The Legend of Charlemagne in Medieval England: The Matter of France in Middle English and Anglo-Norman Literature 1843844729, 9781843844723

Citation preview

The Legend of Charlemagne in Medieval England

Bristol Studies in Medieval Cultures issn 1757-2150

Series Editor Dr Peter Dent Editorial Advisory Board Dr Marianne Ailes Dr Rhiannon Daniels Professor Helen Fulton Dr Emma Hornby Professor Carolyn Muessig Dr Benjamin Pohl Professor Ad Putter Dr Leah Tether Dr Ian Wei Dr Beth Williamson The remit of this peer-reviewed interdisciplinary series is to publish scholarly works on the cultures of the Middle Ages, from late antiquity up to and including the beginning of the sixteenth century. Queries about the series, or proposals for monographs, editions, or collections of essays, should be sent in the first instance to the Director of the Centre for Medieval Studies, who acts as General Editor of the series, at the address below. Centre For Medieval Studies, Faculty of Arts, 3–5 Woodland Road, Bristol, bs8 1tb email: [email protected] Previously published titles are listed at the back of this volume

The Legend of Charlemagne in Medieval England The Matter of France in Middle English and Anglo-Norman Literature

Phillipa Hardman and Marianne Ailes

d. s. brewer

© Phillipa Hardman and Marianne Ailes 2017 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 The right of Phillipa Hardman and Marianne Ailes to be identified as the authors of this work has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988 First published 2017 D. S. Brewer, Cambridge ISBN 978 1 84384 472 3 D. S. Brewer is an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge, Suffolk IP12 3DF, UK and of Boydell & Brewer Inc. 668 Mt Hope Avenue, Rochester, NY 14620–2731, USA website: www.boydellandbrewer.com A catalogue record for this book is available from the British Library The publisher has no responsibility for the continued existence or accuracy of URLs for external or third-party internet websites referred to in this book, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate This publication is printed on acid-free paper

Contents List of Illustrations  vi General Preface Charlemagne: A European Icon  vii Preface  xi Acknowledgements  xiii List of Abbreviations  xiv Introduction Charlemagne in England: Owning the Legend  1 1 Acculturating Charlemagne: The Insular Literary Context  32 2 Charlemagne ‘Translated’: The Anglo-Norman Tradition  110 3 Charlemagne ‘Appropriated’: The Middle English Tradition  156 4 Re-Imagining the Hero: The Insular Roland and the Battle of Roncevaux  221 5 Re-Presenting Otherness: The Insular Fierabras Tradition  264 6 Re-Purposing the Narrative: The Insular Otinel Tradition  346 Conclusion The Insular Afterlife of the Matter of France  402 Appendix

The Corpus: Texts and Manuscripts  412

Bibliography  419 Index  461

v

Illustrations









1 British Library Egerton MS 3028, fol. 80r © The British Library Board Fierabras takes the Passion relics from St Peter’s

127

2 British Library Egerton MS 3028, fol. 81r © The British Library Board Laban flees Rome with the relics

128

3 British Library Egerton MS 3028, fol. 118r © The British Library Board Charlemagne sailing home with the relics

129

4 British Library Egerton MS 3028, fol. 41r © The British Library Board Arthur bears the three lions passant guardant of England Frolle bears France modern: three fleurs de lis

142

5 British Library Egerton MS 3028, fol. 83v © The British Library Board Charlemagne bearing France ancient: semé of fleurs de lis

143

The authors and publishers are grateful to all the institutions and individuals listed for permission to reproduce the materials in which they hold copyright. Every effort has been made to trace the copyright holders; apologies are offered for any omission, and the publishers will be pleased to add any necessary acknowledgement in subsequent editions.

vi

general preface Charlemagne: A European Icon

T

his series of volumes examining the reception of the Charlemagne myth in different linguistic cultures of medieval Europe springs from ‘Charlemagne in England’, a project supported by the Arts and Humanities Research Council to study the literary presence of the emperor in medieval England, an area where the historical Charlemagne never set foot, let alone reigned. The spread of Charlemagne’s myth after his death was even more extensive than was his empire during his life. This larger enterprise, therefore, an investigation of the appropriation of the matter of Charlemagne across Europe, required a network of specialists working on texts written in different languages and different geographical areas. Yet these languages were culturally interdependent: it was largely through the medium of French, with its cultural hegemony, that the legend of Charlemagne spread widely, though Latin was also an important vehicle for texts perceived as historical truth. Furthermore, the same geographical area could also be the ‘host’ for more than one language, the literatures of which would draw upon each other. One particular challenge for the project was thus the question of overlap between geographic areas and cultural or linguistic zones. This is exemplified by the original project focused on Charlemagne in England, a multilingual land of overlapping cultural zones in one geographic area, with Latin, French and English literary cultures and other languages operating in particular areas or social groups. The solution was to allow some overlap in coverage, with, for example, French texts written in England covered from different perspectives in both Charlemagne in England and Charlemagne in Medieval Francophonia, and with awareness that texts in Latin were circulating at the same time and in the same geographic areas as the vernacular narratives. Given the variety of material in different languages and the varying amounts of research that have been produced, the volumes do not all follow the same pattern. Some areas, notably France, have been the object of more than a century of study. Others, such as the Celtic narratives, have received far less vii

viii

general preface • Charlemagne: A European Icon

critical attention; indeed, one of the aims of the project has been to stimulate research in these under-studied areas. The legends and myths of Charlemagne found expression in epics and romances, chronicles and pseudo-chronicles and were alluded to in political and ecclesiastical documents across medieval Europe. Charlemagne was at the same time king of the Franks / the French and ‘empereur d’Occident’. Even in seven volumes we could not aim for encyclopaedic coverage of the Matter of France. We aim rather to address the same research question, a consideration of how the matter of Charlemagne was appropriated in different contexts, whether that exploitation was for political purposes or was more concerned with literary responses. We also limited ourselves in this series to written texts. A similar series of volumes could have been written about the visual representations of Charlemagne. The geographical areas covered include much of Charlemagne’s empire, but also areas beyond its reach, such as England and Scandinavia and medieval Spain, where he was often seen as an aggressor, rather than a heroic king-emperor. We found a mythic emperor whose legend was infinitely malleable and open to (re)interpretation. Even in his own lifetime Charlemagne was the pater Europae, but the phrase no doubt meant something different to the poet of the Paderborn epic from the resonances it has today in a European Union of nation states. In the Middle Ages Charlemagne was often cited to promote local interests and cults, while at the same time he served as an exemplum of Christian unity. Our project will, we trust, shed some light on the many faces of Charlemagne: ‘Karles li rois, nostre emperere magnes’, as he is named in the opening line of the Oxford Chanson de Roland. This interdisciplinary project was made possible through the work of many. We are grateful to the collaborators who have contributed to this, and in particular to the editors of the volumes. We would also like to thank the British Branch of the Société Rencesvals for their support and the opportunities offered to present research papers at both British Branch and International conferences. The Bristol Institute for Research in the Humanities and Arts at the University of Bristol funded a workshop on translating Charlemagne material, as part of a series of workshops on premodern translation, which enabled many of the collaborators to meet and develop the project in its early stages. We would also like to thank colleagues from the University of Bristol’s IT department and the web-designers from Dirty Design for their work on

general preface • Charlemagne: A European Icon

ix

the project website, http://www.charlemagne-icon.ac.uk/. We are also grateful to the Leverhulme Trust for supporting the network financially, giving us opportunities to meet and discuss research findings and to develop a website for the project. Marianne Ailes Philip E. Bennett Project directors

preface Charlemagne in England: The Matter of France in Middle English and Anglo-Norman Literature

