieval Translator 2, ed. by Roger Ellis, Westfield Publications in Medieval Studies 5 (London: University of London, Queen Mary and Westfield College, Centre for Medieval Studies, 1991), pp. 209–29 (pp. 222–3). 38 I am grateful to the Social Sciences and Humanities Research Council of Canada for financial support in the preparation of this article.

79

6 The Middle English and Renaissance Bevis: A Textual Survey JENNIFER FELLOWS

I

n studying the history of the Bevis story in medieval and Renaissance England, one is confronted not by a single literary phenomenon but by several. The various redactions of the story in English must be considered not only in relation to the Anglo-Norman Boeve de Haumtone, their putative source, but also in relation to one another; for the differences between them go beyond the lexical and stylistic to manifest varying attitudes towards the story and varying conceptions as to its meaning. While there are too many missing links in the transmission of the Middle English text for it to be possible to trace a straightforward chronological development of the story in English, certain conceptual tendencies and shifts of emphasis can be detected in the English tradition taken as a whole, these being developed in varying degrees in the different surviving texts. My object in this chapter is to characterize the different verse redactions of Sir Bevis of Hampton, both in manuscript and in print, and to demonstrate that they all have something to tell us about the original version of the story in English and about the complexities of its textual transmission. Most of those who read the Middle English Bevis at all read it in the version represented by the Auchinleck MS (A), since this is the only text that has been published in a readable form, providing as it does the principal base-text for Eugen Kölbing’s edition of the romance. Because it is 

Der anglonormannische Boeve de Haumtone, ed. by Albert Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974) (hereafter Boeve). On verbal corres­ pondences between Bevis and Boeve, see Judith E. Martin, ‘Studies in Some Early Middle English Romances’ (unpub. Ph.D. diss., Univ. of Cambridge, 1968), pp. 109–13, 282–8; Jennifer L. Fellows, ‘Sir Beves of Hampton: Study and Edition’, 5 vols (unpub. Ph.D. diss., Univ. of Cambridge, 1980) (hereafter BHF), V, passim.  A list of manuscripts and editions of the verse Bevis is provided in the Appendix (pp. 104–13 below).  The Romance of Sir Beues of Hamtoun …, ed. by Eugen Kölbing, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) (hereafter BHK). Kölbing’s text of A appears to have been reproduced in the more recent TEAMS edition without recourse to the manuscript: see, e.g., Four Romances of England: ‘King Horn’, ‘Havelok the Dane’, ‘Bevis of Hampton’, ‘Athelston’, ed. by Ronald B. Herzman, Graham Drake and Eve Salisbury, Middle English Texts (Kalamazoo, MI: Medieval Institute, Western Michigan University, for TEAMS in

80



Bevis: A Textual Survey

also considerably the earliest of the extant manuscripts and can be shown to have unique points of contact with Boeve, it has acquired a position of presumed textual authority or ‘correctness’ which, however, is not entirely, or not consistently, justified. By indicating A’s relationship both to Boeve and to the other Middle English redactions of the romance, and by examining those episodes in which it differs most significantly from other versions of the story written in England, I hope to demonstrate how anomalous it in fact is. In order to account for the interrelationships between the extant Bevis texts, one must assume at least one intermediary between A and the Middle English original. Not all the features peculiar to A can, therefore, be ascribed to the Auchinleck redactor himself with 100 per cent certainty. Given that there are close parallels to A in later romances, such as the version of Sir Degaré in Cambridge University Library, MS Ff.2.38, and The Knight of Curtesy (both normally dated to the latter half of the fifteenth century), it does seem likely that the Auchinleck version was not the only one of its kind, even though it is the only one to have survived. There is nothing, however, to suggest that it was not the first of its kind. Any discussion of a peculiarly Auchinleckian version of the romance must focus on those episodes in which A stands alone against Boeve on the one hand and the remaining Middle English redactions on the other. There are five principal episodes of this sort. In order of occurrence they are: Bevis’s encounter with Terri on his journey to Damascus; the dragon fight; Josian’s seven-year separation from Bevis and her twin sons; the single combat against King Yvor; and Bevis’s street battle against the citizenry of London. There are two principal versions in Middle English of the meeting between Bevis and Terri: one represented by A alone, the other by the remaining manuscripts and the printed editions of the romance. Each contains echoes of the Anglo-Norman not found in the other, though by and large A is closer to the Anglo-Norman. It alone, for instance, agrees with Boeve in having



 



association with the University of Rochester, 1999), lines 40 (BHK’s gernede for A’s a ernede), 2814 (BHK’s fraiche for A’s finliche). E.g., the argument in Linda Marie Zaerr, ‘Medieval and modern deletions of repellent passages’, in Improvisation in the Arts of the Middle Ages and Renaissance, ed. by Timothy J. McGee, Early Drama, Art, and Music Monographs 30 (Kalamazoo, MI: Medieval Institute, 2003), pp. 222–46, is built on the premise that passages peculiar to A are omitted from other texts of Bevis in the interests of decency. In fact, as will be seen below, these passages seem to have been added in A, in one of its more substantial departures from the central Bevis tradition. See the stemma at BHF, I, 36; and cf. BHK, p. xxxviii. Cf. Nicolas Jacobs, ‘Sir Degarre, Lay le Freine, Beves of Hamtoun and the “Auchinleck bookshop”‘, Notes and Queries, 29 (1982), 294–301 (p. 299); BHF, V, 113–21, passim. The earliest surviving text of The Knight of Curtesy is a mid-sixteenth-century print by William Copland: see The Oxford Book of Late Medieval Verse and Prose, ed. by Douglas Gray (Oxford: Clarendon Press, 1985), p. 458. BHF, II, 1460–1648; III, 1057–1152; IV, 977–1066 / BHK, pp. 61–6; BHF, II, 3294–699; III, 2231–468; IV, 2235–506 / BHK, pp. 122–32; BHF, II, 4531–985; III, 3275–526; IV, 3230–434 / BHK, pp. 170–83; BHF, II, 5300–423; III, 3845–74; IV, 3809–42 / BHK, pp. 196–8; BHF, II, 5514–889; III, 3967–4160; IV, 3935–4130 / BHK, pp. 203–14.

81



Jennifer Fellows

the hero tell Terri that the man he is seeking has been hanged. At least to begin with, A seems to represent a rewriting of the version of the episode contained in Boeve, while the account contained in the other Middle English texts appears rather as an expansion of that version, tending to fill in the background to events (sometimes with material lifted from later passages in the romance) in a very much more diffuse way. There is no lack of narrative clarity in A’s account, however; indeed, some of the material peculiar to this version seems designed to achieve a more satisfactory integration of original with non-original elements and to effect smoother and more explicit transitions from one episode to the next: it is only here, for example, that we know for certain before Bevis himself does that the palmer he has encountered is none other than Terri in disguise. In my view, A’s version of the Bevis–Terri encounter is more effective in its economy than the more rambling account given elsewhere; but perhaps the most significant point to be made about it is a linguistic one. Lines peculiar to A contain an unusually high proportion of nouns of French derivation, though none of these occurs in the corres­ ponding passage either in the Anglo-Norman Boeve or in the Continental French version of the romance.10 In the space of fewer than forty lines we find medle (‘medlar tree’), the earliest recorded instance of the word in Middle English, male (‘wallet’), flaketes (‘flasks’), foisoun (‘plenty’), boiste (‘box’) and clerk (‘scholar’). It is a reasonable assumption that this lexical feature, which also characterizes some other passages peculiar to this version of Bevis and occurs only where it shows most creativity, reflects the idiolect of its author. This would suggest that, for the bulk of the romance, this author (who may or may not have been the Auchinleck scribe) was working from a Middle English Bevis antedating A. Other episodes in which A differs substantially from other extant versions of the romance seem designed to introduce or emphasize particular thematic concerns. In the Middle English Bevis tradition as a whole, by comparison with the Anglo-Norman, there is an increased emphasis upon specifically Christian values – though these are not, of course, entirely absent in the Anglo-Norman. The attempt to relate Bevis’s prowess and victories to his role as Christian knight operates most successfully within the confines of the single episode: in the boar fight, for instance, which occurs early in Bevis’s career, there is emphasis upon the hero’s antagonist as embodiment of evil, and Bevis’s eventual victory is seen as the direct outcome of prayer.11 The 

This explains why, when Bevis eventually returns to England to reclaim his patrimony, Saber expresses a belief that Bevis had been hanged: BHF, II, 3860 / BHK, p. 141.  For example, C’s lines ‘He ys yn a castell styffe and gode / Closyd wyth the salte flode / In a penakull of the see’ correspond to the bishop of Cologne’s description to Boeve of Sabot’s plight: ‘Sur un rocher de la mer est il herbergé / en un for chastel, ke il ad fundé’ (Boeve 1942–3): cf. BHF, II, 1550–2, 3712–19 / BHK, pp. 64, 135. 10 Boeve 820–66; Der festländische Bueve de Hantone, ed. by Albert Stimming, 5 vols, Gesellschaft für romanische Literatur 25, 30, 34, 41, 42 (Dresden: Niemeyer, 1911–20), lines 2448–524. 11 BHF, II, 898–9 / BHK, p. 39. For further discussion of the function of prayer in this episode, see Corinne Saunders, ‘Desire, will and intention in Sir Beues of Hamtoun’, in The Matter of

82



Bevis: A Textual Survey

same general pattern is followed in the account of Bevis’s battle against the dragon of Cologne – an episode peculiar to the Middle English Bevis and its Irish derivative.12 This battle, however, is treated at much greater length than that against the boar, and a crucial function is performed by a well in which a virgin has bathed and into which Bevis repeatedly falls: the dragon is unable to approach the well, whose waters heal the hero’s wounds and restore his strength. The dragon episode was perhaps introduced primarily in order to equate his status with that of other heroes of secular romance – Lancelot, Guy of Warwick and Wade are mentioned specifically in this context in all Middle English versions, St George only in A;13 but it also has more specifically Christian connotations, which A in particular is at pains to emphasize. Only in A, for example, is Bevis’s encounter with the dragon preceded by a numinous dream which suggests association of the virgin of the well with the Virgin Mary,14 and his eventual victory is effected by his calling upon her name.15 Also peculiar to this version is the description of the effects of the dragon’s venom upon the hero’s flesh: he becomes ‘A foule mesel alse ȝif a were’ (as if he were a filthy leper).16 The use of the ‘mesel’ image, with its connotations of moral disease (leprosy, of course, was commonly regarded as a punishment for sin), suggests an equation between the healing waters of the well and the purifying water of baptism, which was believed to have the power to purge away leprosy. The handling of the episode in A seems, then, as it were to prefigure the fully allegorized treatment of the well motif in canto xi of the first book of Spenser’s Faerie Queene. The dragon fight in Bevis is, in narrative terms, a completely detachable episode which bears no essential relationship to the rest of the story; but, although this and other such episodes are not related to an overarching thematic or structural concept, in A there is within the episode much more than a token gesture in the direction of Christian idealism.17 The same is true of Bevis’s last battle against King Yvor. In this episode it is in fact the printed texts of the romance that stand closest of all to the Anglo-Norman,18 though the manuscripts other than A are broadly in agreement with them. A presents a quite different, and much longer, version of the episode; it occasionally agrees with Boeve against the other Middle English texts, but mostly

12 13 14 15 16 17 18

Identity in Medieval Romance, ed. by Phillipa Hardman (Cambridge: Brewer, 2002), pp. 29–42 (pp. 35–6). See ‘The Irish lives of Guy of Warwick and Bevis of Hampton’, ed. and trans. by F. N. Robinson, Zeitschrift für celtische Philologie, 6 (1908), 9–180, 273–338 (pp. 294–6, 317–18). BHF, II, 3300–5, 3575 / BHK, pp. 122–3, 129. Cf. Judith Weiss, ‘The major interpolations in Sir Beues of Hamtoun’, Medium Ævum, 48:1 (1979), 71–6 (p. 72). BHF, II, 3646 / BHK, p. 131. Bevis also prays to the Virgin Mary in his single combat against Yvor: BHF, II, 5302 / BHK, p. 196. BHF, II, 3596 / BHK, p. 130. For a fuller discussion of the symbolism of this episode, see Jennifer Fellows, ‘ “Dragons two other thre”: the dragon motif in some Middle English romances’ (forthcoming). See BHF, V, 170–2, 245.

83



Jennifer Fellows

in conventional features typical of chanson de geste descriptions of battles.19 A alone parallels lines in the Anglo-Norman which state that Bevis and Yvor both pray to their respective gods before battle commences, and it develops this idea to represent their conflict as one between opposing creeds.20 Only in A is Bevis referred to as a ‘kniȝt of Cristen lawe’.21 In the Anglo-Norman Boeve, the hero’s career after the single combat with Yvor ends in a rather monotonous series of battles against the ­Saracens. The Middle English romance, by contrast, introduces an episode which provides a more striking climax to his career – his street battle against the citizens of London. It has been remarked that ‘English exploits ... dominate the poem’s ending and stamp the ineradicable basic Englishness of Beues and his origins firmly on our minds.’22 The element of ‘Englishness’ is underlined by the abundance of topographical detail, which indicates close knowledge of London and places the hero in an unquestionably English setting; this feature is particularly marked in A, which also shows more interest in practical details. Its description of the battle is more logical and realistic and less conventional and stylized than that attested by other texts, and details such as that of those who have no weapons arming themselves with staves are suggestive of possible acquaintance with the circumstances of actual street battles in thirteenth-century London.23 A’s interest in the character of Josian is also evinced in this episode. At one point a false report reaches her in Putney that Bevis has been killed: whereas all other texts show her advocating flight, in A she urges her sons to go and avenge their father’s death.24 She is thus shown here as a figure of as redoubtable heroism as her menfolk, whereas all other versions emphasize the sons’ valour alone, through their refusal to follow her more timid advice. The Middle English Bevis (and particularly A) seems, indeed, bent on developing Josian’s character in such a way as to make her to some extent a female counterpart to Bevis. The increased emphasis in the Middle English on her virginity until she at last marries Bevis is part and parcel of this: it not only symbolizes her loyalty to the hero but also functions as the feminine equivalent of Bevis’s honour as a knight. Where Bevis prevails through physical strength and courage, Josian shapes her fortunes through ingenuity and guile, which Northrop Frye sees as the feminine counterpart to those qualities in the male-dominated world of romance.25 Her wily resourceful19 20

21 22 23 24 25

See BHF, V, 170–2. A comparable episode, in which a religious debate is conducted in the course of a battle, occurs in Rouland and Vernagu, a Charlemagne romance also contained in the Auchinleck MS (fols 263r–267v). BHF, II, 5381 / BHK, p. 198. Martin, ‘Studies in Some Early Middle English Romances’, p. 154. BHF, II, 5548–9 / BHK, p. 204. See Martin, ‘Studies in Some Early Middle English Romances’, p. 155. BHF, II, 5728–31 / BHK, p. 209. Northrop Frye, The Secular Scripture: A Study of the Structure of Romance (Cambridge, MA; London: Harvard University Press, 1976), pp. 69–70.

84



Bevis: A Textual Survey

ness is particularly apparent in the scene of her wedding-night with Earl Miles,26 which is developed at considerable length in the Middle English – possibly under the influence of the analogous episode in the Nibelungenlied 27 – and where the doomed husband is induced to lock the door of the bridal chamber by a pretence of maidenly modesty on Josian’s part. It is further evinced in the substantial passage, peculiar to A (and again having a particularly high concentration of French-derived vocabulary),28 in which, when the treacherous Ascopard is carrying her back to her husband Yvor immediately after the birth of her twin sons, she reminds him of past obligations and persuades him to let her disappear from view for a moment to ‘do [her] nedes in priuite’.29 Having learned both physic and surgery in her native Ermonie, she immediately seeks out a herb which has the effect of giving her the appearance of a leper – with the result that Yvor wants nothing to do with her and has her kept in a remote castle with Ascopard as her warder. In a slightly later passage, also peculiar to A, Saber rescues Josian and disguises her as a palmer. (In other Middle English versions he anoints Josian with an ointment so that her hue waxes yellow and green.)30 At this point A develops a brief hint in the Anglo-Norman into a substantial passage in which Josiane supports herself and Saber, who has fallen ill, by performing as a minstrel;31 there is no parallel to this elsewhere in the Middle English Bevis. The overall effect of A’s treatment of these episodes is to enhance Josian’s standing as a resourceful and powerful figure in her own right. I should like also to suggest the possibility of influence here from thirteenth-century French romance. In the Roman de Silence, the parents of the eponymous heroine pass her off as a boy in order to get round inconvenient laws of inheritance; like Josian, she spends some time as a minstrel, and she too adopts the expedient at one point of altering her facial appearance through the use of a herb.32 An even closer parallel to the Auchinleck Bevis can be found in Aucassin et Nicolette, where the same combination of motifs occurs, and where Nicolette’s object in altering her appearance is, like Josian’s, to disguise her beauty.33 (Silence, on the other hand, already passes for a boy and wishes to make herself appear of ignoble birth.) There are no close verbal

26 27 28 29 30 31 32

33

BHF, II, 3988–4043 / BHK, pp. 147–9. See Laura Hibbard, ‘The Nibelungenlied and Sir Beves of Hampton’, Modern Language Notes, 26:5 (1911), 159–60. Cf. BHF, V, 148. BHF, II, 4615 / BHK, p. 172. BHF, II, 4909 / BHK, p. 181. Cf. Boeve 2779–80. BHF, II, 4958–77 / BHK, p. 182. Silence: A Thirteenth-Century French Romance, ed. and trans. by Sarah Roche-Mardi, ­Medieval Texts and Studies 10 (East Lansing, MI: Colleagues Press, 1992), lines 2037–58, 3138–43, 2907– 12. Aucassin et Nicolette: Chantefable du xiii e siècle, ed. by Mario Roques, CFMA (Paris: Champion, 1925), xxxviii (p. 36).

85



Jennifer Fellows

Plate 1  Naples, Biblioteca Nazionale, MS XIII.B.29, p. 30. (Reproduced by permission of the Ministero per i Beni e le Attività Culturali.) The tailrhyme passage in column b, where the merchants attempt to sell Bevis, is shared only by the Egerton MS (S). 86



Bevis: A Textual Survey

parallels between either of the French texts and A, but one of them may have provided the germ of an idea for the A scribe or his exemplar.34 Although A is the earliest surviving text of Bevis and at some points stands closer than others to the Anglo-Norman Boeve, there are sufficiently numerous and sufficiently substantial instances of its departure both from Boeve and from other Middle English texts to indicate that we should not regard it as an infallible guide to the readings of a notional ‘Ur-Bevis’.35 All extant texts of the metrical English Bevis, including the printed version taken as a whole, are sometimes alone in agreeing with the Anglo-Norman, and of the manuscripts it is indeed A that contains most of the more substantial narrative departures from its Anglo-Norman forebear and ultimate source. This is not to say, however, that the scribes of other manuscripts of the Middle English Bevis (all of them significantly later than A) do not seem to have felt free to add, expand or abbreviate episodes, to alter thematic emphases or to modify the characterization of the romance’s protagonists. The version represented by the closely related texts in London, British Library, MS Egerton 2862 (S) and Naples, Biblioteca Nazionale, MS XIII.B.29 (N) is perhaps the most conservative: it is a middle-of-the-road version, agreeing more with A in the earlier part of the romance (up until Bevis’s arrival in Cologne), more with Cambridge University Library, MS Ff.2.38 (C) in the second. It does, however, attempt to continue the tail-rhyme stanza beyond the point at which other texts abandon it, introducing a new episode in which the merchants to whom the young Bevis is sold deck him with garlands and silver chains in order to attract buyers (see Plate 1).36 The agreement of N and S in this episode, as well as in palpable errors not shared by any other text,37 suggests that they may well have derived from a common exemplar. That neither derives directly from the other is clear from occasional readings where one of them agrees with another manuscript against the other, and from p. 76 of N, where the scribe has left a four-line space – presumably because his exemplar was illegible or in some way defective at this point (see Plate 2); the missing lines are, however, present in S. Given NS’s agreements with A on the one hand and with C on the other, it seems likely that their common ancestor was, either in itself or in its origins, the product of textual conflation.

34 35 36

For further discussion of the characterization of Josian, see pp. 169–75 below. Cf. BHK, pp. xl–xli. BHF, II, 572–89 / BHK, p. 24. Christian Boje, Über den altfranzösischen Roman von Beuve de Hamtone, Beihefte zur Zeitschrift für romanische Philologie 19 (Halle: Niemeyer, 1909), p. 24, notes a parallel to this episode in the Continental French Bueve. On the change of metre at this point in most Bevis manuscripts, see Ivana Djordjević, ‘Versification and translation in Sir Beves of Hampton’, Medium Ævum, 74:1 (2005), 41–59; Linda Marie Zaerr, ‘Meter change as a relic of performance in the Middle English romance Sir Beues’, Quidditas, 21 (2000), 105–26; BHF, I, 86–90. 37 See, e.g., the nonsensical transposition in BHF, II, 22–3 / BHK, p. 2: ‘And euer he leuyd without wyf / As he was in eche stryf.’

87



Jennifer Fellows

Plate 2  Naples, Biblioteca Nazionale, MS XIII.B.29, p. 76. (Reproduced by

permission of the Ministero per i Beni e le Attività Culturali.) 88



Bevis: A Textual Survey

The textual authority of Cambridge, Gonville and Caius College, MS 175/96 (E) is hard to assess: only about one-third of the text of Bevis survives in this manuscript, and the only passages of sustained correspondence between E and other redactions occur at points where, in the absence of parallels in the romance’s Anglo-Norman source, it is impossible to conclude with any degree of certainty which version stands closest to the Middle English original. In Bevis’s Christmas Day battle against the Saracens, E is in close agreement with C against ANS, and both in the dragon fight and in Bevis’s final battle in the streets of London is closer to NSC than to A. It also agrees with C against A, and less frequently with A against C, in a number of minor variant readings that appear to be non-original.38 On the other hand, it is sometimes alone in sharing with A readings that are almost certainly original: it is, for example, the only manuscript apart from A that ever gives the name Deuoun to Bevis’s stepfather.39 It also agrees with A in making it clear that, when Ascopard abducts Josian after the birth of her children, he is in league with King Yvor;40 this is also a feature of Boeve and can therefore be presumed original.41 Such agreements – in combination with departure from A where that manuscript varies most from other texts (as in the London battle or in the dragon fight), and at least one passage where E is very much closer than any of the other Middle English texts to the Anglo-Norman42 – suggest that, had more of its text survived, E might have been a key witness to the original Middle English form of the Bevis story. MS C is more inclined to agree with NS or with Manchester, Chetham’s Library, MS 8009 (M) than with A and is characterized by a tendency to abridge some episodes and to expand others: thus Bevis’s battle with the lions is here reduced to a mere six lines, as against ANS’s eighty,43 while the account of Bevis’s slaughter of Saracens and destruction of their idols in Damascus is twice as long as A’s, with a proliferation of gory details.44 This expansion, of an episode probably directly influenced by the legend of St George,45 accords with an increased emphasis in the later English versions of the story upon Bevis’s role as Christian knight. This may be due in part to the revival of literary interest in epics of wars against Saracens consequent upon the fall of Constantinople in 1453,46 in part to the influence of the George 38 39 40 41 42 43 44 45

46

Cf. BHF, I, 39 n. 112, 46 n. 157. BHF, II, 3706 / BHK, p. 135. Cf. Doun in Boeve 2008 etc. BHF, II, 4523–62 / BHK, p. 170. Boeve 2655–73. The Trinity and Chetham’s MSS (see below) are in broad agreement with A and E here. BHF, II, 5035–101 / BHK, pp. 186–7, describing the events leading up to Bevis’s first battle against Yvor. For correspondences to Boeve, see BHF, V, 160–2. BHF, II, 3114–93 / BHK, pp. 116–18. This is also a feature of M (see below). BHF, II, 1704–25 / BHK, pp. 67–8. See John E. Matzke, ‘Contributions to the history of the legend of St George …’, PMLA, 17 (1902), 464–535 (pp. 484, 510); Samantha Riches, St George: Hero, Martyr and Myth (Stroud: Sutton, 2000), p. 58. See Margaret Schlauch, Antecedents of the English Novel 1400–1600 (Chaucer to Deloney) (Warsaw: PWN, 1963), pp. 49–50.

89



Jennifer Fellows

legend. At the same time, incidents potentially suggestive of a more passive mode of Christian virtue are introduced in various of the Middle English redactions, arguably in an attempt to redress the balance between the militant and spiritual virtues necessary to the perfect Christian knight. In the Middle English Bevis tradition as a whole, there is more emphasis than in Boeve upon the hero’s sufferings in prison, and some of the motifs that appear here relate the hero to St George in his role as martyr rather than as soldier. A and N, for example, describe how Bevis is bound to a great stone weighing seven quarters of wheat before being cast into prison – in certain versions of the martyrdom of St George a heavy stone, which four men can hardly lift, is placed upon the breast of the saint as he lies tied to the ground in his dungeon.47 The Christian concept of heroism as passive endurance, popular throughout the medieval period, enjoyed a resurgence in the later Middle Ages and during the Renaissance period, finding expression in, for example, Erasmus’ Enchiridion, where the virtue of fortitude is defined as consisting not in the performance of courageous acts or in the slaughter of the heathen, but in patient resignation to the will of God.48 It is entirely consistent with this trend that the two latest manuscripts of Bevis (C and M) and the printed editions of the romance should contain yet more material that relates the account of the hero’s imprisonment to that of the martyrdom of St George. The number of Bevis’s beastly antagonists is increased by the addition of two dragons – George likewise having to encounter a variety of such foes in his prison, according to certain versions of his legend – and (also like George) Bevis is visited in his dungeon by an angel.49 So why is it that variation between Bevis texts is so much greater in some episodes than in others? I would suggest that individual episodes from longer romances (particularly set pieces such as a dragon fight) were excerpted and circulated independently. Indeed, one Bevis manuscript, Cambridge, Trinity College, MS O.2.13 (T), may represent just such a process. Here, a single (and, to some extent, self-contained) episode, beginning with the English prince’s attempt to steal Arundel and ending with the reunion of Saber and Terri in Civile,50 has been detached from its context in the romance and scribbled on the blank leaves at the end of a quire. T appears not to be a fragment of a longer text, since it does not end at the foot of the page (see 47

BHF, II, 1895–6 / BHK, p. 73; cf. Matzke, ‘Contributions’, p. 468. In Boeve 914–15, by contrast, a heavy chain is fastened round the hero’s neck. S’s omission of these lines seems to be the result of scribal error: cf. BHF, V, 68. 48 Desiderius Erasmus, Enchiridion (Amsterdam: Theatrum Orbis Terrarum; New York: Da Capo Press, 1969), sigs [Avi]r–v, Biv–Biiv; cited in Irving L. Zupnick, ‘Saint Sebastian: the vicissitudes of the hero as martyr’, in Concepts of the Hero in the Middle Ages and the Renaissance, ed. by Norman T. Burns and Christopher Reagan (London [etc.]: Hodder & Stoughton, 1975), pp. 239–67 (p. 239). 49 BHF, II, 1911–24; III, 1343–54; IV, 1257–66 / BHK, p. 74; BHF, II, 2097–100; III, 1483–6 / BHK, p. 82. Cf. ‘Vie de Saint Georges’, in Les Œuvres de Simund de Freine, ed. by John E. Matzke, SATF 58 (Paris: Firmin-Didot, 1909), pp. 61–117 (lines 1168–1215). 50 BHF, II, 4431–93 / BHK, pp. 167–83.

90



Bevis: A Textual Survey

Plate 3),51 and the nature of its text would tend to corroborate a hypothesis of oral/memorial transmission: it does not correspond line for line to any other extant Bevis manuscript, and its readings often seem garbled to the point of nonsensicality.52 Quite possibly fairly discrete episodes such as this were transmitted orally: George Puttenham, in The Arte of English Poesie (1589), describes Bevis and Guy of Warwick ‘& such other old Romances or historicall rimes’ as ‘made purposely for recreation of the cōmon people at Christmasse diners & brideales, and in tauernes & alehouses and such other places of base resort’ and says that ‘by breaches or diuisions [they may] be more commodiously song to the harpe in places of assembly, where the company shalbe desirous to heare of old aduentures & valiaunces of noble knights in times past’.53 Oral transmission of this sort, independent of the written tradition, would go some way to explain why it is that two works of the 1590s (Spenser’s Faerie Queene and Richard Johnson’s Seven Champions of Christendom) both provide evidence that their authors knew versions of Bevis other than, and in addition to, the one that would have been most readily available to them in printed form (in one of William Copland’s editions from the 1560s, perhaps, or in Thomas East’s from c. 1582).54 We cannot be certain that what Puttenham describes holds true for the medieval period, though it is not implausible that it does, and it would account for particular types of variation within the manuscript tradition – features that are less plausibly explicable in terms of misunderstanding of a written exemplar.55

51 52

53 54

55

It should be noted, however, that the last page is the only one written by one of three different scribes. E.g. the opening lines of T (‘The kynges son swer by godes crose / That he shuld Aryndell stele / And myth leve & haue his hele’) read like a confused attempt to reproduce the sense of the lines as they appear in E (‘Þe kyngys sone wiþ B was wroþ / And euere more afftyr he was hym loþ / And swor hys oþ For woo or wele / Þat he scholde þat stede stele’): BHF, II, 4449–51 / BHK, p. 167. George Puttenham, The Arte of English Poesie 1589 (Menston: Scolar Press, 1968), pp. 69, 33; my italics. See BHF, V, 116–21; Richard Johnson, The Seven Champions of Christendom (1596/7), ed. by Jennifer Fellows, Non-Canonical Early Modern Popular Texts (Aldershot; Burlington, VT: Ashgate, 2003), pp. 293 (note to p. 106) and 300 (note to p. 156). In C’s version of the dragon episode, there is a passage of ten lines whose substance, while corresponding closely to that of other texts at this point, is differently divided into couplets. Here again, the disparity between C and other extant texts is less plausibly explained as the product of an exclusively written transmission than as the result of a scribe’s working from memory (albeit a pretty accurate memory): see BHF, II, 3427–36 / BHK, p. 126. See also Anne Herlyn, ‘Narrative Strukturen mittelenglischer Romanzen: am Beispiel des Beves of Hamptoun’, in Authors, Heroes, and Lovers: Essays on Medieval English Literature and Language. Selected Papers from the Studientage zum englischen Mittelalter, SEM I & II (Potsdam 1999 & 2000) / Liebhaber, Helden und Autoren: Studien zur alt- und mittelenglischen Literatur und Sprache. Ausgewählte Beiträge der Studientage zum englischen Mittelalter, SEM I & II (Potsdam 1999 & 2000), ed. by Thomas Honegger, Variations 2 (Bern [etc.]: Lang, 2001), pp. 1–26, passim. Herlyn argues that the narrative of Bevis is organized largely in episodes exhibiting a pattern of oral storytelling adapted to a written medium.

91



Jennifer Fellows

Plate 3  Cambridge, Trinity College, MS O.2.13, fol. 152r. (Reproduced by permission of the Master and Fellows, Trinity College, Cambridge.) 92



Bevis: A Textual Survey

Although there is no physical evidence that the fragment of Bevis in Oxford, Bodleian Library, MS Eng. Poet. d.208 (B) – which preserves a poorly written and much garbled version of Saber’s recovery of Arundel after he has been stolen by King Yvor and of Bevis’s single combat against Yvor56 – is not merely the surviving portion of a complete text of the romance, its confusions and illogicalities (as typified in the following extract) are such as might most readily be explained by such a theory of oral/memorial trans­ mission: To gedyr þy ȝedyn anothyr day Sir B smot Iuor wt morglay Þat þe crest to grounde fly He smot hym on þe helm an hy Þt a cantel of hys scheld Fly amydde warde þe feld Kyng Iuor meyte þerby I wete Of B þer was a gode smete Vp styrte Iuor wt gret radun Beuyȝe of hys hors adoun Hys swerd he hente al so hoȝt & at B helm a dynt he smoȝt Wt a swerd was grounde wel kene Vpon þe helm þe dynd was wel sene Sir B wt egyr mod ȝ ȝ Smo\ȝ/t Kyng Iuor þer he stod.  (fol. 49r) But even if B is not itself the direct product of this sort of textual transmission, that is not to say that it might not derive (at however many removes) from a text that was. We know from the other extant texts of Bevis, both in manuscript and in print, that the episode contained in B was subject to largescale variation57 – and this itself, as I have posited, may be symptomatic of excerpting either for manuscript circulation or for purposes of performance. Where the differences between texts are at their most radical in this episode, B is in broad agreement with the central Middle English tradition (as represented by NSC) against A; in smaller-scale variation, however, it is sometimes the only extant manuscript to agree with A: for example, B’s lines ‘Saber seyde þat for thowȝte hym / But i mowe it wynne a ȝen’ (fol. 50r) are paralleled in A alone.58 B thus bears further witness to the diversity and complexity of the manuscript tradition of Bevis in the fifteenth century. The most difficult manuscript text of Bevis to place in relation to others is the Chetham’s MS (M). The first 1314 lines of M and the last 654 (to which I shall refer collectively as Mq) correspond closely to the text of the printed Bevis, while the intervening portion (Ma) agrees rather with the manuscript 56 57 58

Cf. BHF, II, 5206–451 / BHK, pp. 192–200. Cf. pp. 83–4 above. Cf. BHF, II, 5210–11 / BHK, p. 192: ‘Þat Saber seide aþenkeþ me / Boute ȝif ich miȝte winne it aȝe.’ It is likely, however, that E, in a passage where only the first letter or so of each line survives, also agreed with A here (cf. ibid.).

93



Jennifer Fellows

tradition. Ma has certain affinities with C – notably in the omission of the traveller’s tales that Bevis tells to King Yvor, and in the drastically abridged account of his battle with the lions59 – but in general tends to agree rather with one or more members of the group AET against C. Where the witness of T is available, Ma is in close agreement with this manuscript, usually agreeing with E against T only in demonstrably original readings.60 Given that Ma is generally much closer to the later texts of Bevis, it seems likely that where it agrees with A against other manuscripts, the features in question are original. The striking affinities between Mq and the printed tradition (P) are immedi­ ately apparent: both rewrite the stanzaic opening of the romance in rhyming couplets, and there is virtually a line-for-line correspondence between them. It is equally clear, from inconsistencies between Ma and Mq, that M as a whole represents the imperfect conflation of material derived from two distinct exemplars: for example, Ma occasionally lapses into calling Bevis’s stepfather ‘the Emperour’,61 whereas Mq agrees with P in making him the emperor’s brother; at one point this error has been corrected by the Ma scribe (see Plate 4).62 Pace Kölbing, who posits a progressively defective common source for Mq and P in order to account for the shifting relationships between them,63 I would suggest that Ma constitutes the core of M and that the defici­ encies of M’s primary exemplar were made up from a text closely related to P. Although M is a late text, dating from the last quarter of the fifteenth century, and there is evidence that Bevis had been printed at least once before 1498,64 it is unlikely that the source of Mq was itself a printed text: the readings of Mq are frequently more ‘difficult’ than those of P,65 and sometimes demonstrably closer to the Middle English original,66 and Mq shares with ANSEC certain features that do not occur in P.67 Since lines 1467–1826 of P correspond closely to C,68 the evidence favours the supposition that the tradition to which both Mq and P belong is a comparatively late development, antedating the introduction of Bevis into print, from a form of the romance akin to C.69 Since substantial portions of this tradition survive only in P, the 59 60 61 62

63 64 65 66 67 68 69

Cf. BHF, IV, 2000–12, 2135–8 / BHK, pp. 110, 116–18, and cf. n. 43 above. See BHF, I, 39–41. E.g. BHF, IV, 2951. Ibid., 2685. The scribe of Ma does, however, appear to be consistent in correcting some discrepancies between his exemplars: thus Ma agrees with P in the name (Mordoure) given to Bevis’s stepfather. BHK, p. xxxix. See item 1 in the Appendix. Cf., e.g., M kyd coward / P preued a cowarde; M loos / P mede; M courith / P bideth (BHF, IV, 8, 22, 80; III, 8, 22, 80 / BHK, pp. 1, 2, 4. See BHF, V, 251–67, 286–90, passim. E.g. M’s lines ‘I wys modyr thou semyste full well / To be an hore an olde brothell’ have a counterpart in ANSC but not in P: see BHF, IV, 184–5; II, 329–30 / BHK, p. 14. BHK, pp. 81­–101. I am no longer of the opinion that Bevis was rewritten specifically for the press: cf. Jennifer Fellows, ‘Bevis redivivus: the printed editions of Sir Bevis of Hampton’, in Romance Reading on

94



Bevis: A Textual Survey

Plate 4  Manchester, Chetham’s Library, MS 8009, fol. 165v. (Reproduced by permission of Chetham’s Library, Manchester.) The scribal correction in the ninth line suggests the use of more than one exemplar. 95



Jennifer Fellows

early sixteenth-century editions of the romance are as important to a study of the text of Bevis in the fifteenth century as are the manuscripts themselves. The printed Bevis generally continues and develops the trends evident in the latest manuscripts, particularly in its emphasis upon the hero as Christian knight,70 but it also introduces its own modifications of the text. Although the extant printed texts of the metrical Bevis range in date from c. 1500 to 1711, there is little variation between them except at a lexical level, so they can be regarded as representing a single version of the romance. The salient characteristics of this version are the rewriting of the stanzaic opening found in most manuscripts, a tendency to name and provide motivation for minor characters (such as the wicked steward, here named as Sir Brian of Cornwall, who stirs up the citizens of London against our hero, ‘For Beuis had these offices / That some tyme were his’),71 an emphasis on the gruesome details of injury in battle,72 and some modification of the way in which Bevis’s character and his relationship with Josian are portrayed. In Boeve, and even more consistently in the Middle English manuscripts, Bevis’s behaviour towards Josian is harsh, discourteous and domineering, but the later English redactions slightly modify the way in which the relationship between hero and heroine is presented. Although instances of Bevis’s discourtesy towards Josian are still retained here, the printed Bevis also hints at more tenderness between the two and reduces or altogether omits Josian’s expressions of un­requited passion. It even suggests at one point a typically chivalric connection between love and prowess: Beuis loked vp to Iosyan And suche a comfort toke he than That the to lyons gryme and lothe At one stroke he slewe them both.73

P also particularly emphasizes the theme of pryse and honour. This latter is most clearly seen in the episode (shared by C) in which Bevis asks ­Brademond to grant him fair battle instead of condemning him to a dishonourable death and in Bevis’s interior monologue (peculiar to P) in which he decides against killing his stepfather by taking him unawares in his own hall.74 In the printed version, also, the chivalric virtue of largesse is constantly

70 71 72 73

74

the Book: Essays on Medieval Narrative Presented to Maldwyn Mills, ed. by Jennifer Fellows, Rosalind Field, Gillian Rogers and Judith Weiss (Cardiff: University of Wales Press, 1996), pp. 251–68 (p. 254). See, e.g., BHF, III, 484–6, 525–6, 1187–90 / BHK, pp. 29, 31, 68. BHF, III, 3175–6 / BHK, p. 165. E.g. BHF, IV, 515–18, 799–808, 3023–8 / BHK, pp. 30, 48, 160. BHF, III, 2125–8 / BHK, pp. 117–18. The 1580s edition by East, however, and all subsequent prints of Bevis, omit these lines – perhaps because they were felt to be out of keeping with the brusqueness with which Bevis has treated Josian in the immediately preceding passage. BHF, III, 1275–86, 2567–72 / BHK, pp. 72, 139.