T

he heroic literature of medieval England celebrates the acts of three groups of characters: English legendary heroes, King Arthur and his knights, and the Emperor Charlemagne and his Twelve Peers. The first two groups have received major recent scholarly attention, but the texts in the last group have been comparatively neglected, though a series of editions of the Middle English texts for the Early English Text Society in the late nineteenth century gave them a prominent identity as ‘The English Charlemagne Romances’. French-language Charlemagne texts continued to be produced for English readers well into the fifteenth century, and from the fourteenth to the early sixteenth century Middle English verse romances of Charlemagne were composed and copied. Caxton printed a prose life of Charles the Great as well as the Morte Darthur. However, while Arthurian themes persisted in English literature and culture, those of Charlemagne did not. This might seem explicable if he is viewed as the national hero of the French. Yet, to the puzzlement of some critics, the production of most English Charlemagne texts coincided with the Hundred Years’ War. Is it significant, therefore, that the original French texts already existed in Anglo-Norman form? In what sense might Charlemagne have had an established insular identity? The texts traditionally named the ‘Matter of France’ deal with the legendary history of the Emperor Charlemagne, who is presented defending Christendom in campaigns against the Saracens with his Twelve Peers. In Old French the cycle du roi material extends to over fifty chansons de geste, but in the insular tradition, in both Anglo-Norman and Middle English versions, the focus is restricted to a few central, original narratives: the matter of Roncevaux (La Chanson de Roland and the Pseudo-Turpin Chronicle), the Fierabras material (La Destruction de Rome, Fierabras), and Otinel. A chapter of this book is devoted to each of these traditions, but first we put this in a wider context of reception, xi

xii

preface • Charlemagne in England

with discussion of the circulation of the material in England and an examination of the texts in their manuscript contexts. The aim of this study of ‘Charlemagne in England’ is to explore the insular literary tradition with equal focus on the Anglo-Norman and Middle English texts, examining the textual relations between them and the correspondences in narrative form and generic expectations. The book assesses the evidence for the texts’ audiences and reception in a multilingual society, and for their contemporary cultural and political significance. The central conflict represented, between Christians and Saracens, offers parallels to international crusading interests in the earlier Anglo-Norman period, and in the fourteenth and fifteenth centuries mirrors the perceived threat to western security from the Ottoman Turks, while the prominence in so many of the texts of the motif of the chivalrous Saracen indicates, through the fantasy of conversion, a desire for engagement with the Eastern Other. The exploitation of material centred on Charlemagne, a pan-European hero, also raises questions of identity in texts which stress the importance of Christian unity at a time of developing nation states. This complex reflection of cultural and political concerns in medieval England has resonances for contemporary Britain.

Acknowledgements

W

e wish to acknowledge the support of the Arts and Humanities Research Council and the generous award of a Research Grant (2009–12), without which our collaborative research project would not have been possible. We thank the British Branch of the Société Rencesvals and Bristol Studies in Medieval Cultures for their generous support. We also thank our institutions, the Universities of Reading and Bristol, for their joint support of our research. In particular, we are grateful to the members of our schools and departments: the Department of English Literature at Reading and the School of Modern Languages at Bristol, and to our colleagues in the Graduate Centre for Medieval Studies at Reading and the Centre for Medieval Studies at Bristol, for their unstinting support and encouragement. We have been especially fortunate to have had working with us our two AHRC-funded PhD students, the rest of our ‘Team Charlemagne’, Suzanne Leedham and Jade Bailey, and we thank them for their contributions to the project. Preliminary findings from our research have been presented at the Leeds and Kalamazoo Medieval Congresses, and at numerous other conferences, symposia, research seminars and workshops, and we are grateful to all the audiences and participants for their valuable and stimulating questions and comments. We thank especially the members of the Société Rencesvals British Branch, the Romance in Medieval Britain Conference, and the Early Book Society for their continued and enthusiastic support. Many individual scholars and colleagues have generously shared their expertise with us and have enriched our research, and in particular we are grateful to Philip Bennett, Siobhain Bly Calkin, Anne Curry, Rosalind Field, Mark Hutchings, Richard Ingham, Marco Nievergelt, Carol O’Sullivan, Ad Putter, Rebecca Rist, Stephen H. A. Shepherd, Ian Short, Diane Speed, Emily Wingfield, and Jocelyn Wogan-Browne and the French of England project team. Finally, we wish to record our heartfelt gratitude to Adrian and Christopher for their unfailing support and for having to know more about Charlemagne than they ever expected to.

xiii

Abbreviations AND

Anglo-Norman Dictionary

ANTS

Anglo-Norman Text Society

APF

Les Anciens Poètes de la France

BL

British Library

BnF

Bibliothèque nationale de France

CUL

Cambridge University Library

CFMA

Classiques français du moyen âge

EETS, OS, ES Early English Text Society, Original Series, Extra Series MED

Middle English Dictionary

NLS

National Library of Scotland

n.s.

new series

OED

Oxford English Dictionary

PMLA Publications of the Modern Language Association of America SATF

Société des anciens textes français

STS

Scottish Text Society

TNA

The National Archive

A Note on Terminology We use the conventional term ‘Anglo-Norman’ throughout the book to refer to the insular French language of the texts we discuss, following the practice of the Anglo-Norman Dictionary and the Anglo-Norman Text Society.