96

Bevis: A Textual Survey



emphasized, and even the virtue of courtesy is ascribed to the hitherto rather loutish Bevis.75 Although the printed Bevis tradition is remarkably stable from a narrative point of view, the amount of textual variation in comparative minutiae is vast. Only four complete (or nearly complete) editions survive from the Tudor period; these range in date from the early years of the sixteenth century to the 1580s and are the work of Richard Pynson, William Copland and Thomas East.76 In addition, there is a number of extant fragments of Bevis editions from the first three decades of the sixteenth century – some of them attributed to Wynkyn de Worde, others to Julian Notary.77 The relationship between these texts is hardly less complex than that between the Bevis manuscripts. In brief, Pynson’s edition (O) – which, confusingly, Kölbing uses in combination with what I have designated Mq to provide the text in the lower part of the page in his edition – is a somewhat anomalous one, often standing alone against other surviving P texts, and being more radical in its rewritings in the interests of intelligibility; a considerable number of manuscript readings preserved in later editions has been lost from O. The following example, in which O is compared to MS S on the one hand, and to Copland’s 1565 edition (Q) on the other, illustrates both these points: S O Q

Leueþ me now þrouȝoute alle / At hoom y am his parmyngalle There he loued me wel wys. / He was my felawe and I was his For he louith me ouer all / For eyther was other promygall.78

As far as the fragmentary evidence goes, Notary’s texts seem to have corresponded quite closely to O, while William Copland’s two editions (dating from the 1560s) are more inclined to agree with de Worde’s.79 Rather surprisingly, the second of Copland’s editions does not seem to have been set up from his first (Cp). Cp has a number of substantial unique readings,80 as well as sometimes agreeing with O against Q.81 Q, on the other hand, though occasionally agreeing with O in readings that do not appear, from 75 76 77 78 79

80

81

BHF, III, 3163–72 / BHK, p. 165; quoted in Fellows, ‘Bevis redivivus’, p. 255. Nos. 4, 9, 11 and 13 in the Appendix. In addition to the title-page, Pynson’s edition (no. 4) lacks two leaves of text, corresponding to BHF, III, 1005–54, 2765–824 / BHK, pp. 59–61, 148–51. Nos. 2, 3, 5, 6 and 7 in the Appendix. BHF, II, 2805–6; III, 1887–8 / BHK, p. 104. Cf. BHF, II, 1–13, 33–96, 225–88, 509–74, 2831–66, 2872–941, 3191–284; but cf. also n. 81 below. On de Worde’s romance-publishing activities, see further Carol M. Meale, ‘Caxton, de Worde, and the publication of romance in late medieval England’, Library, 14 (1992), 283–98; on Copland’s, see A. S. G. Edwards, ‘William Copland and the identity of printed Middle English romance’, in The Matter of Identity in Medieval Romance, ed. Hardman, pp. 139–47. Cf., e.g., Cp ‘Beuis thrust hym in as it was thunder’ / Q ‘The ouergroine he smote in sunder’; Cp ‘That ye wolde be so good a man’ / Q ‘That ye had made that chafere’ (BHF, III, 666, 1822 / BHK, pp. 39, 101). Cf., e.g. BHF, III, 209, 445, 455. Where Cp agrees with O, the de Worde texts tend to agree with it rather than with Q; and where O, Cp and Q all differ, the de Worde texts tend to be in agreement with Cp: cf., e.g., BHF, III, 10, 56, 58, 3200.

97



Jennifer Fellows

the evidence of manuscript readings, to be those of their ultimate source,82 retains many readings that have manuscript authority but do not survive in O;83 these may well have derived from lost portions of de Worde editions. It therefore seems likely (given the idiosyncracies of both O and Cp) that, of the surviving sixteenth-century editions of Bevis, Q is closest to the exemplar from which they are all ultimately descended. Mostly, the variant readings in the printed Bevis tradition seem intended to make the text more readily intelligible, by the replacement of archaic or ‘difficult’ vocabulary.84 In a few instances, however, even the replacement of a single word seems to be motivated by other considerations. Thus later editions tend to use the title Sir less often and to reduce the number of references to the aristocracy: for example, in Snodham’s text of c. 1610 (V), and in all subsequent editions, Bevis leads not twenty thousand bold barons, but twenty thousand bold yeomen, against Brademond’s army, and even a ‘noble wine’ becomes a ‘pleasant wine’.85 And from the 1626 edition (G) onwards, the angel that appears to Bevis in his Damascus prison is replaced by ‘a Light’86 – perhaps because, in a period of considerable Puritan influence, angels were felt to smack of ‘popery’. As noted above, P as a whole is closer than any of the extant manuscripts to Boeve in the final battle between Bevis and Yvor. This is not the only example of agreement between Boeve and P against the Middle English manuscript texts: for example, P is the only English version of the romance to agree with the Anglo-Norman in indicating that Ascopard is black.87 Furthermore, even some late texts of P have manuscript affiliations not shared by earlier prints: thus East’s edition (K) is closer at one point to NS than to any intervening text: NS Q K

Sir Beuys tolde his eme in a while / Howe the emperour he did bigile Beuis tolde his eame as I you hete / Howe his stepfather and he gan mete Beuis tould his eame there, how he deceiued his stepfather.88

Likewise, lines in Snodham’s edition (V), dating from c. 1610 and in general not far removed textually from K, have peculiar affinities with C:

82 83 84 85 86 87 88

Cf., e.g., Q ‘His malice she sayd she wolde wette’ / O ‘His malice she saide she wolde bete’ / M ‘His malincoly there to abate’ (BHF, III, 582 / BHK, p. 34). Cf., e.g., the example quoted above. Cf. the examples cited in n. 80 above. V, sigs C1v, C2v (BHF, III, 758, 872). G, sig. D3v (BHF, III, 1483). BHF, III, 2220; cf. Boeve 1751. BHF, II, 3866–7; III, 2618 / BHK, p. 141; K, sig. [F.iv]v.

98

Bevis: A Textual Survey

C K V

Of the maystrye y am sewre, / Yf he wyll geue me gode armour And he will [vs] with him holde, we will defend him be he so bould And we will helpe him in that stour, if he will lend my men armour.89

Again, this could be interpreted as suggestive of the influence of oral factors in textual transmission – especially since in most respects the P texts are too closely related to warrant any assumption of independent derivation. The printed editions of the metrical Bevis are mostly small, apparently cheap, quartos.90 In all except James Nicol’s 1711 edition, the main text (though not always the title-page) is printed in black-letter type – itself an indication of cheap production for a mass market. From K onwards, all but the two Scottish editions (of 1630 and 1711 respectively) set out each couplet as a single long line (see Plate 5). Most of these books are illustrated, though here again the Scottish editions are an exception.91 I shall deal with the illustration of Bevis only briefly, since it has been discussed fully elsewhere by Siân Echard.92 Several of the woodcuts in O seem to have been made specifically for Bevis – that of Ascopard carrying Bevis, Josian and Arundel to the ship, for example, or that of Josian holding two horses while Bevis fights Ascopard93 – while others are ‘generic’ (i.e. they deal with stock subjects, such as a wedding or a battle, and can appropriately be used to illustrate a wide variety of texts). Several of them are recycled in later Bevis editions: four of them re-appear in Cp, for instance, and five of them in Q.94 There are no woodcuts in the surviving Notary fragments, and only one in the extant de Worde texts; this occurs in the edition of c. 1533 (W) and shows two knights jousting. Curiously, although Copland’s editions (which are far from being identical to each other in their illustrations) are generally closer textually to the de Worde fragments than to O, and although 89 90

91

92

93 94

BHF, II, 3770–1 / BHK, p. 137; K, sig. [F.iv]r; V, sig. [F4]r. On book prices in the early modern period, see H. S. Bennett, English Books & Readers 1475 to 1557 (Cambridge: Cambridge University Press, 1952), pp. 229–34; William St Clair, The Reading Nation in the Romantic Period (Cambridge: Cambridge University Press, 2004), pp. 62–5, 453–67. The guild system among English printers meant that not only the intellectual-property right to copy and sell a text, but also the associated capital stock, fonts, presses and woodblocks, would have been heritable and tradeable; this would tend to encourage textual stability and the use of a stock set of illustrations. Scottish editions before 1710, however, were produced under a different jurisdiction and may be regarded as offshore printings. They would have been seen in London as ‘piracies’ (although they were entirely legal under Scottish law). Cf. St Clair, The Reading Nation, pp. 104–6. I gratefully acknowledge Mr St Clair’s help on this point. Siân Echard, ‘Of dragons and Saracens: Guy and Bevis in early print illustration’, in Guy of Warwick: Icon and Ancestor, ed. by Alison Wiggins and Rosalind Field, Studies in Medieval Romance (Cambridge: Brewer, 2007), pp. 154–68. I am grateful to Professor Echard for allowing me to see this article before publication. O, sigs G.iiir, G.iv. O, sigs A.vr, B.ivr, G.iv, G.iiir, K.iir; Cp, sigs C.iiiv, [J.iv]r, K.iv, O.iiiv; Q, sigs [A.iv]v, [C.iv]v, K.iv, K.iiir, P.ir.

99

Plate 5 Thomas East’s edition (1582?), sigs [Eiv]v–F[i]r. (Reproduced by permission of the Provost and Fellows, King’s College, Cambridge.) This is the earliest surviving edition to set out each couplet as a single line.



Bevis: A Textual Survey

they both have an illustration with the same subject and in the same position as in W, the woodcut that they use replicates O’s rather than W’s.95 K and V are identical in their scheme of illustration, and it is only in generic wedding pictures that they differ.96 G uses exactly the same woodcuts as V – including a damaged one of the dragon fight.97 With Stansby’s edition of 1630 (H) comes a change in the style of illustration: the subject-matter and placing of pictures remains as in V (except that there is no wedding picture when Bevis’s son Miles marries the king’s daughter), but the idiom is more modern (see Plate 6). This pictorial scheme remains virtually unchanged for the rest of Bevis’s life as an illustrated metrical text.98 Bevis was the only Middle English romance to continue to be printed in its metrical form after the mid-1570s, when an influx of Spanish romances displaced many of the older stories, and it continued to appear in this form until a decline in the vogue of romance in general took place in the midseventeenth century.99 No editions of the metrical Bevis seem to have been printed between 1667 and 1711, though chapbooks of the romance began to appear at about this time and continued until around 1780.100 In these, the style of illustration is different again and has a distinct flavour of the Civil War period. A much-elaborated prose version was produced in 1689, perhaps by John Shurley,101 but by the time this was reprinted, with some additions, in 1775 the story of Bevis was (in the words of its printer, T. Baker of ­Southampton) ‘very little known’.102 No other Middle English romance remained in print in its late medieval form, substantially unchanged, as long as Bevis did. As has been seen, even late texts have something to tell us about the romance’s earliest manifestations in English as well as about changes in ideology and taste during the early

95

96

97 98 99 100

101

102

O, sig. K.iir; Cp, sig. O.iiiv; Q, sig. P.ir. The title-page woodcut in both Cp and Q corresponds closely to one used by Pynson in his 1526 edition of The Canterbury Tales: see Edward Hodnett, English Woodcuts 1480–1535, 2nd edn (Oxford: Oxford University Press, 1973), no. 1943 and fig. 170. They use the same picture for Josian’s wedding to Earl Miles (K, sig. D.iiv; V, sig. D.2v), but V (sig. G[1]r) repeats it for the wedding of Bevis’s son Miles, while K here uses one which is a little closer to Cp’s and to Q’s (K, sig. G.[i]r; cf. Cp, sig. S.iir and Q, sig. [S.iii]r). G, sig. F2r; V, sig. F2r. The last illustrated verse Bevis is A. Ibbitson’s 1667 edition (no. 22 in the Appendix). See Ronald S. Crane, The Vogue of Medieval Chivalric Romance during the English Renaissance (Menasha, WI: Banta, 1919) (hereafter Crane), pp. 7, 28. These versions of the romance are discussed more fully in BHF, I, 111–18, and in Fellows, ‘Bevis redivivus’, pp. 261–4. For a summary of the publishing and reception history of Bevis and Guy of Warwick up to the early years of the nineteenth century, see St Clair, The Reading Nation, pp. 503–5. The Famous and Renowned History of Sir Bevis of Southampton ... (London: for W. Thackeray and J. Deacon, 1689). Shurley’s name does not appear, but the edition is ascribed to him without comment in Arthur Johnston, Enchanted Ground: The Study of Medieval Romance in the Eighteenth Century (London: Athlone Press, 1964), p. 29. The History of the Famous and Extraordinary Sir Bevis of Southampton … (Southampton: T. Baker, 1775), sig. [Ai]v; Baker’s italics.

101



Jennifer Fellows

Plate 6  William Stansby’s edition (1630), sig. E3v, showing a change in the style of illustration (cf. Plate 5 above). (Reproduced by permission of the Folger Shakespeare Library, Washington, DC.)

102



Bevis: A Textual Survey

modern period. Derek Brewer once voiced the hope that in the life to come textual scholars will find in some great celestial research library all those ‘Ur-texts’ and ‘lost exemplars’ that have eluded them in this vale of tears. Satisfying as that would be, a textual tradition such as Bevis’s – with all its missing links, its ambiguities and its inconsistencies – is perhaps all the more fascinating for those very features that make it so hard to pin down.103

103

I am grateful to Professor Maldwyn Mills and to Dr Rhiannon Purdie for commenting on an earlier draft of this chapter.

103



Jennifer Fellows APPENDIX

This appendix aspires to be a complete listing of the extant manuscripts and printed editions of the verse Bevis, from the 1330s until 1711. For a list (undoubtedly incomplete) of chapbooks and other prose editions up until the late eighteenth century, see BHF, I, 32–5. Manuscripts

A  Edinburgh, National Library of Scotland, MS Advocates’ 19.2.1 (­ Auchinleck MS), fols 176r–201r. Date: 1330 × 1340. Title: Sir beues of hamtoun. Bevis is the twenty-fourth of 43 items in this substantial parchment manuscript, which consists of romances, chronicles, saints’ lives and other devotional works and is generally thought to have been produced in London. The manuscript as a whole has been regarded as expressing a sense of English national identity, and as being a ‘women’s manuscript’. Bevis is the work of ‘Scribe 5’, who also copied Reinbrun, the text that immediately precedes it. The dialect of this scribe has been located in south-west Essex. The text of Bevis is complete, except for the loss of a single leaf after fol. 188. Like most of the other items in the manuscript, it is written in double columns; 

 



 



For reproductions of the full manuscript, see The Auchinleck Manuscript: National Library of Scotland Advocates’ MS 19.2.1, introd. by Derek Pearsall and I. C. Cunningham (London: Scolar Press in association with the National Library of Scotland, 1977); The Auchinleck Manuscript, ed. by David Burnley and Alison Wiggins (Edinburgh: National Library of Scotland, 5 July 2003), version 1.1 (http://www.nls.uk/auchinleck/). For descriptions, see Gisela Guddat-Figge, Catalogue of Manuscripts Containing Middle English Romances, Texte und Untersuchungen zur englischen Philologie 4 (Munich: Fink, 1976) (hereafter Guddat-Figge), pp. 121–6; Pamela R. Robinson, ‘A Study of Some Aspects of the Transmission of English Verse Texts in Late Mediaeval Manuscripts’ (unpub. B.Litt. thesis, University of Oxford, 1972) (hereafter Robinson), pp. 120–37; BHF, I, 12–13. The Auchinleck Manuscript, introd. Pearsall and Cunningham, p. vii. Cf. Laura Hibbard Loomis, ‘The Auchinleck Manuscript and a possible London bookshop of 1330–1340’, PMLA, 57 (1942), 595–627; Timothy A. Shonk, ‘A study of the Auchinleck Manuscript: bookmen and bookmaking in the early fourteenth century’, Speculum, 60 (1985), 71–91; Ralph Hanna, London Literature, 1300–1380 (Cambridge: Cambridge University Press, 2005), pp. 104–47. See, e.g., Thorlac Turville-Petre, England the Nation: Language, Literature, and National Identity, 1290–1340 (Oxford: Clarendon Press, 1996), passim; Siobhain Bly Calkin, Saracens and the Making of English Identity: The Auchinleck Manuscript, Studies in Medieval History and Culture (London; New York: Routledge, 2005), passim. Thus Felicity Riddy, in an unpublished paper presented at the third Romance in Medieval England conference (Bristol, 1992). See Angus McIntosh et al., A Linguistic Atlas of Late Mediaeval English, 4 vols (Aberdeen: Aberdeen University Press, 1986) (hereafter LALME), III, 29 (LP 6350). Hiroaki Endo, ‘Middle English dialectology by a machine-readable corpus: with special reference to Sir Beues of Hamtoun’, Osaka daigaku gengo bunkagaku, 10 (2001), 5–23, however, favours a Southampton location. The missing passage corresponds to BHF, II, 2972­–3155 / BHK, pp. 111–17. On the basis of the normal number of lines per column in A, it is possible to calculate that it would have had exactly the same number of lines here as does NS.

104



Bevis: A Textual Survey

there are normally 44 lines per column; ruling is in ink. Paragraph signs, alternating red and blue, are used throughout the text. More significant divisions are marked by blue initials, three to five lines high, with red ornamentation. The initial letter of each line is detached from the rest of the line and picked out with red. Although most of the manuscript’s illuminations have been cut away, a few miniatures survive; the large initial at the beginning of Bevis contains the figure of a knight holding a spear. N  Naples, Biblioteca Nazionale, MS XIII.B.29, pp. 23–79. Date: 1450s. Title: Sir beuys of hamptoñ. The volume, written on paper, contains six items and appears to be the work of two scribes, one of them writing the medical prescriptions with which the manuscript begins (pp. 1–19), the other copying Bevis, ‘Of seint Alex of Rome’, Lybeaus Desconus, Sir Isumbras and Chaucer’s Clerk’s Tale. The manuscript’s dialect forms have been located to Dorset.10 Bevis is written in double columns; there are normally 35–38 lines per column; ruling is usually in ink, though a plummet has occasionally been used in the earlier pages.11 Descenders in the last line of a column have sometimes been elaborated into the forms of fish.12 A full-page drawing of a bell with a finger pointing to it (p. i) has been used to associate the manuscript with the seventeenth-century philosopher Tommaso Campanella, whose personal emblem was a bell,13 but it is not known how the manuscript came to Italy in the first place. E  Cambridge, Gonville and Caius College, MS 175/96,14 pp. 131–56. Date: 1450 × 1475.15 Incipit: Hic incipit beffs De hamptoun. Amen. The manuscript, usually thought to be the work of a single scribe,16 contains texts of Richard Coer de Lion, Sir Isumbras, Athelston and Bevis, as well as devotional works. The parchment is stained in places, and the writing 



10 11 12 13 14

15 16

Described in John M. Manly and Edith Rickert, The Text of the Canterbury Tales, Studied on the Basis of All Known Manuscripts, 8 vols (Chicago, IL: University of Chicago Press, 1940), I, 376–80; Lybeaus Desconus, ed. by M. Mills, EETS, os 261 (London: Oxford University Press for the EETS, 1969), pp. 6–8; Guddat-Figge, pp. 241–2; Robinson, pp. 186–9; BHF, I, 9–11. The evidence afforded by watermarks is insufficient to confirm whether the scribe’s date of 1457, at the end of the Clerk’s Tale (p. 146), applies to the whole manuscript. For discussion of the watermarks, see Lybeaus Desconus, ed. Mills, p. 7; Manly and Rickert, The Text of the ­Canterbury Tales, I, 376. LALME, III, 88 (LP 9490); Manly and Rickert, The Text of the Canterbury Tales, I, 377, however, assign the dialect forms to the East Midlands. Lybeaus Desconus, ed. Mills, p. 7. E.g., on pp. 23, 24, 29. Cf. Lybeaus Desconus, ed. Mills, pp. 7–8. Described in M. R. James, A Descriptive Catalogue of the Manuscripts in the Library of Gonville and Caius College, 3 vols (Cambridge: Cambridge University Press, 1907–14), I, 199–201; Athelston: A Middle English Romance, ed. by A. McI. Trounce, EETS, os 224 (London: Oxford University Press for the EETS, 1951), pp. 1–2; Guddat-Figge, pp. 82–3; Robinson, pp. 102–4; BHF, I, 12–13. Robinson, p. 103, citing Malcolm Parkes. Ibid.; Athelston, ed. Trounce, p. 2. However, Guddat-Figge, p. 82, identifies at least two scribes.

105



Jennifer Fellows

is sometimes partially illegible through wear; the ends of some lines have been obscured by the tightness of the binding. Trounce identifies the dialect forms as East Midland.17 All texts are set out in double columns. Rhymes are bracketed; in the tail-rhyme section of Bevis, some of the tail-lines are written alongside the brackets, while others are linked in pairs by larger brackets.18 Large, decorated initials occur at the beginning of all items and at intervals throughout all but the last (prose) text (De spiritu Gwydonis). In Bevis, these are two to four lines high and written in the same ink and the same hand as the body of the text, with touches of red; in the later part of the romance, some capitals within the line are picked out with red. There are signs in Bevis, of which only about one-third survives, of careful scribal correction. S  London, British Library, MS Egerton 2862 (Sutherland MS / Trentham MS),19 fols 45r–96v. Date: late fourteenth century / fifteenth century.20 Title: Beuous of Hampton. The manuscript is the work of a single scribe and is unusual in consisting entirely of romances: Richard Coer de Lion, Bevis, Sir Degarre, Floris and Blauncheflour, The Batell of Troye and Amis and Amiloun. Both marginalia and dialectal features suggest a Suffolk provenance.21 The first dozen leaves have been badly damaged and stained, so that they are barely (if at all) legible, and many of the other leaves have been heavily cropped. Space has been left for large initials (none of which has been filled in) at the beginning of each item in the manuscript whose opening lines survive and at intervals in the texts of all items. There are running heads throughout the manuscript. The text of Bevis, which lacks the final leaf,22 is written in single columns; there are normally 40 lines to the page; ruling is in ink. In the tail-rhyme section of the romance (fols 45r–49v), each rhyming couplet within the stanza is bracketed and the tail-line written alongside the bracket.

17 18

19

20 21 22

Athelston, ed. Trounce, pp. 45–52; LALME, III, 284 (LP 512) locates them more specifically in Lincolnshire. A similar practice is followed in Athelston, pp. 120–31. Cf. Rhiannon Purdie, ‘The implications of manuscript layout in Chaucer’s Tale of Sir Thopas’, Forum for Modern Language Studies, 41:3 (2005), 263–74 (pp. 263–4 and n. 13). Described in The Seege or Batayle of Troye: A Middle English Metrical Romance, ed. by Mary Elizabeth Barnicle, EETS, os 172 (London: Oxford University Press for the EETS, 1927), pp. xv–xvii; Catalogue of Additions to the Manuscripts in the British Museum in the Years MDCCCCVI–MDCCCCX (London: British Museum, 1912), pp. 238–40; Nicolas Jacobs, ‘The Egerton fragment of Sir Degarre’, Neuphilologische Mitteilungen, 72 (1971), 86–96 (pp. 86–90); E. Kölbing, ‘Vier Romanzen-Handschriften’, Englische Studien, 7 (1884), 177–201; GuddatFigge, pp. 182–3; Robinson, pp. 165–8; BHF, I, 7–9. Guddat-Figge, p. 182; Robinson, p. 167. Jacobs, ‘The Egerton fragment of Sir Degarre’, p. 88; LALME, I, 109; III, 485 (LP 8360). The missing text corresponds to BHF, II, 5900–6000 / BHK, pp. 214–18.

106



Bevis: A Textual Survey

T  Cambridge, Trinity College, MS O.2.13,23 149r–152r. Date: mid- to late fifteenth century.24 Title: none. Bound in a volume with seven other manuscripts, ranging in date from the fifteenth to the seventeenth century, this paper manuscript consists of two sixteen-leaf quires, followed by one twelve-leaf quire containing 245 lines of Bevis;25 the latter lacks six leaves, probably from the beginning.26 The Bevis fragment, apparently included to fill up the blank leaves at the end of the quire, is untidily written by three hands; errors and cancellations occur frequently. There is no ruling or decoration, though the top line of the page is sometimes written in a bolder and more elaborate script. The text is written in single columns, with 34–41 lines per page. Dialectal features in Bevis suggest a possibly East Anglian provenance.27 B  Oxford, Bodleian Library, MS Eng. Poet. d.208,28 fols 49r–50v. Date: mid- to late fifteenth century.29 Title: none. The volume is a guardbook, assembled in the early 1950s by Colonel Moss and containing, among other items, paper fragments of three Middle English texts: the stanzaic Owayne Miles, Syr Tryamour and Bevis. The two leaves of Bevis have been mounted in the wrong order. The Middle English texts are the work of four scribes, one of whom wrote the first page (fol. 50r), and another the remainder, of the Bevis fragment.30 All of them are set out in single columns; in Bevis there are 31–34 lines per column. The first page of Bevis has been heavily corrected by the scribe.31 Dialectal features in Bevis ‘seem to point to training in a band running from central Sussex to southeastern Surrey’.32

23

24 25 26 27 28 29 30 31 32

Described in M. R. James, The Western Manuscripts in the Library of Trinity College, Cambridge, 3 vols (Cambridge: Cambridge University Press, 1900–4), III, 97–9; Guddat-Figge, pp. 86–7; BHF, I, 19–21. I am grateful to Dr Richard Beadle for giving me the benefit of his expertise with regard to this manuscript. Thus Richard Beadle, in a personal communication, on palaeographical grounds. BHF, II, 4431–993 / BHK, pp. 167–83. Guddat-Figge, p. 87 n. 2. E.g. thowth for thought, myth for might: cf. LALME, II, 224. Described in Ralph Hanna, ‘Unnoticed Middle English romance fragments in the Bodleian Library: MS Eng. poet. d.208’, Library, 21:4 (1999), 305–20. Ibid., p. 310. Ibid., pp. 310, 313–14. Cf. ibid., p. 313, where the page is reproduced. Ibid., p. 310.

107



Jennifer Fellows

M  Manchester, Chetham’s Library, MS 8009,33 fols 122r–190v. Date: c. 1470 × 1480.34 Incipit: Here begynnyth A Good Tale of Bevis of Hamtoñ A Good Werriour. Formerly the property of the eighteenth-century antiquarian Richard Farmer,35 this paper manuscript is the work of (probably) nine scribes36 and contains fourteen items, both prose and verse. As well as a number of saints’ lives and other items, it contains three romances: Torrent of Portyngale, Bevis and the stanzaic version of Ipomadon. It is probably of London provenance and may have been compiled for a woman reader.37 Bevis is the work of the scribe who copied about half of the manuscript, including Ipomadon and part of Torrent. The dialect of this text (though not of its scribe) has been characterized as south-western.38 Bevis is written in single columns, with normally 30 lines per page. The incipit and explicit are written in a large semi-textura hand, which also provided the heading to ‘The lyf of Seint katrin’ (fols 30r–47v). C  Cambridge University Library, MS Ff.2.38,39 fols 102v–134r. Date: late fifteenth / early sixteenth century.40 Title: none. This substantial paper manuscript contains forty-three items, including religious and didactic texts, The Seven Sages of Rome and nine Middle English romances: The Erle of Tolous, Sir Eglamour of Artois, Syr Tryamour, the Northern Octavian, Bevis, Guy of Warwick, Le Bone Florence of Rome, Robert of Cisyle and Sir Degare. It is the work of a single scribe, who set out all the texts (both verse and prose) in double columns; in Bevis there are normally 32 lines per column. The scribe’s hand is characteristic of the late period suggested by watermarks;41 marks of abbreviation are used with apparent inconsistency. Dialectal forms seem to suggest a Leicestershire provenance.42 33

34 35 36 37

38 39

40 41 42

Described in N. R. Ker, Medieval Manuscripts in British Libraries, 4 vols (Oxford: Clarendon Press, 1969–92), III, 361–4; Kölbing, ‘Vier Romanzen-Handschriften’, pp. 195–201; Ipomadon, ed. by Rhiannon Purdie, EETS, os 316 (Oxford: Oxford University Press for the EETS, 2001), pp. xviii–xxiii; Guddat-Figge, pp. 238–40; BHF, I, 21–4. Both watermarks and contents (which include an account of a meeting that took place in 1473) support this dating: see Ipomadon, ed. Purdie, pp. xxii–xxiii, xviii. See BHF, I, 24. Ipomadon, ed. Purdie, pp. xx–xxi. See Carol M. Meale, ‘The Middle English romance of Ipomedon: a late medieval “mirror” for princes and merchants’, Reading Medieval Studies, 10 (1984), 136–79 (p. 138); Rhiannon Purdie, ‘Sexing the manuscript: the case for female ownership of MS Chetham 8009’, Neophilologus, 82:1 (1998), 139–48. Ipomadon, ed. Purdie, pp. xli–liv. On the distinction between the dialect of the text and that of its scribe, see ibid., p. xlix. For a facsimile of the full manuscript, see Cambridge University Library MS Ff.2.38, introd. by Frances McSparran and P. R. Robinson (London: Scolar Press, 1979). See also Le Bone Florence of Rome, ed. by Carol Falvo Heffernan, Old and Middle English Texts (Manchester: Manchester University Press; New York: Barnes & Noble, 1976), pp. 1–3; BHF, I, 14–19. Cf. Cambridge University Library MS Ff.2.38, introd. McSparran and Robinson, p. xii. Ibid., p. xiv. LALME, III, 244–5 (LP 531).

108



Bevis: A Textual Survey

Printed texts

For frequently cited works of reference, the following abbreviations are used: EEBO Early English Books on Line (http://eebo.chadwyck.com) Esdaile Arundell Esdaile, A List of English Tales and Prose Romances Printed before 1740 (London: Bibliographical Society, 1912) Hazlitt W. Carew Hazlitt, Hand-Book to the Popular, Poetic and Dramatic Literature of Great Britain: From the Invention of Printing to the Restoration (London: Russell Smith, 1867) Sayle C. E. Sayle, Early English Printed Books in the University Library, Cambridge (1475–1640), 4 vols (Cambridge: Cambridge University Press, 1900–7) STC A Short-Title Catalogue of Books Printed in England, Scotland, & Ireland, and of English Books Printed Abroad, 1475–1640, ed. by W. A. Jackson, F. S. Ferguson and Katharine F. Pantzer, rev. edn, 3 vols (London: Bibliographical Society, 1976–91) Wing Short-Title Catalogue of Books Printed in England, Scotland, Ireland, Wales, and British America and of English Books Printed in Other Countries 1641–1700, ed. by Donald G. Wing, 3 vols (New York: Columbia University Press for the Index Society, 1948–51). Where there is doubt as to an edition’s date and/or printer, I have (unless otherwise stated) followed the attribution made in the relevant union catalogue (STC or Wing) or in the catalogue of the library in which the book is held. Where a particular edition has been discussed above, I conclude by giving the siglum by which I have referred to it. Except where indicated, digital images of these texts are available on EEBO. 1 A printed edition of Bevis, now lost, seems to have been in existence before 1498.43 2 c. 1500 A fragment consisting of two leaves, containing lines 33–96, 225–88.44 Attributed to de Worde.45 Black-letter; 32 lines per page; 2 lines per couplet. Oxford, Bodleian Library. STC 1987; Hazlitt (b). 3 c. 1500. A fragment of the first leaf of the text, containing lines 1–13, 31–43. The large initial at the beginning has been cut away, with consequent losses from 43

See H. R. Plomer, ‘Two lawsuits of Richard Pynson’, Library, 38:10 (1909), 115–33 (pp. 122, 126–8). 44 Line references are to BHF, III. 45 The flyleaf has the following note (in Francis Douce’s hand?): ‘Printed with the same type as that used in the fragments of Robyn Hode and Guy of Warwick, and several others mentioned in the note to the fragment of Robyn Hode, all of which were certainly pr. by Wynkyn de Worde.’

109



Jennifer Fellows

the verso.46 Attributed to de Worde.47 Black-letter; 30 lines per page; 2 lines per couplet. Cambridge University Library. STC 1987.5; Sayle 60. 4 c. 1503. Lacks title-page. Colophon: ¶ Emprynted by Rycharde Pynson in Flete- / strete  at  the  sygne  of  the  George. Quarto, mostly in 6s: A–B8, C–L6, M8. Lacks Ai, Cv, Hvi. Signatures often erroneous. Black-letter; 30 lines per page; 2 lines per couplet; 12 woodcuts. Oxford, Bodleian Library. STC 1988; Hazlitt (a). Siglum: O. 5 c. 1510 (a) A fragment consisting of one leaf (sig. Hi), containing lines 2831–66. Thirty-two(?) lines lost through removal of woodcut. Formerly assumed to be the same edition as no. 6 below. Attributed to Julian Notary. Black-letter; 35 lines per page; 2 lines per couplet. Melbourne, State Library of Victoria. STC 1988.2. Not on EEBO. (b) A fragment consisting of one leaf (sig. Hii), containing lines 2872– 941. Formerly assumed to be the same edition as no. 6 below. Attributed to Julian Notary. Black-letter; 35 lines per page; 2 lines per couplet. San Marino, Henry E. Huntington Library. STC 1988.2. Not on EEBO. 6 c. 1515 A fragment consisting of two leaves (sigs Biii, Civ), containing lines 509– 74, 1037–87. Attributed to Julian Notary. Black-letter; 38 lines per page; 2 lines per couplet. Alexandria, Virginia Theological Seminary. STC 1988.4. Not on EEBO. 7 c. 1533 A fragment consisting of two leaves (sigs Oii, Oiii), containing lines 3191–284. The outer margin of Oii is heavily cropped, so that the lines on the verso are acephalous. Attributed to de Worde.48 Black-letter; 32 lines per page; 2 lines per couplet; 1 surviving woodcut. Cambridge University Library. STC 1988.6; Sayle 212. Siglum: W. 8 Licensed to Thomas Marshe between 10 July 1558 and 10 July 1559.49

46

Cf. E. Gordon Duff, Early English Printing: A Series of Facsimiles of All the Types Used in England during the XVth Century ... (London: Paul, Trench, Trübner, 1896), plate xii. 47 See Henry Bradshaw’s manuscript note accompanying the fragment. 48 See Henry Bradshaw, Collected Papers of Henry Bradshaw, Late University Librarian (Cambridge: Cambridge University Press, 1889), p. 349. 49 Edward Arber, A Transcript of the Registers of the Company of Stationers of London 1554–1640 ad (London: privately printed, 1875–94) (hereafter Arber), I, 95. See Crane, p. 36.

110



Bevis: A Textual Survey

9 1560?50 Title-page:  Syr Beuys of / Hampton / [woodcut]. Colophon: Imprynted at lon- / don in the vinetre vpon the thre / Crane wharfe, by William / Coplande. Quarto in 4s: A–S4. Black-letter; 32 lines per page; 2 lines per couplet; 8 woodcuts. San Marino, Henry E. Huntington Library. STC 1988.8; Hazlitt (c). Siglum: Cp. 10 Licensed to John Tysdale, 11 May 1561.51 11 1565? Title-page:  Syr Beuys of / Hampton / [woodcut]. Colophon: ¶ Imprented at London, in Lothburye by / Wyllyam Copland. Quarto in 4s: A–S4. Black-letter; 32 lines per page; 2 lines per couplet; 12 woodcuts. London, British Library. STC 1989. Siglum: Q. 12 Licensed to John Alde between 22 July 1568 and 22 July 1569.52 13 1582?53 Title-page: SYR BEVIS OF / Hampton. / [woodcut] / Printed by Thomas East / dwelling in Aldersgate streete. Colophon: Imprinted at London by Thomas / East, dwelling in Aldersgate streete, / at the signe of the black / horse. Quarto in 4s: A–I4. Black-letter; 38 lines per page; 1 line per couplet; 17 woodcuts. Oxford, Bodleian Library; Cambridge, King’s College. STC 1990; Hazlitt (d). Siglum: K. 14 The inventory of the Edinburgh printer Robert Gourlaw (d.1585) includes an edition of Bevis, almost certainly in verse.54 15 c. 1610 Title-page lacking.55 Sig. A2r begins: THE HISTORIE OF / Beuis of Hampton. Attributed to T. Snodham. Quarto in 4s: A–I4. Black-letter; 38 lines per page; 1 line per couplet; 17 woodcuts. London, British Library. STC 1992; Hazlitt (f ). Siglum: V.

50 51 52 53

See Crane, p. 37. Arber, I, 156. See Crane, p. 37. Arber, I, 389. See Crane, p. 38. Thus STC; but cf. Sayle, I, 317, who states that East started to print at the sign of the Black Horse in 1589. 54 Cf. Priscilla Bawcutt, ‘English books and Scottish readers in the fifteenth and sixteenth centuries’, Review of Scottish Culture, 14 (2001/2002), 1–12 (pp. 2, 8); Rhiannon Purdie, ‘Medieval romance in Scotland’, in A Companion to Medieval Scottish Poetry, ed. by Priscilla Bawcutt and Janet Hadley Williams (Cambridge: Brewer, 2006), pp. 165–7 (pp. 169–70). I am grateful to Dr Purdie for these references. 55 The manuscript title-page is apparently copied from a later edition (no. 16 below).

111



Jennifer Fellows

16 1626? Title-page: SYR BEVIS OF / Hampton. / [woodcut] / LONDON / Printed by G. W. for W. Lee.56 Quarto in 4s: A–I4. Black-letter; 38 lines per page; 1 line per couplet; 17 woodcuts. Oxford, Bodleian Library. STC 1993. Siglum: G. 17 c. 1630 Title-page: BEVIS OF / HAMPTON. / Newly Corrected and amended. / [woodcut] / LONDON. / Printed by William Stansby. Quarto in 4s: A–H4, I2. Black-letter; 38 lines per page; 1 line per couplet; 17 woodcuts. Washington, Folger Shakespeare Library; San Marino, Henry E. Huntington Library. STC 1994; Hazlitt (e). 18 1630 Title-page: The Historie of / Sir Bevis of South- / Hampton. / [printer’s ornament] / Printed in ABERDENE / by Edvvard Raban, For / David Melvill. 1630. Octavo in 8s: A–I8; 144 pp.; some errors in pagination. Black-letter; 30 lines per page; 2 lines per couplet; no woodcuts. London, British Library. STC 1995; Hazlitt (h). 19 1639? Title-page: SIR / BEVIS OF / HAMPTON / Newly Corrected and amended. / [woodcut] / LONDON / Printed by Richard Bishop. Quarto in 4s: A–H4, I2. Black-letter; 38 lines per page; 1 line per couplet; 16 woodcuts. Oxford, Bodleian Library; Washington DC, Folger Shakespeare Library. STC 1996; Hazlitt (i). 20 c. 1654? Title-page: THE / HISTORY / OF / The famous and renowned Knight / Sir Bevis of Hampton. / [woodcut] / LONDON, / Printed by G. D. for Andrew Crook, and are to be sold at the sign / of the Green Dragon in Pauls Church-yard. Quarto in 4s: A–H4. Black-letter; 42 lines per page; 1 line per couplet; 16 woodcuts. Cambridge University Library; Washington DC, Folger ­Shakespeare Library. Not in Wing. Not on EEBO. 21 1662 Title-page: THE / HISTORY / OF / The famous and renowned Knight / Sir Bevis of Hampton. / [woodcut] / LONDON, / Printed by G. D. for Andrew Crook, and are to be sold at the sign of / the Green Dragon in Pauls Churchyard. 1662. Quarto in 4s: A–H4. Black-letter; 42 lines per page; 1 line per couplet; 16 woodcuts. Oxford, Bodleian Library. Wing H2160;57 Hazlitt (k). Not on EEBO. 56 57

The first edition of STC attributes the edition to C. Wright, the second to G. Wood. According to Wing, copies of this edition are also held by the Folger Shakespeare Library and

112



Bevis: A Textual Survey

22 1667 Title-page: THE / HISTORY / OF / The Famous and Renowned Knight / Sir BEVIS of Hampton. / [woodcut] / LONDON, / Printed by A. Ibbitson, for Andrew Crook. 1667. Quarto in 4s: A–H4. Black-letter; 42 lines per page; 1 line per couplet; 16 woodcuts. Washington DC, Folger Shakespeare Library. Not in Wing. Not on EEBO. 23 [1711]58 Title-page: THE MOST FAMOUS / HISTORY / OF THE VALIANT / AND RENOWNED / CHAMPION, / BEUIS / OF / HAMPTON. / With a Narration of his strange / Adventures, Fights and Battels. / Very delectable to be Read. / [single rule] / ABERDEEN, / Printed by JAMES NICOL, Printer to / the TOWN and UNIVERSITY. 120 pp. [A]8, B4, C8, D–E4, F2, G4, H2, I4, K2, L4, M2, N4, O2, P4, Q4 (including 2 fly-leaves). Roman type; 36 lines per page; 2 lines per couplet; no woodcuts. Oxford, Bodleian Library. Hazlitt (n); Esdaile, p. 164. Not on EEBO.

by Lambeth Palace Library. However, the Folger copy is in fact no. 20 above, and there is no evidence that Lambeth Palace Library has ever held this item. I am grateful to Jennifer Higham, Assistant Librarian at Lambeth Palace Library, for her help on this point. 58 The title-page of the Bodleian copy is undated, but the other works with which this edition is bound both bear the date 1711. Both Hazlitt and Esdaile likewise date this Bevis edition to 1711.