xiv

introduction Charlemagne in England: Owning the Legend

T

he idealized figure of Charlemagne (742–814), ‘Christian Emperor, mighty conqueror and patron of learning’, 1 has long been associated with a European sense of identity. A succession of recent scholarly and popular histories of Charlemagne stress this concern in their subtitles – ‘Father of a Continent’, ‘The Formation of a European Identity’2 – or make it explicit in their introductory pages.3 Charlemagne’s modern role as embodying the idea of European integration is seen in diverse invocations, from the name of the building that houses the European Commission (the Charlemagne building) to the prize awarded for ‘distinguished service in the cause of Europe and European unification’ (the International Charlemagne Prize),4 though behind this focus on Charlemagne as icon of Europeanness is a history of conflicting attempts to appropriate his founding status for both German and French national identities.5 Charlemagne’s relation to England has seemed less obvious: indeed, on 9 July 1960, 1 Rosamond McKitterick, Charlemagne: The Formation of a European Identity (Cambridge: Cambridge University Press, 2008), p. 5. 2 Alessandro Barbero, Charlemagne: Father of a Continent, trans. Allan Cameron (Berkeley: University of California Press, 2004); McKitterick, Charlemagne: The Formation of a European Identity. 3 See, for example, Hywel Williams, Emperor of the West: Charlemagne and the Carolingian Empire (London: Quercus, 2010); Derek Wilson, Charlemagne: The Great Adventure (London: Hutchinson, 2005). Both are popular histories that highlight Charlemagne’s role in forming an idea of a distinctly European identity and culture. 4 Walter Eversheim, ‘A Citizens’ Prize for Distinguished Service on Behalf of European Unification’, http://www.aachen.de/de/stadt_buerger/ pdfs_stadtbuerger/pdf_karlspreis/karlspreis_03/eversheim03_en.pdf [accessed 23/08/2016]. 5 For a brief summary, see Joanna Story, ‘Charlemagne’s Reputation’, in Charlemagne: Empire and Society, ed. J. Story (Manchester: Manchester University Press, 2005), pp. 1–4.

1

2

Charlemagne in Medieval England

contemplating the reasons for a possible British application to join the Common Market, Harold Macmillan confided to his diary his fear of a future ‘caught between a hostile (or at least less and less friendly) America and a boastful, powerful “Empire of Charlemagne” – now under French but later bound to come under German control’.6 Here the exclusively continental identity of a political alliance imagined in terms of Charlemagne’s empire seems logically opposed to British involvement. Half a millennium earlier, Caxton, though celebrating both as Christian Worthies, set the native Englishness of Arthur, ‘kyng and Emperour of the same [royame of Englond]’, against the alterity of Charlemagne’s identity as ‘kyng of fraunce & emperour of Rome’;7 and one modern scholar has noted as an ‘amusing irony’ (in view of ‘the age-old and perennial animosity between France and its cross-Channel neighbours’) the fact that the most famous literary Charlemagne text, ‘the French national epic’, is known from its English provenance as ‘the Oxford Song of Roland’,8 while another ponders the ‘paradox’ inherent in ‘the production of the Middle English Charlemagne romances … during a period of prolonged Anglo-French hostility’. 9 However, this 6 Harold Macmillan, Pointing the Way, 1959–61 (London: Macmillan, 1972), p. 316. 7 Caxton’s prefaces to Morte Darthur (1485) and Charles the Grete (1485). 8 Ian Short, ‘Literary Culture at the Court of Henry II’, in Henry II: New Interpretations, ed. C. Harper-Bill and N. Vincent (Woodbridge: Boydell Press, 2007), pp. 335–61 (p. 350), on Oxford, Bodleian Library MS Digby 23. For an edition, see La Chanson de Roland: The French Corpus, ed. J. J. Duggan et al., 3 vols (Turnhout: Brepols, 2005), I: The Oxford Version, ed. I. Short. This edition has been chosen as the reference edition for this book because it allows easy comparison of the different Old French redactions; the edition is also the basis for a translation into English of both the Oxford text and the Châteauroux-Venice 7 version: The Song of Roland: Translations of the Versions in Assonance and Rhyme of the ‘Chanson de Roland’ , trans. Joseph J. Duggan and Annalee C. Rejhon (Turnhout: Brepols, 2012). In what follows, we give citations to the edition with the name of the editor of the particular MS in the form Short/Duggan; for ease of comparison the Duggan and Rejhon translation is used unless otherwise stated. 9 Robert Warm, ‘Identity, Narrative and Participation: Defining a Context for the Middle English Charlemagne Romances’, in Tradition and Transformation in Medieval Romance, ed. Rosalind Field (Cambridge: D.

Introduction

3

sense of strangeness fluctuates according to historical distance and political perspective. Fifty-odd years earlier than Caxton, Lydgate, in his ‘Ballade to King Henry VI upon his Coronation’, addressed Henry as a descendant of both St Edward and St Louis, and immediately invoked the parallel paired kings, ‘knightly’ Arthur and ‘Charlles of gret prys’ (l. 13), as his twin secular patrons.10 And as Ian Short argues, not only was the earliest extant version of La Chanson de Roland copied and preserved in England, but it seems it was designed to draw a parallel between William the Conqueror and Charlemagne, who also, ‘we are told, crossed the sea and conquered England, holding it by personal tenure’;11 its revision can thus be seen as an act of appropriation, adapting La Chanson de Roland for an insular audience. This unique text may be considered the first significant landmark in tracing the literary career of Charlemagne in England.12 No such special focus on England occurs in any other version of the Chanson de Roland (although England is mentioned prominently among Charlemagne’s conquests in the Pseudo-Turpin Chronicle), and the passages in question may represent localized innovations by the Anglo-Norman copyist, S. Brewer, 1999), pp. 87–100 (p. 87). The span of the Hundred Years’ War coincided with the composition and/or copying of the majority of the extant English Charlemagne texts. 10 The Minor Poems of John Lydgate, ed. H. N. MacCracken, vol. II: Secular Poems, EETS OS 192 (1934), pp. 624–30. 11 Short, ‘Literary Culture at the Court of Henry II’, p. 351, referring to two brief passages in the text (ll. 370–3 and 2331–2). Short rejects arguments that this shows a ‘Norman bias’ or ‘Insular influence’ in the original composition of the Chanson (p. 356), but there is certainly a case for Anglo-Norman adaptation in the manuscript copy as we have it. For further discussion of ll. 370–3 and the link with William I, see Melissa Furrow, ‘Chanson de geste as Romance in England’, in The Exploitations of Medieval Romance, ed. Laura Ashe, Ivana Djordjević and Judith Weiss (Cambridge: D. S. Brewer, 2010), pp. 57–72 (pp. 60–1); for further discussion, see Chapter 4. 12 For discussion of documentary evidence for actual contemporary perceptions of Charlemagne’s authority in England, see Joanna Story, ‘Charlemagne and the Anglo-Saxons’, in Charlemagne: Empire and Society, ed. Story, pp. 195–210: ‘From the Frankish perspective, the Anglo-Saxons lay within the boundaries of Charlemagne’s authority when it came to the proper performance of Christian duties’ (p. 202).