113

7 For King and Country? The Tension between National and Regional Identities in Sir Bevis of Hampton ROBERT ALLEN ROUSE

T

he discourse of national identity in medieval England has been the subject of much critical debate in the past decade. The publication in 1996 of Thorlac Turville-Petre’s England the Nation established the study of medieval English nationalism as a vibrant and important field of study, and numerous additions to the debate over the origins, development and nature of medieval notions of Englishness have appeared since. Important studies by scholars such as Siobhain Bly Calkin, Geraldine Heng, Jeffrey Jerome Cohen and Kathy Lavezzo illustrate the degree to which the study of nationalism has become embedded within the practice of medieval scholarship. This chapter seeks to examine the narrative of English identity found in Sir Bevis of Hampton, reconsidering it in the light of two important geographical foci of the romance – the region of Hampshire and the lands of the East – in order to highlight the complexities of identity that are suggested by Bevis’s continual geographical relocation within the romance. Bevis is, as Turville-Petre has argued, a text that is deeply concerned with the construction of Englishness. In describing the Auchinleck MS ­(Edinburgh, National Library of Scotland, MS Advocates’ 19.2.1), the home of the bestknown and oldest extant Middle English version of Bevis, as ‘a handbook of



See Siobhain Bly Calkin, Saracens and the Making of English Identity: The Auchinleck Manuscript (London; New York: Routledge, 2005), pp. 7–10; Geraldine Heng, The Empire of Magic: Medieval Romance and the Politics of Cultural Fantasy (New York: Columbia University Press, 2003), pp. 6–8; Jeffrey Jerome Cohen, Medieval Identity Machines (Minneapolis, MN: University of Minnesota Press, 2003); Kathy Lavezzo, Imagining a Medieval English Nation (Minneapolis, MN: University of Minnesota Press, 2004), pp. vii–xxxiv; Thorlac Turville-Petre, ‘Afterword: the Brutus prologue to Sir Gawain and the Green Knight’, ibid., pp. 340–6; Robert Allen Rouse, The Idea of Anglo-Saxon England in Middle English Romance, Studies in Medieval Romance (Cambridge: Brewer, 2005), pp. 70–5; and, for a counter-argument, Derek Pearsall, ‘The idea of Englishness in the fifteenth century’, in Nation, Court and Culture: New Essays on FifteenthCentury English Poetry, ed. by Helen Cooney (Dublin: Four Courts Press, 2001), pp. 15–27.

114



National and Regional Identities in Sir Bevis of Hampton

the nation’, Turville-Petre argues that the manuscript’s narrative of England, written in English, ‘does not simply recognise a social need but is an expression of the very character of the manuscript, of its passion for England and its pride in being English’. The individual romances within the manuscript contribute to the construction of the manuscript’s idea of Englishness – as Heng’s analysis of Richard Coer de Lyon has demonstrated. While the historical royal figure of Richard I is one embodiment of the English nation in the Auchinleck MS, in Bevis we witness another important narrative space in which Englishness is constructed. In this respect the Middle English Bevis differs from its antecedent, the Anglo-Norman Boeve de Haumtone: as Susan Crane notes, ‘Sir Beues of Hamtoun undertakes an important development, whose beginnings are barely discernible in Boeve, from the perception of the baronial family as a political unit owing personal allegiance to rulers on the basis of reciprocal support, to a wider perception of national identity and the importance of national interests.’ The degree to which the Auchinleck Bevis (hereafter A) has been refocused towards this ideological task can be seen in the additions to the text that were made during its redaction: a number of unique additions to the Auchinleck romances can be seen as a direct result of the manuscript’s attempt to ‘English’ the texts. One of the unique aspects of A is a reference to St George that occurs during Bevis’s battle with the dragon of Cologne. Weary and battered from his day-long combat, Bevis finds refuge at a well in which a virgin has once bathed, the virtue of whom imbues the water with a holiness that repels the dragon. Refreshing himself with the blessed water, Bevis calls upon St George as he girds himself to face the dragon once more: ‘A nemenede sein Gorge, our leuedi kniȝt, / And sete on his helm, þat was briȝt’ (p. 129). Bly Calkin observes that the manuscript identifies Bevis with St George, who, by the date of the compilation of the Auchinleck MS, had become established as a figure associated with England. From the legends of his appearance  



 





Thorlac Turville-Petre, England the Nation: Language, Literature, and National Identity, 1290– 1340 (Oxford: Clarendon Press, 1996), p. 112. Ibid., p. 138. The Auchinleck MS’s ‘Englishing’ of the romances even goes as far as including a prologue to the two romances of French heroes (Roland and Vernagu and Otuel a Knight), which introduces their exploits but neglects to inform the reader that they are in fact French. Geraldine Heng, ‘The romance of England: Richard Coer de Lyon, Saracens, Jews, and the politics of race and nation’, in The Post-Colonial Middle Ages, ed. Cohen, pp. 135–72 (an expanded version of which is to be found in chapter 2 of Heng’s Empire of Magic). Susan Crane, Insular Romance: Politics, Faith, and Culture in Anglo-Norman and Middle English Literature (Berkeley, CA: University of California Press, 1986), p. 59. For a discussion of the national character of the Auchinleck MS, and an examination of Englishness within a number of other miscellany manuscripts, see Phillipa Hardman, ‘Compiling the nation: fifteenth-century miscellany manuscripts’, in Nation, Court and Culture, ed. Cooney, pp. 50–69. All references to the text of A are by page number to The Romance of Sir Beues of Hamtoun …, ed. by Eugen Kölbing, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) (hereafter BHK). For a discussion of the literary connections between Bevis and the English St George legends, see Jennifer Fellows, ‘St George as romance hero’, Reading Medieval Studies, 19 (1992), 27–54.

115



Robert Allen Rouse

to the Crusaders at the siege of Antioch in 1098, St George had become a popular saint in England. Utilized as a national symbol by Edward I in Wales and Scotland during the late thirteenth century, St George’s is a name with which the manuscript seeks to conjure an aura of Englishness. Bly Calkin reads this addition to the narrative in terms of her wider argument as to the text’s flirtation with notions of cultural slippage.10 Figuring the reference to St George as recuperative strategy by which the text seeks to minimize anxieties about hybridity, she argues that the naming of the saint represents ‘an explicitly anti-Muslim corrective to Bevis’s moments of assimilation into Saracen culture’.11 However, the episode in which Bevis ‘nemenede sein Gorge’ is far removed both geographically and temporally (in terms of the romance’s internal chronology) from his encounters with the Saracens, and this lack of immediacy acts to minimize such significance as may be ascribed to it. Rather, the primary function of the reference, which occurs in the direct context of Bevis’s battle with the dragon near Cologne, is to associate Bevis with St George – and Englishness – in a more straightforward manner, amplifying Bevis’s reputation as an English dragon-slayer: it occurs soon after a list of other famous English dragon-killers including Lancelot, Wade and Guy of Warwick.12 The reference to St George thus signals A’s amplification of national sentiment in the Bevis narrative – a process that, as we shall see, lies behind a number of the additions and variations to the story that are to be found in A.13 While Bevis has been identified as a romance that engages with a nascent sense of Englishness in the early fourteenth century, less attention has been paid to the complexities and apparent contradictions of the identity politics in the text. Bevis is, in many ways, a strange and unsettling example of an English knight. Comparison to a less problematic avatar of Englishness, such as Guy of Warwick, reveals Bevis’s awkward position with regard to the



10 11 12

13

See Bly Calkin, Saracens and the Making of English Identity, p. 58, for a discussion of the ­devel­oping importance of St George in England during the medieval period. The increasing popularity of St George in England is witnessed in many aspects of medieval life, from royal appropriation to church art to prayers and popular lyrics. The epithet ‘Our Lady’s Knight’ has been described as ‘the most hackneyed epithet of St George’ in The Early English Carols, ed. by Richard Leighton Greene, 2nd edn (Oxford: Clarendon Press, 1977), p. 419. On its significance, see Samantha Riches, St George: Hero, Martyr and Myth (Stroud: Sutton, 2000), pp. 68–100; also, with specific reference to Bevis, Jennifer Fellows, ‘ “Dragons two other thre”: the dragon motif in some Middle English romances’ (forthcoming). Bly Calkin, Saracens and the Making of English Identity, pp. 13–60. Ibid., p. 58. BHK, pp. 122–3. One might note that Lancelot is not technically English, but in a manuscript that presents the narratives of Charlemagne’s Twelve Peers as English (or at least not as French), the conflation of Arthur’s foremost knight with Englishness seems perfectly understandable. Judith Weiss, ‘The major interpolations in Sir Beues of Hamtoun’, Medium Ævum, 47 (1978), 71–6 (p. 72), notes that the dragon-slaying episode is an addition to the Middle English Bevis tradition that is not found in the Anglo-Norman Boeve, and observes that we ‘are reminded that it is an English knight who conquers the dragon’.

116



National and Regional Identities in Sir Bevis of Hampton

English.14 Bevis, the son of an English earl, is exiled as a boy and brought up in a Saracen kingdom; he marries a Saracen princess, and – after a brief and turbulent return to the land of his birth – ultimately abandons England for the lands of the East. Bevis thus stands as a complex narrative of Englishness, drawing our attention to some of the many difficulties and anxieties within the fantasy of English national identity during the medieval period. There are two chief complicating factors that impact upon the coherence of the medieval fantasy of Englishness in this romance. First, there is the question of Bevis’s connections with the Saracen East – his upbringing, his marriage, his horse and his kingdom – connections which, I argue, raise serious concerns about the cultural identity of Western knights who spend a prolonged period of time in the Orient. To encounter the Other physically is to enter into what has been termed a ‘contact zone’,15 and such proximity brings with it risks of external contamination of one’s own culture. Secondly, there is the problem of Bevis’s own decidedly regional origins and geographical focus. A, a version with an inherently regional narrative focus that has been co-opted by its manuscript context for a national and cosmopolitan audience,16 stands as a complex manifestation of a sense of Englishness, containing an internal set of tensions between a constructed national English identity centred around ‘the symbolizing potential of the king’, and the powerful regional identities that were an important aspect of medieval English culture.17 In a recent reading of Bevis through the lens of postcolonial theory, Kofi Campbell argues: ‘this text functions as an early example of narrating the nation. It seeks to educate its audience as to what comprises Englishness and, equally importantly, what does not … The Saracens are there to make clearer the bounds of England and Christianity.’18 For Campbell, Bevis is primarily a text that concerns itself with the delineation and the territorial expansion of 14

15

16

17

18

For a detailed analysis of the role of Guy of Warwick within the identity politics of Englishness, see Robert Allen Rouse, ‘An exemplary life: Guy of Warwick as medieval culture-hero’, in Guy of Warwick: Icon and Ancestor, ed. by Alison Wiggins and Rosalind Field (Cambridge: Brewer, 2007), pp. 94–109, and ‘Expectation vs experience: encountering the Saracen Other in Middle English romance’, SELIM, 10 (2002), 125–40. A ‘contact zone’ is a space of cultural encounter, in which peoples geographically and historically distanced come into contact and establish ongoing relations: cf. Marie Louise Pratt, Imperial Eyes: Travel Writing and Transculturation (London; New York: Routledge, 1992), p. 7. The Auchinleck MS has been thought to have been produced in London for either a member of the merchant classes, a wealthy female reader, or a member of the crusading gentry: see Alison Wiggins, ‘History and Owners’ (http://www.nls.uk/auchinleck/editorial/history.html). Regional identities were important and powerful in medieval England. Turville-Petre notes that ‘within England there were marked regional differences, and for many purposes the country could still be divided at the Humber … with a northern territory that had its own traditions, its great lordships with their shifting allegiances to the crown and relationships with the Scots, and where royal control was far looser than it was in the south.’ (England the Nation, p. 7). For an extended discussion of regional identities in medieval English romance see Rouse, The Idea of Anglo-Saxon England in Middle English Romance, pp. 85–92. Kofi Campbell, ‘Nation-building colonialist-style in Bevis of Hampton’, Exemplaria, 18.1 (2006), 205–32 (p. 232).

117



Robert Allen Rouse

Englishness. Campbell points out the theoretical commonplace that romances such as Guy of Warwick, The Sowdon of Babylon and Richard Coer de Lyon utilize the figure of the Saracen in an early form of Orientalism, situating in the racial and religious Other all those things against which medieval Christen­dom was defined.19 However, in working hard to reconcile the difficulties of the narrative with a postcolonial reading of the text, Campbell too readily sutures the inherent fractures within the rhetoric of identity with which Bevis presents its readers. Colonial expansion, the fantasy of which Campbell rightly identifies as an important ideological concern in Bevis, brings its own fears and anxieties to the populace of the imperial homeland. While it is indeed true that encounters with the Saracen Other are an important part of the medieval poetics of otherness, there are also concomitant fears of cultural infection and miscegenation. While Campbell argues that Bevis after his fashion makes the East English, we must remind ourselves that acculturation is rarely unidirectional, and even the dominant culture in the process is itself changed through colonial and other forms of cultural interaction.20 Bly Calkin highlights medieval anxieties over the potential for cultural contamination amongst the Christian settlers of the Holy Land.21 Contem­ porary commentators expressed concerns about the dangers of sexual inter­mingling, the taking-on of Eastern customs, and the degeneration of Christian morals. Citing the Itinerarium peregrinorum, Bly Calkin points out that such fears were often associated with the failure of the Crusades themselves: the capture of Jerusalem in 1187 was considered to be a direct consequence of the debasement and contamination of the Western Christian culture of the Crusader states.22 Such admonitions, articulating the ‘fear that western Christians … might lose their sense of proper mores and become too similar to their Muslim opponents’23 through contact with the Saracen East, were common. Jean de Joinville, in his Life of Saint Louis, records on a number of occasions his anxieties about the dangerous influence of Eastern culture on his fellow-Christians. At the end of his account of the appearance and 19

20

21

22 23

The use of the Saracen as a medieval cultural Other has become the subject of increasing critical debate: important studies include Dorothee Metlitzki, The Matter of Araby in Medieval England (New Haven, CT; London: Yale University Press, 1977); Diane Speed, ‘The construction of the nation in medieval romance’, in Readings in Medieval English Romance, ed. by Carol M. Meale (Cambridge: Brewer, 1994), pp. 135–57; Jeffrey Jerome Cohen, Of Giants: Sex, Monsters, and the Middle Ages (Minneapolis, MN: Minnesota University Press, 1999), and ‘On Saracen enjoyment’, in his Medieval Identity Machines, pp. 188–221. The two-way street of acculturation is also evident in other medieval romances. For a discussion of this phenomenon in Havelok, see Rouse, ‘In his time were gode lawes: romance, law, and the Anglo-Saxon past’, in Cultural Encounters in Medieval English Romance, ed. by Corinne ­Saunders, Studies in Medieval Romance (Cambridge: Brewer, 2005), pp. 69–83. The historical context of such concerns over the cultural identity of the Christian settlers in the Holy Land is addressed in Alan V. Murray, ‘Ethnic identity in the Crusader states: the Frankish race and the settlement of Outremer’, in Concepts of National Identity in the Middle Ages, ed. by Simon Forde, Lesley Johnson and Alan V. Murray, Leeds Texts & Monographs (Leeds: Leeds Studies in English, 1995), pp. 59–74. Bly Calkin, Saracens and the Making of English Identity, p. 55. Ibid., p. 54.

118



National and Regional Identities in Sir Bevis of Hampton

customs of the Bedouin, Joinville highlights the potential risks of such cultural and theological infection: ‘In our own country, since I returned from the land overseas, I have come across certain disloyal Christians who follow the Bedouin faith in holding that no man can die except on the appointed day. This belief is in effect a denial of our religion …’.24 Joinville views the potential for cultural contamination not only as a danger to those Christians in the East, but also as a feared contagion that may spread to the body of Christendom itself. The dangers to Christians living in the East were not confined to the temptations and heresies that might be transmitted from the cultures with which they came into contact; rather, geography itself posed real and present dangers to such dislocated Westerners. Suzanne Conklin Akbari recounts the connections between national identity and geography that are drawn by Isidore of Seville: ‘the physical qualities of a nation are altered by the climate its members inhabit’.25 The thirteenth-century Bartholomæus Anglicus is even more insistent on the geographical and climatic effects on the character and nature of a people, arguing that ‘climate is the only factor that determines the characteristics of a nation’.26 To the medieval geographical mind, the most significant climatic feature is the sun, which influences the nature of man in two ways: first, the degree of exposure to the sun historically created the races of men – both the bold and hardy white-skinned men of Europe, and the cowardly sun-blackened races of Africa and the East.27 Secondly, the sun’s heat has a more immediate influence, altering the humoral composition of those living under its influence, thus changing their behaviour and character. Geography, with the attendant climatological humoral effects upon both character and behaviour, underpins the cultural anxiety over the dangers to Western Christians of prolonged habitation of the Saracen lands of the East. When we read Bevis in the light of these contemporary concerns regarding the cultural integrity of transposed European Christians, episodes of apparent identity confusion are revealed as articulating anxieties about cultural hybridity, the fear that Christian English identity might be dangerously similar to that of the Saracens. Cohen observes that ‘when pagan and Christian subjectivities seem close enough almost to touch, violence erupts to redraw the faltering self/ other boundary’.28 One such moment of violent reaction to identity crisis is encountered during Bevis’s early years in Ermonye. While Bevis is out riding with King Ermin’s men, he is informed by a Saracen knight that it is Christmas, and is encouraged by the knight to follow his Christian traditions 24 25 26 27 28

Jean de Joinville, ‘The Life of Saint Louis’, in Chronicles of the Crusades, trans. by M. R. B. Shaw (Harmondsworth: Penguin, 1963), p. 228. Suzanne Conklin Akbari, ‘From due east to true north: Orientalism and orientation’, in The PostColonial Middle Ages, ed. Cohen, pp. 19–34 (p. 25). Akbari, ‘From due east to true north’, p. 25. Ibid., p. 29. Cohen, ‘On Saracen enjoyment’, p. 205.

119



Robert Allen Rouse

and honour his God as the Saracen honours his own. Despite the Saracen knight’s framing of Christmas in terms of religious tolerance, the reference to the Christian holy day has an opposite effect on Bevis:29 Beues to þat Sarasin said: ‘Of Cristendom ȝit ichaue a-braid, Ichaue seie on þis dai riȝt Armed mani a gentil kniȝt, Torneande riȝt in þe feld With helmes briȝt and mani scheld; And were ich alse stiþ in plas, Ase euer Gii, me fader, was, Ich wolde for me lordes loue, Þat sit hiȝ in heuene aboue, Fiȝte wiþ ȝow euerichon, Er þan ich wolde hennes gon!’  (p. 29)

Bevis associates Christmas not with the Mass or with religious observance, but with the tournaments that he remembers witnessing in his childhood. This martial remembrance stirs the boastful youth to declare that he could defeat all fifty of Ermin’s knights if he had a mind to, for the love of his God. The Saracen takes offence at Bevis’s boasting, and mocks him in front of the other knights, who try to teach ‘þe ȝonge cristene hounde’ (p. 29) a lesson. Bevis fights back and slays all fifty of them. Bevis’s first violent contact with the Saracens is thus contrived through religious difference – a point that is emphasized by its setting on Christmas Day. Bevis seeks to remedy his own lack of Christian self-identity through a bloody martial baptism. Denied the community of fellow-Christians, and threatened by the seductive religious tolerance of the Saracen knight, Bevis inscribes the inviolability of his Christian identity upon the bodies of the Saracens. This performance of his Christian faith is a scene readily understandable by his English readers, and is facilitated through the only means he has available to him – violent conflict with the religious Other. While Bevis martially reaffirms his own threatened Christian – and English – identity in the Christmas Day episode, there are other moments within the text where the self/other boundaries again threaten to dissolve. One such revealing moment of identity confusion occurs during Bevis’s escape from Brademond’s prison. After seven years in the dungeon, an opportunity finally arises for him to escape. Hearing him praying to Christ and Mary to grant him either death or freedom, one of his gaolers becomes enraged and enters

29

In his reaction to the Saracen knight’s knowledge of Christianity and its holy days, Bevis expresses an outrage that is shared by other medieval commentators such as Jean de Joinville. Far from taking comfort in a Saracen familiarity with Christian belief, this knowledge appears to Bevis as yet another instance of identity confusion, only adding to his anxieties as to the separateness of his own Christian identity.

120



National and Regional Identities in Sir Bevis of Hampton

his prison to threaten him. Bevis seizes the moment, beating the Saracen to the ground with his bare fists. However, still trapped within the twenty-footdeep dungeon, Bevis now has to rely upon his wits, rather than his fists, to facilitate the second part of his escape. Raising his voice in an impersonation of the first of his gaolers, he successfully dupes the second by feigning that he requires aid: For þe loue of sein Mahoun, Be þe rop glid bliue adoun And help, þat þis þef wer ded!  (p. 84)

If this act of verbal impersonation is all that is required for a Christian knight to masquerade as a Saracen, then the difference between Bevis and the Saracen Other is narrow and complex indeed. Bly Calkin figures this act of ‘passing’ as a moment which ‘provoke[s] consideration about the relative permanence and veracity of an asserted identity’.30 Raising questions that threaten the entire premise of the Christian desire to convert the East, Bevis’s linguistic sleight-of-hand illustrates the essentially performative nature of religious affiliation and therefore cultural identity. If this similarity does, as Campbell suggests, play a part in the rhetoric of the possibility of conversion of the Saracens, it must also be kept in mind that Saracens are not the only objects of the conversionary impulse. Accounts of the Crusades are littered with examples of the forced (or otherwise) conversion of Christians to Islam, and in Bevis’s vocal duping of his prison guards through the invocation of the Saracen deity we can perhaps see alluded to, in the comic safety of the romance genre, a cultural anxiety regarding the two-way street of conversion in the Middle East. While Campbell sees Bevis’s interactions with Saracen culture, and his flirtations with Saracen identity, as exemplifying the ‘colonialist desire of the text’,31 in that his cultural hybridity is the vehicle for the expansion of English control over the lands that Bevis and his sons conquer and convert, we must also recognize that such colonial endeavours take a toll on the colonizers themselves. Bly Calkin notes that Bevis and Josian – ‘the convert-Christian couple’ – seem ill-suited to life in Christian England, but must instead depart to ‘a kingdom close to Saracen lands and interests’.32 Bevis stands within his romance as a complex manifestation of hybrid English–Eastern identity, seemingly never comfortable upon his return to the land of his birth, and continually forced back to the East in order to carve out a new Christian kingdom that can accommodate him and his converted bride. While the Englishness embodied by Bevis can be seen to be subject to anxieties relating to external contamination, this is not the only problematic 30 31 32

Bly Calkin, Saracens and the Making of English Identity, p. 79. Campbell, ‘Nation-building colonialist-style’, p. 229. Bly Calkin, Saracens and the Making of English Identity, p. 85.

121



Robert Allen Rouse

aspect of national identity in the romance. The homogeneity of Englishness is also at issue within the bounds of England itself. Like other Matter of England romances, Bevis is intimately connected with its own regional locale; as Crane comments, ‘the place of national spirit in the romance is uneasy’.33 Bevis, a narrative with an avowedly regional focus, manifests an anxiety concerning centralized power in the body of the king and the locus of London. Through this regional narrative of identity, the text is forced to negotiate the claims of group identities other than that of simply ‘English’. Rosalind Field’s statement that medieval romance operates ‘to create a history for a country, a family, a city’ at once identifies the broad scope of the romance mode of narrative history and alerts us to the competing historiographical voices that such romances may contain.34 While English ‘historical’ romances such as Bevis and Guy of Warwick participate in the telling of a history of England, they are also regional narratives, closely connected with their settings and the origins of their protagonists. The importance of an appreciation of the tension between national and regional voices within medieval historiography is clear. Michelle Warren has highlighted the importance of recognizing the regional origins of writers such as Geoffrey of Monmouth, Robert of Gloucester and Laȝamon, to name but three.35 The existence of regional voices in these texts raises the question of whether Englishness in Bevis is constructed as a homogeneous whole, or whether regional tensions lurk within. Bevis’s sobriquet ‘of Hampton’, I would suggest, signals a powerful regional discourse that runs through much of this romance – one that manifests itself in Bevis’s repeated conflicts with centralized royal authority in the form of King Edgar. In making her case for the possibility of medieval national identities, Heng argues that medieval nationalism is not the same as its modern counterpart: That nation is not, of course, a modern state: among the distinguishing properties of the medieval nation – always a community of the realm, communitas regni – is the symbolizing potential of the king, whose figural status allows leveling discourses and an expressive vocabulary of unity, cohesion, and stability to be imagined, in a language functioning as the linguistic equivalent of the nation’s incipient modernity.36

In her analysis of Richard Coer de Lyon, she argues that the figure of the king occupies a centrally symbolic place within the articulation of the medieval English nation. In Bevis, however, the figure of King Edgar is anything but a unifying symbol of English identity: he does not act when Bevis’s father is murdered; he does nothing to return Bevis’s usurped lands and title until 33 34

Crane, Insular Romance, p. 86. Rosalind Field, ‘Romance in England’, in The Cambridge History of Medieval English Literature, ed. by David Wallace (Cambridge: Cambridge University Press, 1999), pp. 152–76 (p. 162). 35 Michelle Warren, History on the Edge: Excalibur and the Borders of Britain, 1100–1300 ­(Minneapolis, MN: University of Minnesota Press, 2000). 36 Heng, ‘The romance of England’, p. 139.

122



National and Regional Identities in Sir Bevis of Hampton

Bevis confronts him over his inaction; he intemperately and illegally attempts to prosecute Bevis for Arundel’s killing of his larcenous son; and he stands by impotently while the people of London attempt to murder Bevis at the instigation of one of his vassals. An examination of the interactions between Bevis and King Edgar illustrates the inherent tensions between region and centre that exist in Bevis. These regional concerns are most clearly articulated near the end of the romance, in an episode that does not occur in Bevis’s Anglo-Norman source.37 After his return from exile to restore his foster-father Saber’s heir, Robaunt, to his lands in Hampshire, Bevis is involved in a street battle against the inhabitants of London. As the conflict rages, Bevis is confronted by a particularly fierce Londoner: Þar was a Lombard in þe toun, Þat was scherewed & feloun; He armede him in yrene wede And lep vpon a sterne stede And rod forþ wiþ gret randoun And þouȝte haue slawe sire Beuoun.  (p. 211)

Lombards were a particularly visible alien community in London during this period (Lombard Street had an Italian community of long standing), and ill-feeling and violence towards foreigners were common: from the Flemish merchants murdered in the riots of 1381 to the riots against the presence of aliens (especially Italians and Lombards) in 1456 and 1457, there were frequent outbursts of anti-foreign sentiment.38 London is here constructed as a cosmopolitan, immigrant city, full of the kinds of foreigners that are dangerous to Bevis and to his regional Englishness. Despite the memorable figure of the Lombard, London’s alterity is in fact downplayed to a degree in A by comparison to the version of the romance represented by the printed texts and, in part, by Manchester, Chetham’s Library, MS 8009. As Derek Pearsall notes, in this version (hereafter P), ‘the London opposition to Bevis is dominated by Lombards from the start’.39 In P, Bevis first defeats the treacherous steward, and then is immediately beset not by one, but by ‘meny Lumbardis’.40 Not only is London represented as being even more alien in P; there is also 37

Weiss, ‘The major interpolations’, p. 73, notes: ‘though the English writer could not change these final details, he played down the conquest of Monbrant, and added a dramatic last fight in the city of London’. 38 Claire Sponsler, ‘Alien nation: London’s aliens and Lydgate’s mummings for the mercers and goldsmiths’, in The Postcolonial Middle Ages, ed. Cohen, pp. 229–42 (pp. 230–1). 39 Derek Pearsall, ‘Strangers in late-fourteenth-century London’, in The Stranger in Medieval Society, ed. by F. R. P. Akehurst and Stephanie Cain Van D’Elden (Minneapolis, MN: University of Minnesota Press, 1997), pp. 46–62 (p. 53). 40 Jennifer L. Fellows, ‘Sir Beves of Hampton: Study and Edition’, 5 vols (unpub. Ph.D. diss., Univ. of Cambridge, 1980) (hereafter BHF), IV, 4102.

123



Robert Allen Rouse

another tension, absent from A, evident within P’s construction of English identity. In A, Edgar’s troublesome steward is presented as a stereotypical romance steward, envious of Bevis’s fame and corrupt in his use of his position. However, in P there is further evidence of his perfidious character: here he is not simply ‘þe stiward of þe halle … þe worste frend of alle’ (p. 203), as in A, but is named as Sir Brian of Cornwall.41 P here makes use of a wellestablished geographically internal Other in its characterization of Bevis’s foe.42 The anxiety regarding London and its inhabitants that is apparent in Bevis has provoked a number of interpretations. Crane has argued that the text manifests concerns regarding the relationship between the Crown and the barony;43 while Pearsall has observed that it seems ‘almost as if the translator became aware that he needed to “alienize” this native opposition to an English hero’.44 However, I would suggest another reading, in which the emphasis is placed upon the contrast between Bevis, as provincial English, and the cosmopolitan, non-English nature of London. In terms of English identity, the figure of the Lombard looms large, characterizing London as being not at the centre of Englishness, but at the margins. Bevis seems to exhibit a communal anxiety regarding the increasing influence of London’s immigrant populace over affairs of state, and of the very identity of the city. London, as the locus of royal power and as a major point of contact with the foreign, is constructed in the regional discourse of this romance as a site of both the contestation and the contamination of English identity. The social discourse that underlies this anxiety is perhaps difficult to ascertain, although one might point out the increasing centralization of the royal court, as well as the numbers of foreign aliens during the thirteenth and fourteenth centuries, as likely sources of apprehensiveness. This concern about the relationship of London to English identity, as strong today as it seems to be in Bevis, conveys a powerful criticism of the capital as unrepresentative of England and the English.45 However, the romance does not leave this apparent fracture in its fantasy of Englishness unaddressed. After the London episode, Edgar and Bevis (nation and region) are reconciled, and the king offers Bevis’s son Miles the hand of his daughter in marriage, making him his heir.46 This merging of Bevis’s family with that of the royal house of England, seen by ­Campbell as 41 42

43 44 45 46

Brian seems in part to be motivated by professional jealousy in his enmity towards Bevis, as he once held the offices that Bevis now holds: see BHF, III, 3175–6, quoted p. 96 above. On the role of the Cornish as an Insular Other for the English, see Rouse, The Idea of AngloSaxon England, pp. 85–8. Numerous Cornish knights demonstrate their innate treachery in medieval English romances, including Medyok, the lord ‘of al Cornewaile’ in Guy of Warwick, Godrich the earl of Cornwall in Havelok, and, of course, the Arthurian figure of King Mark. Crane, Insular Romance, p. 218. Pearsall, ‘Strangers in late-fourteenth-century London’, p. 53. A concern that is echoed in the London populace’s frequent outbursts of violence against the alien communities: cf. Sponsler, ‘Alien nation’, pp. 230–1. BHK, p. 215.

124



National and Regional Identities in Sir Bevis of Hampton

‘uniting under a single name’ England and Bevis’s conquests in the East,47 attempts within A to reconcile Bevis’s family – and vicariously the audience – with the communitas regni of the English nation. Interestingly, this union of the regional and the national takes place not in London – the symbolic location of centralized royal English power thus far in the narrative – but in Nottingham.48 No rationale for this is given in the text, and we are left to speculate as to A’s variation from other versions of Bevis, where the wedding takes place in London. Perhaps in the location of the marriage in Nottingham we might read the return of royal ceremony and power to the regional stage. Again we can see the Auchinleck narrative as being particularly concerned with a rapprochement between the regional and national discourses within the text. It is notable that while A clearly states that Miles becomes Edgar’s heir, not all versions of Bevis do so. P recounts a different version of the marriage, holding it in London and making no mention of Miles being in line to inherit the crown (perhaps assuming that Edgar’s daughter is not next in line to the throne) but instead giving him the earldom of Cornwall;49 it also, like all the versions of the narrative except A, places the wedding in London, maintaining the royal privilege of the chief city of the English kings. Bevis illustrates the inherent distrust in regional discourse towards the centralized royal and bureaucratic power that is characterized by cosmopolitan London. It demonstrates a questioning of the cultural and national centre – a characteristic element of regional voices both within the medieval period and outside it. The concern with the foreign nature of London in Bevis suggests a questioning of the nature of Englishness that is as import­ant as the development of Englishness itself. In Bevis we find an imagined England constructed as a space in which tensions between competing regional discourses of identity can be played out in a simplified and secure past, rather than in the complex everyday world of the audience; for if the Anglo-Saxon English, despite the regional tensions that are evident in the romance narratives of the pre-Conquest past, could operate within a wider English national identity, then the rural and the urban, the regional and the central, could all see a way of identifying with an inclusive national fantasy of Englishness. Bevis dies and is buried far from England, in his newly won kingdom of Mombraunt. His burial signifies none of the unambiguously colonial conflation of the heroic and the national body that we might expect from ­Campbell’s reading of the romance. Rather, Bevis’s death and burial in the exotic East act only to reinforce his own troublesome relation to English identity. Bevis may well have converted Mombraunt by the sword but, as was all too evident to Bevis’s fourteenth-century audience, such conquests had proved transitory

47 48 49

Campbell, ‘Nation-building colonialist-style’, p. 229. BHK, p. 215. BHF, III, 4171–2.

125



Robert Allen Rouse

and were already part of a vanishing legacy of European conquest within the Saracen world. Bevis’s death, and the troublesome hybrid nature of his identity during his life, can be read as representing the internal tensions and external anxieties that were important concerns for the nascent fantasy of English identity during the Middle Ages. If we can derive one important conclusion from these complexities, it is perhaps that we should be careful not to ascribe to the medieval English national identity portrayed in Middle English romances such as Bevis the monolithic homogeneity that we have come to expect from the forms of nationalism prevalent in the modern age.

126

8 Defining Christian Knighthood in a Saracen World: Changing Depictions of the Protagonist in Sir Bevis of Hampton Siobhain Bly Calkin

O

ne of the most popular romances to circulate in Europe, Sir Bevis of Hampton stands out for its many depictions of Saracen characters and settings. In narrating the exploits of an English hero, it repeatedly situates him, and develops his identity, in relation to the world in which he lives after having been sold to Saracens. Saracens raise Bevis, knight him and arm him; they communicate his universal desirability as son, lover and warrior; they serve under him and help him to reclaim his heritage and avenge his father’s murder; they also afford Bevis opportunities to manifest his exemplary ­Christianity. In short, Saracens define Bevis’s status as a heroic ­ Christian knight. This is not to say, however, that the notion of what it means to be a Christian knight remains static across the centuries of the text’s circulation. In fact, while the broad outlines of Bevis’s encounters with Saracens generally stay the same, specific details vary in different versions of the romance. This article examines how some of Bevis’s engagements with Saracens are variously represented in versions of the romance that circulated in England between the thirteenth and the sixteenth century. Specifically, this paper considers the Anglo-Norman Boeve de Haumtone, the fourteenth-century Middle English version (hereafter A) found in Edinburgh, National Library of Scotland, MS Advocates’ 19.2.1 (the Auchinleck MS), the fifteenth-century version (hereafter C) found in Cambridge, University Library, Ff.2.38, and the version (hereafter P) represented by the sixteenth-century printed editions of the romance and by parts of Manchester, Chetham’s Library, MS 8009. While these texts depict many of the same encounters between Bevis and



See Charles W. Dunn, ‘Romances derived from English legends’, in A Manual of the Writings in Middle English 1050–1500, vol. 1, Romances, ed. J. Burke Severs (New Haven, CT: Connecticut Academy of Arts and Sciences, 1967), p. 26; and cf. pp. 2–6 above.  The Auchinleck MS is usually dated to the 1330s: cf. p. 104 above.  Scholars have assigned dates to this manuscript that range from the mid-fifteenth century to the very early sixteenth century. Most place it within the late fifteenth century: cf. p. 108 above.

127



Siobhain Bly Calkin

Saracens, they use these cross-cultural encounters to offer subtly different definitions of what it means to be a heroic Christian knight. Since other chapters in this collection discuss the encounters between Bevis and Josian, and those between Bevis and Ascopard, I shall concentrate here on Bevis’s experiences with Saracens while in the service of the Saracen King Ermin. These encounters constitute a significant portion of the narrative and offer an extensive portrait of Bevis’s conduct as a Christian hero living in a Saracen political world. Bevis’s first substantial encounter with a Saracen occurs early in the tale when Bevis is sold as a child to Ermin, ruler of Armenia, and Ermin enquires about Bevis’s origins. Although Bevis is too young to have been dubbed a knight at this point, the encounter signals his possession of characteristics that will make him an admirable Christian knight later on. In Boeve, the king says: Emfes ... di moi dount tu es; par Mahun mon dieu! jeo ne vi unkes mes enfaunt de ta beuté de loins ne de pres.  (380–2) (Child … tell me where you come from. By my god Mahomet, I never saw a child as beautiful as you before, whether near or far.)

With these words, he affirms Boeve’s innate comeliness, a characteristic shared by most heroes of medieval romance, and indicates a cross-cultural valorization of such beauty. Hermine’s request also allows Boeve to announce his noble origins, yet another essential attribute of the exemplary romance knight. Hermine then asserts Boeve’s desirability as a retainer when he offers Boeve his daughter’s hand in marriage and future possession of his realm if Boeve will convert to the Saracen faith. Now that Boeve’s aristocratic status and worthiness of future knighthood have been established, his exemplary Christianity begins to manifest itself. Boeve roundly refuses Hermine’s offer: ‘Rois’, ceo dist l’emfes, ‘vus parlez de folie; ke pur tut la tere ke est en paenie ne pur ta file ov tut, ke taunt est colorie, ne vodrai reneier Jhesu, le fiz Marie. Mahun ne put taunt fere con la formie,



The realm is identified as Egypt in the Anglo-Norman version: see Der anglonormannische Boeve de Haumtone, ed. by Albert Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974) (hereafter Boeve), line 362. All references to Boeve are by line number to this edition.  All translations of Boeve are quoted from ‘Boeve de Haumtone’ and ‘Gui de Warewic’: Two Anglo-Norman Romances, trans. by Judith Weiss, French of England Translation Series (Tempe, AZ: Medieval and Renaissance Texts and Studies, 2008).

128



Christian Knighthood in Sir Bevis of Hampton ke la formie mut, e si ne fet il mie. Honi seit de son cors ki en Mahun se afie!’  (399–405) (‘King’, said the child, ‘you talk foolishly. Not for all the land in heathen parts nor for your daughter with it, rosy-cheeked as she is, would I renounce Jesus son of Mary. Mahomet can’t even do as much as an ant, for an ant can move and he can’t. Shame on him who trusts in Mahomet!’)

Although the broad outlines of this encounter remain unchanged in the four versions studied here, the hero’s rejection of the king’s offer varies. Here, there is no attempt to be diplomatic, and Boeve insults both Hermine himself and his religion. Hermine’s response, however, is to praise Boeve for his estable quer (‘steadfast heart’, 406) and to promise that he will knight Boeve when Boeve comes of age. Hermine thus identifies steadfastness of heart as the virtue that demonstrates Boeve’s suitability for future dubbing, and suggests that neither rude speech nor religious difference will impede Boeve’s advancement in the Saracen world. In A too, Ermin notes Bevis’s comeliness, allows him to state his aristocratic origins and provokes Bevis’s first assertion of his Christianity. There is, however, a change in tone, A’s Bevis refusing Ermin’s offer thus: ‘For gode!’ queþ Beues, ‘þat I nolde For al þe seluer ne al þe golde, Þat is vnder heuene liȝt, Ne for þe douȝter, þat is so briȝt: I nolde for-sake in none manere Iesu, þat bouȝte me so dere: Al mote þai be doum and deue, Þat on þe false godes be-leue!’  (p. 26)

Gone are the personal attack on the king’s wisdom and the comment that Mahoun has less power than an ant. What remains is a rather general reviling of those who believe in ‘false godes’. This still encompasses Ermin but does not insult him as directly as do the words of the Anglo-Norman Boeve. Interestingly, Bevis’s description of what he is renouncing has also changed somewhat. While Boeve refuses land and a royal bride, A’s Bevis states that he is refusing money and a bride, even though Ermin has actually offered land and a bride (p. 26). Bevis’s speech and Ermin’s combine to depict the Saracen world as a place to acquire both land and money. The text’s dual emphasis on wealth and land meshes well with the rhetoric of texts advocating the renewal of crusade in the fourteenth century – there crusade was often touted  

Boeve 409–10. References to the text of A are by page number to The Romance of Sir Beues of Hamtoun …, ed. by Eugen Kölbing, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) (hereafter BHK).

129



Siobhain Bly Calkin

as an opportunity to acquire commercial wealth as well as territory. The Saracen king’s response to Bevis’s words also changes somewhat. Whereas the admirable trait in Boeve is a steadfast heart, in A Ermin admires Bevis for his fearlessness (‘For him ne stod of noman sore’) and decides that such fearlessness makes Bevis worthy to carry his banner in the future when Bevis will have been ‘dobbed kniȝt’ (p. 27). Thus, while the broad outlines of this encounter remain the same, the hero’s tone, his definition of the Saracen world’s attractions, and the specific virtue proving his worthiness of future knighting vary. Another variation can be traced in C. Although this text closely resembles A at this point, it adds one significant detail to Bevis’s refusal of Ermin’s offer: ‘All be they brente to dethe, / Þat on odur false godes beleuyth!’ The idea that non-Christians should be burned to death is a motif that occurs elsewhere in C,10 which in this respect is unique among the versions studied here. C dates from the late fifteenth or early sixteenth century, by which time the idea that heretics – and, by extension, non-believers – deserve to be burned was familiar to English writers. From 1401 onwards, Lollards, including William Sautre, John Badby and John Oldcastle, suffered this fate.11 The idea of the burning misbeliever thus gained a cultural potency in the fifteenth century that makes its appearance in C understandable. The addition of references signalling Bevis’s willingness to burn non-believers marks the romance’s ability to update its image of exemplary Christian virtue; whereas burning non-believers may not have been a task for the model Christian warrior in thirteenth- or fourteenth-century England, it has clearly become one by the fifteenth century. In P, the encounter with Ermin emphasizes yet different aspects of Bevis’s character and heritage. Ermin responds to Bevis’s declaration of his lineage by saying ‘of Guy of Hampton I haue heard tel, / Many a Paynime and



Consider, for example, Pierre Dubois’s words in his treatise dated to roughly 1306: ‘Quibus pre­missis sic per Dei graciam ordinates, catholici concordes possidebunt totam ripam maris Mediterranei, ab ejus occidente usque ad orientem versus septentrionem, et meliorem partem tangentem Terram Promissionis versus meridiem; ita quod bene vivere etiam corporaliter non poterunt Arabes, nisi communicent catholicis commercia rerum suarum; et idem in orientalibus populis, et de ipsis’ (‘When these projects have, by the grace of God, been accomplished, ­Catholics of the same mind will be in possession of the whole Mediterranean coast, from the west all the way to the east on the north side, and the greater part touching the Land of Promise on the south. The Arabs will then be unable to prosper materially unless they share with the Catholics the commerce in their products. This will also be true in the case of oriental peoples and their ­products’). Pierre Dubois, De recuperatione terre sancte: Traité de politique générale, ed by Ch.-V. Langlois (Paris: Durand, 1891), p. 89. Translation from Pierre Dubois, The Recovery of the Holy Land, trans. by Walther I. Brandt (New York: Columbia University Press, 1956), pp. 156–7.  Jennifer L. Fellows, ‘Sir Beves of Hampton: Study and Edition’, 5 vols (unpub. Ph.D. diss., Univ. of Cambridge, 1980) (hereafter BHF), II, 632–3. 10 E.g. ibid. II, 1722–5, 1734. 11 Paul Strohm, England’s Empty Throne: Usurpation and the Language of Legitimation 1399–1422 (New Haven, CT: Yale University Press, 1998), pp. 36, 38–45, 53–8.

130



Christian Knighthood in Sir Bevis of Hampton

sarasine, he hath slaine’ (sig. Biiv).12 P thus suggests that Bevis belongs to a renowned dynasty of Saracen-killing Christian knights, and his sale to Saracens is recast as an act which situates him in a world familiar to his father; here, therefore, Bevis follows a charted paternal path of Christian heroism, whereas the other texts discussed imply that Bevis enters a completely foreign world. Boeve, A and C convey a greater sense of adventure, while P communicates a sense of ongoing and familiar struggle between English knights and the Saracen world. Bevis’s refusal of Ermin’s offer also takes a slightly different form in P. Here, Bevis says: Sir … that will I nought, for all things that euer was wrought Neither for gift that may be, nor for thy daughter that is so free, I did my selfe great dishonour, if I should forsake my creatour. (sig. Biiv)

Bevis refuses Ermin’s offer politely, while stating that acceptance would compromise his honour. As Jennifer Fellows notes, this focus on honour distinguishes P from manuscript versions of Bevis.13 For example, in a dis­quisition found later in P, Bevis reflects that killing his stepfather while visiting under an assumed identity would link him (Bevis) to ‘villany’, ‘treason’ and ‘cowardice’ (sig. Fiiiv) and the act is therefore unacceptable. The encounter with Ermin recounted in P thus manifests early on the concern with honour that becomes a central part of Bevis’s exemplary knighthood in this text. Ermin’s response to Bevis’s refusal also contributes to P’s distinctive portrait of the exemplary Christian. Ermin admires Bevis for his ­‘stedfast … creuence’ (sig. Biiv), a phrase that evokes the steadfastness of heart praised in Boeve, but explicitly links steadfastness to religious belief. Such an emphasis seems rather appropriate for a text that circulated throughout the sixteenth century when adherence to one faith in changing political contexts was certainly an act of bravery. As different versions of Bevis recount the protagonist’s initial encounter with Ermin, basic events remain the same, but specific emphases differ. Similar variation is manifested in Bevis’s encounters with King Brademond of Damascus. Bevis’s journey to Damascus, his interactions with ­Brademond, and his imprisonment in Brademond’s dungeon all shape distinctive images of the hero and of the Saracen world, and these encounters convey a particularly strong sense of how ideals of Christian knighthood, and of its relationship to Eastern Others, changed during the fourteenth and fifteenth centuries. 12

All references to P are by signature to Thomas East’s edition of c. 1582 (hereafter K): cf. no. 13 at p. 111 above. For an account of the printed texts of the metrical Bevis, see pp. 96–101 above. 13 Jennifer Fellows, ‘Bevis redivivus: the printed editions of Sir Bevis of Hampton’, in Romance Reading on the Book: Essays on Medieval Narrative Presented to Maldwyn Mills, ed. by Jennifer Fellows, Rosalind Field, Gillian Rogers and Judith Weiss (Cardiff: University of Wales Press, 1996), pp. 251–68 (p. 255).

131



Siobhain Bly Calkin

In all four texts under discussion, as Boeve/Bevis travels to Brademond’s city, he happens upon his cousin (or foster-brother) Terri, who has been seeking him for seven years. The two exchange information about Bevis in a conversation that charts a fascinating shift in understandings of the range of relationships a Christian knight can have with Saracens. In Boeve and in A, Bevis tells Terri that Bevis has been hanged. In mendaciously affirming his own death, Bevis severs his ties to his English associates, unmoved by the fact that Terri weeps, swoons, and laments his cousin’s loss.14 Terri then asks about Bevis’s letter, stating (with remarkable romance prescience) that it might order Bevis’s death. The hero of Boeve replies by saying, ‘moun seignur ne le freit pur treis cent citez’ (‘my lord wouldn’t do that for three hundred cities’, 861), while A’s Bevis asserts: He, þat me tok þis letter an honde, He ne wolde loue me non oþer, Þan ich were is owene broþer.  (p. 66)

In Boeve, the hero disavows his family in favour of a seigneurial relationship with Hermine; he cuts his ties to England and affirms an unshakeable trust in his Saracen overlord. In A, Bevis radicalizes his secession from his family by affirming that he and Ermin are like brothers, a statement that replaces his Christian family with a Saracen one. Other texts excise these statements disowning Bevis’s English roots and affirming his cherished position in the Saracen world: in both C and P, Bevis promises to speak to Bevis about travelling to England once Ermin’s message has been delivered, and there is no suggestion whatsoever that he might permanently cut his ties to England. In the later versions, Bevis articulates no personal sense of relationship to the Saracen king: he does not even mention Ermin but instead emphasizes his task. Close relationships between a Christian knight and a Saracen ruler may be envisaged in the thirteenth and fourteenth centuries, but by the fifteenth this is no longer the case. This shift, I would suggest, reflects political changes in England and the East. England’s engagement in crusading endeavours of the thirteenth and early fourteenth centuries gradually diminished as rulers such as Edward I, who led a Crusade to the East, were replaced by leaders such as Edward III and Henry V, who never took Crusading vows, never travelled to the East, and were more concerned with prosecuting wars against France than with attempting to retake Jerusalem and the Crusader kingdoms of Outremer.15 As

14

This incident in A is discussed in more detail in Siobhain Bly Calkin, ‘The anxieties of encounter and exchange: Saracens and Christian heroism in Sir Beves of Hamtoun’, Florilegium, 21 (2004), 135–58 (pp. 140–3). 15 See Simon Lloyd, English Society and the Crusade, 1216–1307 (Oxford: Clarendon Press, 1988) for a thorough portrait of England’s involvement with Crusades to the East during the thirteenth and early fourteenth centuries. On this topic, see also Christopher Tyerman, England and the

132



Christian Knighthood in Sir Bevis of Hampton

time passed after the fall of Acre in 1291, and as hopes of imminent repossession of the Holy Land decreased, memories of a time when Westerners ruled parts of the Middle East became dim.16 As a result, the ability to envisage Saracen–Christian geographical and political co-existence decreased. The rise of the Ottoman Empire also contributed to this shift. In the fifteenth and sixteenth centuries, Europeans who looked East did not see Western Christian kingdoms, however small and beleaguered, scattered about within a Muslim and Orthodox Christian East; instead, they saw a rising and unified political Muslim threat as the Ottoman Empire consolidated its holdings and expanded into Europe.17 It is not surprising, therefore, that texts about an English knight in the East from the thirteenth and fourteenth centuries can imagine him becoming close to a Saracen ruler, while such an idea seems much stranger in texts that circulated in the fifteenth and sixteenth centuries. If Bevis’s encounter with Terri raises the issue of a Christian knight’s relationship to a Saracen world, his arrival at Damascus extends consideration of this issue. Boeve spends nine lines laying out the architectural glory of Damascus: the buildings are covered in silver or gold, and Brademound has commissioned statues of gold and gems so bright that they shine like the sun.18 With the wealth and impressiveness of the city made clear, the text next describes how Boeve enters the city, hears people worshipping Mahoun, and responds by attacking this ceremony. He enters the temple, beats the idol of Mahoun, and breaks the neck of a Saracen ‘priest’.19 In this version, the cultural attractions of the Saracen world are evoked – the wealth, the artistic achievements – and then the problematic aspect of that world (from a Christian perspective) is exhibited and destroyed. The Christian knight may perceive and appreciate the cultural wonders of the Saracen East, but he must also work to destroy certain elements of that world. In A, this encounter changes a little. The cultural attractions of the Saracen world are not depicted. Instead, this version states: Forþ him wente sire Beuoun, Til a com to Dames toun; Aboute þe time of middai Out of a mameri a sai Sarasins come gret foisoun,

16 17

18 19

Crusades 1095–1588 (Chicago, IL: University of Chicago Press, 1988), pp. 86–132, 152–228. For the situation in the later fourteenth and fifteenth centuries, see Tyerman, pp. 229–58. Tyerman, England and the Crusades, pp. 229–30, 308–9, 321–2. As Tyerman puts it, ‘By the 1460s the Ottomans were on the Danube and the Adriatic; in 1480 they were in Italy.’ The sixteenth century saw further gains for the Ottomans, who advanced to take Hungary and, in 1529, to besiege Vienna. See Tyerman, England and the Crusades, pp. 320, 346; see also Jonathan Riley-Smith, The Crusades: A Short History (New Haven, CT: Yale University Press, 1987), pp. 190–1. Boeve 868–75. Boeve 881–3.

133



Siobhain Bly Calkin Þat hadde anoured here Mahoun. Beues of is palfrei aliȝte And ran to her mameri ful riȝte And slouȝ here prest, þat þer was in, And þrew here godes in þe fen. (pp. 66–8)

Here, the important event is Bevis’s attack on the Saracen ‘mameri’. The narrative foregrounds the issue of religious difference and excises any suggestion of artistic admiration for a Saracen city. Bevis may perceive Ermin as a Saracen ‘broþer’ in A, but readers are not encouraged to linger over the cultural attractions of a Saracen realm; rather, their attention is directed towards the Christian knight’s role as a disrupter of Saracen worship. As I have argued elsewhere,20 the reason for this may be the fact that A’s Bevis is already so radical in his desire to occupy a familial space in the Saracen world that he periodically has to re-manifest his right to the appellation of Christian English knight. The Damascus incident marks one such re-­manifestation and hence must be unclouded by any further reference to admirable or desirable elements of the Saracen world. In both Boeve and A, the hero explicitly targets Saracen idols and ‘priests’; his enmity does not extend to mere Saracen worshippers – a fact which suggests the possibility of some sort of peaceful co-existence. Religious difference is a problem for the Christian knight in these two texts, but there is a sense in which the Saracen world consists of more than religious practices – Saracens may also be desirable overlords, wealthy and gifted city-builders, and surrogate family members. The problem for the protagonist is not the larger Saracen world per se, but the Saracen religion in particular, and up to a point these can be differentiated. This is not the case in C or P. C appears at first glance to paint an impressive portrait of Saracen architectural achievement and wealth. In a lengthy passage (1647–1704),21 it catalogues the beauty and luxury of Damascus. The houses are made of crystal, pillars are made of gold, and each roof is made of lead, pointed with gold. In short, as Bevis says, ‘Soche a cyte sawe y neuer noon!’ (1703). The description, however, also suggests that Bevis’s perspective is less that of an admiring onlooker than that of a prospective enemy. The double walls of the city are noted, as are its protective moats and the number of gates and entry ways: ‘Ther be a-bowte sixty gatys, ywys, / And ii brygges and ii portcolys’ (1676–7). The text even explicitly states ‘Soche a cyte was noon vndur sonne – / Hyt was neuer, nor schall be, wonne’ (1674–5). The description of wealth and clever craftsmanship is thus seasoned with references that remind us of Bevis’s role as a warrior entering an enemy city. This perception is sharpened when Bevis notices the ‘tempull’ and immediately draws his sword to attack the Saracens therein: 20 21

Bly Calkin, ‘The anxieties of encounter and exchange’, pp. 135–6, 138–43, 145–8. References to C are to BHF, II.

134



Christian Knighthood in Sir Bevis of Hampton And some he smote of the hedde, And some he woundyd to the dedde – Some the leggys & some the arme: To sle them he thought no harme! Besyde the tempull he made a fyre And threwe þer-yn dam & syre: Sone & doghtur all he brent, And all þer goddys, or he went.  (1718–25)

Bevis’s enmity towards the Saracen world, hinted at in the description of Damascus, here becomes explicit. For the Bevis of C, there is no distinction between Saracen worshippers and their idols and ‘priests’: he attacks all and sundry. Killing Saracens is accorded even more narrative importance than the destruction of idols, as the human slaughter is recounted first and in greater detail. Where the Bevis of A perceives Ermin as a Saracen brother, his counter­part in C sees Saracen families as targets for attack. His task is to root out any future possibility of Saracen worship by destroying all Saracens. Enmity unmitigated by compassion or admiration is the defining characteristic of Bevis’s encounter with the Saracens of Damascus in C. P’s Bevis differs somewhat from his precursors. He, too, enters a Damascus linked to architectural wonders and wealth; the description, however, is much shorter than in C, and the twelve lines present Brademond’s palace as indicative of Saracen wealth and engineering ‘ginne’: the sense of pervasive wealth found in other descriptions of Damascus is lost. Impressive Saracen architecture becomes tied more closely to kingship and perhaps is less exotic for that. Nevertheless, Bevis clearly admires what he sees: ‘of that place meruailed he’ (sig. Dir). However, when he comes into the city Bevis finds himself in the midst of a Saracen festival. Saracen worship is here not confined to a temple but permeates the city, and Bevis’s admiration turns to consternation as he demands of the revelling Saracens ‘what deuill of hell doe yee? / Why make you Mahound this present, and dispise God Omnipotent?’ (sig. Dir). Bevis then leaps ‘unto Mahound, and t[akes] him right by the crowne, / And cast[s] him amidst the mire’ (sig. Dir). As in Boeve and in A, Bevis concentrates his attack on the religious centre of Saracen worship rather than on all Saracens. In P, however, the Saracen worshippers fight back, and this changes the situation. Rather than killing Saracens indiscriminately, like C’s Bevis, the Bevis of P kills only those 200 Saracens that attack him. Nevertheless, he shares with his counterpart in C an ultimate inability to differentiate Saracen worship from a larger Saracen world. Interestingly, however, P represents this expansive course of action as something for which Saracens themselves are equally responsible. In P, the Saracens fill the city with acts of worship rather than confining them to a temple, they attack Bevis, and they claim that he ‘will destroy all the citie’ (sig. Dir). Saracen acts and words make Bevis’s attack on Mahound an attack on the entire city of Damascus. The Saracens themselves insist that there be no separation between religion and the larger 135



Siobhain Bly Calkin

Saracen world, and they thereby deepen the divide between themselves and the Christian knight. Thus, C and P make both Saracen worship and the larger Saracen civilization objects of their exemplary knight’s aggression, insisting on an entirely inimical relationship between the Christian knight and the Saracen world in which he finds himself. When Bevis finally reaches Brademond, his words of greeting also manifest a distinctive sense of how a Christian knight should interact with the Saracen world. In Boeve, the hero responds to Brademound’s enquiry about the purpose of his mission with the words Par mun chef ! … jeo vus frai ben saver, lisez moi ceo bref toust saunz demorrer, ou je vus couperai la teste o mun espeie de ascer.  (898–900) (By my head! … you will soon know. Read me this letter without delay, or I will cut off your head with my steel sword.)

As in his first encounter with Hermine, Boeve speaks very directly, indeed rudely, and wastes no time on diplomatic wordplay. The text also emphasizes Boeve’s physical ferocity: he threatens Brademound with beheading, and the Saracen king trembles with fear.22 This encounter stresses Boeve’s martial prowess, his disregard for political niceties, and his ability to inspire fear. Religious difference does not figure in the exchange. In the Middle English versions, however, it has a much bigger role. In A, Bevis greets Brademond thus: ‘God, þat made þis world al ronde, Þe saue, sire king Brademond, And ek alle þine fere, Þat I se now here, And ȝif þat ilche blessing Likeþ þe riȝt noþing, Mahoun, þat is god þin, Teruagant & Apolin, Þe blessi and diȝte Be alle here miȝte! …’ Beues kneuled & nolde nouȝt stonde & ȝaf vp is deþ wiþ is owene honde.  (pp. 69–70)

As the scene is fleshed out in more detail, different aspects of Bevis’s identity come to the fore. He offers a diplomatic greeting, but one which highlights the divisions between him and Brademond’s court by consisting of first a Christian blessing and then a Saracen one. This politically correct Bevis

22

Boeve 901–2.

136



Christian Knighthood in Sir Bevis of Hampton

differs from his Anglo-Norman predecessor in his courtesy (he even kneels to Brademond) and in his concern to address the fact of religious difference. Clearly A’s ‘kniȝt of cristene lawe’ (p. 198) is expected to be diplomatic and to insist upon his Christian identity at the same time. The Bevis of C, by contrast, uses his greeting to express implacable hostility: Ihesu, that made þe planettes vii And all the worlde vndur heuyn, And made þys worlde wyde & rome, Geue the, Syr Kyng, Hys malyson! And Mahounde, that ys god thyn – Bothe Mahounde and Apolyn – The blesse now might, And all thy men right!  (1765–76)

The ‘malyson’ of God is invoked upon the Saracens, in contradistinction to their own gods’ blessing, and C thereby emphasizes religious difference in a way that insists upon the enmity between religions. Diplomacy still appears, as Bevis does not confine his greeting to a curse, but his words are tinged with antagonism in ways not present in A. There is also no reference to Bevis kneeling or assuming a subservient position in any way. Again, C insists that the Christian hero manifest open hostility to the Saracen world even as he functions within its political system. This hostility is explicitly religious and clearly differs from the physical hostility voiced by Boeve. Religious hostility also characterizes P’s narration of this encounter. In his role as diplomatic messenger, Bevis kneels to deliver the letter to ­Brade­mond, but uses his words rather less diplomatically:             Jesus that I vnderstand, That shaped al ye world so round, now giue thee sorrow king Bradmound. But Mahound and Apolyne, and Termagaunt which be Gods thyne: Let them giue thee now their blessing, that short life, and euill ending. (sig. Div)

In addition to incorporating the Christian malediction found in C, P uses the Saracen blessing to insult Brademond’s religion by averring that the blessing of the Saracen gods consists of ‘short life, and euill ending’. The outward forms of diplomacy are maintained, but the words constitute a very undiplomatic condemnation of the Saracen religion. This Christian knight leaves no opportunity to attack Saracen religious beliefs unexploited, an aspect of his comportment that is emphasized in P as Ermin’s letter warns Brademond, ‘if [Bevis] passe away, he will destroy all your lay’ (sig. Div). Of the versions studied here, only P includes this comment suggesting that Bevis threatens the very Saracen religion with destruction. P thereby depicts Brademond’s imprisonment of Bevis as an important act of religious self-defence. P again 137



Siobhain Bly Calkin

emphasizes Christian–Saracen religious animosity above all else and makes this animosity a mutual affair. Both C and P, then, suggest that a Christian knight ideally manifests unmitigated enmity towards the Saracen world. Religious difference is a pervasive concern of these texts, while Boeve emphasizes physical aggression, and portrays for readers the attractions of the Saracen world, implying that it is a venue within which a Christian knight can function both as a knightly retainer and as a Christian. A hovers somewhere between the other versions, stressing Bevis’s Christian links more than Boeve does, but sharing with it the sense that an exemplary Christian knight can exist in a Saracen world without trying to destroy it altogether. Boeve and A thus envisage a relationship between a Christian knight and the Saracen world that encompasses religious confrontation but consists of more than that. It is interesting that as the Bevis story becomes further removed from a time of Western Christian presence in the Middle East, the protagonist’s scope for engagement with non-Christians narrows. As geography divides Western Christians and Muslims more definitively, and as Muslim forces press on the edges of Europe, Bevis depicts only one role for a knight in the Saracen world: as the attacker of non-Christians. Although the hero’s arrival in Damascus foregrounds the task of religious confrontation, a Christian knight must also manifest religious faith in other ways. For example, he should have a spiritual relationship with God. Once Brademond has imprisoned the hero, these spiritual considerations come to the fore, as Boeve/Bevis uses the seven years he spends in a Saracen dungeon to reflect on his predicament and engineer his escape. In all four texts considered here, this involves a role for God, although the precise nature of that role differs markedly. In Boeve, after briefly praying to God for help,23 the hero spends eight lines lamenting his betrayal by Hermine. As befits the text’s emphasis on the knightly over the godly, Boeve focuses on this world, not the next, when he assesses his predicament. Boeve’s reflection also concentrates on inter­ personal relationships rather than on those between religions; in his lament, the sense of a retainer having been betrayed by his overlord overrides any sense that Boeve has been the victim of Saracen enmity: Jeo sui ci trai mult felunment, jeo ne le ai pas deservi, si cum jeo entent, que il me dust trair si tres ledement; une reaume li conquis ov le espeie trenchaunt.  (960–3) (I am most wickedly betrayed here. I believe I have not deserved his betraying me so badly: I conquered a kingdom for him with my sharp sword.)

23

Boeve 954.

138

Christian Knighthood in Sir Bevis of Hampton



The emphasis on the kingdom Boeve won for Hermine through battle ­characterizes Hermine’s act as a betrayal of the knight’s martial endeavour and the rewards it should bring; Hermine’s Saracenness is not even mentioned. Boeve eventually turns more of his attention to religious matters, however, as after seven years in gaol it becomes apparent that he will need God’s help to escape. He prays that he either die or escape rather than languish any longer in the gaol.24 His Saracen gaolers hear this and decide to oblige Boeve by killing him – an act which leads to their deaths at Boeve’s hands and to Boeve’s subsequent starvation for three days as no one feeds him and he remains bound in irons and unable to effect his own escape. Boeve then prays to God again, and is enabled to escape: par la vertu deu, ke roi est de pité, si sunt ly liens tretuz depescé; e Boefs le veyt, unkes ne fu si le: de joie saili en haut quinze pez mesurez, en une voute saili, ne se est de rien dotez. (1089–93) (his bonds were quite shattered, through the power of God, king of compassion. And Boeve saw this and never before had he been so happy: he jumped fifteen feet high for joy, I reckon.)

Boeve’s protagonist thus develops his relationship to God through prayer, and escapes through God’s help. Boeve’s religious credentials receive a boost, although God’s involvement is limited to the loosing of Boeve’s irons; the miraculous jump seems to be the product solely of Boeve’s own joy. Once again, Boeve ensures that the hero’s physical prowess is not overshadowed by his religious faith. As a whole, the imprisonment narrative in Boeve reveals more of Boeve’s religious faith, but balances this with comments on retainer– overlord interactions and with acts of physical prowess. Religion and spiritual development figure in this text, but general knightly endeavour receives more emphasis. A also uses Bevis’s imprisonment to explore his spiritual state. This version lacks any of the retainer lamentation found in Boeve, focusing instead on Bevis’s prayer to God after seven years in prison. A deems this prayer worthy of twelve lines, twice as many as the prayer receives in Boeve, thus giving increased emphasis to the protagonist’s relationship with his God. The effect of the prayer on Bevis’s gaolers is the same as in Boeve, and he once more endures three days without food after he kills them. As in Boeve, this state of affairs leads Bevis to pray to God again, and his prayer is immediately answered:

24

Ibid. 1040–5.

139



Siobhain Bly Calkin So ȝerne he gan to Iesu speke, Þat his vetres gonne breke And of is medel þe grete ston. Iesu Crist he þankede anon.  (p. 85)

Bevis then climbs the gaolers’ rope to freedom, no joyous leap being necessary. Here, Bevis’s escape becomes more closely tied to divine assistance: God both loosens his fetters and removes a huge stone from his middle. There is no immediate juxtaposition of divine favour with exceptional physical prowess; rather, Christ receives all the credit for the escape. But though A thus emphasizes the knight’s religious attributes over his physical ones, it does not dwell on them exclusively, as it spends a number of lines recounting the martial details of Bevis’s battles with reptiles in the dungeon.25 While the reptile battles in A parallel the dungeon experiences of St George battling dragons in some narratives, these hagiographic parallels are not explicitly developed at this point in A as the reptiles are not called dragons and no reference is made to St George.26 In later versions of Bevis, however, the protagonist’s battles in Brademond’s dungeon acquire much more explicit hagiographic colouring as part of a narrative elaboration of this scene. For example, as C lengthens its account of Bevis’s incarceration, it has him fight two dragons in addition to other reptiles. While this dragon-fighting undeniably demonstrates Bevis’s physical prowess, it also means that his experiences in a Saracen prison begin to resemble more clearly those of St George and other saints. This hagiographic resemblance becomes even more pointed later on in C when Bevis is bitten by an adder and prays for assistance.27 In response, he receives a divine visitation: Wyth that come an angell downe, And appered to Syr Befowne; And, thorow þat angell & þat lyght, That fowle worme loste hys syght, And there before hym all to-braste.  (2097–101)

Bevis experiences a miracle, and the nature of this miracle likens his adventures in a Saracen prison to those of St Margaret, St Juliana or St George during their respective imprisonments. C’s Bevis is indeed an exemplary Christian, so favoured that God sends an angel to do battle for him.

25 26

BHK, p. 81. A instead emphasizes Bevis’s links to St George later on, when Bevis does battle with a dragon in the Christian realm of ‘Coloyne’ and calls on St George for help (BHK, p. 129). For detailed discussions of the connections between Bevis and St George, see John E. Matzke, ‘The legend of St George: its development into a roman d’aventure’, PMLA, 19 (1904), 449–78; Jennifer Fellows, ‘St George as romance hero’, Reading Medieval Studies, 19 (1993), 27–54. See also pp. 89–90 above. 27 BHF, II, 2092–6.

140



Christian Knighthood in Sir Bevis of Hampton

In C, Bevis’s exemplary Christianity also includes less martial behaviour. After his battle against the dragons, Bevis displays an admirable Christian willingness to endure woe with gratitude: he ‘þanked god, þat all may wynne, / Of the care that he was ynne’ (1929–30). The idea behind these lines is that suffering is desirable because it circumvents one’s ability to indulge in sin and fosters an awareness of how unimportant earthly life is in relation to spiritual life and the heavenly afterlife. The text further engages this concept when it states that after seven years in prison ‘Ffull symple was Sir Befyse pryde’ (2062), a phrase which indicates that Bevis has overcome a deadly sin. C also increases the number of prayers made by Bevis. Although A includes the battle with the adder at the start of Bevis’s imprisonment, and Boeve refers to poisonous snakes in the dungeon, neither text has the hero utter a prayer about an adder. In C, however, Bevis prays      Ihesu, my Lorde, mercy! Thys eddur hath beton me sore – Ffro lyfe y wolde that y were! Helpe, Lorde, yf hyt be Thy wylle: Let me not yn preson spylle!  (2092–6)

In addition to increasing the number of Bevis’s prayers, C also depicts Bevis’s prison experiences in terms that imply they are religiously formative. For example, C describes Bevis’s three-day enforced fast after he kills his gaolers as a ‘penance’ – a term absent from the earlier versions, and one that characterizes Bevis’s hunger as part of his religious development. In C, then, Bevis’s time in a Saracen prison becomes an opportunity to depict the Christian knight both as a quasi-saintly hero who does battle with dragons and receives angelic visitations, and as a model of more imitable Christian deportment such as prayer and penance. C, however, does not neglect to include references that also emphasize the model Christian knight’s implacable enmity towards Saracens. Following the adder’s explosion, Bevis prays, Thes false Sarsyns worche me woo: Venge me, Lorde – y am ther foo, And sende me grace heuyn to wynne, Owt of þe preson þat y am ynne.  (2119–22)

Unlike Boeve, who attributes perfidy specifically to Hermine, C’s Bevis makes falseness a characteristic of Saracens generally. His words in line 2120 also highlight that fierce enmity that is so much part of C’s depiction of Bevis’s conduct in Damascus. Interestingly, the Saracen gaolers pick up on this antagonism and generalized condemnation: ‘They seyde, “Heryst þou þat felon – / How he dyspysyth owre nacion?” ’ (2129–30). The sequence of events that leads to Bevis’s escape thus begins, in C, with statements on both sides expressing a generalized enmity between Christians and Saracens. 141



Siobhain Bly Calkin

Unsurprisingly, then, Bevis’s escape is characterized here as a battle between the two faiths, with emphasis on the divine aid available to a Christian knight. Bevis kills the second gaoler ‘thorow Godys sonde’ (2180), and leaps to freedom through God’s grace (2201).28 C’s narration of Bevis’s imprisonment emphasizes repeatedly the divine, occasionally miraculous, assistance available to a model Christian knight in his martial endeavours while simultaneously depicting less martial acts, such as prayer and acceptance of suffering, that also constitute exemplary Christian behaviour. P resembles C closely in its portrayal of Bevis’s imprisonment. As in C, Bevis kills two dragons and thus displays saintly martial prowess. P reduces the length of the dragon fight from three days to one, but the impressiveness of the protagonist’s feat is undiminished. P’s Bevis also receives angelic assistance against an adder, and thus, like C’s, approaches sainthood. In P, as in C, Bevis also displays more imitable Christian virtues: he prays, he loses his pride, and he welcomes suffering. P emphasizes this suffering by describing Bevis’s existence as one of ‘povertie and great strife’ (sig. Diir). P also uses Bevis’s experiences in prison to develop its portrait of Saracen–Christian hostility. For example, Bevis prays: ‘The Sarasines do me much woe, have mercy on me O Lord do so / Now give me grace heaven to winne, and out of prison that I am in’ (sig. Diiiv). Unlike C, P does not emphasize Saracen perfidy in this prayer: the Saracens simply work woe generally as part of their ongoing animosity towards Christians. The Saracen guards’ reaction to Bevis’s prayer further communicates this animosity:      hearest thou this felowne, how he dispiseth our Mahound? He weeneth that his God may him saue, and succour thinketh he to    haue. But by Mahound it shall not gaine, I shall not rest till he be slaine. (sig. Diiiv)

In C, the ‘despite’ was done to the gaoler’s nation; here, Bevis’s prayer offends Mahound directly, but the escape that follows is characterized in both texts as a battle between gods. In P, Bevis cuts the rope carrying the second gaoler ‘through Gods hand’ (sig. Divr), and when he suffers what is again termed the ‘penance’ of three days without food, he prays for aid and ‘by Gods might, Bevis skipped he was ful light. / And gate the rope in his hand, and came up’ (sig. Divr). The spectacular leap that was the product of the hero’s physical prowess in Boeve has become in P, as in C, an act of God as the hero’s spiritual state and religious exemplarity receive more elaboration in these texts. Bevis’s adventures in Brademond’s dungeon remain similar in all versions of the romance studied here, but specific details change in ways that indicate

28

C does not mention either a stone or fetters as obstacles to be overcome in Bevis’s escape; thus his challenge is the height of the dungeon entrance alone.

142



Christian Knighthood in Sir Bevis of Hampton

an increasing religiosity. Boeve depicts the protagonist’s appeals to God, but keeps them brief and complements them with lamentations about an overlord’s perfidy and with acts of physical prowess that work in tandem with divine aid to effect the knight’s escape. A increases the length of the appeals to God and paves the way for more divine intervention, but C and P add details about Bevis’s spiritual state and elevate his experience to a hagio­ graphic level. The texts considered here thus chart an increasing concern with the spiritual state of the Christian knight. The fifteenth and sixteenth centuries in England were periods of increasing attention to lay spirituality and religious practices, as manifested in the spread of lay devotional texts, of Lollardy and, eventually, of Protestantism. Versions of Bevis that circulated in this period are clearly adapted in ways that speak to this concern, even though they maintain the narrative’s emphasis on knightly action in battle. Interestingly, however, the texts that do engage more ideas about Bevis’s spiritual state also depict a pervasive religious hostility between Saracens and Christians. As Bevis circulates in times of increasing dissent about faith practices within English Christendom, internal dissent is pointedly ignored in favour of emphasizing hard-and-fast lines of division and enmity between Christians and Saracens. While the broad outlines of the Bevis story do not change in versions circulating between the thirteenth century and the sixteenth, specific details vary. As different texts recount Bevis’s encounters with Saracens before he reclaims his heritage, distinctive portraits of the ideal Christian knight emerge. Boeve depicts its protagonist as a martial figure whose encounters with Saracens suggest that religious difference must be of concern to a ­ Christian knight, but that other factors can play a part in cross-cultural relationships. Personal treachery, knightly admiration, retainer–overlord duties, and cultural achievements all play a role in Boeve’s experience of the Saracen world, and it is clear that one can separate the knight from the Christian. A, produced in the 1330s, emphasizes the Christianity of its ‘kniȝt of cristene lawe’ rather more, but it does not portray this Christianity as inherently inimical to the Saracen world or equate the Saracen world with religious difference. ­ Saracens and Christians interact in multifaceted ways in this text, and Bevis is able to be knightly in a Saracen context without attacking that context indiscriminately. C, dated to the late fifteenth or early sixteenth century, reduces interfaith interaction to enmity. Encounters between Saracens and this text’s protagonist suggest that to be a Christian knight is to be constantly on the lookout for opportunities to end religious difference, preferably by wholesale destruction. The text also depicts its hero as a model Christian who embraces hardship as spiritually generative and who receives angelic assistance in response to prayer. The same combination of religious aggression and divine favour characterizes the Bevis of P. This text, however, differs from C in making enmity between Saracens and Christians a more mutual affair, and in emphasizing honour as a knightly attribute. In all four versions, then, Bevis’s encounters with Saracens while in Ermin’s service remain largely the same, 143



Siobhain Bly Calkin

but the details of those encounters reveal very different heroes. In this way, the heroism of Bevis adapts to the passage of time and remains an ideal of Christian knighthood even as specific ingredients of that ideal change with historical developments.29

29

I would like to express my gratitude to the editors of this collection for their generosity and careful attention to detail, and to Carleton University and the Social Sciences and Humanities Research Council of Canada for grants supporting the research for this article.

144

9 Ascopard’s Betrayal: A Narrative Problem MELISSA FURROW

A

scopard, the Saracen giant sent to retrieve Josian and kill Bevis, appears more than half-way through the Middle English Sir Bevis of Hampton and is an intermittent presence for only about one-third of the whole. But as Bevis’s ungainly page he is one of the memorable features of Bevis’s story: it is he whose painting was paired with Bevis’s at Southampton’s Bargate. Whether he was remembered as an amusing marvel (like the horse Arundel) or a sinister threat (like Guy of Warwick’s long-remembered giant opponent Colbrand) is unclear but would in large part have depended on the slant given the story by particular tellers. Ascopard commits a terrible crime, a betrayal of trust, when he abandons his allegiance to Bevis, returns to the service of Josian’s first husband, King Yvor of Mombraunt, and arranges to kidnap Josian just after she has given birth. Yet he is also an engaging comic foil to the hero, and he captures Bevis’s murderous stepfather when Bevis himself has been unable to do so. Escopart’s betrayal of Boeve is a shocking dissonance in the source text, the Anglo-Norman Boeve de Haumtone, and the Middle English versions use several strategies to try to cope with that dissonance. The betrayal creates a problem of consistency for the Middle English versions that is frankly unresolved in Boeve as we have it. The change in behaviour is so unprepared for that it produces the impression not of psychological complexity, but merely of incomprehensibility. Cut the poem in two after the restoration of Boeve’s lands and his marriage to Josiane, and we have two utterly different conceptions of Escopart. Judith Weiss, in her doctoral dissertation, makes a persuasive case on stylistic grounds that the 

In the Auchinleck MS version he suddenly appears at line 2505 (p. 118), although he is mentioned at line 2352 (p. 114) as being sent by Garcy. He dies at line 3887 (p. 181), and the romance ends at line 4620 (p. 218). References to the text of Bevis are to The Romance of Sir Beues of Hamtoun …, ed. by Eugen Kölbing, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) (hereafter BHK).  See Adrian Rance, ‘The Bevis and Ascupart panels, Bargate Museum, Southampton’, Proceedings of the Hampshire Field Club and Archaeological Society, 42 (1986), 147–53, for a description of the panels and discussion of the dating. The present panels, much repainted, probably go back to no earlier than 1594 (ibid., p. 149) but, if so, they replaced earlier panels with the same subjects.

145



Melissa Furrow

original Anglo-Norman poem did end with Boeve’s reclaiming of Haumtone and his marriage to Josiane, and that the rest of the poem as we have it is an addition by a different author. If we accept this argument, as I believe we should, the original conception of Escopart is a happy one: he enters Boeve’s service, converts to Christianity and serves Boeve well despite his animallike features and hideous appearance, and his story culminates in his triumphant capture of Boeve’s stepfather, the Emperor Doun. The continuation, however, disposes of Escopart with shocking abruptness, and the betrayal is not adequately motivated. Boeve is exiled from England and leaves his lands to his mestre, Sabaoth. When Josiane asks Boeve whom they will take with them, Sabaoth immediately suggests his own son and heir, Terri. Then come lines in which Escopart (in this passage for the first time referred to by his baptismal name, Guy) asks what is to happen to him: ‘Sire,’ dist Guy, ‘que avez enpensé? Menerez moi o vus ou ci me lerrez?’ ‘Amy,’ dist Boves, ‘o Sabaoth remeyndrez. Large vus durrai o deus cens chevalers.’ E cil respont: ‘Merci en eyez.’ Puis s’en torne dolent e irez. Passe la jur, la nuit est serrez, e li pautoner est el chimin entrez …  (2649–56) (‘My lord’, said Guy, ‘what is your plan? Will you take me with you or leave me here?’ ‘My friend’, said Boeve, ‘stay with Sabaoth. I will [treat you generously] with two hundred knights.’ ‘Thank you’, said he, then turned away, miserable and angry. The day passed, the night grew dark and the scoundrel took to the road …)

The passage is filled with reminders of Boeve’s care for Escopart: the name Guy, which was after all Boeve’s father’s, the address Amy (‘Friend’); the promise that Boeve will give him great wealth and two hundred knights. But Escopart’s reaction is grudging thanks, sorrow and anger, and an immedi­ate narratorial conversion to li pautoner (‘the scoundrel’, ‘the mercenary’). By the next morning he is on the road to Monbrant and King Yvori. It is not spelled out why Escopart reacts as he does, and it is up to the listener or reader to see the betrayal either as the inherent treachery of a Saracen giant who has been only nominally Christianized or as the pique of a servant who thinks himself insufficiently valued as a companion by his master. Boeve has just been exiled for refusing to abandon his horse Arundel when the latter is 

Judith E. Martin, ‘Studies in Some Early Middle English Romances’ (unpub. Ph.D. diss., Univ. of Cambridge, 1968), pp. 95–105.  All references to the text of Boeve are to Der anglonormannische Boeve de Haumtone, ed. by Albert Stimming, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974) (hereafter Boeve). Translations are quoted from Weiss’s translation (cf. p. 11 n. 9 above), departures from which are indicated by square brackets.