4

Charlemagne in Medieval England

aiming to absorb the story of Charlemagne into the history of Britain (echoing Brutus and prefiguring William I), to produce a pattern of conquerors who made England their own. In one passage, much corrected by a subsequent reviser, Roland lists the countries he has won for Charlemagne, culminating in the British Isles: ‘jo l’en cunquis e Escoce e \Uales/ Irlande | e Engletere’ (2331–2);13 the reviser’s addition of Wales here gives an authentic insight into the reception of the text in Anglo-Norman Britain, suggesting a desire to assimilate the text’s representation of Charlemagne’s world to the audience’s own. Three further landmarks present themselves: the proliferation of insular Charlemagne texts in the fourteenth and fifteenth centuries (both Middle English adaptations and Anglo-Norman copies); the publication by the Early English Text Society, beginning in 1879, of a series of editions of Middle English texts under the general title The English Charlemagne Romances;14 and the increasing critical interest in these texts at the present time. Our concerns in this book are primarily to contextualize the occurrence of the late-medieval texts and to explore the timeliness of modern critical responses; but the extraordinary nature of the EETS series, the only body of romances to be distinguished in this way among the society’s publications, cannot go unrecognized. Sidney J. H. Herrtage, the initial editor, offers no rationale for embarking on the project, but a hint may be found in Emil Hausknecht’s introduction to Part V. This describes with nineteenthcentury enthusiasm the ‘glorification’ of Charlemagne’s reputation for military power through ‘the sublime figure of Charlemagne’, the ‘valiant champion of Christendom’ with his ‘nimbus of majestic grandeur’, in imaginative literature of universal appeal: he was, wrote Hausknecht, ‘celebrated in song by almost all European nations’.15 The Middle 13 Short tentatively dates the reviser’s hand ‘well into the thirteenth century’ and notes that by adding Wales, the reviser has ‘complet[ed] the twelfthcentury Anglo-Norman kingdom of Britain’; the interpolation is not included in the edited line (Oxford Version, pp. I/103, I/312). For further discussion, see Chapter 4. 14 See Marianne Ailes and Phillipa Hardman, ‘How English are the English Charlemagne Romances?’, in Boundaries in Medieval Romance, ed. Neil Cartlidge (Cambridge: D. S. Brewer, 2008), pp. 43–55. 15 The Sowdone of Babylone, ed. Emil Hausknecht, EETS ES 38 (1881), pp. v–vi. Hausknecht imagines Charlemagne’s Europe in terms of

Introduction

5

English romances of Charlemagne, though very late in the medieval tradition of the Matter of France, nevertheless preserve the optimism of this early literary image, eschewing those later portrayals which show, as Hausknecht puts it, ‘the splendour of Royalty tarnished and debased’. 16 This may have been their late-nineteenth-century appeal: a body of English texts that represent in an apparently unproblematic way ‘the great Christian hero’ of all Europe, and which could be brought to public attention by being grouped together under this title, with its claim that Charlemagne has a recognizable identity in English literary culture. While Hausknecht’s focus is on the extent of Charlemagne’s conquests in Western Europe, the medieval legendary tradition attributes to him a huge sphere of authority across the Mediterranean world, south into Spain, and eastwards to Jerusalem, and this wider context supports a different emphasis in studies of Charlemagne’s image. Besides his role as ‘Father of Europe’, Charlemagne’s reputation as champion of Christendom and the widespread legend of his journey to the East saw him cast in the role of ‘Crusader’,17 as in the celebrated thirteenth-century Charlemagne window at Chartres, where pictorial scenes from the legendary traditions represent his conquests in Spain and the Holy Land. This constructed image of Charlemagne, the European hero defeating the opposed powers of Eastern rulers, is reflected throughout the insular Matter of France tradition, with its repeated representation of conflict between Charlemagne, his peers and forces (an ‘international brigade’ from a wide range of European nineteenth-century nation-states. 16 Hausknecht cites Guy de Bourgoyne as an example of the later tradition. 17 See, for example, Nancy Bisaha, ‘Crusade and Charlemagne: Medieval Influences’, in her Creating East and West: Renaissance Humanists and the Ottoman Turks (Philadelphia: University of Pennsylvania Press, 2004), pp. 13–42; Jace Stuckey, ‘Charlemagne as Crusader? Memory, Propaganda, and the Many Uses of Charlemagne’s Legendary Expedition to Spain’, in The Legend of Charlemagne in the Middle Ages: Power, Faith, and Crusade, ed. Matthew Gabriele and Jace Stuckey, The New Middle Ages (New York: Palgrave Macmillan, 2008), pp. 153–68; Matthew Gabriele, An Empire of Memory: The Legend of Charlemagne, the Franks, and Jerusalem before the First Crusade (Oxford: Oxford University Press, 2011).

6

Charlemagne in Medieval England

locations)18 and Saracen armies drawn from all across the East.19 Yet the concept of ‘crusade’, at least in the narrow sense of the journey to the East epitomized in the OED definition of ‘crusade’ (a ‘military expedition made by Europeans to recover the Holy Land from the Muslims’) and as reflected in the stereotypical ‘good deaths’ of many medieval romance heroes, fighting for God in the Holy Land, needs further interrogation in relation to these texts.20 In the reopening in 2010 of Britain’s oldest public museum, the Ashmolean at Oxford, the collections were reorganized to present a new approach to the understanding of the past: entitled ‘Crossing Cultures Crossing Time’, it was intended to reveal ‘how the civilisations of the east and west have developed as part of an interrelated world culture’. 21 This twenty-first-century concern to recognize the commonalities and interdependencies of global human history is tellingly expressed in terms specifically indicating the expectation of difference, of opposition, that it seeks to counter, in the formulation of the world as a construct of two halves, east and west. Stabilized and reified as ‘the East’ and ‘the West’, these geographical distinctions carry far-reaching and longstanding cultural associations, and attempts to interrogate the relations between them have understandably proliferated since 2001. The editors of Cultural Encounters between East and West: 1453–1699, for instance, introduce their volume with the claim that ‘a radical reappraisal of the relationship between “East” and “West” is currently under way’ which questions previous critical approaches that had assumed ‘a binary 18 For this feature in the Oxford Roland, see Gabriele, An Empire of Memory, p. 138. 19 See Dorothee Metlitzki, The Matter of Araby in Medieval England (New Haven, CT: Yale University Press, 1977), p. 119. 20 Although sometimes described as ‘crusade-romances’, most of the insular Charlemagne texts deal with conflicts between Christians and Saracens that are not ‘journeys to the East’; only one (Roland and Vernagu, based on the Old French ‘Johannes’ version of the Pseudo-Turpin Chronicle) includes an account of Charlemagne liberating the Holy Land. 21 Ashmolean Museum leaflet (2010), p. 2; in the latest version, the theme ‘Crossing Cultures’ is designed to ‘show how civilisations developed as part of a connected world culture’, and introduces collections presented as ‘East Meets West’, http://www.ashmolean.org/assets/images/Plan/ FloorPlan.pdf [accessed 23/08/2016].