146



Ascopard’s Betrayal

charged with murdering the king’s son. His words on that occasion, ‘e ki ad bon serf, ne le deit guerpir’ (‘whoever has a good servant should not abandon him’, 2595) endorse his loyalty to his horse and within sixty lines underline that the same loyalty does not extend to Escopart. When he is introduced, Escopart is seen to have many of the subhuman traits of monstrous Saracens in chansons de geste, such as Agrapars in Aliscans, or Algalafre in Fierabras. He is distinguished not only by his size but by animal features: a forehead as big as an elephant’s backside, a horned nose, hair as long as a horse’s mane, teeth as long as a boar’s, with fingernails hard enough to scratch through any wall in Christendom; he bays when he talks, like a mastiff. He is described in class terms: veleyn (‘churl’), lede sergant (‘hideous fellow’), and once (with irony) lede bacheler (‘ugly young [knight]’). He is likewise described in terms of racial difference: ‘plu neyr ou la char ke n’est arement’ (‘his skin was blacker than ink’). Already, then, he embodies differences that the other Saracens Boeve has dealt with tend not to display – or, if they do, only in one or two ways: Rudefoun, ­Brademound’s standard-bearer, ‘plus estoit velu ke nul porc o tusun’ (‘was hairier than any pig with a fleece’, 572), and Brademound has a brother who is a giant (1289ff.) and inhospitable, but not otherwise remarkable. Although Boeve moves through racially very mixed territory around the ­Mediterranean when he travels from Egypt to Damascus, from Damascus to Jerusalem, from Jerusalem to ‘Monbrant’ (which is apparently in North Africa, as his informant mentions Carthage in the directions how to get there), almost no one’s skin colour or facial features are worthy of mention except Escopart’s. ­Escopart lacks even the dignity of individuality: usually, though not always, he is not ‘Escopart’ but ‘l’Escopart’ – a member of a race rather than a named person.10 So Escopart is the embodiment of difference, and he obligingly

 

   10

Ibid. 1743–70. Cf. Aliscans, ed. by Claude Régnier, 2 vols, CFMA (Paris: Champion, 1990), lines 6260–3: ‘Rois Agrapart fu de lede façon: / Lons a les crins desi que au menton, / Les elz ot rouges ausi come charbon, / Ongles aguz ausi come grifon’ (‘King Agrapart was ugly: he has hair down to his chin, he had red eyes just like embers, sharp nails just like a gryphon’); but note that Agrapart is unusually short rather than unusually tall. Cf. the description of Algalafre in Fierabras, ed. by Marc Le Person, CFMA (Paris: Champion, 2003), lines 4890–905: ‘Li paiens estoit grans, hydeusement formez: / El hasterel deriere avoit les euz tornez, / Plainne paume out de goule et demi pié de nez’ (‘The pagan was big, hideously formed: he had his eyes turned to the back of his head, a full palm’s width was his gob and half a foot his nose’). Boeve 1744–75. Ibid. 1376–7. The exception is Josiane, whose complexion is very briefly described: ‘plus fut ele colouré ke rose en umbrage’ (‘she was more fresh of colour than a rose in the shade’, 373). The Escopartes or Ascopartes were thought to be an African people, and are mentioned ‘dans une quarantaine de chansons de geste’, according to Romaine Wolf-Bonvin, ‘Escopart, le géant dépérissant de Beuve de Hantone’, in La Chrétienté au péril sarrasin: Actes du colloque de la section française de la Société Internationale Rencesvals (Aix-en-Provence, 30 septembre – 1er octobre 1999), Senefiance 46 (Aix-en-Provence: Centre universitaire d’études et de recherches médiévales d’Aix, Université de Provence, 2000), pp. 249–65 (p. 265 n. 29).

147



Melissa Furrow

swears by Tervagaunt and by Mahoun to underscore that that difference is also religious. The Anglo-Norman poet’s originality comes in combining these monstrous Saracen traits with traits of the good Saracen giant. There are two particularly famous admirable Saracen giants of chanson de geste – Rainoart from the cycle of Guillaume d’Orange, and Fierabras – who share the character­ istics of being inordinately tall and strong, handsome,11 powerful fighters and, most importantly, converts to Christianity.12 Although hideous, Escopart is, like them, tall and strong, a powerful fighter and a convert. At Josiane’s prompting, he cries for mercy when overcome in battle with Boeve and asks to be christened: L’Escopart comença donc a crier, ke tretut le boys fet a resoner: ‘Boves, ne me tuez mye, jeo me voile cristiener.’  (1831–3) (The Escopart began to roar so that the whole wood rang: ‘Boeve, don’t kill me, I want to be a Christian.’)

The poet thus creates a challenging hybrid: an alien-looking monster brought into the fold, playing for the home team. That combination of the terrifyingly alien exterior and the converted interior produces an incongruity that the poet exploits as humour. Hurrying to rescue Josiane from burning, for example, Escopart has to cope with merchants who mistake him for a devil and jump out of their ship rather than carry him to shore, and a shepherd who is too terrified to give him information.13 Even his willingness to convert has no dignity: elsewhere, such willingness is portrayed as coming from concession to the heroic opponent, after a long and equal struggle (Fierabras to the wounded Olivier), or from identification with Christian patrons (Rainoart with William and Guiborc). Escopart, comically, just does not want his head cut off. Even before his conversion he is so different, in fact, that he fits in nowhere. Boeve’s laughter at Escopart is preceded by mockery in Escopart’s own country for being so small (everyone calls him dwarf because he is only nine feet tall) and such great shame at being different from the others at home that he runs away to serve Yvori of Monbrant. Big as he is, he ought to be a significant threat 11 12

Cf. Aliscans 3593; Fierabras 1291ff. In considering Rainoart as a giant, I follow Jeffrey Jerome Cohen, Of Giants: Sex, Monsters, and the Middle Ages, Medieval Cultures 17 (Minneapolis, MN: University of Minnesota Press, 1999), pp. 166–72. But it should be noted that, though Rainoart is clearly very large and extravagantly strong, he is not called a giant, and he is described as only about a foot taller than the tallest barons: see Aliscans, ed. Régnier, lines 7891–3: ‘N’i a baron, s’il velt vers lui aler, / Que son chief puist a s’espale adeser; / Envers lui puet un enfant resembler’ (‘There is not a baron, if he wishes to go towards him, whose head can reach his shoulder; compared to him he could look like a child’). Fierabras, at 15 feet tall, is not called a giant either, and is said to be only half a foot taller than Olivier (ibid. 604, 1198). 13 Boeve 2087–91, 2153ff.

148



Ascopard’s Betrayal

to Boeve. But, unlike the lions Boeve has just insisted on defeating singlehanded, unaided by Josiane, Escopart does not after all turn out to be a threat worthy of defeat at Boeve’s hands alone. Instead, as they fight, the wonderful horse Arundel intervenes when he feels his master is in trouble, places his hooves on Escopart’s chest, knocks him down, and stamps on him till Boeve is ready to cut off his head. The humour, however, serves to make the monstrous Escopart more engaging; it also makes his betrayal more dissonant in the poem as we have it, a version seamlessly combining the work of the original author, who is sympathetic to Escopart, and that of the continuator, who rejects him. Sometime around 1300, a translator turned the extended Boeve into a Middle English romance and had to cope with the narrative split in Escopart. That new Middle English version is lost to us in its original form but is represented, to a greater or lesser extent, in nine surviving manuscripts and twenty-two extant printed texts:14 these versions reflect and sometimes expand upon the translator’s efforts to make a more coherent character of Ascopard. The Middle English versions all have an Ascopard who is more extravagantly tall (30 feet or, in C, 24 feet, rather than the nine feet of the AngloNorman poem), but otherwise much more human. The only animal metaphor in Middle English versions other than A is the observation that his hair was like the bristles of a sow (C), or his beard like the bristles of a swine (NSM), or that he himself was bristled like a sow (P).15 Even this last, rather ordinary, metaphor is absent from A: ‘His berd was boþe grete & rowe’ (p. 118). There are no class terms used to describe him, and his skin colour is not mentioned. All versions agree that he looks loathsome, but there is a paring-out of much of the description. P is the most detailed of the English versions, though the details are new ones, not translations: His lyppes were great & hanged syde, His iyen were holowe, his mouth was wyde. Lothely he was to loke on than And lyker a deuyl than a man.  (p. 119)

The other versions do not give these details. P does not give the comic, but incidentally rather touching account, of his rejection in his own country as a dwarf, but the other Middle English versions do include it. In the first encounter with the hero, the Middle English versions reduce the degree to which Ascopard is an alien threat. Bevis wonders, but does not laugh, when he sees him, and, rather than pointing out that he is very ugly and asking if everyone in his country is so hideous and big, asks merely if

14 15

Cf. pp. 104–13 above. BHK, p. 118. See the list of sigla at pp. ix–x above; P, which denotes the printed editions of Bevis collectively, is equivalent to Kölbing’s O, which he intermixes with passages from M in the text at the foot of the page.

149



Melissa Furrow

people in his country are as big as he is. Ascopard does not begin by roaring insults and orders, as does Escopart, but tells Bevis his orders from Garcy: to take back the queen and to kill (or in P to take back) Bevis. P, like Boeve, gives a protracted description of the battle between Bevis and Ascopard – though again this is newly invented material rather than derived from the Anglo-Norman. But in all the other Middle English versions the encounter is quite simple: Bevis, mounted on Arundel, charges and strikes Ascopard in the shoulder with his spear; Ascopard tries to retaliate with his club, misses, ‘And fel wiþ his owene dentte’ (p. 120). Bevis is then about to cut off his head when interrupted by Josian’s plea, which is not contingent on conversion (as it is in Boeve): ‘Let him liuen & ben our knaue!’ (ibid.). In fact, all religious references are absent here from the Middle English texts: Boeve’s charge to Escopart ‘pur icel deu en quey estes creant’ (‘by the god you believe in’, 1777), the oaths by Tervagant and Mahoun, Boeve’s use of the term Paien (‘Pagan’, 1782) to address Escopart, Josiane’s advice to Escopart that he become Boeve’s man and embrace Christianity, Boeve’s oath ‘par cele deu, ke dey honurer / e ke ceo lessa en croiz morer!’ (‘by that God I should revere, whom this man allowed to die on the cross’, 1827–8), Escopart’s cry ‘jeo me voile cristiener’ (‘I want to be a Christian’, 1833) and Josiane’s oath by ‘Sen Pere’ (‘St Peter’, 1835) – all are absent from the Middle English versions.16 Only Bevis’s resistance ‘Þourȝ godes help’ (p. 119), and Josian’s advice to Bevis ‘so god þe saue’ (p. 120), put references to God into the scene in the Middle English versions (except for P), and these do not draw attention to a distinction between Christianity and the medieval romance notion of Islam. This diminishing of Ascopard’s difference, however, might serve only to make the narrative rupture worse, reducing even the force of the stereotype of the monstrous Saracen that might justify his treachery. But they are accompanied by other changes that serve to mitigate Ascopard’s guilt and the shock of his betrayal. The baptism scene in Cologne is key to the characterization of Ascopard. In Boeve it is comic, as it is in Bevis, but the central change is the baptism itself. In Boeve, Escopart complains that the ‘shepherd’ (the bishop of Cologne) is trying to drown him and says that being a Christian is too much for him:        … malveis velen berger, mey volez vus en cest ewe neyer? Trop su jeo crestien, lessez moi alez.  (1971–3) (Base, wicked shepherd – do you want to drown me in this water? Let me go; I’ve had enough of being a Christian.)

But he has jumped into the font of his own volition and has been successfully baptized: ‘si fu en la funte Guy nomez’ (‘and was given the name Guy in the

16

Cf. Boeve 1777, 1784, 1793, 1824, 1782, 1794, 1824, 1827–8, 1833, 1835.

150



Ascopard’s Betrayal

font’, 1967). In Bevis, Ascopard is never christened; in fact, he never says he wants to be christened – it is Bevis’s idea to add the christening of Ascopard to that of Josian. Again a special font has to be made for the giant, but here he firmly refuses to get in it, jumps on to a bench and, in A, proclaims ‘Icham to meche te be cristine!’ (‘I am too big to be Christian’, p. 122). The scene is comic, but it has huge implications. In the Anglo-Norman continuation, Escopart will become an apostate when he reverts to Yvori’s service. It will be twice pointed out when he defects that he has been baptized – indeed, that he owes his baptism to Boeve – once by the narrator just before he makes his decision, and once by Josiane just before he is killed: A tant estevus l’Escopart le fer, ke Boves fist baptiser e lever e a Coloyne Gui fu nomé. (2646–8) (At that moment the fierce Escopart appeared, whom Boeve had baptized and christened and in Cologne had given the name of Gui.) ‘Di moi, dame,’ dist Sabaoth, ‘sont il Sarzins?’ ‘Oyl, bel sire, veez le pautoner, ke Boves fist baptiser e lever.’ just le oy, mort li abati.  (2761–5) (‘Tell me, my lady’, said Sabaoth, ‘are these Saracens?’ ‘Yes, my fair lord, there’s the scoundrel, Gui, whom Boeve had baptized and christened.’ Sabaoth seized his pilgrim’s staff, struck the traitor near the eye, and knocked him down dead.)

He has betrayed Christianity, he has betrayed Boeve who had him baptized, and suddenly he is very easily defeated by an old man with a pilgrim’s staff. Rather than making Ascopard worse by having him refuse baptism, the effect of the refusal in the Middle English versions is to diminish his guilt. When Ascopard dies, he is attacked by a group of twelve armed knights plus Saber, who abandon their pilgrim disguise, rush at him and cut him to pieces with their pilgrim staves made of sharpened steel.17 The various Middle English versions differ slightly in the details, but all agree in making the battle not simply between Saber and Ascopard, and not a case of an unarmed pilgrim effortlessly defeating a Saracen giant – an effortlessness which in Boeve, of course, reinforces the righteousness of Sabaoth’s cause and the concomitant wickedness of Escopart’s. It is harder to kill the Middle English Ascopard. In Bevis, the burden of ingratitude is not so firmly planted on Ascopard’s shoulders. Boeve, after all, gives Escopart a rich reward for his services: elevation to leadership, with two hundred knights to follow him, and wealth to support his new status. But the English Ascopard has not been thus elevated. 17

BHK, pp. 180–1.

151



Melissa Furrow

The idea to take Terri as companion comes from Bevis himself, and the latter makes no provision for Ascopard, who is merely left behind, in a rapid scene in which Bevis tells Josian of the reasons for his exile ‘Ascopard fore’ (‘in front of Ascopard’, p. 169). Ascopard’s motivation is addressed in tantaliz­ ingly ambiguous phrasing: Aȝen to Mombraunt he gan schake, To be-traie Beues, as ȝe mai se, For he was falle in pouerte, For, whan a man is in pouerte falle, He haþ fewe frendes wiþ alle.  (p. 170)

A great deal depends on the antecedent of he in the third line: is it Bevis, the nearest referent, or is it Ascopard, the previous referent of the last he? For the redactor of P, the guilt is clearcut: This Ascaparde, fals was he, For Beuys was fallen in pouerte: Whan a man in pouerte is fal, Fewe frendys meteth he wythal.  (p. 170)

M and E do not address the motivation, but NSC has wording similar to A’s here. From the context it is easier to apply the he to Ascopard:18 he has lost his role as Bevis’s page and is stranded – whereas Bevis, though he has had to give up his newly claimed lands, still has the support of Saber, Josian and Arundel, and with Terri will shortly gain new lordship in Aumbeforce. An Ascopard abandoned by Bevis has much better reason to look for service on the enemy side. An Ascopard relieved of the apostasy and ingratitude shown by Escopart is already a much softened villain. His portrayal as perpetually childlike further mitigates his villainy. One aspect of this portrayal is that Ascopard expresses no sexual interest in anyone at any time. This would be utterly unremarkable in a subordinate character if he were not a Saracen giant. Jeffrey Jerome Cohen’s Of Giants: Sex, Monsters, and the Middle Ages is a subtle study of many different genres of giants of medieval literature, from Beowulf, the chansons de geste, romance and chronicle. Of Harpin de la Montagne, a giant from Chrétien’s Chevalier au lion (Yvain), he remarks: ‘In the copious body of Harpin de la Montagne is condensed the sexual perversity (sexual expression is sexual perversion in the romances), anarchic violence, disregard for authority and its world-ordering hierarchies, and the gross, boundary-breaking appetites that characterize the giant of romance.’19 The particular manifestation of the grossness of Harpin’s appetites that we see in Le ­Chevalier au lion

18 19

Cf., e.g., Cohen, Of Giants, p. 175. Ibid., p. 77.

152



Ascopard’s Betrayal

we see again in Bevis: like Harpin, a suitor in Bevis seeks a woman (Josian) in marriage and, when he is denied, threatens to capture her by force, rape her and then give her to his lowest-ranking servant or servants for further sexual degradation. But in Bevis the threat is not a giant, but the Saracen King Brademond. Saracens too are stereotypically associated with sexual excess by Christian writers. John Tolan cites the thirteenth-century English writer Roger Bacon: ‘The Saracens, under the influence of Venus, “are plunged in the passions of lust”.’ Tolan goes on to say: ‘The association between the Saracen religion and the cult of Venus is evident elsewhere (in Petrus Alfonsi and other writers); more generally, the portrayal of Saracen libidinousness is commonplace.’20 As a Saracen giant, Ascopard is doubly determined: the stereotype of sexual predator is readily available to characterize him. And yet (though Cohen would be the last to suggest that his description of Harpin fits all the giants he examines, from Grendel to Galehaut) surely what is most noticeable about the ‘gross, boundary-breaking appetites that characterize the giant of romance’ is the degree to which the characterization does not match Ascopard. We never see him eat or drink; he has plenty of regard for hierarchy, never seeking to overturn his role as a servant, but only moving from one service to another; he does not disregard authority, except perhaps that of the bishop of Cologne: his violence is all in the service of restoration of order (from one or another point of view); and, most strikingly, particularly given his status as a Saracen, he never shows sexual appetite. Ascopard’s lack of interest in sexual matters is unlikely to be insignificant in a romance as much concerned with the regulation of sexuality as Bevis. Female sexual integrity is extravagantly prized in the romance: Josian cannot marry her beloved unless she is a virgin because the Patriarch of Jerusalem commands Bevis he must only marry one. The lions cannot harm her because she is one. She has protected herself from the consummation of her seven-year-long marriage to Yvor by magic. Bevis himself fights a loathsome dragon and is healed of the fleshly corruption it inflicts on him in a well sanctified by a virgin’s bathing – this aspect of the story, enhancing its emphasis on virginity, is in the Middle English versions only.21 Josian pleads womanly shamefastness in order to get rid of unwanted witnesses in her marriage chamber when she is forced to marry Miles, and then she kills him. It is genuine shamefastness that causes Josian to send away Bevis and Terri and leave herself unprotected as she goes into labour; and it is by a pretence of such shamefastness that she tricks Ascopard, in one of A’s major additions to the story. A’s added episode makes Ascopard’s lack of sexual involvement in other

20

John V. Tolan, Saracens: Islam in the Medieval European Imagination (New York: Columbia University Press, 2002), p. 226. His citation from Roger Bacon is from the Opus maius pars septima seu moralis philosophia, ed. by Eugenio Massa, Bibliotheca scriptorum Latinorum mediae et recentioris aetatis (Turin: In Ædibus Thesauri Mundi, 1953), pp. 189–90. 21 Cf. p. 83 above.

153



Melissa Furrow

versions of the story into a virtue (or at least the pronounced absence of a vice) in this one. The episode itself is sexually highly charged. Josian has just given birth to twins, and has done so after making her declaration that having a man present at all, even her own husband, would represent a violation of sexual decency: God for-bede for is pite, Þat no wimman is priuite To noman þourȝ me be kouþe.  (p. 171)

A heightens her degradation by having her captured and beaten by forty Saracens: Þai made hire wiþ hem to go, And gret scorning of hire þai maked And bete hire wiþ swerdes naked.  (pp. 171–2)

Here, Josian asks Ascopard to remember that she once saved his life and persuaded Bevis not to kill him: Þar fore i praie, on me þe rewe And ȝeue me space a lite wiȝt, For wende out of þis folkes siȝt, To do me nedes in priuite, For kende hit is, wimman te be Schamfaste and ful of corteisie, & hate dedes of fileinie.  (p. 172)

Ascopard grants her request to preserve her womanly shamefastness. Once concealed by the bushes, she picks out herbs that will make her appear leprous and then returns to her captors. They lead her on to King Yvor, who does not recognize her at first because of the herbs’ effects. Leprosy implicitly is read by Yvor as a venereal disease: Josian keeps herself from sexual involvement by feigning sexual contamination. Disgusted with both her and Ascopard, Yvor sends them away to a nearby castle where she will eventually be found by Saber and his band of rescuers disguised as pilgrims. In several ways the Auchinleck reviser has picked up traits found elsewhere in the story and used them to make it more coherent. As in the episode of Miles, Ascopard shows himself to be too trusting for his own good; while Josian a second time exploits her own recurrent and real anxiety about sexual exposure. As far as our reading of Ascopard is concerned, the episode has two effects: it establishes him as remembering his obligation to Josian for her protection when he was captured by Bevis, and as being in turn protective of her now, thus mitigating the moral horror of his betrayal; but, significantly, it also reinforces his stand on the side of sexual innocence: he is the Saracen giant 154



Ascopard’s Betrayal

who protects the heroine from sexual humiliation rather than threatening to impose it. This is a romance in which the hero grows from little boy to fully fledged chivalric manhood, regaining his land and identity, marrying, and founding a family. As Romaine Wolf-Bonvin has shrewdly pointed out about the AngloNorman poem, ‘Humilié chez lui, contraint tout petit à l’exil, Beuve ne se retrouvait-il pas au service du roi Hermine dont il devenait le champion? En ce sens, la trajectoire du colosse redouble celle du héros.’22 In Bevis too Ascopard is one of the hero’s foils. But, paradoxically, the giant is a boy who never grows up. Like Bevis, Ascopard is an exile as a child – from a place unspecified in Bevis, from among the Escopartes of Egypt in Boeve. He finds service with Yvori in Boeve, with Garcy in Bevis, then with Bevis himself. His enormous size makes him useful, but the work he does is that of a page or knave. Both terms were identifiers for a male child and for a menial domestic servant. Pages (or knaves) fetched and carried in great households, but the position was ideally a temporary one on the way to higher stages of chivalric identity. Geoffrey Chaucer’s wealthy merchant father scored a success in getting the boy Geoffrey placed as a page in a royal household, that of Prince Lionel; from there Geoffrey became a squire and went on to military service in France, before graduating into a distinguished civil service career. ­Chaucer’s doubly fictional Philostrate (the knight Arcite in the Knight’s Tale, in disguise as an unknown outsider in Athens) graduates from being a page in Emelye’s chamber to being a squire in Theseus’ service. As page ‘Wel koude he hewen wode, and water bere’; 23 in this labour he manifests that he is of ‘so gentil a condicioun’ that everyone at court agrees:         it were a charitee That Theseus wolde enhauncen his degree, And putten hym in worshipful servyse, Ther as he myghte his vertu exercise.24

But there is no enhancing of the Middle English Ascopard’s degree. Until the end of his service to Bevis, he is a page, a boy. The work Ascopard does is non-chivalric. His life is spared at Josian’s urging so that he can become their page, taking over service from the late Bonefas, freshly devoured by lions. But he does not have Bonefas’s status – as Sir Bonefas and Josian’s chamberlain.25 Ascopard acts as a prodigious labourer, carrying horse, mule and people aboard ship, he sails a ship to Cologne, he fights with a club on Bevis’s and Josian’s behalf, and his crowning achievement is to seize Bevis’s stepfather, horse and all, and carry them off for Bevis. But he never 22 23

Wolf-Bonvin, ‘Escopart, le géant dépérissant de Beuve de Hantone’, p. 258. The Canterbury Tales, i.1422 (The Riverside Chaucer, ed. by Larry D. Benson (Boston, MA: Houghton Mifflin, 1987), p. 44). 24 Ibid., i.1433–6 (pp. 44–5). 25 BHK, pp. 111, 112, 115, 54.

155



Melissa Furrow

undergoes the chivalric development that Cohen traces in the comic Saracen Rainoart of Aliscans.26 Rainoart is serving as scullion in the kitchen of the Emperor Louis. Cohen argues: The giant’s violence, strength, and appetite are the very traits that distinguish him as a site of raw potential to the visiting William. Aroused by the young man’s fair figure beneath his besmirched skin, William decides to indoctrinate Rainoart into the ways of chivalry by teaching him how to channel his drives and prowess toward a set of goals. The process of the giant’s education under William’s tutelage creates a narrative space in which an underlying ideology of embodiment moves into the open through an extended explication. William and the other Christian knights gently instruct Rainoart in the care of the self, enjoining him to impose demarcating limits upon his body, allowing him to produce himself as chivalric subject.27

In short, he is trained to become a knight, and Rainoart au tinel (Rainoart with the yoke) learns to control his yoke as a club rather than simply smash with it, tries riding a horse, then learns to wield a sword instead of a club, and ends Aliscans rewarded with lands, married to the emperor’s daughter and, when she dies in childbirth, nevertheless a father and a founder of a noble line. All this is striking by comparison to Ascopard, whom nobody educates. Ascopard fights with a club, first against Bevis, then against the mariners, and then in the climactic battle against the emperor. He never achieves chivalric skill: his greatest service involves picking up the emperor and his horse and carrying them into Bevis’s castle. He does not have strategic skills, as he demonstrates when he falls for the stratagem of Earl Miles and hurries to an island where he can be imprisoned because, in his eagerness to please, he obeys a false message purporting to be from Bevis. Not surprisingly, though he is put at the head of one-third of the army in the battle against the emperor, he is not really in charge: Bevis tells him what to do in terms that emphasize their intellectual inequality and Ascopard’s assumed motivation (money, not chivalric glory): Ascopard, take riȝt gode hede: Þemperur rit on a whit stede; Þin hure i schel þe ȝilde wel, Wiþ þat þow bringe him to me castel! (p. 160)

As Ruth Mazo Karras has argued, becoming a man for members of the knightly classes in the late Middle Ages involved gaining land: ‘For knights, marriage might not be possible until a man had inherited his patrimony or 26

Cohen, Of Giants, p. 167, calls Aliscans ‘the foremost poem of the Guillaume d’Orange cycle and the favorite chanson de geste of the twelfth century’, but I do not know on what grounds. 27 Ibid., pp. 168–9.

156



Ascopard’s Betrayal

had acquired a substantial landholding in some other way, but it was the inheritance rather than the marriage that signified his coming of age.’28 As for the ‘love of ladies’, it was ‘a necessary part of aristocratic masculinity ... Knights felt compelled to demonstrate to other men not their “sexual orientation”, which was not a medieval concept, but their superiority to other men in the competition for women.’29 Like Bevis an exiled child, Ascopard has no chance to inherit a patrimony, and he is altogether outside the competition for women. He leaves Bevis’s service once the latter has established his own fully adult masculinity and is replaced by Terri, who becomes Bevis’s ‘swain’, or in NS his ‘squyer’ (p. 169) instead of his page. The second part of the romance unfolds with the hero’s companion being someone on the track to knighthood, a European Christian of normal size who can marry the second-string bride and take over the territory won with her. When Bevis proposes to Terri that they take part in the tournament for the hand and land of the lord of Aumbeforce’s daughter, Terri’s response is the appropriately mature chivalric one: ‘Wile we tornaie for þat leuedy?’ ‘Ȝe, sire,’ a seide, ‘be sein Thomas of Ynde! Whan wer we woned be by-hinde? We scholle lete for non nede, Þat we ne scholle manliche forþ vs bede.’  (p. 177)

Terri’s hearty manliness here is in sharp contrast to the earlier comic moment added in Bevis when Ascopard is invited by Bevis to help him fight the dragon: at first agreeing ‘Bleþeliche’ (p. 127), Ascopard soon backs down, despite his great size and strength unwilling to approach the dragon once he hears its roar. In A, he even whines: ‘Icham weri, ich most haue reste; / Go now forþ and do þe beste!’ (p. 128). The Middle English Bevis, and particularly A, emphasizes Ascopard’s unmanliness, making him comically childish. Our first introduction to him plays on the paradox of his size: a giant compared to Bevis, he is ‘lite’ and ‘meruȝ’ (‘small’ and ‘delicate’, p. 119), in his own race, driven away from his home town by bullying and name-calling.30 Even in Boeve, Escopart leaves home because he cannot bear the taunting of those who call him dwarf. His likeness in this to the child Boeve goes only so far. Boeve is insulted and taunted by the porter at Hampton castle: ‘Fuez de ci, fiz a putein, ribaud, vistement! / Mult estes petit e si estes fort truaunt’ (‘Get out of here fast, son of a whore, scum! You are very small and thus a great good-for-nothing’, 275–6), but his reaction is not, like Escopart’s, to run away from home;

28

Ruth Mazo Karras, From Boys to Men: Formations of Masculinity in Late Medieval Europe (Philadelphia, PA: University of Pennsylvania Press, 2003), pp. 144–5. 29 Ibid., p. 51. 30 Only P omits the story of his comparative delicacy.

157



Melissa Furrow

instead, he clubs the porter to death.31 Little Escopart is less manly than little Boeve. But the later Escopart is not systematically developed as ­Ascopard is towards childishness. True, Escopart fights with a club, as Ascopard does, and as the seven-year-old Bevis does against the porter and his father’s killer, and Escopart is several times called a garsoun (1892, 1923, 1924, 1937), equivalent to Middle English ‘page’ or ‘knave’; nevertheless, he is also described as carrying ‘a son geron un bon branc trenchant’ (‘by his side a good sharp sword’, 1748) – the weapon of a knight, not a boy. He is invited by Josiane to become Boeve’s home (‘man’, 1823), rather than his ‘knave’, as in the Middle English texts. True, Escopart is, like Ascopard, outside the possibility of sexual involvement. Boeve trusts him to protect Josiane against the suitors that she is worried about attracting in Boeve’s absence;32 there is never any whiff of suggestion that he himself could be a sexual danger. Wolf-Bonvin calls him ‘un chien de garde, lâché contre tous ceux qui tentent de ravir la femme’.33 But in Boeve there is less explicit emphasis on shamefastness than in the Middle English Bevis in general (it is not mentioned in the Miles episode), and particularly than in A (which alone includes the postpartum herb-gathering episode). Escopart cries out to have his life saved; A’s ­Ascopard makes childish excuses to mask his fear. The more boyish the giant, the less he can be held accountable for his actions, and the less shocking is his moral failure. The English texts are not uniform in their treatment of Ascopard or in their strategies to mend the narrative split in his character. From the betrayal onwards, P is peppered with terms of denigration for him (‘that false thefe’, pp. 169, 180; ‘On treason thought the thefe aye’, p. 172) and with reminders of his reversion to being Saracen (‘by Tarmagant’, p. 170; ‘he sware by Tarmagant’, p. 172; ‘Iesu Christ [b]e hyr warrant’, p. 181). We have already noted that even in the introduction of Ascopard, P brings in more description of ugly features. In that description P also suggests that he is ‘lyker a deuyl than a man’ (p. 119) – a comparison like that voiced by the wicked emperor, and therefore thoroughly undercut, in the other English versions: Erþliche man semeþ he nouȝt, Ne noman of flesch ne felle, Boute a fend stolen out of helle.  (p. 155)

And P also suppresses the potentially touching details of Ascopard’s rejection in his home country that are present in all other versions, Anglo-Norman as well as Middle English. In short, P is unlike other Middle English versions

31

Of the Middle English versions, only NS (403) preserves the combination of the porter’s other insults with his remarking that Bevis is ‘lyte’. The hero’s parallel with Escopart is thus here clearer in the Anglo-Norman version than in most of the Middle English ones. 32 Boeve 1984–90. 33 Wolf-Bonvin, ‘Escopart, le géant dépérissant de Beuve de Hantone’, p. 260.

158



Ascopard’s Betrayal

in moving Ascopard more towards a stock stereotype of the wicked Saracen giant that would make his betrayal no surprise. The other English texts are more sparing of abuse after the betrayal. Saber’s wife reveals Ascopard’s treason by interpreting Saber’s dream, and in E calls him ‘þat fals traytour’ (p. 180); in M he is ‘that fals man’ (p. 182), and in E again he is called ‘þat wykkyd traytour’ (p. 181). But the other versions attribute Bevis’s loss of wife or child to Ascopard and his guile or wrong-doing without abusive language. Where A blames Ascopard for treason, it does so from the interested perspective of Bevis’s new swain, Terri, whose sympathetic, attentive behaviour and bitter cursing of Ascopard together suggest an argument that Bevis was right to have changed followers, despite the lack at the time of any apparent reason to do so: Beues fel þar doun and swouȝ; Terri wep and him up drouȝ, And koursede biter þat while Ascopard is tresoun & is gile.34 (p. 173)

Not until his end, when sympathy has to be engaged for Saber as Josian’s champion, does A remind the reader that Ascopard is Saracen: ‘& be Mahoun a swor his oþ / To deþe a scholde Saber diȝte’ (p. 181). One could thus arrange the English texts on a continuum, according to their degree of sympathetic engagement with or denigration of Ascopart, from A to P: A → NSC→ M→ E→ P.  A, and to a lesser degree NSC, attempt to bridge the gap between the comic, ungainly servant and his treachery by a series of mitigations of his guilt; P instead reinforces his inherent wickedness. In sum, the character of Escopart demands a readjustment in the light of the clash produced by the endearing qualities invented in the shorter AngloNorman version (his parallels to Boeve, his eagerness to please, his comic incongruities, his conversion to Christianity, his defeat of the most hated and unprincipled of Boeve’s enemies), and the viciousness of the betrayal introduced by the continuator (his ingratitude, his apostasy, his failure to renounce his fealty to Boeve openly, his engineering of the brutal kidnapping of his protector Josiane just after she has given birth, and his willingness to curry favour with Yvori by returning Josiane to his control). The Middle English versions work to reduce that dissonance by various narrative strategies, including making Ascopard a comically oversized boy. At the same time he is less than a man and so cannot be held fully responsible for his moral decisions; and, stuck in a perpetual childishness, a pre-chivalric state, he cannot be carried forward as Bevis’s companion into fully adult life, where the romance 34

NSC does not have any version of the last two lines; E has Bevis realize that ‘Ascopard hadde don hym gyle’, and M has both Bevis and Terri fulminating: ‘And sore they warryed the while, / Of Ascopard and of his gile.’

159



Melissa Furrow

hero needs a friend who can participate in the winning and giving of women. The plot demands Ascopard’s death – despite his odd childish innocence, indeed because of it – and the story continues with the pallid and unmemor­ able Terri as his much more conventional replacement.

160

10 Gender, Virtue and Wisdom in Sir Bevis of Hampton CORINNE SAUNDERS Sir Bevis of Hampton looks in two directions: it is rooted in convention and nostalgia for the past and its ideals, yet it also offers new, often original, perspectives on and treatments of romance materials. It translates and adapts the twelfth-century Anglo-Norman Boeve de Haumtone but is also firmly situated within the tradition of English romance: one of the group of celebrated romances in the Auchinleck MS, it was widely disseminated and remained popular into the eighteenth century. While Bevis reflects a growing nationalism in England, it has a more universal appeal, as is evident in its popularity on the Continent. The oppositions of Bevis are powerful ones: good and evil, Christian and Saracen, human and monster, chastity and lust, consent and force. The narrative is structured around archetypal romance motifs: the hero’s exile and return, employed twice; the conflict between Christian and pagan; the challenge of love, as Bevis overcomes obstacles to his union with the pagan princess Josian, defeating a series of rivals who threaten enforced marriage; and the performance of a sequence of heroic feats, most memorably Bevis’s battles against enemies of all kinds: the emperor of Almaine, Saracens, the people of London, the giant Ascopard, and a series of wild beasts – boar, lions and dragon. Bevis’s identity is intimately bound up with all these chivalric achievements, as well as with regaining his lands and later protecting them from the treacherous English King Edgar for his chosen heirs, while he rules the kingdom of Mombraunt with his wife Josian. This romance, then, seems founded on chivalric action and on the attendant gender stereotypes. Yet Bevis not only depends upon but also ­ reinvents the conventional motifs of romance, including those intimately associated with ideas of gender – the knight-hero, the courtly lady, love, marriage and the family. Its hero is, indeed, defined by a series of colourful adventures: the narrative can seem episodic rather than sophisticated, robust rather than ­intellectual. Derek Brewer places Bevis as ‘folk-hero’ rather than as chivalric 

This essay focuses on the Auchinleck version (A); all references are by page number to The Romance of Sir Beues of Hamtoun …, ed. by Eugen Kölbing, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) (hereafter BHK). While recognizing the crucial relation between Anglo-Norman and English versions of Bevis, I focus here on the nuances of the Middle English romance.

161



Corinne Saunders

knight. Yet, like Guy of Warwick, Bevis is also marked by Christian didacticism and an extended engagement with the spiritual. As Susan Crane notes, the English Bevis places new emphasis on Christianity, so that ‘the hero himself gains the Christian awareness and solemnity of an ideal Crusader knight, “þe kniȝt of cristene lawe” ’. As I have discussed elsewhere, the romance is dominated by the idea of riȝt – the right of heritage, lost and regained by Bevis, the right of marriage, the giving and taking of right counsel, right chivalric behaviour, and right Christian belief. Bevis’s battles for Christian virtue and social order overlap with his pursuit of love and his defence of his lady, and the private and public aspects of chivalry mesh as Bevis comes to represent ‘an ideal of personal fulfilment and social service’. Thus to some extent, despite the active emphasis of the narrative, the generic conventions of romance interweave in Bevis with those of hagiography and moral tale, and in this interweaving gender stereotypes and assumptions are complicated. Masculinity and heroism are placed in the context of service to God, and the testing of Bevis’s surpassing valour merges with his miraculous preservation in the face of insurmountable odds – and the accompanying illumination of his inner, spiritual virtue. In the portrayal of Josian too, gender stereotypes are complicated. Josian both is and is not the romance lady familiar from French romance: while she is defined by courtly virtues and qualities, including her appearance, she fits the more active ‘wooing woman’ type found in Anglo-Norman romance and chanson de geste. She also recalls the protagonists of legends of holy women, empowered through virginity and chastity. Josian’s Christian virtue, however, is interwoven with her ‘otherness’, her Saracen origin, which plays an important role in rendering acceptable her remarkable powers of medicine, healing and protection, as well as her proactiveness in wooing and in preserving her chastity. She is further elaborated through her difference from Bevis’s mother. Throughout the work, gender roles are nuanced through their 

 

 

Derek Brewer, ‘The popular English metrical romances’, in A Companion to Romance: From Classical to Contemporary, ed. by Corinne Saunders, Blackwell Companions to Literature and Culture (Oxford; Malden, MA: Blackwell, 2004), p. 48. Judith Weiss, ‘Anglo-Norman romances’, ibid., pp. 33, 34, characterizes Boeve as ‘comic, vigorous, and even crude’, and writes of Boeve himself: ‘though described as le curteis guerrer, [he] deserves only the last word: arrogant, stupid, rude, and rash, his nobility and dignity are too often abrogated in favour of vigorous, even brutal words and acts, whose details destroy idealism’. Susan Crane, Insular Romance: Politics, Faith, and Culture in Anglo-Norman and Middle English Literature (Berkeley, CA: University of California Press, 1986), p. 105. Cf. pp. 143–4 above. This essay engages with and develops material in Corinne Saunders, ‘Desire, will and intention in Sir Beues of Hamtoun’, in The Matter of Identity in Medieval Romance, ed. by Phillipa Hardman (Cambridge: Brewer, 2002), pp. 29–42. The progression of Bevis, from taking evil counsel to receiving and giving good counsel, is also traced in Geraldine Barnes, Counsel and Strategy in Middle English Romance (Cambridge: Brewer, 1993), pp. 81–90. W. R. J. Barron, English Medieval Romance, Longman Literature in English Series (London; New York: Longman, 1987), p. 85. See further Judith Weiss, ‘The wooing woman in Anglo-Norman romance’, in Romance in Medieval England, ed. by Maldwyn Mills, Jennifer Fellows and Carol M. Meale (Cambridge: Brewer, 1991), pp. 149–61.

162



Gender, Virtue and Wisdom in Bevis

intersections with a complicated network of cultural and literary influences. The portrayals of Bevis and Josian depend upon the elaboration of their individual virtue and wisdom as much as upon the expected romance roles of knight and lady in situations of love and war. The opening sequence of Bevis, which leads to the hero’s exile, establishes a number of crucial patterns, but in particular the opposition between positive and negative gender models, as well as the more general opposition between good and evil. Bevis’s mother procures the death of her elderly husband, Earl Guy, at the hands of her lover and, on being challenged by Bevis, strikes him and orders that he be killed, in egregiously callous terms: Let sle me ȝonge sone Bef, Þat is so bold! Let him an-hange swiþe hiȝe, I ne reche, what deþ he diȝe, Siþþe he be cold!  (p. 15)

When Bevis returns to attack his stepfather, death is replaced by exile: she has Bevis sold ‘in to heþenesse’ (p. 23), a form of living death. The ­supposedly Christian princess is rapidly revealed to be evil, betraying both husband and son, and perverting the ideals of chaste wife and loving, nurturing mother that are so central to romance, and more generally to Western cultural mores and understandings of gender. This exceptionally negative model will be opposed by Josian, the pagan princess, who will prove to be active in her virtue, choosing and living out the laws of Christianity. Adultery is relatively rare in romance: when it does occur, it tends to be associated with tragedy, as in the instances of Iseult and Guenevere. Marie de France treats the theme poignantly in her sequence of lais of malmariées; Malory’s Morte Darthur devastatingly shows the threat posed by adultery to social order. Lusty, adulterous women who escape tragedy belong to the low-life world of the fabliau more than to that of romance. If the adulterous woman is unusual, the wicked mother is still more so. The folk motif of the wicked stepmother, by contrast, is widespread. The pattern of betrayal and exile of children also recurs in romance. Frequently, the child is exiled with the parent, who is the victim of false accusation, as in the Constance story, Octavian and The Tempest. The Winter’s Tale varies this motif in depicting the exile of the child alone. As Ivana Djordjević notes of Boeve/Bevis, ‘Exileand-return tales tend to rely on a conventional usurper figure, typically an



Ivana Djordjević, ‘Original and translation: Bevis’s mother in Anglo-Norman and Middle English’, in Cultural Encounters in the Romance of Medieval England, ed. by Corinne Saunders (Cambridge: Brewer, 2005), pp. 11–26, discusses in highly illuminating detail the differences between the Anglo-Norman and Middle English in this episode.  It occurs, e.g., in William of Palerne, in Gower’s Tale of Florent and in Floris and Blanchefleur, where Floris’s mother orchestrates the exile of her son’s beloved Blanchefleur to protect him from marriage to a Christian.

163



Corinne Saunders

outsider, to get the plot going. … The evil mother is quite an oddity.’ The blood relation between mothers and children, indeed, seems sacrosanct in romance. The action of Bevis’s mother is triply shocking as she takes a lover, plots the death of her husband and casts off her child. Jennifer Fellows argues: ‘It is in Bevis that the mother is seen most explicitly as a threat to patri­ archal values’, although she notes the ‘darker and more threatening aspects of maternity’ hinted at in Sir Degaré, with its suggestion of incest, and in Sir Perceval de Galles and Lybeaus Desconus, which illuminate the dangers of maternal over-protectiveness.10 Bevis’s mother, however, is much more extreme: she is the antithesis of the idealized wife and mother of romance, defined instead by adultery and the firm design to murder her own child. She fits more readily with the examples of clerical antifeminist writings, in which the notion of female promiscuity was fuelled by the role assigned to Eve in the Fall. 11 The exploits of Bevis’s mother would fit well among the stories of the wicked wives who figure in the book owned by the Wife of Bath’s sixth husband, Jankin. In such a misogynist collection of cautionary tales, she might be set alongside Clytemnestra, adulteress and murderess of her husband, or Medea, whose revenge against Jason took the form of killing her children. The combination of adultery, murder of husband and attempt to kill the child, however, surpasses even these examples, and is certainly not typical of romance. Yet the portrayal of Bevis’s mother is also complicated by a realism that renders her more than an extreme antifeminist stereotype. The narrative carefully creates a sense of motivation, by revealing that before her marriage to Sir Guy, she was ‘loued paramur’ (p. 3) by the emperor of Almayne, who repeatedly asked her father, the king of Scotland, to consent to their marriage. Much is made too of the age of Sir Guy, despite his stalwart heroism in defending himself.12 The unhappy outcome of unequal marriages, particularly of young women to old husbands, is a recurrent subject of romance.13 Marie de France’s lais of malmariées, Guigemar, Yonec, Laüstic, Milun and Chevrefoil, all depict unhappy young wives married to jealous, usually older, husbands – and to an extent Iseult also fits this pattern. In Bevis too, the young woman cannot be controlled by her older husband, and is led by her sexual desire and promiscuity to betray him, but in this case her intentions and the consequences are far darker. Although the portrait of the countess is  10

Djordjević, ‘Original and translation’, p. 14. Jennifer Fellows, ‘Mothers in Middle English romance’, in Women and Literature in Britain, 1150–1500, ed. by Carol M. Meale, Cambridge Studies in Medieval Literature 17 (Cambridge: Cambridge University Press, 1993), pp. 41–60 (pp. 52, 51). 11 Fellows, ibid., p. 53, characterizes the episode as one of ‘gratuitous misogyny’. 12 BHK, p. 4. 13 Djordjević, ‘Original and translation’, pp. 17–24, offers an interesting discussion of the pro­cesses by which Sir Guy and the countess become the literary types of the senex amans and the malmariée, contrasting this treatment to the emphasis in Boeve on Gui’s need to marry in order to gain an heir and on his wife’s disloyalty.

164



Gender, Virtue and Wisdom in Bevis

certainly not sympathetic, the context of unequal marriage does offer some explanation of her evil. It is against this extraordinarily negative paradigm of unhappy marriage, then, that the marital and maternal ideal of Bevis is constructed: Josian, the actively wooing woman, chooses her own husband and is active in his preservation. The resonant motif of enforced marriage recurs across the narrative, pointing up Josian’s active chastity and her faithfulness to Bevis. Bevis too is defined through opposition to the evil deeds of his mother: he lives out and perfects the chivalric model of his father. Sir Guy’s loyalty, virtue and chivalry are elaborated, the injustice of his death establishing the grounds for Bevis’s later mission of revenge. Though only seven years of age at the start of the romance, Bevis is already a chivalric figure, who charges his mother with both adultery and murder.14 Saved by Saber, Bevis refuses to remain in disguise as a shepherd, returning to kill the porter and challenge the emperor, striking him unconscious. In retaliation, his mother orders her knights to sell the child to Saracen merchants from ‘heþenesse’ (p. 23). The initial opposition between good and evil is replaced by that between pagan and Christian, as Bevis finds himself in the Saracen kingdom of Ermonie. Carol F. Heffernan argues persuasively that his sojourn on the Continent engages creatively with the idea of the Crusades in terms both of warfare and of economic and cultural exchange. The Saracen is both ‘infidel enemy in battle’ and ‘trader in goods, some of which were Christian slaves’.15 The landscape of the East in the period of the Crusades is realistically evoked: trade and culture intersect with religious warfare and individual pilgrimage. Later, as Heffernan observes, when Bevis, disguised as a palmer, is asked ‘Whar is pes and whar is werre’ (p. 110), he lists various Crusading territories; he also visits Jerusalem in order to consult the Patriarch.16 In this Saracen world, Bevis follows the pattern of the Fair Unknown, loved despite his Christianity by the pagan King Ermin, appointed his chamberlain and promised the office of ensign-bearer after being knighted. Foremost in chivalric prowess, Bevis’s motto might be the lines spoken much later to Terri, as they set out to tourney for a king’s daughter: ‘We scholle lete for non nede, / Þat we ne scholle manliche forþ vs bede!’ (p. 177). The making of Bevis’s masculine chivalric identity, however, is intimately bound up with his assertion of Christianity and defence of virtue. There is no suggestion that Bevis himself is engaged in the mission to win back the Holy Land, but his battles, often undertaken on behalf of Josian, are repeatedly characterized in terms of the opposition of Christian and Saracen. He proves his honour and Christian identity, first by refusing to convert despite the offer of Josian’s hand; next by defeating the Saracens who taunt him with his failure 14 15

BHK, p. 14. Carol F. Heffernan, The Orient in Chaucer and Medieval Romance, Studies in Medieval Romance (Cambridge: Brewer, 2003), p. 15. 16 BHK, pp. 110, 97–8.

165



Corinne Saunders

to remember that it is Christmas Day. Here, the patterning of Bevis on his father is made explicit:    were ich alse stiþ in plas, Ase euer Gii, me fader, was, Ich wolde for me lordes loue, Þat sit hiȝ in heuene aboue, Fiȝte wiþ ȝow euerichon.  (p. 29)

Bevis, though weaponless, succeeds in taking a sword and killing all his attackers: his strength promises to surpass even that of Guy, so that the son will perfect the father. Successive battles against the pursuers of Josian (King Brademond, King Yvor and his allies) re-enact the victory of Christian over pagan, and establish Bevis as a ‘kniȝt of cristene lawe’ (p. 198). Bevis eventually urges the defeated King Yvor to renounce his gods and embrace Christianity in order to avoid the pains of Hell; when Yvor refuses, Bevis beheads him amid general Christian rejoicing, and the Saracens flee, cut down from behind.17 Bevis’s role as Christian avenger is still more overt when, in Damascus, he kills a priest who has been leading a crowd of Saracens in sacrifice to ‘Mahoun’.18 His actions are violent, but they also echo Christ’s anger at the money-changers in the Temple; as Heffernan notes, ‘The matter-of-fact tone of the narrator’s description of this massacre suggests that he views Beves as merely doing the work of a crusader knight.’19 Eventually, at Ermin’s death, Bevis and his son Guy cause the whole of Armenia to be christened, and Bevis acquires the Saracen kingdom of Mombraunt.20 Bevis’s role as Christian knight in the world of the East is balanced by his two returns to England in order to regain and then protect his lands. Even the ‘wonder-gret cheualrie’ (p. 154) of the king of Scotland alongside the emperor of Almayne cannot defeat the might of the army led by Bevis, Saber and Ascopard. As is typical of this narrative, extreme violence is justifiable on the side of riȝt: Sir Guy’s death is redressed as the emperor meets a hideous death in boiling pitch, brimstone and lead. The narrative avoids the unnaturalness of taking revenge on the mother, however: in her grief she falls and breaks her neck. Bevis rejoices but also states his innocence, ‘Damme, for-ȝeue me þis gilt, / I ne ȝaf þe noþer dent ne pilt!’ (p. 162): her death seems an act of divine retribution. The emphasis of Bevis’s battles against the Saracens is sustained in the reference to the ‘lawe & riȝt vsage’ (p. 163) of Bevis’s succession and to the thanks he offers to God for his vengeance. The focus shifts from familial to national when Bevis, who has been exiled when King Edgar’s son is killed in attempting to steal Arundel, returns to confront Edgar and preserve his lands when they are withheld from his 17 18 19 20

Ibid., p. 199. Ibid., p. 68. Heffernan, The Orient in Chaucer and Medieval Romance, p. 16. BHK, p. 200.

166



Gender, Virtue and Wisdom in Bevis

chosen heirs by the king. This second return becomes, as Geraldine Barnes notes, a social and moral mission, ‘to save England … from the internal threat of royal tyranny’.21 In this case the extreme violence of the battle in London, which causes the Thames to run red with blood, is followed by concord between Bevis and the king. With the marriage of Bevis’s son Miles to Edgar’s daughter, the stability of the English Crown and the justice of rule are assured, and Bevis may return to Mombraunt with his wife. The active pattern of Bevis’s masculinity, his military defence of Christi­ anity and justice, is complemented by an emphasis on spiritual devotion and prayer in need. His survival in impossible situations asserts the power of God against the heathens, and places him as recipient of God’s grace, affirming his virtue in that he merits divine assistance. Thus, close to starving after deceiving his guards and killing his gaoler, when imprisoned by King ­Brade­mond, Bevis is miraculously freed: ‘So ȝerne he gan to Iesu speke, / Þat his vetres gonne breke’ (p. 85). In his battle with the pagan King Grander, sent after him by Brademond, Grander’s shield ‘fleiȝ awei þourȝ help of Crist’ (p. 190). A further miracle occurs when, after praying to Jesus, he swims across the entire sea on Grander’s horse, Trenchefis, to escape ­Brademond’s pursuing army.22 In Bevis’s final battle against King Yvor, the opposition between Christian and pagan gods is made especially clear: when Bevis strikes Yvor’s left hand, ‘Hit fleȝ awei þourȝ help of Crist’ (p. 198), and ‘þourȝ þe miȝt of heuene king’ (ibid.) he cuts off Yvor’s right arm and shoulder. By contrast, Yvor’s prayers to his gods to rescue him do not avail, and his assertion that Islam is ‘wel þe beter lawe’ (ibid.) is shown to be hollow. In Bevis’s English battles too the role of God is emphasized:23 he is God’s agent, fighting on the side of right, and he is victorious only through God’s ‘so gode sokour’ (p. 212), so that violence is repeatedly represented as justifiable. Interwoven with Bevis’s battles against the Saracens and for justice in England and the East are his fights with the great boar and the dragon. Yet although these seem more secular in their focus on individual chivalric excellence, they too emphasize Bevis’s identity as Christian knight. The chivalric action of battle against the monsters is vividly detailed, and Bevis’s prowess dramatically realized, but spiritual excellence and divine intervention are crucial to Bevis’s success. Bevis’s battle against the boar, elaborated by the English romancer, appears to be just the kind of lone, apparently impossible, challenge fundamental to the shaping of the individual knight’s identity in chivalric romance. Like Chrétien’s Yvain, who sets out secretly to meet the challenger of the magical spring, Bevis undertakes the adventure of the boar alone, in order to test his prowess:

21 22 23

Barnes, Counsel and Strategy, p. 82. BHK, pp. 91–2. Ibid., p. 207.

167



Corinne Saunders [A] þouȝte, a wolde keþen is miȝt Vpon þat swin him self one, Þat noman scholde wiþ him gone. (p. 36)

Bevis succeeds, however, only after, exhausted and near death, he prays for divine assistance.24 This pattern of human frailty and the need for divine intervention echoes that of Bevis’s battles against his heathen enemies. Most striking is Bevis’s defeat of the dragon, an episode added by the English romancer, perhaps in imitation of other Middle English romances such as Sir Degarre and Guy of Warwick (the narrator refers to Guy expli­ citly, along with Lancelot and Wade).25 The monstrous, perverted quality of sin is literally written on the dragon, once a violent and tyrannical king, now punished with the pains of Hell, and it is appropriate that defeat of the dragon should also offer a moral lesson. Bevis is afforded rest in battle only when he drinks from the holy water of a well in which a virgin has bathed; the spring subsequently heals him when he falls into it covered with fatal, burning venom. The epic battle most of all demonstrates the miracu­ lous power of God, and the purity of Bevis, who undergoes a kind of ritual baptism in the holy well, which the dragon is afraid to approach. On hearing Bevis’s prayer to Christ and Mary, the creature flees and is wounded by Bevis. Once again, however, Bevis’s human frailty is emphasized: his body becomes that of a ‘foule mesel’ (p. 130) and his armour breaks; it is made clear that the venom would have killed him but for the miraculous water. Extraordinary strength remains necessary; even Bevis must strike a hundred blows to cut off the dragon’s head. Virtue, however, is even more crucial than prowess. The battle illustrates that God’s power surpasses even that of the best knight in ­Christendom and that chivalric excellence must be underpinned and enhanced by virtue and by recognition of the need for God’s help (voiced through prayer). The principle that might is right, that the victor in single combat enacts the will of God, is rewritten in Bevis: here the victory depends upon divine assistance, and the structures of romance are thus interwoven with those of hagiography and moral didactic narrative. The theme is reiterated in Bevis’s relation with the giant Ascopard, who orginally serves King Garcy, and fights to recover Josian for King Yvor. Defeated and made Bevis’s squire, Ascopard becomes both a comic and a dark counterpart to Bevis.26 He too is an exile, scorned by other giants for his smallness, but his service to Bevis is occasioned by force and he is not reli­ able. Though Ascopard agrees to become a Christian, he proves too large for the baptismal font and reverts to paganism. He flees from the dragon despite his boasts, and eventually chooses material reward over loyalty, betraying

24 25 26

Ibid., p. 39. Ibid., pp. 122–3. See p. 155 above.

168



Gender, Virtue and Wisdom in Bevis

Bevis in his second exile by abducting Josian for Yvor.27 Ascopard’s oath ‘be Mahoun’ (p. 181) fails to preserve him as Saber and his men cut him to pieces. Bevis’s positive role as servant of chivalry and Christianity, which will end in holy death, is illuminated by Ascopard’s perverted model of service, with its disastrous end. Bevis is defined less by prowess, and more by virtue and spiritual need, than we might expect of ideal masculinity in this romance of challenge and battle. Ideal femininity is more active. That Josian frequently transgresses conventional gender roles is in part consistent with her origins in AngloNorman romance: her active wooing of Bevis, writes Helen Cooper, is ‘true to form for Anglo-Norman heroines and their descendants’.28 The traditional power relation between knight and lady is reversed as Bevis becomes the desired object. Josian is startlingly assertive, defending the integrity of Bevis to her father against false accusations of treachery: Whan þe child, that is so bold, His owene tale haþ itolde, And þow wite þe soþ, apliȝt, Who haþ þe wrong, who haþ riȝt, Ȝef him his dom, þat he schel haue, Whaþer þow wilt him slen or saue! (p. 32)

Her articulate intervention on the side of justice is complemented by her explicitly physical desire for Bevis.29 It is Josian who demands Bevis’s love:     Beues, lemman, þin ore! Ichaue loued þe ful ȝore, Sikerli can i no rede, Boute þow me loue, icham dede, And boute þow wiþ me do þe wille. (p. 52) Her outrage when Bevis rejects her is swiftly replaced by her offer to convert to Christianity, which plays on the proverbial stereotype of the shortlived anger of women.30 As ‘wooing woman’, Josian actively rewrites the adulterous model established by Bevis’s mother. The purity of Josian’s desire 27

See the discussion of Ascopard in Jeffrey Jerome Cohen, Of Giants: Sex, Monsters, and the Middle Ages, Medieval Cultures 17 (Minneapolis, MN: University of Minnesota Press, 1999), pp. 172–8; and see further pp. 145–60 above . 28 Helen Cooper, The English Romance in Time: Transforming Motifs from Geoffrey of Monmouth to the Death of Shakespeare (Oxford: Oxford University Press, 2004), p. 247. In Boeve, as Weiss remarks (‘The wooing woman’, p. 153), Josiane plays a central role in the romance structure: ‘she ensures that the marriage will take place (in the first half of the romance) and that the family will survive, to found a dynasty (in the second)’. 29 Cf. BHK, pp. 37, 53. See further Corinne Saunders, ‘Erotic magic: the enchantress in Middle English romance’, in ‘Parfit blisse of love’: A Search for the Erotic in Medieval Britain, ed. by Amanda Hopkins and Cory Rushton (Cambridge: Brewer, 2007), pp. 38–52. 30 BHK, pp. 56–7.

169



Corinne Saunders

for Bevis is proved by her wish to legitimize her love in marriage and by her willingness to adopt Christianity. While Josian’s open wooing of Bevis is treated positively, the reversal of traditional gender roles not only provides comedy, but also contributes to the recurrent emphasis on Bevis’s chastity.31 Although his strength is not, like that of the Grail knights, dependent on his own virginity, he is repeatedly associated with the virginal and with purity, as in his dream before the dragon-fight that ‘a virgine / Him brouȝte out of al is pine’ (p. 126) and his healing by the water of the virgin’s well. While Bevis insists from the start on Josian’s Christianity, the patriarch of Jerusalem also instructs him that he must marry only a ‘clene maide’ (p. 98): Josian’s purity rewrites the faithlessness in marriage of Bevis’s mother. Bevis’s chivalry is repeatedly motivated by the need to protect Josian’s virginity. The link between virginity and marvellous preservation is reiterated when the two lions who attack them cannot harm ‘a kinges doughter that maid is’ (p. 115). When Josian offers to hold one lion, however, so that Bevis can defeat the other, he refuses: I myȝt ȝelp of lytel prys, There y had a lyon quelde, Þe while a woman a nother helde!  (p. 116)

Battle is explicitly and severely gendered: Josian’s protective power signifies her virtue, but Bevis must win the combat alone, through military means and ‘Þourȝ godes grace and is vertu’ (p. 118). Bevis’s rejection of Josian’s help becomes further proof of his virtue and prowess, while her virginal empowerment proves her excellence and her suitability as a match for Bevis. The theme of unnatural marriage so uneasily explored in relation to Bevis’s mother recurs in Josian’s repeated experiences of enforced marriage. Josian proves as active as Bevis’s mother in her increasingly dramatic responses to the threat of violation. She is first pursued by King Brademond, who ‘þouȝte winne wiþ meistrie / Iosiane, þat maide briȝt’ (p. 43): if he is not given Josian’s hand, he threatens to ‘winne hire in plein bataile’ (ibid.), destroy Ermin’s lands and kill him, rape Josian and then give her to a wainwright.32 In this instance she is rescued by Bevis, but when Ermin forces her to accept marriage to King Yvor, Bevis is not present: she preserves her virginity by means of magic. Virginity, with its marvellous powers, is protected by marvellous means. Josian thus remains a fitting prize for Bevis, while at the same time demonstrating her remarkable powers of self-preservation and resistance to force. Josian’s active preservation of her virginity is most dramatically treated 31

Weiss, ‘The wooing woman’, p. 154. Lee C. Ramsey, Chivalric Romances: Popular Literature in Medieval England (Bloomington, IN: Indiana University Press, 1983), p. 67, argues, less persuasively, that ‘Bevis has violated his warrior’s purity by the kiss’– hence his imprisonment. 32 See also the discussion of Josian’s marriages in Corinne Saunders, Rape and Ravishment in the Literature of Medieval England (Cambridge: Brewer, 2001), pp. 199, 204–6.

170



Gender, Virtue and Wisdom in Bevis

in the episode of Earl Miles. Josian delays her violation by demanding that Miles legitimize his desire through marriage. In a sequence of events that recalls the story of Judith and Holofernes, Josian encourages her newly wed husband’s revelry, and then, when they are alone, strangles Miles and hangs him by the curtain rail with a towel (or, in other versions, by means of a slipring in her own girdle, emblem of her chastity).33 As in some hagiographic works where holy virgins are threatened with rape, a strain of black humour runs through the narrative: Miles complains, for instance, that he has never before had to remove his own shoes.34 Josian, however, as Cooper remarks, plays a more active role in her own preservation than her counterparts in lives of holy women.35 As Cooper also notes, among the lives of female saints, perhaps the narrative that most recalls Josian’s exploits is that of the English Christina of Markyate, who is remarkably active in her resistance to the bishop of Durham’s desire to rape her, bolting him into the bedroom while she escapes. Christina later eludes the attentions of her intended husband, first by verbal argument and then by superhuman efforts of flight and concealment.36 Crucially, Christina escapes from but does not harm her enemies, whereas Josian takes on the male role of vengeance, her action dramatically opposing the attempt to possess her body. Fellows argues that the dubious act of murder, like Josian’s other assertive gestures, is justified by her ‘Bevis-centredness’.37 Barnes and Ramsey also remark on the misogyn­ istic strain of the narrative, suggesting that Josian is ‘clearly punished for her first two marriages’, ‘unwanted and unconsummated though they are’, by being threatened with burning and through her capture after the birth of her sons.38 Yet, like the death of Miles, these episodes seem to add to the heroic stature of Josian; they also, in the threat of torture and physical testing, sustain the hagiographic resonances of the work. Josian’s powers extend beyond the verbal: throughout the romance her knowledge of herbs and medicine is emphasized, and she plays an important role as healer.39 She is also skilled as a minstrel, employing her skills on the fiddle, ‘Here sostenaunse for to winne’ (p. 182), while her protector, Saber, lies ill. The narrative emphasizes the marvellous quality of Josian’s medical skills, but avoids portraying her as a practitioner of medical magic of the kind

33

34

35 36 37 38 39

For the Judith story and the Miles episode, see further ibid., pp. 138–9, 206. On parallels between Bevis and the Nibelungenlied, see Laura A. Hibbard, ‘The Nibelungenlied and Sir Beves of Hampton’, Modern Language Notes, 26:5 (1911), 159–60. Lalia Phipps Boone, ‘Criminal law and the Matter of England,’ Boston University Studies in English, 2 (1956), 2–16 (p. 12), notes that Josian is the only woman in the ‘Matter of England’ to be tried for treason. Cooper, The English Romance in Time, p. 247. Ibid., p. 248; and see further Saunders, Rape and Ravishment, pp. 147–50. Fellows, ‘Mothers in Middle English romance’, p. 53. Ramsey, Chivalric Romances, p. 59; Barnes, Counsel and Strategy, p. 89. Josian’s medical arts are discussed in similar terms in Corinne Saunders, ‘Bodily narratives: illness, medicine and healing in Middle English romance’, in Boundaries in Medieval Romance, ed. by Neil Cartlidge, Studies in Medieval Romance (Cambridge: Brewer, 2008), pp. 175–90.

171



Corinne Saunders

widely condemned by theologians from Augustine onwards. Josian is not an enchantress, but possesses sophisticated medical learning: While ȝhe was in Ermonie, Boþe fysik and sirgirie Ȝhe hadde lerned of meisters grete Of Boloyne þe gras and of Tulete, Þat ȝhe knew erbes mani & fale, To make boþe boute & bale.  (p. 172)

Partonope of Blois offers a similar description of the healing arts of Melior: in this case, however, Melior’s knowledge of herbs, roots and spices complements her magical knowledge.40 By including details of medical learning but isolating medical from magical arts, the poet of Bevis suggests that Josian’s knowledge is respectable. The figure of the woman healer is not unrealistic. The earliest medical faculty had been established at Salerno in the mid-tenth century and flourished particularly from the late eleventh to the early thirteenth century; it was especially associated with women because of the reputation of Trotula. Records survive of actual women practitioners; although such women were rare, their existence offers an intriguing context for Josian. The particular places associated by the poet with medical learning are realistic. From the early thirteenth century, ‘Boloyne þe gras’ (Bologna la grassa) was indeed Italy’s great centre of medical learning, while ‘Tulete’ (Toledo) evokes the Arabic learning that flourished in Spain during the period of Muslim rule.41 While the Anglo-Norman Josiane is princess of Egypt, the change to Armenia in the Middle English places her within the Muslim empire, but in a country closer to Spain and Italy than Egypt is. She is imagined as having access to the ancient, especially Arabic, traditions of learned medicine, rather as she has learned the art of minstrelsy. Josian is characterized as confident and knowledgeable in healing Bevis: ‘Lemman’, ȝhe seide, ‘wiþ gode entent Ichaue brouȝt an oyniment, For make þe boþe hol & fere.’  (p. 34)

At the instigation of her father, she prepares for him ‘riche baþes’ (p. 35) which soon make him ‘boþe hol and sonde’ (p. 36). Baths, employing traditional herbal remedies, recur in romance, presented particularly as an aspect of female medical knowledge: Iseult and her mother in most versions of the Tristan legend treat Tristan’s wounds with healing baths, while Elaine of Astolat is sent out to gather herbs for Lancelot’s baths. Later, Josian acts as 40

The Middle-English Versions of Partonope of Blois, ed. by A. Trampe Bödtker, EETS, es 109 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1912 for 1911), lines 5921–7, 5915, 5918. 41 See further Nancy G. Siraisi, Medieval & Early Renaissance Medicine: An Introduction to Knowledge and Practice (Chicago, IL: University of Chicago Press, 1990) pp. 29–30.

172



Gender, Virtue and Wisdom in Bevis

her own midwife at the birth of her twins, having sent Bevis and Terri out hunting away from her ‘paines’ (p. 171) in order to preserve her womanly modesty. Her herbal knowledge produces marvellous effects, most strikingly when she is seized by Ascopard: she picks a herb which transforms her appearance to that of a leper, and thus repels Yvor, preserving her chastity.42 When rescued, Josian swiftly applies ‘an oiniment’ that restores her beauty.43 Josian is powerful enough to feign an illness that so often stands as a terrifying reminder of God’s powers, yet the emphasis on her use of herbs just keeps her abilities within the bounds of credible medical knowledge. Still more remarkable is Josian’s use (in A) of a magic ring to preserve her chastity from King Yvor.44 Stones were commonly employed, along with herbs, to protect the body from illness. Although theologians argued against the superstitious use of amulets, stones of special vertu (most often set in rings) could be viewed differently: like herbs, stones, instilled with power through the hidden forces of the stars and planets, were a material sign of God’s grace, a token of the beneficent pattern of the universe. Repeatedly, romances mention stones ‘of swich vertu’ that they offer marvellous protection, usually from wounds and other kinds of harm, as in King Horn, Floris and Blaunchefleur, Ywain and Gawain and Sir Perceval de Galles. The learning of a physician such as Josian would include such knowledge of the virtues of plants and stones. Josian’s ring, with its powers of natural healing, contrasts with that of the old king, Garcy, in whose charge Yvor leaves her. Garcy, who ‘muche can of Nygremancy’ (p. 111), possesses a gold ring that allows him to see ‘What any man dooth in alle þing’ (ibid.). He is eventually deceived by the natural magic of the herb placed in his ‘reynessh wyne’ (ibid.) by Josian’s chamberlain, Bonefas, which puts him to sleep for a day and a night, allowing the lovers to escape. Crucially, Josian’s ring is not counted as ‘Nygremancy’, and thus cannot be associated with the black, potentially demonic, magical arts. Rather, its ‘vertu’ is an extension of her own virtue, a token of her wisdom in natural, medical magic. It seems likely that this detail of the ring was not in the earliest English translation, for not only Boeve but also all versions of Bevis except for A and the printed tradition (P)45 describe instead a magic girdle. As Cooper notes, the detail of Josian’s marvellous device could be seen as irrelevant;46 yet different redactors evidently engage differently with the issue of magic. The manuscript versions other than A emphasize the fantastic, ignoring any prob-

42 43

BHK, pp. 172, 173. Generydes employs the same motif, the hero transforming himself into a leper with an ­‘oyntement’, and later washing away the effects with ‘a water’: cf. Generydes: A Romance in Seven-Line Stanzas, ed. by W. Aldis Wright, EETS, os 55 (London: Trübner for the EETS, 1878), lines 4274, 4315. 44 BHK, p. 77. 45 P is equivalent to Kölbing’s O, which he intermixes with passages from M in the text at the foot of the page. 46 Cooper, The English Romance in Time, p. 142.

173



Corinne Saunders

lematic aspect of representing Josian as an enchantress. Little detail is given, but the girdle functions as a kind of chastity belt. Although the chastity belt was a post-medieval, possibly Renaissance, invention, given a fictitious medieval history, this particular use of a magical girdle is apparently unique in romance and does seem to have been suggested by a connection between the idea of the girdle and that of chastity. In this tradition, the enchanted girdle is echoed and answered in the most unmagical girdle that Josian employs to strangle Miles. While an earlier English redactor made the change from girdle to ring, in P the protective object is changed once again, becoming a wrytte. The episode gains a completely new legal resonance: I shal go make me a wrytte Thoroughe a clerke wyse of wyt, That there shal no man haue grace, Whyle that letter is in place, Agaynst my wyl to lye me by Nor do me shame nor velany! (p. 77)

Here, the female enchantress is transformed into the male clerk who assists Josian, perhaps through occult and legal as well as theological knowledge. The writ seems to function as an amulet when fastened around the neck, implying the near-magical power afforded a legal oath. By contrast, the redactor of A (or his earlier English source) seems to have been eager to present Josian not as an enchantress figure but, rather, as an independent wise woman with the knowledge of natural magic, in particular the medical arts of the East. Her learning in the arts of medicine, magic and music balances Bevis’s physical prowess, while both are marked by astuteness as well as courage. As Barnes writes, ‘[Josian’s] outstanding quality is her talent for gyn’.47 Like Bevis, Josian is distinguished from the rest of her sex, but this distinction is primarily the result of her learning and actions, rather than of prayer and miraculous protection. The parallel with lives of holy women, however, is also maintained: after Bevis and she have ruled for twenty years in Mombraunt, Josian in true hagiographic fashion foresees her own death, and Bevis dies on the same day as her, having discovered Arundel dead in the stable. The honour with which their deaths are commemorated affirms the holiness of both. The role of Josian both echoes and functions in counterpoint to that of Bevis. She is carefully constructed to subvert conventional medieval notions of women as naturally frail and passive and, in particular, to counteract the model of Bevis’s mother. She resists too the model of the courtly lady, object of desire and trophy of battle, and though she is repeatedly threatened by 47

Barnes, Counsel and Strategy, p. 89; cf. Jennifer L. Fellows, ‘Sir Beves of Hampton: Study and Edition’, 5 vols (unpub. Ph.D. diss., Univ. of Cambridge, 1980), I, 79–80.

174



Gender, Virtue and Wisdom in Bevis

enforced marriage, she never becomes the damsel in distress. Her virtue is proved through the startlingly active features of her pursuit of love, preser­ vation of chastity and role as healer. Likewise, Bevis does not conform to typical romance patterns of masculinity. He far exceeds all others in prowess as defender of Christianity, justice, love and chastity, but his virtue is specially evident in his recognition of his own frailty, and his need for God’s help. The pattern of the chivalric knight is refracted through the lens of devotional literature, and Bevis becomes an exemplar of Christian virtue and the need for divine intervention. The holy deaths of Bevis, Josian and Arundel, all occurring at once, are ‘the culminating marvel of their story’, a final assertion of spiritual virtue that takes the narrative into the realm of hagiography.48 The portrayal of both Bevis and Josian, elaborated through an emphasis on virtue and wisdom, as well as by their actions and interactions across the work, complicates traditional romance ideas of gender and lends interest, depth and colour to the narrative of the romance.

48

Cooper, The English Romance in Time, p. 361.

175

11 Sir Bevis of Hampton: Renaissance Influence and Reception Andrew King

S

 ir Bevis of Hampton presents an astonishing instance of persistence and later influence. Translated from the Anglo-Norman c. 1300, the poem was printed, in its essentially medieval text and as part of a continuous tradition of non-scholarly reading, from the incunable period until 1711. It was also read in manuscript during the sixteenth and seventeenth centuries, as the marks and annotations of readers attest. Bevis is therefore clearly part of the historical depth of the literary culture of sixteenth- and seventeenth-century England. Understanding how the text was received – and, in the context of its assimilation into new works, transformed – entails an important sense of the complex continuities between the medieval and Renaissance periods. Norfolk’s wry observation in Shakespeare’s Henry VIII, that the exaggerated feats of chivalry at the Field of Cloth of Gold ‘got credit / That Bevis was believed’, indicates only one possible response to the legend: bemused, possibly scornful, recognition of its appealing but essentially puerile fantasies. Roger Ascham and Thomas Nashe are the figures quoted most often as representative of the Humanist and Protestant objection to a work (and its romance tradition in general) that was seen to originate in ‘monkish’ culture. John Selden, in providing a historian’s commentary on Michael Drayton’s verse ‘Songs’ celebrating the British landscape and its legends in Poly

Bevis survives in nine manuscripts from the medieval period and was printed at least nine times in the sixteenth century, with continued printing in the seventeenth century: see pp. 109–13 above. On sixteenth- and seventeenth-century annotations in medieval manuscript anthologies containing Bevis, see The Auchinleck Manuscript: National Library of Scotland Advocates’ MS 19.2.1, introd. by Derek Pearsall and I. C. Cunningham (London: Scolar Press in association with the National Library of Scotland, 1977), pp. x–xvi; Cambridge University Library MS Ff.2.38, introd. by Frances McSparran and P. R. Robinson (London: Scolar Press, 1979), p. xvii.  All Is True (Henry VIII), i.i.37–8, in The Complete Works, gen. eds Stanley Wells and Gary Taylor, 2nd edn (Oxford: Clarendon Press, 2005). Buckingham’s response to Norfolk is ‘O, you go far!’ (i.i.38). All references to the works of Shakespeare are to the Wells and Taylor edition.  Roger Ascham, English Works, ed. by William Aldis Wright, Cambridge English Classics (Cambridge: Cambridge University Press, 1904), p. 231; The Works of Thomas Nashe, ed. by Ronald B. McKerrow, 2nd edn, rev. by F. P. Wilson, 5 vols (Oxford: Blackwell, 1958), I, 11. See further Robert P. Adams, ‘ “Bold bawdry and open manslaughter”: the English New Humanist attack on medieval romance’, Huntington Library Quarterly, 23 (1959–60), 33–48.

176



Bevis: Renaissance Influence and Reception

Olbion, writes: ‘And it is wished that the poeticall Monkes in celebration of him [Bevis], Arthur, and other such Worthies had containd themselves within bounds of likelyness.’ Many more instances of the disparaging reactions of Humanist and Protestant readers could be cited. But Bevis continued to be read. These comments deserve to be seen as testimonies to that fact rather than indications of waning popularity. What Selden’s comment above really alerts us to is two major concerns preoccupying early modern authors (and by implication readers) who sought to make a more positive response to the insistent presence of Bevis than, say, Nashe, with his airy dismissal of ‘worne out absurdities’. One problem for writers and readers in a post-Reformation context was how to validate ‘monkish’ literature. This was part of a larger project extending well beyond romance. John Foxe’s Actes and Monuments (or Book of Martyrs, as it came to be known), first published in England in 1563, provides one of the most strategically formulated and influential examples, though the problem preoccupied a wide range of writers and works, such as John Bale and, of course, Spenser, Drayton and Bunyan. Foxe constructs a narrative for medieval history and literature in which writers such as Ælfric, Chaucer and Gower are redeemed as proto-Protestants and the foreseers of the changes that have, by the Elizabethan era, come into place. This view is contained within a larger, more powerful narrative concerning Christ’s true Church, which existed for Foxe in Britain since Apostolic times – well attested in medieval Arthurian romance’s accounts of Joseph of Arimathea and his evangelization of Britain. Throughout the later Middle Ages, narrates Foxe, the true Church, which is the body of the elect souls, began to be covered in darkness, lost from vision because of the corrupt ‘Romish’ Church, ‘so shining in outward beauty’ yet substantially false (I, xix). Nevertheless, the true Church was occasionally manifested, in England in particular, in writers such as Chaucer as well as religious reformists and martyrs, affording glimpses of the hidden radiance of Christ’s Church. Foxe’s narrative gains unquenchable authority through its alliance with biblical apocalyptic imagery: the Woman Clothed with the Sun fled into the wilderness, challenged by the Whore of Babylon, until such time as she could return. The glosses of the Geneva Bible identified these figures

     

The Works of Michael Drayton, ed. by J. William Hebel et al., 5 vols (Oxford: Blackwell for the Shakespeare Head Press, 1961), IV, 46. For a recent account, see Alex Davis, Chivalry and Romance in the English Renaissance, Studies in Renaissance Literature 11 (Cambridge: Brewer, 2003), pp. 1–18. See further Helen Cooper, The English Romance in Time: Transforming Motifs from Geoffrey of Monmouth to the Death of Shakespeare (Oxford: Oxford University Press, 2004), pp. 36–9. The Works of Thomas Nashe, ed. McKerrow, I, 26. The Acts and Monuments of John Foxe, ed. by George Townsend, 8 vols (London: Seeley & Burnside, 1837–41), I, xix–xx. Revelation xii.1–6. See further John N. King, English Reformation Literature: The Tudor Origins of the Protestant Tradition (Princeton, NJ: Princeton University Press, 1982), pp. 66, 68; Andrew King, ‘The Faerie Queene’ and Middle English Romance: The Matter of Just Memory (Oxford: Oxford University Press, 2000), pp. 157–8.