Introduction

7

opposition between a civilized Christian “West” and the encroaching barbarity of an infidel “East” ’, although they point out the anachronism of the governing idea: ‘In this period there is generally no sense of a clear or rigid demarcation between what are ordinarily thought of as “East” and “West”.’ 22 This may need a little qualification: while agreeing that the modern binary ‘East’ and ‘West’ was not standard usage in the Middle Ages, Suzanne Conklin Akbari points to the construction, from the late fourteenth century onwards, of the opposed terms ‘Occident’ and ‘Orient’ as denoting distinct geographical areas and their inhabitant peoples.23 There is no doubt, however, that an opposition between geographically defined peoples characterized as either Christian or ‘infidel’ is very widely inscribed in medieval and early modern texts and culture, and it is the consequences of this faith-defined division that are also largely the focus of current interrogations of East and West.24 Prominent in both medieval texts and modern analyses is the 22 Cultural Encounters between East and West: 1453–1699, ed. Matthew Birchwood and Matthew Dimmock (Newcastle: Cambridge Scholars Press, 2005), p. 1. 23 Suzanne Conklin Akbari, Idols in the East: European Representations of Islam and the Orient, 1100–1450 (Ithaca, NY: Cornell University Press, 2009), pp. 3, 46. Akbari cites, for example, John Gower’s retelling the story of the repopulation of the earth after Noah’s flood, which very precisely aligns the whole of Asia (everywhere east of the Nile as far as Paradise) with the term ‘Orient’: ‘And schortly for to speke it so, | Of Orient in general | Withinne his bounde Asie hath al’ (Confessio Amantis, 7: 572–4). The ‘Occident’, meanwhile, covers the rest of the world ‘Westward’ (7: 582, 576), setting a cold ‘Occident’ against a hot ‘Orient’ (7: 582–3). See also Gower’s account of Julius Caesar’s conquering nations: ‘Noght al only of thorient | Bot al the Marche of thoccident’ (Prologue, 719–20). 24 For instance, in his sweepingly titled Worlds at War: The 2,500-Year Struggle between East and West (Oxford: Oxford University Press, 2008), Anthony Pagden chronicles the successive manifestations over millennia of the ‘ancient struggles between an ever-shifting West and an equally amorphous East’ (p. xix), but centres his survey on the bitter wars between the two universal religions, Christianity and Islam, as the most potent and influential aspect of that persistent conflict and its modern expressions.

8

Charlemagne in Medieval England

phenomenon and afterlife of the crusades, and while Christopher Tyerman concludes his study of the historiography of the crusades with a sharp critique of the more extreme manifestations of the ‘clash of civilizations’ theory and their unhistorical collapsing of past and present, it is fundamental to his own argument that ‘one of the oldest features of crusade historiography had been its relation to contemporary cultural and political attitudes’.25 Moreover, the interpenetration of past and present could be seen as characteristic of some crusade thinking from the beginning: a ‘war of liberation’, to free the Christian people of the eastern Churches from the tyranny of Muslim rule,26 was at the same time, with the innovative ritual of ‘taking the Cross’, a return to the world of the New Testament for those who heard the call to fight as a summons to avenge the insult to Christ and to reclaim the place of his death and burial.27 Thus it is not altogether surprising to find the language and attitudes of the crusades imaginatively projected back on to the conflicts of Christian heroes of earlier centuries, Charlemagne and the legendary Arthur, in epic and romance.28 Surveying the functions of representations of Charlemagne over the last 1,200 years, Rosamund McKitterick calls attention to ‘the degrees to which a people communicates with the past in order to form or to inform its own contemporary concerns, to heighten its sense of identity and cultural affiliations and to shape its political purpose’.29 In this book we aim to trace the relation of the insular Matter of France texts, with their focus on the battles between Charlemagne’s Christian knights and enemy Saracen forces, and their oblique reflection of the history of crusades to the East, to the cultural and political concerns of their own times, and to ask what and how legends of an imagined Carolingian past contributed 25 Christopher Tyerman, The Debate on the Crusades (Manchester: Manchester University Press, 2011), pp. 247, 217. 26 As urged by Urban II before the First Crusade: Jonathan Riley-Smith, The Crusades: A History, 2nd edn (London: Continuum, 2005), pp. 4–8. 27 Ibid., pp. 23–5, 10–11; Tyerman speaks of a ‘religio-historical nostalgia’ (Debate on the Crusades, p. 22). 28 An instructive instance from the late fifteenth century is William Caxton’s project of publishing the histories of Charlemagne and Arthur as companion pieces to his history of Godeffroy of Boloyne, based on William of Tyre’s account of the First Crusade. 29 McKitterick, Charlemagne: The Formation of a European Identity, p. 6.

Introduction

9

to the self-image of fourteenth- and fifteenth-century readers in England (while we are, of course, conscious of our own inescapable situatedness in a world newly sensitized to the representation of conflict between Muslims and Christians, East and West). Crusades historian Jonathan Phillips points out that the Old French Chanson de Roland was composed almost immediately after the First Crusade to the Holy Land (1096–9), alongside texts dealing directly with the crusade: Latin chronicles of the Holy War and other chansons de geste that celebrate the heroic deeds of the Christian forces combating the Muslim inhabitants of the East, such as the Chanson de Jérusalem and the Chanson d’Antioche; he sees this as evidence of how deeply the memory of that crusade and its spectacular victories had entered into the ‘political and spiritual culture of western Europe’.30 The fervour attending the preaching and preparations for the First Crusade, its astonishing success in liberating the holy places and establishing the crusader kingdom of Jerusalem, and the joy with which the returning crusaders were welcomed home as heroes and ‘models for future generations to emulate’,31 all make it somewhat surprising that these stirring epics of the crusade cycle were not appropriated for an insular audience. French texts on the subject were available in Britain, for a copy of the gestes was borrowed from the Master of the Temple in 1250 for Eleanor of Castile, who had murals of the subject painted in the ‘Antioch Chamber’ at Westminster,32 but there is no evidence of any attempt to produce a cultural adaptation of the crusade cycle in either insular vernacular. One little-known poem, surviving only in AngloNorman manuscripts and based on a chronicle account of the First Crusade, is potentially the only extant insular poetic treatment of this glorious chapter of Christian history;33 indeed, there are very few insular 30 Jonathan Phillips, Holy Warriors: A Modern History of the Crusades (London: Bodley Head, 2009), p. 78. 31 Ibid., p. 27. 32 Peter Brieger, English Art, 1216–1307, Oxford History of English Art (Oxford: Clarendon Press, 1957), p. 133. 33 An edition has been published of part of the text only: La Chanson de la Première Croisade en ancien français d’après Baudri de Bourgueil: Edition et analyse lexicale, ed. Jennifer Gabel de Aguirre (Heidelberg: Universitäetsverlag Winter, 2015); see also Paul Meyer, ‘Un récit en vers français de la première croisade fondé sur Baudri de Bourgueil’, Romania

10

Charlemagne in Medieval England

chansons de geste or romances directly concerned with the topic of crusade to the East at all.34 The lack is particularly surprising given the fact that, as Tyerman observes, ‘militant Christianity, enshrined in tales of Charlemagne or warrior saints, was in fashion in the Anglo-Norman world’. 35 One exceptional example is the Middle English romance of Richard Coeur de Lion, a fantasized account of the part played by Richard I in the Third Crusade (1189–92). This text had considerable success, judging by the large number of extant versions,36 and it offers a useful paradigm of insular taste. There are obvious attractions in its bizarre combination of relatively recent history and supernatural fiction, with its central role for an insular crusading hero-king, and Geraldine Heng argues for its power as a model of English national identity;37 but perhaps an additional source of interest the text provides is the extended confrontation between Richard and his opponent, the charismatic Muslim leader Saladin. Although Richard is obviously the heroic central 5 (1890), 1–63. For further discussion, see Chapter 1. Two chronicles, one Norman, one Anglo-Norman, are devoted to the Third Crusade: The History of the Holy War: Ambroise’s ‘Estoire de la Guerre Sainte’, ed. Marianne Ailes and Malcolm Barber, 2 vols (Woodbridge: Boydell Press, 2003); The Crusade and Death of Richard I, ed. R. C. Johnston, ANTS 17 (Oxford, 1961). 34 Crusade to the East features frequently in insular narrative (for example in Guy of Warwick and Bevis of Hampton) as a subsidiary topic, the ultimate test of Christian knighthood. 35 Christopher Tyerman, England and the Crusades, 1095–1588 (Chicago: University of Chicago Press, 1988), p. 14. 36 Eleven distinct medieval texts in manuscript and print survive (dated from c. 1330 to 1528) – more than for either Guy of Warwick or Bevis of Hampton, for example. No Anglo-Norman original for the Middle English romance is known, but a contemporary Anglo-Norman chronicle account of Richard I’s exploits in the crusade, based on Roger of Howden, exists in The Crusade and Death of Richard I. 37 Geraldine Heng, ‘The Romance of England: Richard Coeur de Lyon, Saracens, Jews and the Politics of Race and Nation’, in The Post-Colonial Middle Ages, ed. Jeffrey J. Cohen (New York: St Martin’s Press, 2000), pp. 135–72; rev. as Chapter 2 in Geraldine Heng, Empire of Magic: Medieval Romance and the Politics of Cultural Fantasy (New York: Columbia University Press, 2003), pp. 63–113.

Introduction

11

figure of the romance, receiving assistance on different occasions from angelic advisers and from St George, while Saladin is correspondingly portrayed as confounded, still, the repeated focus on the two leaders, the several episodes of actual or projected single combat (on one occasion specifically to test the power of their respective faiths), and the eventual three-year truce they agree produce an effect of something more like equal combat between the great commanders, against a background of dissent among French and English in the Christian camp and of reported unrest at home in England. The subject of Richard and Saladin was of immediate and long-lasting interest: as Phillips observes, ‘eyewitnesses on both sides recognised that the two main players in the Third Crusade were exceptionally charismatic men’, 38 and half a century later their combat was depicted in wall paintings at Henry III’s Clarendon Palace, as well as in the celebrated Chertsey floor tiles, possibly designed for the palace of Westminster.39 What this may suggest is that there was a particular insular taste for a version of the opposition between East and West, Muslim and Christian, figured almost symbolically in the image of one-to-one confrontation between heroic representatives of each. Such a taste in insular culture could help to explain an apparent anomaly in the statistics relating to the selection of Charlemagne texts for copying and translating in Britain, statistics that show multiple versions of a surprisingly small number of source texts. As Rosalind Field has remarked, ‘The Middle English Matter of France romances can be seen as translations derived from the already established selection of chanson material available as insular texts in Anglo-Norman from the twelfth century into the mid-fourteenth century’, a precedent that ‘may account for the otherwise odd concentration on Ot[in]el and Fierabras material amongst the Middle English works’. 40 To summarize briefly: of the considerable range of Old French chansons de geste dealing with the Matter of France, Anglo-Norman rewriters appear largely to have 38 Phillips, Holy Warriors, p. 136. 39 Age of Chivalry: Art in Plantagenet England, 1200–1400, ed. Jonathan Alexander and Paul Binski (London: Royal Academy of Arts, 1987), p. 204. 40 Rosalind Field, ‘Patterns of Availability and Demand in Middle English Translations de romanz’, in The Exploitations of Medieval Romance, ed. Ashe et al., pp. 73–89 (p. 81).

12

Charlemagne in Medieval England

restricted their choice to a small group of texts concerning Charlemagne and his closest peers, Roland and Oliver, in direct conflict with Saracen enemies, often with named individual Saracen champions such as Otinel and Fierabras. And from this selection of texts, Middle English translators and adaptors further refined their choice,41 to produce three distinct retellings of the combat between Firumbras and Oliver, and three versions of Otuel’s fight with Roland. The central image of the two opponents locked in single combat is sometimes reflected in the titles given to the Middle English romances by their scribes: Duke Rowlande and Sir Ottuell of Spayne; Off Rowland and off Otuel; and Off Firumbras de Alisavndre and Syr Olyuer. It is also worth noting the construction of a similar, unique Middle English romance, Roland and Vernagu, created from incidents in the Johannes Turpin, a thirteenth-century French adaptation of the Pseudo-Turpin Chronicle.42 In each case, the Christian knight battles a Saracen giant and makes an attempt at converting him before the Saracen is eventually defeated. This represents a remarkable concentration on narratives that follow the pattern of opposed champions of the two faiths, like Richard the Lionheart confronting Saladin in popular imagination (though they never met in historical reality), as depicted in Richard Coeur de Lion and on the Chertsey tiles. Indeed, one might see almost an insular fixation on the persistent motif of the individual Christian hero in contention with the potentially convertible Saracen.43 One of our concerns in this book is to trace 41 A notable exclusion is the Chanson d’Aspremont, of which there are numerous Anglo-Norman copies. Aspremont is discussed in Chapter 1, but as it was never translated into Middle English, a full consideration of its place in insular culture has been reserved to the Anglo-Norman chapter of the Charlemagne in Medieval Francophonia volume (ed. Marianne Ailes and Philip E. Bennett) in the series ‘Charlemagne: A European Icon’. 42 The Johannes Turpin includes in its first chapter an abbreviated version of the legend of Charlemagne’s journey to the East and recovery of the relics of the Passion (Descriptio qualiter Karolus Magnus clavum et coronam Domini a Constantinopoli Aquisgrani detulerit). See The Old French Johannes Translation of the Pseudo-Turpin Chronicle, ed. R. N. Walpole, 2 vols (Berkeley: University of California Press, 1976). 43 ‘Wishful’ traditions grew up around the historical Saladin, presented as a potential Christian convert, who received knighthood from a Christian