177



Andrew King

as the true Church and the Church of Rome respectively.10 So too Foxe’s and also Bale’s presentation of a select group of medieval texts is a comparable narrative, involving the initial exile and later resuscitation and re-dressing in appropriately gleaming vestments of an older, true literature that had been besmirched or misunderstood in a corrupt age.11 Bevis becomes an ancillary part of this elaborate project of re-reading the medieval literary past in terms of the religious, cultural and literary claims of the present in a number of authors, including Spenser and Bunyan. Selden’s comment above reveals another challenge facing early modern writers who sought to draw medieval English romance into the sphere of post-Reformation literary interest and historical understanding: what he calls ‘likelyness’ was considered an integral aspect of valuable literature, since such texts do not mislead readers into morally dangerous assumptions concerning the nature of human experience and a just and natural society.12 Selden continues in his reactions to Drayton’s handling of the Bevis story, developing further this aspect: The sweet grace of an inchanting Poem … often compels beliefe; but so farre have the indigested reports of barren and Monkish invention expatiated out of the lists of Truth, that far from the intermixed and absurd fauxeties hath proceeded doubt; and, in some, even deniall of what was truth. His [Bevis’s] sword is kept as a relique in Arundell Castle, not equalling in length (as it is now worne) that of Edward the thirds at Westminster.13

Selden’s reaction is ambivalent, furthering the complex relationship of support and denial that his commentary offers in relation to Drayton’s text. Here he subverts the historicity, and hence the moral value, of the Bevis story; and yet in an abrupt shift he advances Bevis’s still extant sword as a ‘relique’ that should serve as a proof of his historicity, linking the narrative world of the romance to the world known to contemporary readers. The device of a narrator highlighting some object or feature of landscape that has its origins in the legend and that is ‘still there’ for readers to see is a well-­established topos in Middle English romance, historical writings and (as Selden’s word ‘relique’ possibly suggests) saints’ lives.14 The Auchinleck MS text of Bevis (as well as later versions) participates in this topos implicitly when it describes Bevis’s battle in London with a topographical accuracy that would allow a contempo-

10 11 12 13 14

The Geneva Bible: A Facsimile of the 1560 Edition, introd. by Lloyd E. Berry (Madison, WI: University of Wisconsin Press, 1969). John Bale, Scriptorvm illustriū Maioris Brytannie … catalogus (Basle: apud I. Oporinum, 1557– 9), p. 474. See further William Nelson, Fact or Fiction: The Dilemma of the Renaissance Storyteller (Cambridge, MA: Harvard University Press, 1973), passim. The Works of Michael Drayton, ed. Hebel, IV, 46. King, ‘The Faerie Queene’ and Middle English Romance, pp. 48–52, 161–2.

178



Bevis: Renaissance Influence and Reception

rary reader to visit sites that featured in the romance narrative.15 The object or feature that is ‘still there’, common to both the textual world and the reader’s experience, sets up the text as a mirror for contemporary self-understanding. However obviously flawed the logic, the ‘still there’ object – whether it be a sword, a tomb, a place-name, or a feature of landscape – becomes the ‘proof ’ that the textual events really happened and have left their mark on the reader’s world. So Bevis’s sword in Arundel Castle (according to the romance, named after Bevis’s horse) becomes a means by which the truth of the story may be asserted.16 Selden’s employment of the topos here seems half-hearted at best.17 But it highlights the challenge that early modern responses to the Bevis narrative face: the need to establish the truth of the story, if not literally then at least in a mythopoeic, morally or culturally edifying sense. The rest of this chapter will look at a number of sixteenth- and seventeenth-century uses of the Bevis legend that illustrate varied responses to these issues. It will highlight how writers negotiate the challenges of dealing with Bevis in a post-Reformation context. Furthermore, the chapter will consider Bevis’s diminishing stock as respectable middle-class literature in late sixteenth-century England, asking how early modern writers might respond to this situation strategically, turning liability into literary capital. Spenser, The Faerie Queene, Book i: the dragon fight Spenser’s use of Bevis in Book i – the book of The Faerie Queene that focuses on Holiness – illuminates in spectacular fashion the challenges facing the Protestant reception of pre-Reformation literature. The appearance of Bevis in Redcrosse’s dragon fight is certainly provocative. The assumption here is that Elizabethan readers had the capacity to recognize the palimpsest of Bevis in Spenser’s text – even that this could be part of Spenser’s putative strategy.18 The fact that Richard Johnson similarly uses Bevis’s dragon fight in his account of St George (see below) adds credibility to this view. Also significant is the fame of Bevis, well attested through the popularity of the text. The dragon fight (which is one of several passages original to the Middle English version) seems to have been particularly impressive to medieval as

15 16

Cf. BHK, pp. 202–3, 206, 210. A figure of Bevis used to exist on the Bargate at Southampton: cf. Jennifer Fellows, ‘Sir Bevis of Hampton in popular tradition’, Proceedings of the Hampshire Field Club and Archaeological Society, 42 (1986), 139–45; Adrian B. Rance, ‘The Bevis and Ascupart panels, Bargate Museum, Southampton’, ibid., 147–53. 17 Compare John Leland’s Assertio inclytissimi Arturii regis Britanniae (1544; trans. Richard Robinson, 1582), an intensely committed presentation of the landscape near Glastonbury and elsewhere in Britain as bearing the vestigial marks of Arthur’s presence. Text in Christopher Middleton, The Famous Historie of Chinon of England, EETS, os 165 (London: Oxford University Press for the EETS, 1925). 18 Cooper, The English Romance in Time, p. 31, concurs with this point.

179



Andrew King

well as early modern readers; its details are borrowed substantially for the dragon fight in Sir Degare.19 Thomas Warton was the first to note that Spenser draws upon Bevis for his dragon fight.20 The relevant passages need to be quoted at some length. The following is the dragon fight from Thomas East’s print of Bevis from c. 1582,21 the closest surviving print in date antecedent to The Faerie Queene:           but Beuis rode foorth and sought, And when the Dragon that foule is, had a sight of sir Beuis, He cast vp a loud cry, as it had thundred in the sky. He turned his belly toward the sun, it was bigger then any tunne, His scales wes brighter then the glasse, & harder they wer then any   brasse, betweene his shoulder and his tayle, was forty foote without fayle … What for wery and what for faint, sir Beuis was neere attaint, The Dragon sewed on Beuis so hard, yt as he would haue fled   backw[a]rd, There was a well as I weene, and he stumb[l]ed right therein … Then was the well of such vertue, through the might of Christ Iesu. For sometime dwelled in that lond, a Virgin full of Christs sond, That had bene bathed in that well, that euer after as men tell, Might no venimous worme come therein, by the vertue of that Uirgin, Nigh it seauen foote and more, then Beuis was glad therefore, When he sawe the Dragon fell, had no power to come to the well, Then was he glad without fayle, and rested a while for his auayle, And dranke of that water his fill, and then he lept out with good will, And with Morglay his brand, assayled the Dragon I vnderstand.22 (sigs Fiiv–Fiiir)

Bevis falls into the healing well a second time after the dragon nearly kills him with an onslaught of venom: His armour burst in that stound, and he fell dead to the ground, There was no life on him seene, he lay as a dead man on the greene, The Dragon smot him without faile, that he tourned top and tayle, but thereoff he tooke no keepe, he lay as a dead man on sleepe, He smote Beuis as I you tell, the dint smote him into the well, That was of great vertue that time, for it would suffer no venime, 19

Nicolas Jacobs, ‘Sir Degarré, Lay le Freine, Beves of Hamtoun and the “Auchinleck bookshop” ’, Notes and Queries, 29 (1982), 294–301 (pp. 297–300), and ‘The second revision of Sir Degarre: the Egerton fragment and its congeners’, Neuphilologische Mitteilungen, 85 (1984), 95–107 (pp. 101–4). 20 Thomas Warton, Observations on the Fairy Queen of Spenser, 2nd edn, 2 vols (London: Dodsley, 1762), I, 46–54. Warton’s discussion of this borrowing is quoted in part in Edmund Spenser, Works, ed. by Edwin Greenlaw, Charles Grosvenor Osgood, and Frederick Morgan Padelford, 9 vols in 10 (Baltimore, MD: Johns Hopkins Press, 1932–49), I, 66–75. 21 Cf. p. 111 above (no. 13). All references to the text of Bevis are to this edition. 22 I follow East’s layout in printing each couplet as a single line.

180



Bevis: Renaissance Influence and Reception Through vertue of the Uirgin, that once was bathed therin, When Beuis was at the ground, the water made him whole & sound, And quenched all the venim away, this well saued Beuis that day, When he feelt him whole and light, and knew that well of great might Then was he a ioyfull man, he was as fresh as when he began, He kneeled downe in that place, to Iesu Christ he called for grace, That he would send him might, to slay the Dragon in fight.  (sig. Fiiir)

Whenever Bevis felt weakened in the battle from this point, ‘he went to ye wel & washed him thore’, emerging ‘as fresh as when he began’ (sig. Fiiir). Spenser’s Redcrosse similarly stumbles unwittingly into a healing well at a crucial moment in his encounter with a dragon: It fortuned (as fayre it then befell,) Behynd his backe vnweeting, where he stood, Of auncient time there was a springing well, From which fast trickled forth a siluer flood, Full of great vertues, and for med’cine good. Whylome, before that cursed Dragon got That happie land, and all with innocent blood Defyld those sacred waues, it rightly hot The well of life, ne yet his vertues had forgot.  (i.xi.29)23

Redcrosse emerges from the well with renewed strength. Later during the fight, he is driven backwards by the dragon and falls against a restorative tree: There grew a goodly tree him faire beside, Loaden with fruit and apples rosie red, As they in pure vermilion had beene dide, Whereof great vertues ouer all were red: For happie life to all, which thereon fedd, And life eke euerlasting did befall: Great God it planted in that blessed sted With his almightie hand, and did it call The tree of life, the crime of our first fathers fall … From that first tree forth flowd, as from a well, A trickling streame of Balme, most soueraine And daintie deare, which on the ground still fell, And ouerflowed all the fertill plaine, As it had deawed bene with timely raine: Life and long health that gratious ointment gaue, And deadly woundes could heale, and reare againe 23

All references are to Edmund Spenser, The Faerie Qveene, ed. by A. C. Hamilton, Hiroshi Yamashita and Toshiyuki Suzuki, 2nd edn (Harlow; New York: Longman, 2001) (hereafter FQ), i.xi.29–30.

181



Andrew King The senselesse corse appointed for the graue. Into that same he fell: which did from death him saue.  (i.xi.46, 48)

Obviously the striking parallel between the two scenes is, in Warton’s phrase, the heroes’ ‘miraculous manner of healing’ (I, 49). Both Bevis and Redcrosse fall into a restorative well. The two falls in Bevis have a structural counterpart in Redcrosse’s single fall into the well with another fall against the Tree of Life. Both heroes gain victory on the third attempt, in time-honoured fashion. Spenser’s Tree of Life is still rooted as much in the imagery of the well in the Bevis scene as it is in the Book of Revelation: in The Faerie Queene, a ‘trickling streame of Balme’ flows from the Tree of Life ‘as from a well’ (i.xi.48; my emphasis). Spenser’s dragon cannot approach this well-like emission from the Tree: For nigh thereto the euer damned beast Durst not approch, for he was deadly made, And all that life preserued, did detest.  (i.xi.49)

In Bevis, an area around the well is similarly inviolable.24 When Spenser came to write the ‘legend’ of Holiness, or St George, English precedents already existed as a descriptive basis for George and the dragon fight, namely in Barclay, Lydgate and (via Caxton) de Voragine.25 Why, then, did Spenser go outside the literature of St George, to Bevis? If recognized by his readers, the borrowing might risk subverting this moment of high seriousness: Redcrosse’s defeat of the dragon is a definitive moment in his realization of his identity as St George and in the demonstration of the nature of Holiness. Both the readily mocked popular appeal of Bevis and its pre-Reformation origins threaten to undermine this climax. But in fact deliberate subversion is exactly what the scene demands at this point. What we have above all is a radical reconfiguring of romance according to a new understanding; and that sense of ‘new understanding’ goes hand in hand with Foxe’s apocalyptic narrative concerning the Church – the Woman Clothed with the Sun, for a time fled into the wilderness and now brought forth revealed in her true glory. Una provides an image of this transformation, or reformation, when she lays aside ‘her mournefull stole’ (i.xii.22) and wimple and displays her true radiance: ‘The blazing brightnesse of her beauties beame, / And glorious light of her sunshyny face’ (i.xii.23). So too in this scene Bevis and his romance world are, through Spenser’s reception

24 25

Sig. F.iiir. For more verbal parallels, beyond those mentioned by Warton, cf. BHF, V, 113–21. Alexander Barclay, The Life of St George, ed. by William Nelson, EETS, os 230 (London: Oxford University Press for the EETS, 1955); Jacobus de Voragine, The Golden Legend, or Lives of the Saints, trans. by William Caxton, ed. by F. S. Ellis, 7 vols (London, 1900); John Lydgate, The Minor Poems of John Lydgate, ed. by Henry Noble MacCracken, 2 vols, EETS, es 107, os 192 (London: Trench, Trübner & Co. for the EETS, 1911, 1934), I, 145–54.

182



Bevis: Renaissance Influence and Reception

and reformation, given their ‘true’ definition, revealed finally in terms of their real value, what they always seemed to be trying to say. That strategy of the reader’s recognition of how the familiar and even the potentially despicable have now come into their own is at the heart of Spenser’s engagement with Bevis in Book i. For example, the well in Bevis is arguably not the baptismal symbol that it becomes in Spenser’s text.26 True, it restores Bevis, making him ‘whole & sound’ when ‘he lay as a dead man’. The source of the well’s intrinsic virtue is that a virgin ‘full of Christs sond [message]’ bathed in it. But no further details concerning the virgin or her message (which could enhance a symbolic reading of the well) are offered. Furthermore, the well revives Bevis at this specific stage in his adventures, but we are given no sense that its efficacy will continue throughout his life or that this incident is a pivotal moment in Bevis’s spiritual journey. Calvin attacks the Catholic sense of baptismal and holy water as intrinsically powerful in their physical substance: Paul did not mean to signify that our cleansing and salvation are accomplished by water, or that water contains in itself the power to cleanse, regenerate, and renew: nor that here is the cause of salvation, but only that in this sacrament are received the knowledge and certainty of such gifts.27

But the first sense here is surely the well in Bevis – ‘water [that] contains in itself the power to cleanse, regenerate, and renew’. Recognition of this ‘monkish’ palimpsest of Bevis must be accompanied by a realization of its radical transformation in Spenser’s reception. For one, even though Bevis gains crucial assistance from his well, he still embodies the heroic selfsufficiency of the romance hero. Bevis accrues personal glory and chivalric renown for defeating his dragon: ‘Then was Beuis name in honour, euery man had him fauour, / In euery land is spred, sir Beuis price and laud’ (sig. Fiiiv). But such an act of individual glory and self-aggrandisement is impossible in the narrative world of Book i of The Faerie Queene: Ne let the man ascribe it to his skill, That thorough grace hath gained victory. If any strength we haue, it is to ill, But all the good is Gods, both power and eke will.  (i.x.1)

The Calvinist emphasis here on the inefficacy of the fallen human will and the need for God’s grace, symbolized in the baptismal well, challenges 26

For a different view of the symbolic significance of the well in Bevis, see Jennifer Fellows, ‘ “Dragons two other thre”: the function of the dragon fight in some Middle English romances’ (forthcoming). 27 Jean Calvin, Institutes of the Christian Religion, ed. by John T. McNeill, trans. by Ford Lewis Battles, 2 vols (Philadelphia, PA: Westminster Press, 1960), iv.xv.2 (p. 1304).

183



Andrew King

the validity of the heroic romance mode. In the dragon fight, Spenser seems to rewrite this famous romance encounter with a reformed sense of the necessity of God’s grace in all human achievement. Bevis’s well becomes in ­Spenser’s reception something like a literal narrative element that is only fully understood when unlocked with the interpretative key or allegorical gloss of Spenser’s text. Bevis’s well, Spenser’s use implies, points us to the perfectly realized image of God’s grace as ongoing replenishment and sustenance, as embodied in his text. Spenser’s strategy in making visible the presence of Bevis in his dragon fight aligns with Foxe’s redefinition of chosen medieval texts as fulfilled ‘prophecies’. In the process, a popular medieval native romance is resuscitated and given new signification. At the same time, Spenser’s account of Holiness gains the authority that comes from historical depth: the story is familiar to us, indeed had probably been part of our childhood reading, and now with a sense of revelation its full meaning comes into consciousness. Spenser’s St George is clearly not a saint in the old Catholic sense; nor (like Foxe’s Church) is he a newly constructed entity. Rather, as one of the saintly elect, he gains through Bevis the familiarity and Englishry of a beloved folk-hero. But the sense of Bevis coming into its ‘proper’ meaning in Spenser’s handling extends beyond the dragon fight to the whole narrative structure of Book i, which is an equally ingenious and startling revision of this popular medieval narrative. Redcrosse, Bevis and displaced youths Bevis presents one of the most influential and popular of all narrative structures in Middle English romance: the displaced youth. Havelok, Arthur, Lybeaus Desconus, Florent (in Octavian), and Perceval are all medieval examples; in modern romance narratives, Harry Potter and Luke Skywalker are perfect instances.28 The crucial elements in this narrative tradition are a youth of special, generally aristocratic, birth. In infancy, the youth is kidnapped or in some manner exiled from his homeland, his social position and his patrimony. This displacement often occurs at such a young age that the youth matures in the foreign context, unaware of who he really is. The context into which the youth has been displaced is inevitably degenerate in moral, social and cultural terms by comparison with his rightful place. So displaced youths typically find themselves removed from courts and palaces to wildernesses, rustic dwellings, impoverished serfdom or even pagan lands: Perceval is raised in a forest, Florent by a Parisian butcher, and Havelok by a poor fisherman. The youth’s upbringing in this context can have calamitous effects 28

See further Paul R. Rovang, Refashioning ‘Knights and Ladies Gentle Deeds’: The Intertextuality of Spenser’s ‘Faerie Queene’ and Malory’s ‘Morte Darthur’ (Madison, WI; London: Fairleigh Dickinson University Press, 1996), pp. 23–38; King, ‘The Faerie Queene’ and Middle English Romance, pp. 145–53.

184



Bevis: Renaissance Influence and Reception

on his maturation, leaving him ignorant of religion, manners and courtly society; a variation, though, is that the youth miraculously retains the courtly or Christian qualities, even though his context has provided no means for him to acquire them. Whether the chivalric and patrilinear aspirations are imperfectly realized in the youth or fully in place, the crucial element in this kind of narrative is that the youth strives against the grain of his rustic, degenerate or pagan upbringing, feeling within himself a sense that this is not his proper sphere – that he belongs in another world, occupying a very different position in that world. Bevis is a prime example of this narrative type: the displaced youth’s struggle to regain his patrimony and re-enter his proper social sphere. The intertextuality of Spenser’s Book i is developed through an ingeni­ous reworking of Bevis’s narrative of the displaced youth. Both Bevis and Redcrosse are separated from their ancestral contexts, privileges and birthrights: Bevis’s father is murdered by his mother and her lover, and he is exiled to pagan lands; Redcrosse, whose true identity is St George and who is descended ‘from ancient race / Of Saxon kinges’, was stolen by fairies and raised by a ploughman in a deprived and rustic context.29 Both Bevis and Redcrosse feel out of place in their exile contexts, yet each also retains an imperfect sense of what his normal pattern of living should be. Bevis is taunted by pagan knights for not knowing the significance of Christmas Day, yet also demonstrates (however brutally) an incipient Christian identity in his antagonistic response to these knights (sigs Biiv–iiiv); Redcrosse’s efforts at chivalry are similarly clumsy and ill-judged from the start. Redcrosse appears in the ‘Letter to Raleigh’ in borrowed armour and with rustic behaviour, yet aspiring to knighthood – details that recall other displaced youths such as Lybeaus Desconus, Malory’s Bewmaynes, and Florent in Octavian, whose borrowed armour is ‘sutty, blakk and unclene’.30 Every effort that Redcrosse makes at chivalric conquest leads towards humiliation or embarrassing dependence on outside agency, and the sense is strong that his exile has had a detrimental effect on his chivalric aptitude, however much knighthood remains in his blood. The experiences of both Bevis and Redcrosse culminate in a regaining of patrimony and a return from exile; Bevis has come to know who he is gradually, whereas for Redcrosse it is a revelation, given to him on Contemplation’s mount. Both heroes may seem to have earned or struggled towards their leading places in their worlds, but at a deeper level the texts are insistent that the heroes only seem to achieve a position that is foremost the gift of birth. King Edgar recognizes Bevis’s birthright and title near the end of the text, after Bevis asserts not his travails in overcoming exile, but the simple fact of his parentage; Edgar restores Bevis to his rightful place,

29 30

FQ, i.x.65–6. Octavian, line 885, in Six Middle English Romances, ed. by Maldwyn Mills, Everyman’s University Library (London: Dent; Totowa, NJ: Rowman & Littlefield, 1973), p. 99.

185



Andrew King

making him marshall.31 So too Redcrosse’s ‘ancient race / Of Saxon kinges’ seems to be a necessary condition for his eventual triumph. Spenser’s adaptation of this story is more than just a lively reworking of its narrative details. What his use of the displaced youth narrative suggests is a perception that the narrative elements of this story type provide a striking correspondence to the Calvinist account of human history and salvation. The displaced youth such as Bevis is in exile from his homeland, removed appropriately to non-Christian lands; more than just exile from his Saxon kings, Redcrosse suffers the deeper exile of the Fall – from his heavenly homeland and his heavenly father, God. Calvin emphasizes the depravity of humans after the Fall and our inability to achieve any positive or good action by our own strength.32 The clumsiness of the displaced youth as he seeks to overcome exile becomes in Spenser’s Calvinist reconceiving the mark of Fall and Original Sin. In Redcrosse’s armour ‘old dints of deepe woundes did remaine’ (i.i.1). His near failure with Error and his abysmal defeat at the hands of Orgoglio both dramatize the effects of the Fall through the traditional displaced youth’s untutored ineptness, stemming from his rustic exile or deprived upbringing. Birth is the crucial element in Bevis; in Spenser’s alignment of the displaced youth narrative with Calvinism, the element of noble birth translates into God’s election of the individual for salvation – a condition decreed indeed before the individual’s birth and, as much as an aristocratic birthright, entirely outside the individual’s control. Una reminds Redcrosse that he has been ‘chosen’ (i.ix.53), just as Bevis was born to his opportunities and could never have achieved his position as earl of ­Southampton, however strenuous his efforts, without that gift of birth. So too Redcrosse, as Contemplation informs him, has the gift of God’s Predestination; however heroic his efforts may seem, they run counter (more intensely than in Bevis) to the crucial fact of his birth – not here in a social or aristo­ cratic sense, but in terms of the elect of God. Of course Redcrosse feels within him the promptings to get back to his homeland and father – God in Heaven – demonstrating Calvin’s sense of the imago Dei and ‘true and lively Faith’.33 The inclination that the displaced youth of romance has towards his birthright, even if he is ignorant of it, becomes in Spenser’s text the hero’s response to vocation. Spenser’s strategy is surely as provocative and rhetorical as Foxe’s. Foxe presents certain medieval writers such as Wyclif and Chaucer as ‘proto­Protestant’ – in effect, ‘prophets’ attaining their true meaning only in the present age of fulfilment. Even more daringly, Spenser presents an old and 31 32 33

Sig. G.ivv. Calvin, Institutes of the Christian Religion, ed. McNeill, iii.xi.2 (pp. 726–7). The doctrine of imago Dei is rooted in Genesis i.26. See further Institutes of the Christian Religion, ed. McNeill, i.xv.3 (pp. 186–9). Article xii, ‘Of Good Works’, of the Thirty-Nine Articles stresses that good works ‘cannot put away our sin’; nevertheless, because we are in imago Dei, good deeds will ‘spring out necessarily of a true and lively Faith’ and may be taken as signs of election.

186



Bevis: Renaissance Influence and Reception

often derided popular narrative as having held a mystery for all these years, only now revealed. Bevis and other narratives of displaced youths can now be understood in terms of the most intense definitions given to the narrative in Spenser’s reworking: it is a narrative about God the Father, about the Fall, about recovery not through individual good works but through the grace and gift of the Father. Just as Foxe presented the true Church in terms reminiscent of the displaced youth as well as of patterns from Revelation – wandering for a time in exile and obscurity until restored to its rightful place – so in Spenser’s revisionist treatment we can see that Bevis as a text has, like its eponymous hero, finally come home to its ‘true’ meaning and heritage. Richard Johnson, The Seven Champions of Christendom Spenser’s coordination of Bevis and St George recurs within the decade in Part i of Richard Johnson’s The Most famous History of the Seauen Champions of Christendome … (1596).34 The very fact that this text draws closely upon the dragon fight in the process of relating the life of St George suggests indebtedness to Spenser.35 The possibility of Spenserian influence is made more likely by the fact that Johnson’s St George gains strength during the battle by stumbling under a tree – not into the well found in Bevis. Spenser of course had both the Well of Life and, crucially, the Tree of Life. ­Johnson’s tree resists any possibility of theological or symbolic reading, being (rather surprisingly) an orange tree. This restorative citrus fruit, which cures the consumer ‘of all manner of diseases and infirmities whatsoeuer’ (p. 13), removes Johnson’s text from any hint of the baptismal or sacramental. Just to make sure that the tree is allowed no associations with Christian imagery, the oranges are also used to revive George’s horse. Johnson’s response to the potentially Catholic elements in Bevis is more iconoclastic than Spenser’s: simple erasure rather than provocative realignment. Even though Johnson is ostensibly writing the life of a saint, he consistently seeks to minimize the religious emphasis of Bevis. Before killing the dragon, George does indeed pray ‘that God would send him (for his deare sonnes sake) such strength and agillity of body as to slay the furious and tirrable monster’ (p. 14). But Bevis’s details about the well (not tree) acquiring miraculous 34

References are to Richard Johnson, The Seven Champions of Christendom (1596/7), ed. by Jennifer Fellows, Non-Canonical Early Modern Popular Texts (Aldershot; Burlington, VT: Ashgate, 2003); links between Johnson’s work and Bevis and Spenser in the dragon fight are explored at pp. 268–9 (notes to pp. 12–15). See also Jennifer Fellows, ‘St George as romance hero’, Reading Medieval Studies, 19 (1993), 27–54. 35 M. Pauline Parker, The Allegory of ‘The Faerie Queene’ (Oxford: Clarendon Press, 1960), pp. 101–2, is wrong to state that Spenser follows Johnson’s Seven Champions in his description of the dragon fight. Johnson is of course later than Spenser, and Parker’s confusion must stem from the fact that Johnson, like Spenser, draws upon the dragon fight in Bevis. Her mistake is interesting in that it demonstrates how readers – both modern and early modern – might read any of these responses to Bevis with the full intertextual influence of the tradition.

187



Andrew King

efficacy through the bathing of the virgin with God’s ‘sond’, or the account of the dragon prior to that moment, opposed by a man ‘full of Gods sond’ (sig. Fiir) who prayed ‘for his mothers loue’ that God remove the dragons, are absent. Instead, Johnson’s text places stronger emphasis on St George as an English hero, seeking to invoke in his readers a patriotic response. Even more than Spenser’s text, Johnson’s reception of Bevis attempts to resuscitate Bevis as a beloved native hero whose continuance is ensured through his merging into the nation’s patron saint. For Johnson’s initial readers, recognizing that Bevis has, in effect, ‘become’ St George is a moment (outside the text) comparable to Redcrosse’s own realization that he too is in fact the nation’s saint. The implication of Johnson’s layered text is that the moment of recognition of identity has been transferred from the central character to the reader; at the moment of recognition, the reader strengthens the historical depth and intertextuality of Johnson’s figure, confirming this George’s rich heritage in Middle English romance. Michael Drayton, Poly-Olbion Johnson’s use of Bevis for St George seeks to minimize the religious potential of the narrative, but the figure of Bevis in Drayton’s Poly-Olbion (1612) takes in a new direction Spenser’s recasting of the displaced-youth narrative as that of a Christian exile in a distinctly Calvinist universe. Drayton’s Bevis remains a representative Christian for the reader’s emulation, but he is a far more active and self-aware figure than either the medieval Bevis or Redcrosse. Or, in contrast to Redcrosse, his muscular, bellicose Christianity is here the correct response – rather than the token of his deluded pride and presumed self-sufficiency. When Bevis in the medieval text is taunted by pagan knights for not knowing the significance of Christmas Day, he reveals the damage inflicted by his exile on his self-awareness; Drayton’s Bevis, however, seems fully cognizant of his Christian identity and religion. When he overhears ‘damn’d Panims … who in despightfull sort / Derided Christ the Lord’ (ii.314–15),36 he acts immediately and effectively in Christ’s defence:               … for his Redeemers sake He on those heathen hounds did there such slaughter make, That whilst in ther black mouthes their blasphemies they drue, They headlong went to hell.  (ii.315–18)

If Redcrosse had acted with such impulsive aggression (as he does with Sans ioy at the House of Pride), he would only have been exiled further into delusion and error. In Drayton’s text, though, such militancy is the proper response and a demonstration that the Christian Everyman can indeed act 36

Cf. n. 4 above; all references to Poly-Olbion are to Hebel’s edition.

188



Bevis: Renaissance Influence and Reception

with moral purpose and effectiveness – by contrast to Calvinism. Altogether, Drayton’s Bevis is far less dependent on God’s saving grace than either Redcrosse or, indeed, the Middle English Bevis. When, for example, he escapes from prison in Damascus (ii.359), the Middle English romance’s details of Bevis praying to Jesus and his chains miraculously falling off are omitted.37 Drayton emphasizes the intrinsic heroic virtue and ability of this earl’s son, despite the potential damage inflicted by exile. When Bevis is presented to the King of Armenia, the English hero a ‘wondrous mixture shew’d of grace and majestie’, with ‘more then man-like shape, and matchlesse stature’ (ii.304–5); Josian ‘Admir’d the god-like man’ (ii.311). Drayton’s Bevis is closer than earlier versions to that variation in the displaced-youth story, seen for example in Guiderius and Arviragus in Cymbeline, where the innate royalty of the displaced youths has not been tarnished by exile.38 The greater emphasis in Drayton’s Bevis on individual human struggle must have coincided with certain elements in the narrative to create a unique resonance for the text’s initial Jacobean readership. Bevis’s mother ‘Most wickedly seduc’t by the unlawfull love / Of Mordure … / [his] Sire to death disloyally had done’ (ii.262.4). George Buchanan’s account of King James’s mother, Mary Queen of Scots, offered a precisely parallel narrative, of a queen plotting and executing the murder of her husband (Lord Darnley) with her lover (the earl of Bothwell).39 Hamlet’s accusation to Gertrude – ‘almost as bad, good-mother, / As kill a king and marry with his brother’ (iii.iv.27–8) – could almost be Bevis’s words (omitting ‘brother’) to his mother – or, indeed, James’s to his. Even as Bevis exemplifies Drayton’s sense of the history embedded in the landscape, the text has the potential to resonate powerfully for its early seventeenth-century readership. John Bunyan, The Pilgrim’s Progress The intertextual relations between Bunyan’s Pilgrim’s Progress (Part One, 1678) and Bevis are not the final instance of post-medieval response to the romance, but only a suitable conclusion for an examination of some key sixteenth- and seventeenth-century creative adaptations. Later than Bunyan’s work, Bevis was still being printed and read in something remarkably close to its medieval form, as well as appearing in new prose versions.40 These texts combine basic fidelity to the original narrative with stylistic and descriptive embellishments in line with the development of seventeenth-century prose romance. The Famous and Renowned History of Sir Bevis of Southampton 37

The scene from Bevis possibly finds its analogue in FQ in Redcrosse’s imprisonment in Orgoglio’s dungeon, until rescued by Arthur, emblematic of grace. 38 Cf. Cymbeline, iii.iii.79–86. 39 G[eorge] B[uchanan], Ane Detectioun of the duinges of Marie Quene of Scottes (London, 1571), sigs D.iiv–E.iv. 40 Cf. p. 101 above.

189



Andrew King

(1689) presents new characters, such as Rosalinda, daughter of the sultan of Babylon and in love with her father’s enemy, Sir Miles. The dragon fight in this text presents Bevis as more self-aware and ingenious: rather than stumbling accidentally into the well, this Bevis ‘perceive[s] the Monster to make a stop some paces from it’ (p. 38), and he realizes that the well, ‘by reason of a Saintlike Virgins washing in it’ (ibid.), has restorative power. The 1691 version is more squeamish regarding the potential Catholic elements in the well, restricting itself to the cautious observation ‘it seems the Well was holy water’ (sig. Cir). Bevis was clearly available to Bunyan’s readers in a variety of forms and could serve as a context or intertext within which his own work could strategic­ally be placed. Spenser’s adaptation created a tension between, on the one hand, the romance mode as a certain kind of representation of human experience and, on the other, the nature of Holiness that lies at the end of the knight’s quest. Bunyan too employs Bevis and romance motifs in general ironically, invoking them only to encourage with great subtlety their repudiation. In ‘The Author’s Way of Sending Forth His Second Part of The Pilgrim’, which prefaced Part ii in 1684, one of the objections cited in relation to Part i hints at the complex relationship of the work with romance: ‘But some love not the method of your first [part], / Romance they count it.’41 This constructed devil’s advocate who opposes romance fiction outright has ironically reached a conclusion that the text supports, though he has done so without the insightful process that Bunyan’s handling of romance motifs invites. In A Few Sighs from Hell (1658), an extended, homiletic exposition of the parable of Dives and Lazarus, Bunyan confesses his youthful preference for romance over Scripture: ‘the Scriptures thought I, what are they? a dead letter, a little ink and paper, of three or four shillings price. Alas, what is the Scripture, give me a Ballad, a Newsbook, George on horseback, or Bevis of Southhampton.’42 ‘George on horseback’ – probably St George in a chapbook version of Johnson’s work43 – is juxtaposed to Bevis, echoing the association of the two dragon-slayers in both Spenser’s and Johnson’s work. Bunyan repents his proclivities as a juvenile reader: we were all of one spirit, loved all the same sins, slighted all the same counsels, promises, incouragements, and threatenings of the Scriptures; and they are still as I left them, still in unbelief, still provoking God, and rejecting good counsel, so hardned in their wayes, so bent to follow sin.44

This passage suggests that the romance elements in The Pilgrim’s Progress 41

John Bunyan, The Pilgrim’s Progress, ed. by N. H. Keeble, World’s Classics (Oxford: Oxford University Press, 1984), p. 139. 42 John Bunyan, A Few Sighs from Hell; or, The Groans of a damned Soul (London, 1658), pp. 156–7. 43 Cf. The Seven Champions of Christendom, ed. Fellows, p. xxvi. 44 Bunyan, A Few Sighs from Hell, p. 157.

190



Bevis: Renaissance Influence and Reception

are not simply a descriptive facilitator, but ensnarements that might lead the reader to ‘follow sin’, away from a reading experience that would be commensurate with Christian’s own progress towards the City of Zion. Like Spenser’s, Bunyan’s use of Middle English romance involves not simply the borrowing or even the adaptation of a literary tradition, but the location of romance elements within a Christological scheme that interrogates, and sometimes subverts, the romance tradition. Just as, in A Few Sighs from Hell, Bunyan presents Middle English romances (including Bevis) as enticingly attractive experiences that nevertheless lead to moral apathy and negligence, so too the landscape of The Pilgrim’s Progress is littered with romance elements that threaten to hinder Christian’s – and later, in Part ii, Christiana’s – progress. Among other things, The Pilgrim’s Progress is an account of the proper moral navigation of romance – ultimately a renunciation of or escape from the form – and each romance element in the text overcome by Christian can be likened to Bunyan’s own rejection of his youthful reading. The ‘inchanted Arbor, upon which, if a Man sits, or in which if a man sleeps, ’tis a question … whether ever they shall rise or wake again in this World’ (p. 248), reminiscent of the insidious trees and forests of Breton lais such as Sir Orfeo, as well as of Ormond’s garden in The Seven Champions, offers the same perilous delights of reading romance, which can lull the reader into a drowsy forgetfulness of the moral vigilance and muscular Christianity required in a fallen world. Apollyon is another monster indebted to Bevis’s dragon – ‘cloathed with scales like a Fish … Wings like a Dragon, feet like a Bear, and out of his belly came Fire and Smoak’ (p. 47). Like the other romance elements in the text, it must be overcome not just because it is a monster, but because it belongs to a mental and narrative landscape – romance – that is in opposition to the pilgrim’s proper sense of God’s intervention in and governance of all human affairs. The one romance element that remains supremely positive in Bunyan’s text is, of course, the notion of the quest.45 Even more than Redcrosse, Christian is on a quest with a strong sense of linear progression and forward momentum. But Bunyan’s use of the quest is not so much the revision of a romance motif as it is an assertion that all quests and journeys (in romance and in other forms) are appropriations or shadowy versions of the archetypal quest of the Christian life. Like Spenser, Bunyan employs Bevis and the romance tradition in general with the profound and ultimately subversive insight that this literary mode is a half-expressed version of the Christian narrative in which all readers are displaced youths – removed from their true home and seeking, however ineptly, to return to it.

45

See Cooper, The English Romance in Time, pp. 45–105, for a brilliantly insightful relating of quest and pilgrimage – vital for understanding Bunyan’s complex relation to the romance tradition.