Introduction

13

and interrogate this motif in the Anglo-Norman and Middle English Charlemagne tradition.44 Unlike King Richard and Saladin, however, the heroes of these Charlemagne romances confront their Saracen foes not in the Holy Land in the East, but in Western Europe, either at the frontier of Islamic Spain, or in Christian territories: Rome, Paris, and Lombardy.45 The events of Fierabras and its English versions take place initially on the border with Islamic Spain, and then, as Charlemagne’s forces pursue the Saracens with their French captives, the action moves into the heart of the sultan’s Spanish empire, and ends with the conquest of Spain and the establishment of Christian rule. But the antecedents of the story, as told in the chanson de geste, La Destruction de Rome, and incorporated into the Middle English romance of The Sowdone of Babylone, explain that Charlemagne’s expedition to Spain was an act of continuing revenge and repossession after the atrocities previously carried out by the sultan and his son Fierabras in Rome, where they sacked the city, killed the inhabitants – including the pope – and stole the relics of Christ’s Passion. The story of Otinel begins in France, where a messenger from the Saracen king of Spain brings a similar report: the Saracens have sacked Rome, and demand that Charlemagne renounce Christianity and submit to the Saracen king as his sovereign. When Charles refuses, the action eventually moves to Lombardy, where the Saracen king has made his stronghold; and again, the story ends with conquest of the Saracen-held lands and the establishment of a Christian kingdom. The main events of Roland and Vernagu take place in Spain, which Charlemagne conquers, city by city, and where the Saracen giant comes lord: see Margaret A. Jubb, ‘Enemies in the Holy War, but Brothers in Chivalry: The Crusaders’ View of their Saracen Opponents’, in Aspects de l’épopée romane: Mentalités, idéologies, intertextualités, ed. Hans van Dijk and Willem Noomen (Groningen: Forsten, 1995), pp. 251–9; Sarah Lambert (citing Orderic Vitalis, Historia Ecclesiastica), ‘Heroines and Saracens’, Medieval World 1 (1991), 3–9 (p. 9). The idea can be found widely in connection with virtuous non-Christians such as, for example, the Saracen Sir Baltirdam in The Beauchamp Pageant, ed. Alexandra Sinclair (Donington: Watkins, 2003), p. 88. 44 In the Appendix we give details of the manuscripts and editions of all the insular Charlemagne texts discussed in this book. 45 The action of Aspremont takes place in Calabria.

14

Charlemagne in Medieval England

to challenge him; the action of The Sege of Melayne, as indicated in the title, is set in Lombardy and climaxes with the siege of the Saracen-held city of Milan – but as the text is incomplete, we can only guess that it most likely concludes with the reconquest of the city and the restoration of Christian rule. There is evident replication between these various texts, and, not surprisingly, it has been argued both that Otinel reworks material from Fierabras,46 and that The Sege of Melayne is a compilation of well-known motifs from Otinel and elsewhere.47 This only strengthens the impression that insular audiences had an insatiable appetite for fiction adhering to a particular, recognizable formula: triumphal narratives of familiar Christian heroes defending or reconquering European lands in the face of Saracen aggression. We explore the issue in this book by placing it in the context of recurrent concerns for the Christian homeland in the later medieval period that may have provoked repeated insular engagement with these specific Charlemagne texts. The vivid portrayal of hostile relations between English and French contingents in the crusader forces in Richard Coeur de Lion indicates another site of concern around the selection of texts for insular appropriation: the issue of language and identity. Questions raised in the past about the production of English-language Charlemagne texts, particularly in the fourteenth and fifteenth centuries during the period of the Hundred Years’ War, tended to assume a narrative of developing national and linguistic identity, in which war-time English adaptations of French-language narratives celebrating Charlemagne and the peers of ‘douce France’ would seem to present an anomaly.48 However, this assumption has been profoundly questioned by much subsequent work on the complex patterns of language use and identity construction in

46 The nineteenth-century critic Léon Gautier described Otinel as ‘servilement calquée sur la légende de Fierabras’: Les Épopées françaises: Étude sur les origines et l’histoire de la littérature nationale, 4 vols (Paris: Société Générale de Librairie Catholique, 1878–92), III, p. 398; Marianne Ailes, ‘Otinel: An Epic in Dialogue with the Tradition’, Olifant 27 (2015), 9–39. 47 W. R. J. Barron, English Medieval Romance (London: Longman, 1987), p. 96. 48 Warm, ‘Identity, Narrative and Participation’, p. 87.

Introduction

15

medieval Britain and France.49 Michael Bennett, focusing on relations between England and France in the early years of the war, stresses the vitality of cross-Channel Francophone cultural exchange: ‘In the late 1350s, the court of Edward III could claim to be the centre of the Frenchspeaking world.’50 Ardis Butterfield offers a striking insight into the anxieties surrounding the ownership of language and its relation to identity in response to the course of the war: ‘The realization from the other side, once the English soldier started to dominate at Crécy and Poitiers, that “French” as a political and linguistic category now meant “English” (at least to the English) was one of the crises of the period.’ 51 Marisa Libbon, reading the earliest copy of Richard Coeur de Lion in the Auchinleck MS (NLS MS Adv. 19.2.1), argues that details in the romance create a parallel between Richard and Charlemagne which, together with other ‘textual inventions’ in the manuscript, reinvent the French canon as ‘England’s rightful inheritance’. 52 Applying a similar insight to BL MS Royal 15 E vi, a manuscript that was shaped by the outcome of the war as a presentation volume for Margaret of Anjou, Andrew Taylor proposes that its collection of French Charlemagne texts for a new queen of England could be read ‘as an act of cultural appropriation, a translatio studii that reinforced a translatio imperii, laying claim to