191

Bibliography of Bevis Scholarship The bibliography lists editions, translations and studies of the versions of the Bevis story under discussion in this book, from 1838 to the present day. Texts and translations English Burnley, David, and Alison Wiggins (eds), The Auchinleck Manuscript ­(Edinburgh: National Library of Scotland, 5 July 2003), version 1.1 (http:// www.nls.uk/auchinleck/mss/ beues.html) Fellows, Jennifer L., ‘Sir Beves of Hampton: Study and Edition’, 5 vols (unpub. Ph.D. diss., Univ. of Cambridge, 1980). (Abstract in Index to Theses Accepted for Higher Degrees in the Universities of Great Britain and Ireland, 31 (1982), 285) Herzman, Ronald B.; Drake, Graham; Salisbury, Eve (eds), Four Romances of England: ‘King Horn’, ‘Havelok the Dane’, ‘Bevis of Hampton’, ‘Athelston’, Middle English Texts (Kalamazoo, MI: Medieval Inst., Western Michigan Univ., for TEAMS in assn with the Univ. of Rochester, 1999) Kölbing, Eugen (ed.), The Romance of Sir Beues of Hampton …, EETS, es 46, 48, 65 (London: Kegan Paul, Trench, Trübner & Co. for the EETS, 1885–94) McSparran, Frances, and P. R. Robinson (introd.), Cambridge University Library MS Ff.2.38 (London: Scolar Press, 1979) Pearsall, Derek, and I. C. Cunningham (introd.), The Auchinleck Manuscript: National Library of Scotland Advocates’ MS 19.2.1 (London: Scolar Press in association with the National Library of Scotland, 1977) Turnbull, W. B. D. D. (ed.), Sir Beves of Hamtoun: A Metrical Romance …, ­Maitland Club 44 (Edinburgh, 1838; repr. New York: AMS Press, 1973) Anglo-Norman Stimming, Albert (ed.), Der anglonormannische Boeve de Haumtone, Bibliotheca Normannica 7 (Halle: Niemeyer, 1899; repr. Geneva: Slatkine, 1974) Weiss, Judith (trans.), ‘Boeve de Haumtone’ and ‘Gui de Warewic’: Two AngloNorman Romances, French of England Translation Series (Tempe, AZ: ­Medieval and Renaissance Texts and Studies, 2008) Icelandic Cederschiöld, Gustaf (ed.), Fornsögur Suðrlanda: Magus saga jarls, Konraðs saga, Bærings saga, Bevers saga, Lunds universitets årsskrift 15 (Lund: [s.n.], 1884), pp. 209–67 193



Bibliography of Bevis Scholarship

Djurhuus, N. (ed.), ‘Bevusar tættir’, in Fóroya kvæði, Corpus Carminum Færoensium 5 (Copenhagen: Akademisk Forlag, 1968), pp. 309–12 Sanders, Christopher (ed.), Bevers saga, Stofnun Árna Magnússonar á Íslandi 51 (Reykjavík: Stofnun Árna Magnússonar á Íslandi, 2001)

Irish Robinson, F. N. (ed. and trans.), ‘The Irish lives of Guy of Warwick and Bevis of Hampton’, Zeitschrift für celtische Philologie, 6 (1908), 9–180, 273–338 Welsh Watkin, Morgan (ed.), Ystorya Bown de Hamtwn: cyfieithiad canol y drydedd ganrif ar ddeg, o, La geste de Boun de Hamtone (Cardiff: University of Wales Press, 1958) Williams, Robert (ed. and trans.), Selections from the Hengwrt MSS Preserved in the Peniarth Library, 2 vols in 3 (London: Richards, 1876–92), II, 119–88, 518–65 Studies Barnes, Geraldine, Counsel and Strategy in Middle English Romance (Cambridge: Brewer, 1993) Baugh, Albert C., ‘Convention and individuality in the Middle English romance’, in Medieval Literature and Folklore Studies: Essays in Honor of Francis Lee Utley, ed. by Jerome Mandel and Bruce A. Rosenberg (New Brunswick, NJ: Rutgers University Press, 1970), pp. 123–46 —— ‘Improvisation in the Middle English romance’, Proceedings of the ­American Philosophical Society, 103 (1959), 418–54 —— ‘The making of Beves of Hampton’, Library Chronicle, 40 (1974), 15–37 Becker, Philipp August, Beuve de Hantone, Berichte über die Verhandlungen der Sächsischen Akademie der Wissenschaften zu Leipzig, philologisch– ­historische Klasse 93:3 (Leipzig: Hirzel, 1941) Best, Debra E., ‘Villains and monsters: enacting evil in Beves of Hamptoun’, Medieval Perspectives, 13 (1998), 56–68 Bly Calkin, Siobhain, ‘The anxieties of encounter and exchange: Saracens and Christian heroism in Sir Beves of Hamtoun’, Florilegium, 21 (2004), 135–58 —— Saracens and the Making of English Identity: The Auchinleck Manuscript (London; New York: Routledge, 2005) Boje, Christian, Über den altfranzösischen Roman von Beuve de Hamtone, Beihefte zur Zeitschrift für romanische Philologie 19 (Halle: Niemeyer, 1909) Boone, Lalia Phipps, ‘Criminal law and the Matter of England’, Boston University Studies in English, 2 (1956), 2–16 Borrajo, Edward M., ‘Bevis de Hampton’, Notes and Queries, 8th series, 11 (1897), 258 Brantley, Jessica, ‘Images of the vernacular in the Taymouth Hours’, English Manuscript Studies, 1100–1700, 10 (2002), 83–113 194



Bibliography of Bevis Scholarship

Brownrigg, Linda, ‘The Taymouth Hours and the romance of Beves of Hampton’, English Manuscript Studies, 1100–1700, 1 (1989), 222–41 Brugger, E.: see Jordan, Leo Campbell, Kofi, ‘Nation-building colonialist-style in Bevis of Hampton’, ­Exemplaria, 18:1 (2006), 205–32 Chesnutt, Michael, ‘Aspects of the Faroese traditional ballad in the nineteenth century’, Scandinavian Yearbook of Folklore, 48 (1992), 247–59 —— ‘Bevussrímur and Bevusar tættir: a case study of Icelandic influence on Faroese balladry’, in Opuscula, vol. XII, ed. by Britta Olrik Frederiksen, Bibliotheca Arnamagnæana 44 (Copenhagen: Reitzels, 2005), pp. 399–437 Cohen, Jeffrey Jerome, Of Giants: Sex, Monsters, and the Middle Ages, Medieval Cultures 17 (Minneapolis, MN: University of Minnesota Press, 1999) Cooper, Helen, The English Romance in Time: Transforming Motifs from ­Geoffrey of Monmouth to the Death of Shakespeare (Oxford; New York: Oxford University Press, 2004) Crane, Ronald S., The Vogue of Medieval Chivalric Romance during the English Renaissance (Menasha, WI: University of Wisconsin Press, 1919) Crane, Susan, Insular Romance: Politics, Faith, and Culture in Anglo-Norman and Middle English Literature (Berkeley, CA: University of California Press, 1986) —— see also Dannenbaum, Susan Crawford, S. J., ‘Sir Bevis of Hamtoun’, Wessex, 1:3 (1930), 46–57 Curry, John T., ‘Arundel Castle legend’, Notes and Queries, 10th series, 8 (1907), 390 Dalrymple, Roger, Language and Piety in Middle English Romance (Cambridge: Brewer, 2000) —— ‘The literary use of religious formulae in certain Middle English romances’, Medium Ævum, 64 (1995), 250–63 Dannenbaum, Susan, ‘Anglo-Norman romances of English heroes: “ancestral romances”?’, Romance Philology, 35:4 (1981/2), 601–8 —— see also Crane, Susan Djordjević, Ivana, ‘Mapping medieval translation’, in Medieval Insular Romance: Translation and Innovation, ed. by Judith Weiss, Jennifer Fellows and Morgan Dickson (Cambridge: Brewer, 2000), pp. 1–23 —— ‘Mapping Medieval Translation: Methodological Problems and a Case Study’ (unpub. Ph.D. diss., McGill University, 2003) —— ‘Original and translation: Bevis’s mother in Anglo-Norman and Middle English’, in Cultural Encounters in the Romance of Medieval England, ed. by Corinne Saunders, Studies in Medieval Romance (Cambridge: Brewer, 2005), pp. 11–26 —— ‘Rewriting divine favour’, in Boundaries in Medieval Romance, ed. by Neil Cartlidge, Studies in Medieval Romance (Cambridge: Brewer, 2008), pp. 161–73 —— ‘Translating courtesy in a Middle English romance’, Studia Neophilologica, 76 (2004), 140–51 —— ‘Versification and translation in Sir Beves of Hampton’, Medium Ævum, 74:1 (2005), 41–59 Echard, Siân, ‘Of dragons and Saracens: Guy and Bevis in early print illustration’, in Guy of Warwick: Icon and Ancestor, ed. by Alison Wiggins and 195



Bibliography of Bevis Scholarship

Rosalind Field, Studies in Medieval Romance (Cambridge: Brewer, 2007), pp. 154–68 Edwards, A. S. G., ‘William Copland and the identity of printed Middle English romance’, in The Matter of Identity in Medieval Romance, ed. by Phillipa Hardman (Cambridge: Brewer, 2002), pp. 139–47 Endo, Hiroaki, ‘Middle English dialectology by a machine-readable corpus: with special reference to Sir Beues of Hamtoun’, Osaka daigaku gengo bunkagaku, 10 (2001), 5–23 Fellows, Jennifer, ‘Author, author, author …: an apology for parallel texts’, in A Guide to Editing Middle English, ed. by Vincent P. McCarren and Douglas Moffat (Ann Arbor, MI: University of Michigan Press, 1998), pp. 15–24 —— ‘Bevis redivivus: the printed editions of Sir Bevis of Hampton’, in Romance Reading on the Book: Essays on Medieval Narrative Presented to Maldwyn Mills, ed. by Jennifer Fellows, Rosalind Field, Gillian Rogers and Judith Weiss (Cardiff: University of Wales Press, 1996), pp. 251–68 —— ‘Editing Middle English romances’, in Romance in Medieval England, ed. by Maldwyn Mills, Jennifer Fellows and Carol M. Meale (Cambridge: Brewer, 1991), pp. 5–16 —— ‘Mothers in Middle English romance’, in Women and Literature in Britain, 1150–1500, ed. by Carol M. Meale, Cambridge Studies in Medieval Literature 17 (Cambridge: Cambridge University Press, 1993), pp. 41–60 —— ‘On the iconography of a carving in King’s College Chapel, Cambridge’, Journal of the Warburg and Courtauld Institutes, 39 (1976), 262 —— ‘Romance among the choir-stalls: Middle English romance motifs on English misericords’, in Profane Images in Marginal Arts of the Middle Ages, ed. by Elaine C. Block and Malcolm Jones (Turnhout: Brepols, forthcoming) —— ‘St George as romance hero’, Reading Medieval Studies, 19 (1993), 27– 54 —— ‘Sir Bevis of Hampton in popular tradition’, Proceedings of the Hampshire Field Club and Archaeological Society, 42 (1986), 139­–45 —— (ed.), The Seven Champions of Christendom (1596/7), Non-Canonical Early Modern Popular Texts (Aldershot; Burlington, VT: Ashgate, 2003) Field, Rosalind, ‘Romance in England, 1066–1400’, in The Cambridge History of Medieval English Literature, ed. by David Wallace (Cambridge: Cambridge University Press, 1999), pp. 152–76 —— see also Wadsworth, Rosalind Fleming, J. B., ‘Bevis de Hampton’, Notes and Queries, 8th series, 11 (1897), 258 Galent-Fasseur, Valérie, ‘Un médiateur de la Providence: le personnage de Sabaoth dans la version anglo-normande et la version en prose de Beuve de Hantone’, Littérales, 22 (1998), 25–38 —— ‘La tentation sarrasine de Beuve de Hantone’, in La Chrétienté au péril sarrasin: Actes du colloque de la section française de la Société Internationale Rencesvals (Aix-en-Provence, 30 septembre – 1er octobre 1999), Senefiance 46 (Aix-en-Provence: Centre universitaire d’études et de recherches médiévales d’Aix, Université de Provence, 2000), pp. 27–39 Gerritsen, Willem P., and Anthony G. van Melle (eds), A Dictionary of Medieval Heroes: Characters in Medieval Narrative Traditions and Their Afterlife in 196



Bibliography of Bevis Scholarship

Literature, Theatre and the Visual Arts, trans. by Tanis Guest (Woodbridge: Boydell Press, 1998), pp. 62–4 Greenfield, B. W., ‘The heraldry and exterior decorations of the Bargate, ­Southampton’, Papers and Proceedings of the Hampshire Field Club and Archaeological Society, 4:2 (1900), 139­–45. (With architectural notes by R. M. D. Lucas) Guy, John, Castles in Sussex (Chichester: Phillimore, 1984), pp. 20–9 Hanna, Ralph, III, London Literature, 1300–1380, Cambridge Studies in Medieval Literature (Cambridge: Cambridge University Press, 2005) —— ‘Unnoticed Middle English romance fragments in the Bodleian Library: MS Eng. poet. d.208’, Library, 21:4 (1999), 305–20 Herlyn, Anne, ‘Narrative Strukturen mittelenglischer Romanzen: am Beispiel des Beves of Hamptoun’, in Authors, Heroes, and Lovers: Essays on Medieval English Literature and Language. Selected Papers from the Studientage zum englischen Mittelalter, SEM I & II (Potsdam 1999 & 2000) / Liebhaber, Helden und Autoren: Studien zur alt- und mittelenglischen Literatur und Sprache. Ausgewählte Beiträge der Studientage zum englischen Mittelalter, SEM I & II (Potsdam 1999 & 2000), ed. by Thomas Honegger, Variations 2 (Bern [etc.]: Lang, 2001), pp. 1–26 Hibbard, Laura A., ‘Jacques de Vitry and Boeve de Haumtone’, Modern Language Notes, 34 (1919), 408–11 —— Mediæval Romance in England: A Study of the Sources and Analogues of the Non-Cyclic Metrical Romances (New York: Oxford University Press, 1924), pp. 115–26 —— ‘The Nibelungenlied and Sir Beves of Hampton’, Modern Language Notes, 26:5 (1911), 159–60 Horrent, Jacques, ‘Bueve de Hantone et la Condesa traidora’, Lettres romanes, 34:1 (1982), 41–57 Hoyt, Prentiss C., ‘The home of the Beves saga’, PMLA, 17 (1902), 237–46 Hunt, Stephen, ‘Further translation errors in Bevers saga’, Notes and Queries, 32 (1985), 455–6 Hussey, Arthur, ‘Bevis de Hampton’, Notes and Queries, 8th series, 11 (1897), 207 Jacobs, Nicolas, ‘Sir Degarré, Lay le Freine, Beves of Hamtoun and the ­“Auchinleck bookshop” ’, Notes and Queries, 29 (1982), 294–301 Jones, Nesta, ‘Copi ychwanegol o ddechrau Ystorya Bown de Hamtwn’, Bulletin of the Board of Celtic Studies, 23:1 (1968), 17–26 Jordan, Leo, Über Boeve de Hanstone, Beihefte zur Zeitschrift für romanische Philologie 14 (Halle: Niemeyer, 1908). Reviewed by E. Brugger in Zeitschrift für französische Sprache und Literatur, 34:2 (1909), 25–40 King, Andrew. ‘The Faerie Queene’ and Middle English Romance: The Matter of Just Memory, Oxford English Monographs (Oxford: Clarendon Press, 2000) Kölbing, E., ‘Die Alliteration in Sir Beues of Hamtoun’, Englische Studien, 19 (1894), 441–53 Lister, Wilfred, ‘Sir Bevis of Hampton: A Study of the Vogue and Successive Transformations of This Popular English Romance from the Middle Ages to the Present Day’ (unpub. M.Phil. diss., Univ. of Southampton, 1971). (Abstract in Index to Theses Accepted for Higher Degrees in the Universities of Great Britain and Ireland, 21 (1970/71), 340) 197



Bibliography of Bevis Scholarship

Lower, Mark Antony, ‘Sir Bevis of Hampton and his horse “Arundel” ’, Sussex Archaeological Collections, 4 (1851), 31–6 Martin, Jean-Pierre, ‘Beuve de Hantone entre roman et chanson de geste’, ­Littérales, 31 (2003), 97–112 Martin, Judith E., ‘Studies in Some Early Middle English Romances’ (unpub. Ph.D. diss., Univ. of Cambridge, 1968), pp. 91–187 —— see also Weiss, Judith Matzke, John E., ‘The legend of Saint George: its development into a roman d’aventure’, PMLA, 19 (1904), 449–78 —— ‘The oldest form of the Beves legend’, Modern Philology, 10 (1912/13), 19–54 Meale, Carol M., ‘Caxton, de Worde, and the publication of romance in late medieval England’, Library, 14 (1992), 283–98 Mehl, Dieter, The Middle English Romances of the Thirteenth and Fourteenth Centuries (London: Routledge & Kegan Paul, 1968), pp. 211–20 Menuge, Noël James, ‘A few home truths: the medieval mother as guardian in romance and law’, in Medieval Women and the Law, ed. by Noël James Menuge (Woodbridge: Boydell Press, 2000), pp. 77–103 —— ‘The wardship romance: a new methodology’, in Tradition and Transformation in Medieval Romance, ed. by Rosalind Field (Cambridge: Brewer, 1999), pp. 29–43 Metlitzki, Dorothee, The Matter of Araby in Medieval England (New Haven, CT; London: Yale University Press, 1977) Mills, Maldwyn, ‘Structure and meaning in Guy of Warwick’, in From Medieval to Medievalism, ed. by John Simons, Insights (Basingstoke: Macmillan, 1992), pp. 54–68 Nolsøe, Eftir Mortan, ‘Ein “rímnaflokkur” í føroyskari tungulist?’, ­Fróðskaparrit, 24 (1976), 46–70 Peacock, Edward, ‘Bevis de Hampton’, Notes and Queries, 8th series, 11 (1897), 396 Poppe, Erich, ‘Adaption und Akkulturation: narrative Techniken in der mittel­ kymrischen Ystorya Bown de Hamtwn’, in Übersetzung, Adaptation und Akkulturation im insularen Mittelalter, ed. by Erich Poppe and Hildegard L. C. Tristram, Studien und Texte zur Keltologie 4 (Münster: Nodus, 1999), pp. 305–17 —— ‘Beues of Hamtoun in Welsh bardic poetry’, Cambrian Medieval Celtic Studies, 43 (2002), 49–57 —— ‘Codes of conduct and honour in Stair Bibuis’, in Ogma: Essays in Celtic Studies in Honour of Próinséas Ní Chatháin, ed. by Michael Richter and JeanMichel Picard (Dublin; Portland, OR: Four Courts Press, 2002), pp. 200–10 —— ‘The Early Modern Irish version of Beves of Hamtoun’, Cambridge Medieval Celtic Studies, 23 (1992), 77–98 —— ‘Narrative structure of medieval Irish adaptations: the case of Guy and Beues’, in Medieval Celtic Literature and Society, ed. by Helen Fulton (Dublin: Four Courts Press, 2005), pp. 205–29 —— ‘Owein, Ystorya Bown and the problem of “relative distance”: some methodological considerations and speculations’, Arthurian Literature, 21 (2004),73–94 198



Bibliography of Bevis Scholarship

—— ‘The progressive in Ystorya Bown de Hamtwn’, in Yr Hen Iaith: Studies in Early Welsh, ed. by Paul Russell, Celtic Studies Publications 7 (Oxford: Oxbow, 2003), pp. 145–69 —— ‘Stair Nuadit Fin Femin: eine irische Romanze?’, Zeitschrift für celtische Philologie, 49/50 (1997), 749–59 —— and Regine Reck, ‘A French romance in Wales: Ystorya Bown o Hamtwn: processes of medieval translations’, Zeitschrift für celtische Philologie, 55 (2006), 122–80 —— and Regine Reck, ‘Ein mittelalterlicher bestseller in Wales: zu Erforschung von Prozessen der literarischen Übersetzung und des transkulturellen Texttransfers’, Marburger UniJournal, 12 (2002), 37 Prideaux, W. F., ‘Bevis Marks’, Notes and Queries, 8th series, 11 (1897), ­385–6 Ramsey, Lee C., Chivalric Romances: Popular Literature in Medieval England (Bloomington, IN: Indiana University Press, 1983) Rance, Adrian B., ‘The Bevis and Ascupart panels, Bargate Museum, ­Southampton’, Proceedings of the Hampshire Field Club and Archaeological Society, 42 (1986), 147–53 Reck, Regine, ‘Heiligere Streiter und keuschere Jungfrauen: religiöse Elemente in der kymrischen Adaption des anglo-normannischen Boeve de Haumtone’, in Übersetzung, Adaptation und Akkulturation im insularen Mittelalter, ed. by Erich Poppe and Hildegard L. C. Tristram, Studien und Texte zur Keltologie 4 (Münster: Nodus, 1999), pp. 289–304 Robinson, J. N., ‘Celtic versions of Sir Beues of Hamtoun’, Englische Studien, 24 (1898), 463–4 Rouse, Robert, ‘English identity and the law in Havelok the Dane, Horn Childe and Maiden Rimnild and Beues of Hamtoun’, in Cultural Encounters in the Romance of Medieval England, ed. by Corinne Saunders, Studies in Medieval Romance (Cambridge: Brewer, 2005), pp. 69–83 —— ‘Expectation vs experience: encountering the Saracen Other in Middle English romance’, SELIM, 10 (2000), 125–40 —— The Idea of Anglo-Saxon England in Middle English Romance, Studies in Medieval Romance (Cambridge: Brewer, 2005) Sanders, Christopher, ‘Bevers saga et la chanson anglo-normande Boeve d’Haumtone’, Revue des langues romanes, 102:1 (1998), 25–44 Saunders, Corinne, ‘Bodily narratives: illness, medicine and healing in Middle English romance’, in Boundaries in Medieval Romance, ed. by Neil Cartlidge, Studies in Medieval Romance (Cambridge: Brewer, 2008), pp. 175–90 —— ‘Desire, will and intention in Sir Beues of Hamtoun’, in The Matter of Identity in Medieval Romance, ed. by Phillipa Hardman (Cambridge: Brewer, 2002), pp. 29–42 —— ‘Love and loyalty in Middle English romance’, in Writings on Love in the English Middle Ages, ed. by Helen Cooney, Studies in Arthurian and Courtly Cultures (Basingstoke: Palgrave Macmillan, 2006), pp. 45–61 —— Rape and Ravishment in the Literature of Medieval England (Cambridge: Brewer, 2001) Schendl, Herbert, ‘ME. randon in Sir Bevis of Hampton’, Anglia, 102:1/2 (1984), 101–7 Seaman, Myra, ‘Engendering genre in Middle English romance: performing the feminine in Sir Beves of Hamtoun’, Studies in Philology, 98:1 (2001), 49–75 199



Bibliography of Bevis Scholarship

Shimogasa, Tokuji, ‘Popular expressions in The Romance of Sir Beues of Hamtoun’, in Medieval Heritage: Essays in Honour of Tadahiro Ikegami, ed. by Masahiko Kanno et al. (Tokyo: Yushodo Press, 1997), pp. 349–67 Skeat, Walter W., et al., ‘Arundel Castle legend’, Notes and Queries, 10th series, 8 (1907), 434–5 Spector, Sheila Abbye, ‘Studies in the Bovo Buch and Bevis of Hampton’ (unpub. Ph.D. diss., Univ. of Maryland, 1976). (Abstract in Dissertation Abstracts International, 38 (1977), 3523a.) Surridge, Marie E., ‘The number and status of Romance words attested in Ystorya Bown de Hamtwn’, Bulletin of the Board of Celtic Studies, 32 (1985), 68–78 Wadsworth, Rosalind, ‘Historical Romance in England: Studies in Anglo-Norman and Middle English Romance’ (unpub. D.Phil. diss., Univ. of York, 1972) —— see also Field, Rosalind Watkin, Morgan, ‘Albert Stimming’s Welsche Fassung in the Anglonormannische Boeve de Haumtone: an examination of a critique’, in Studies in French Language and Mediæval Literature Presented to Professor Mildred K. Pope, by Pupils, Colleagues and Friends (Manchester: Manchester University Press, 1939), pp. 371–9 Weiss, Judith, ‘The Anglo-Norman Boeve de Haumtone: a fragment of a new manuscript’, Modern Language Review, 95 (2000), 305–10 —— ‘ “The courteous warrior” ’: epic, romance and comedy in the making of Boeve de Haumtone’, in Boundaries in Medieval Romance, ed. by Neil Cartlidge, Studies in Medieval Romance (Cambridge: Brewer, 2008), pp. 149–60 —— ‘The date of the Anglo-Norman Boeve de Haumtone’, Medium Ævum, 55:2 (1986), 237–41 —— ‘Insular beginnings: Anglo-Norman romance’, in A Companion to Romance: from Classical to Contemporary, ed. by Corinne Saunders, Blackwell ­Companions to Literature and Culture (Oxford: Blackwell, 2004), pp. 26–44 —— ‘The major interpolations in Sir Beues of Hamtoun’, Medium Ævum, 48:1 (1979), 71–6 —— ‘The power and the weakness of women in Anglo-Norman romance’, in Women and Literature in Britain, 1150–1500, ed. by Carol M. Meale, Cambridge Studies in Medieval Literature 17 (Cambridge: Cambridge University Press, 1993), pp. 7–24 —— ‘The wooing woman in Anglo-Norman romance’, in Romance in Medieval England, ed. by Maldwyn Mills, Jennifer Fellows and Carol M. Meale (Cambridge: Brewer, 1991), pp. 149–61 —— see also Martin, Judith E. Wilson, Anne, Plots and Powers: Magical Structures in Medieval Narrative (Gainesville [etc.], FL: University Press of Florida, 2001) Wolf-Bonvin, Romaine, ‘Escopart, le géant dépérissant de Beuve de Hantone’, in La Chrétienté au péril sarrasin: Actes du colloque de la section française de la Société Internationale Rencesvals (Aix-en-Provence, 30 septembre – 1er octobre 1999), Senefiance 46 (Aix-en-Provence: Centre universitaire d’études et de recherches médiévales d’Aix, Université de Provence, 2000), pp. 249– 65 Zaerr, Linda Marie, ‘Medieval and modern deletions of repellent passages’, in Improvisation in the Arts of the Middle Ages and Renaissance, ed. by Timothy 200



Bibliography of Bevis Scholarship

J. McGee, Early Drama, Art, and Music Monographs 30 (Kalamazoo, MI: Medieval Institute, 2003), pp. 222–46 —— ‘Meter change as a relic of performance in the Middle English romance Sir Beues’, Quidditas, 21 (2000), 105–26 Zenker, Rudolf, Boeve–Amlethus: das altfranzösische Epos von Boeve de Hamtone und der Ursprung der Hamletsage, Literarische Forschungen 32 (Berlin; Leipzig: Felber, 1905)

201

Index

Ælfric  177 Æthelberht of Kent  34–5 Akbari, Suzanne Conklin  119 Alde, John  111 Aliscans  147, 156 Andrea da Barberino I reali di Francia  5 Antioch, siege of  116 Ari the Learned  60 Arthur, King  184 Arundel Castle  179 Arundel, earls of (d’Albini family)  2, 35–6 Arundel(a) (horse)  33, 53, 146–7, 149, 166, 174, 175, 179 Ascopard/Escopart (character)  19, 54, 85, 89, 145–60, 168–9, 173 Ascham, Roger  176 Aucassin et Nicolette  85, 87 Bachur, Elia Levita  5 Bacon, Roger  153 Bale, John  177, 178 baptism: see under Christianity Barclay, Alexander  182 Bargate, Southampton  145 Barnes, Geraldine  57, 167, 171, 174 Bede Historia ecclesiastica  34–5 Bergr Sokkason  63 Bevers (character): see Bevis/Boeve/ Bown/Bibus (character) Bevers saga  4, 46 n. 38, 48, 51–66 dating  52, 65 dialogue  54–5 genre  56 manuscripts  64–5 narrator  65–6 style  52–3, 62–3 translation  52–3, 58 Bevis/Boeve/Bown/Bibus/Bevers (character) as Christian knight  39, 55, 84, 89–90, 96, 120, 127, 142, 161, 165–7, 175



as folk hero  1, 161–2 characterization, 54, 96, 175 English identity  3, 34–6, 84, 114, 116, 117–18, 120, 121–2, 124, 125, 188 Bevis story European scholarship  2–3 folk motifs  163 genre  55 narrative motifs  161 oral dissemination  12, 91, 93 versions Anglo-Norman: see Boeve de Haumtone Dutch  4 English: see Sir Bevis of Hampton Faroese: see Bevusar tættir / Bevusar ríma French (Continental)  4, 9, 23–4; see also Bueve de Hamtone Icelandic: see Bevers saga Irish: see Stair Bibuis Italian: see Buovo d’Antona Romanian  5 Russian  6 Welsh: see Ystorya Bown o Hamtwn Yiddish: see Bovo-Buch Bevusar tættir / Bevusar ríma  4 Bibus (character): see Bevis/Boeve/Bown/ Bevers (character) Bishop, Richard  112 Bly Calkin, Siobhain  114, 115, 116, 118, 121 Boeve (character): see Bevis/Bown/Bibus/ Bevers (character) Boeve de Haumtone  3, 9–24, 25, 32, 38–44, 46 n. 38, 54, 61, 67–79, 80, 81–4, 115, 127, 128–9, 132, 133, 135, 139, 141, 143, 145–9, 157–8 as ancestral romance  9 date  9–10, 25 epithets  76, 77–9 genre  10, 13–14 humour  53, 54, 150 laisse structure  12, 13, 14–17, 24 manuscripts  10 n. 5

203



Index colonialism  121 Constance story  163 Cooper, Helen  169, 171, 173 Copland, William  1, 91, 97–8, 99, 101, 111 Crane, Susan  9, 10, 115, 122, 124 Cronin, Michael  39 Crook, Andrew  112–13 Crusades  116, 118, 121, 132–3, 162

prosody  75–6 rhetoric  9, 12, 13, 15–17, 19, 24 Böldl, Klaus  59 Bonefas (character)  155, 173 Boulton, Maureen B. M.  9–10 Bovo-Buch  5 Bown (character): see Bevis/Boeve/Bibus/ Bevers (character) Bown o Hamtwn: see Ystorya Bown o Hamtwn Brad(e)mo(u)nd, King (character)  20, 71, 72, 78, 131, 132, 133, 136–7, 138, 147, 153, 166, 169, 170 Brewer, Derek  161–2 Buchanan, George  189 Bueve de Hamtone  82 Bunyan, John  177, 178 A Few Sighs from Hell  190–1 The Pilgrim’s Progress  189–91 Buovo d’Antona  4–6 operatic adaptation  6 Calvinism  183–4, 186, 188–9 Campbell, Kofi  117, 118, 121, 124, 125 Caxton, William  182 Recuyell of the Historyes of Troye  45 chanson de geste  9–24, 52, 75 Chanson de Roland  14, 37, 57, 64 Charlemagne  37, 45 Karlamagnúss saga  57, 62, 63–4, 66 Pelèrinage de Charlemagne  37 Chaucer, Geoffrey  155, 177, 186 The Canterbury Tales Knight’s Tale  155 ‘Sir Thopas’  69–70, 75 Wife of Bath’s Prologue  164 Chevalerie Ogier de Danemarche, La  22 Chrétien de Troyes Erec et Enide  57 Perceval  57, 58 Yvain  11, 57, 152–3, 167 Christianity  24, 54, 83, 117, 118, 119, 121, 146, 148, 150, 163, 165, 169, 185, 188–9, 191 baptism  150–1, 168, 183 Christian Church  177–8 prayer  20–1, 71, 82, 139, 141, 142, 167, 168, 189 Protestantism  143, 179 Christina of Markyate  171 Cicero  56 Civile (place) 72–3 Cohen, Jeffrey Jerome  114, 119, 152, 156

d’Albini family: see Arundel, earls of Daniel of Beccles Liber urbanus  27 Dares Phrygius  37, 45 Daurel et Beton  3 n. 15 de Worde, Wynkyn  1, 97–8, 99, 101, 109–10 Dean, Ruth J.  9–10 Djordjević, Ivana  68 n. 3, 163–4 Doun, Emperor (character)  148 Drayton, Michael Poly-Olbion  176–7, 188–9 Duby, Georges  26, 32 East, Thomas  91, 97, 98, 100, 111 Echard, Siân  99 Edgar, King (character)  30, 122–3, 124, 125, 161, 166–7, 185–6 Eneborc (character)  35 Erasmus, Desiderius Enchiridion  90 Erle of Tolous, The  74 Ermin(e)/Hermine/Erminríkr, King (character)  30, 54, 78, 128–9, 130–1, 155, 165, 169, 170 Escopart: see Ascopard/Escopart (character) Fellows, Jennifer  46, 67, 68, 131, 164, 171 Field, Rosalind  122 Fierabras  22, 24, 147–8 Floris and Blaunchefleur  173 Fouke le Fitz Waryn  10 Foxe, John  178, 182, 186, 187 Actes and Monuments  177 Frye, Northrop  84 Gaimar, Geffrei  35 Garcy (character)  173 George, St  83, 89–90, 115–16, 140, 179, 182, 185, 188, 190 Glossa ordinaria  56 Goldoni, Carlo  6

204



Index

Gourlaw, Robert  111 Gower, John 177 Gra(u)nder/Grainnder (character)  47, 167 Gruffudd Bola  44 Gui de Warewic  10, 25, 31–3 Guillaume d’Orange  14, 22, 148 Guy, Earl (Bevis’s father) 163, 164 Guy of Warwick  1–2, 45, 91, 116–17, 118, 122, 162, 168 Humanist/Puritan attacks, 1 Guy story  145 versions Anglo-Norman: see Gui de Warewic English: see Guy of Warwick Irish  3 Gvion (Bevers’ son) 55 hagiography  140, 142, 143, 162, 171, 174 Hartman, A. Richard  24 Haveloc: see Lai d’Haveloc Havelok the Dane  74, 184 Heffernan, Carol F.  165, 166 Heinemann, Edward A.  14 n. 22 Heng, Geraldine  114, 115, 122 Henry of Huntingdon Historia Anglorum  35 Hermine: see Ermin(e) Horn: see King Horn; Roman de Horn Hue de Rotelande Ipomedon  25, 28–9, 32 Protheselaus  30 Hunt, Tony  11 Ibbitson, A.  113 Isidore of Seville  119 Islam  54, 150, 167; see also Saracens Itinerarium peregrinorum  118 Ivorius: see Yvor(i)/Ivorius (character) Jacobus de Voragine  182 James I and VI, King of England and Scotland  189 Jean de Joinville  118 John of Worcester Chronicle  35 Johnson, Richard The Seven Champions of Christendom 91, 179, 187–8, 190, 191 Jones, Catherine  14 n. 22 Josian(e)/Josivena/Sisian (character)  19, 33, 45, 53, 76–7, 84–5, 89, 96, 121, 148, 149, 150, 153, 158, 162–3, 169–75 Judith story  171

Juliana, St  140 Kalinke, Marianne  57–8 Karlamagnúss saga: see under Charlemagne Karras, Ruth Mazo  156–7 King Horn  173 Knight of Curtesy, The  81 Kölbing, Eugen  80, 94 Kramarz-Bein, Susanne  58 Lai d’Haveloc  10 Lancelot  116, 172 Lavezzo, Kathy  114 Lee, W.  112 Legge, M. Dominica  9, 24 Lise Præstgaard Andersen  62 n. 43 Lollards  130, 143 Lombards  123 London  123–4, 125 Lybeaus Desconus  74, 164, 184, 185 Lydgate, John  182 Mabinogion  40, 41 Mac an Leagha, Uilliam  45–6, 47 Malory, Thomas, Sir Morte Darthur  163, 185 manuscripts Aberystwyth, National Library of Wales MS Peniarth 4–5 (Llyfr Gwyn Rhydderch / White Book of Rhydderch)  38–9 Cambridge, Gonville and Caius College MS 175/96  89, 105–6, 152 Cambridge, Trinity College MS O.2.13  90–1, 92, 107 Cambridge University Library MS Ff.2.38  46, 81, 87–90, 91 n. 55, 108, 127, 130, 132, 134–7, 141, 143 Copenhagen, Arnamagnæanske Samling 544 4to (Hauksbók)  64–5 764 4to 64–5 Dublin, Trinity College MS 1298 (olim H.2.7)  45 Edinburgh, National Library of Scotland MS Advocates’ 19.2.1 (Auchinleck MS)  80–5, 104–5, 114–15, 127, 135, 139–40, 143, 154, 161–75, 178–9

205



Index

Firmin Didot MS (passed from possession of Paris bookseller Firmin Didot, via Librairie Labitte and Hermann Suchier, to University of Louvain before Second World War; subsequently destroyed by enemy action)  10 n. 5 London, British Library MS Add. 30512  45 MS Egerton 2862  87, 106 Manchester, Chetham’s Library MS 8009  46, 89–90, 93–5, 108, 123, 127, 152 Naples, Biblioteca Nazionale MS XIII.B.29  86–8, 105 Oxford, Bodleian Library MS Eng. Poet. d.208  93, 107 Oxford, Jesus College MS 11 (Llyfr Coch Hergest / Red Book of Hergest)  38–9 Paris, Bibliothèque Nationale MS Nouv. acq. fr. 4532  10 n. 5 Trinity College Dublin MS 1298  45 Margaret, St  140 Marie de France  163, 164 marriage  22–3, 46–7, 170, 175 malmariée  23, 163, 164 Marshal, William  26–7 Marshe, Thomas  110 Martin, Jean-Pierre  12 Mary of Egypt, St  37 Mary, Queen of Scots  189 Maugis d’Aigremont  14, 22 Melvill, David  112 Miles (Bevis’s son)  124, 125, 167 Miles, Earl/Milis, Iarla (character)  45, 76, 153, 156, 171 Nashe, Thomas  176–7 New Philology  50 Nibelungenlied  171 n. 33 Nicol, James  99, 113 Notary, Julian  97, 99, 110 Octavian  184 Ogier the Dane: see Chevalerie Ogier de Danemarche, La Old Norse literature biblical exegesis  55–6, 61 biblical translation  60–1 bishops’ sagas  61 hagiography  61 historiography  58–60

Íslendigasögur  59 kings’ sagas  59, 61 pseudo-historical sagas  62, 66 riddarasögur  51, 57, 61–3, 65, 66 translations from Latin  52 Ottoman Empire  133 Parcevals saga  58 Partonope of Blois  172 Pearsall, Derek  123, 124 Pelèrinage de Charlemagne: see under Charlemagne Poppe, Erich  65 postcolonialism  118 prayer: see under Christianity Preben Meulengracht Sørensen  59 Protestantism: see under Christianity pseudo-Turpin Historia  37 Puttenham, George The Arte of English Poesie  91 Pynson, Richard  97–8, 110 quest narrative  191 Raban, Edward  112 Ramsey, Lee C.  171 Reck, Regine  65 regionalism  122, 125 Revelation, Book of  182, 187 Richard Coer de Lyon  115, 118, 122 Roman de Horn  10, 22, 24, 25, 27–8, 31, 32 Roman de Silence  85, 87 Rychner, Jean  11, 12, 13–14 Saber/Sabot/Sabaoth (character)  25, 29–35, 78, 85, 148, 165 Sabert/Sæberht, King  34 Saracens  19–20, 24, 117, 118, 120–1, 126, 127, 136–7, 148, 150, 152, 153–5, 159, 163, 165 Selden, John  176–7, 178, 179 Shakespeare, William Cymbeline  189 Hamlet  189 King Henry VIII  176 The Tempest  163 The Winter’s Tale  163 Short, Ian  35 Shurley, John  101 Sir Bevis of Hampton  25, 33, 47, 48, 80–113, 114–26, 127–44, 149–60, 161–75, 176–91

206



boar fight  82, 167 chapbooks  101 Christmas Day battle  89, 185, 188 Damascus episode  89, 133–43, 166 dragon fight  71, 83, 91 n. 55, 115– 16, 153, 157, 167, 168, 179–84 epithets  77–9 formulae  69–72, 75, 79 French vocabulary  82, 85 gender roles/stereotypes  161, 162, 167, 169, 170, 174–5 Humanist/Puritan attacks  1, 176–7 lions 72, 89, 169, 170 London battle  84, 167, 178–9 magic  173 manuscripts  80–95, 104–8 medicine  171–3 mother (Bevis’s)  163, 164, 166 nationalism  3, 125, 126, 161 printed texts  83, 94–101, 109–13, 123, 127, 130, 131, 132, 134–6, 137–8, 143, 158 woodcuts  99–101 prose versions  101, 189–90 text  80–113, 159 translation  67–79 Sir Degaré  81, 164, 168 Sir Eglamour of Artois  74 Sir Orfeo  191 Sir Perceval of Galles  164, 173, 184 Sisian: see Josian(e)/Josivena/Sisian (character) Snodham, T.  98, 111 Snorri Sturluson  59–60 Southern, R. W.  60 Sowdon of Babylon, The  118 Spenser, Edmund  178, 191 The Faerie Queene  83, 91, 177, 179–84, 187 Stair Bibuis  3–4, 45–50, 83 dialogue  48 hospitality 46, 47 manuscript  45 marriage  46–7 narrative structure  48–50 narrator  48 rhetoric 47 source  46 style  47–8 texts  45 translation  49–50 women  46–7

Index Stair Nuadat Find Femin  45 Stansby, William  101, 102, 112 Sturla Þorðarson  60 Suard, François  9, 13–14, 21 Terri (character)  31, 81–2, 132, 133, 152, 157, 159, 160 Theodoricus monachus  59 Thomas Tristan  57 Tolan, John 153 Toury, Gideon  69 Traetta, Tommaso  6 Tristan story  164, 172 Turville-Petre, Thorlac  114–15 Tysdale, John  111 Virgil Aeneid  45 virginity  153, 170–1, 174, 175 Wace Roman de Brut  33 Roman de Rou  27, 31 Wade  116 Waldef  10, 32 Warren, Michelle  122 Warton, Thomas  182 Warwick, earls of  2 Weiss, Judith  10 n. 5, 52, 145–6, 162 n. 2 William of Malmesbury Gesta regum Anglorum  35 William of Palerne  45 Wolf-Bonvin, Romaine  155, 158 women  46, 169, 172 Würth, Stefanie  62 Wyclif, John  186 Ystorya Bown o Hamtwn  4, 37–45, 48, 49–50 manuscripts 38–9 narrative structure  39 narrator  41 style  38–40, 43 texts  38 translation  39–45 Yvor(i)/Ivorius (character)  19, 53, 54, 71–2, 73, 74, 76, 83, 89, 148, 153, 154, 166, 167, 170, 173 Ywain and Gawain  173

207