49 See, for example, the ‘French of England’ programme at Fordham University, NY, and the University of York; and the Leverhulme-funded international ‘Multilingualism in the Middle Ages’ project network. For a brief response in relation to Middle English romances, see Thomas H. Crofts and Robert Allen Rouse, ‘Middle English Popular Romance and National Identity’, in A Companion to Medieval Popular Romance, ed. Raluca L. Radulescu and Cory James Rushton (Cambridge: D. S. Brewer, 2009), pp. 79–95. 50 Michael Bennett, ‘France in England: Anglo-French Culture in the Reign of Edward III’, in Language and Culture in Medieval Britain: The French of England, c. 1100–c. 1500, ed. Jocelyn Wogan-Browne et al. (York: York Medieval Press, 2009), pp. 320–33 (p. 327). 51 Ardis Butterfield, The Familiar Enemy: Chaucer, Language and Nation in the Hundred Years’ War (Oxford: Oxford University Press, 2009), p. xxvii. 52 Marisa Libbon, ‘The Invention of King Richard’, in The Auchinleck Manuscript: New Perspectives, ed. Susanna Fein (York: York Medieval Press, 2016), pp. 127–38 (p. 137).

16

Charlemagne in Medieval England

the glories of French heritage’. 53 We believe that the same project of cultural appropriation may be seen as lying behind the repeated acts of translation and rewriting that produced the Middle English Charlemagne romances. Texts celebrating Charlemagne and his peers would be as much the legitimate heritage of English-speaking subjects of Plantagenet kings as of their French-speaking fellows, and indeed this consciousness might have been felt especially keenly during those phases of the Hundred Years’ War when Edward III and later Henry V explicitly laid claim to the crown of France. Cultural appropriation as an aspect of translation has been an issue in recent developments in translation theory, especially, as applied to medieval texts, in combination with a renewed emphasis on manuscript transmission and multilingualism.54 Indeed, given the emphasis in medieval writing on the process of rewriting, of adapting an authoritative text for a new context, it is possible to consider ‘translation’ in its widest sense as a major impulse in medieval literary creation. Nevertheless, in the past translations were often dismissed as being derivative and therefore of little interest. The English Charlemagne romances have suffered from this tendency, a tendency sufficiently marked that our texts do not appear at all in William Calin’s extensive study of The French Tradition and the Literature of Medieval England.55 Even when credit is given to the creativity of translators, some critical judgements of the Middle English Charlemagne romances have been harsh. Rosalind Field argues that ‘the most intriguing cultural translatio gives rise to one of the least satisfactory groups of romances, the ME romances of the Matter of France. The quality of the surviving ME versions of Matter of France material is generally considered

53 Andrew Taylor, ‘The Self-Presentation of an English Mastiff: John Talbot’s Book of Chivalry’, in Language and Culture in Medieval Britain, ed. Wogan-Browne et al., pp. 444–56 (p. 453). 54 See, for example, the studies of Scandinavian material in Sif Rikhardsdóttir, Medieval Translations and Cultural Discourse: The Movement of Texts in England, France and Scandinavia (Cambridge: D. S. Brewer, 2012), and Stefka Georgieva Eriksen, Writing and Reading in Medieval Manuscript Culture (Turnhout: Brepols, 2014). 55 William Calin, The French Tradition and the Literature of Medieval England (Toronto: University of Toronto Press, 1994).

Introduction

17

unimpressive.’56 This follows her more detailed judgement of their quality as ‘derivative and feeble versions’ of the original chansons de geste, which ‘displace their matter into an exotic, distancing, romance mode in which it can easily topple into absurdity or banality’.57 This view reflects the majority critical opinion of the Charlemagne romances in the twentieth century, and the sense that these texts are of limited value because of their status as ‘derivative’ translations has permeated much discussion.58 What lies behind such dismissive judgements of the Middle English translations can be demonstrated with reference to the varied assessments of Caxton’s prose translations. N. F. Blake, in his overview of Caxton, compares the very different judgements made by two editors of Caxton’s work: The difference between the views of Sommer and Kellner quoted above is the result not so much of the different quality of the two translations, as the different expectations and outlooks of the two modern editors. The mistakes in the two works are comparable, but Kellner viewed them in relation to other fifteenth-century translations, whereas Sommer looked at them with a modern eye.59 Reviews of modern translations generally evaluate them in terms of a balance between fidelity to the original and fluency in the target language,60 but if such criteria were applied to Caxton’s Charlemagne 56 Rosalind Field, ‘Romance’, in The Oxford History of Literary Translation in English, ed. Roger Ellis (Oxford: Oxford University Press, 2008), I, pp. 296–331 (p. 312). 57 Rosalind Field, ‘Romance in England, 1066–1400’, in The Cambridge History of Medieval English Literature, ed. David Wallace (Cambridge: Cambridge University Press, 1999), pp. 152–76 (p. 172). 58 See, for example, H. M. Smyser, ‘Charlemagne Legends’, in A Manual of the Writings in Middle English, 1050–1500, vol. I: Romances, ed. J. B. Severs (New Haven: Connecticut Academy of Arts and Sciences, 1967), p. 80; D. B. Sands, Middle English Verse Romances (New York: Holt, Rinehart and Winston, 1966), p. 2. 59 N. F. Blake, Caxton and his World (London: André Deutsch, 1969), p. 126. 60 Fidelity to the original is often the main concern with academic translations, which may be used as an aid for the ‘semi-languaged’. For a more detailed discussion of fluctuating attitudes to the fidus interpres,

18

Charlemagne in Medieval England

romances they would be found wanting. However, in recent years the development of a more general academic study of translation practice and theory has had a beneficial effect on approaches to medieval translation,61 and the theory and practice of translation in the Middle Ages have, over the last three decades, been a subject for investigation in their own right.62 Situating the practice of medieval translation in the context of medieval rhetorical practice has revolutionized recent approaches to medieval translations. This principle received its fullest and most systematic treatment in a major study by Rita Copeland,63 while both Copeland herself and Karen Pratt had previously applied the same

see Luis Kelly, The True Interpreter (Oxford: Blackwell, 1979); Lawrence Venuti, in The Translator’s Invisibility: A History of Translation (London: Routledge, 2008), with conscious and clearly articulated polemical purposes, stressed that fluency has been a major criterion in the evaluation of translations since the seventeenth century. 61 There is no room here for a comprehensive bibliography. Influential scholarship includes: The Manipulation of Literature: Studies in Literary Translation, ed. Theo Hermans (London: Croom Helm, 1985); Susan Bassnett and André Lefevere, Constructing Cultures (Clevedon: Multilingual Matters, 1998); Walter Benjamin, ‘The Task of the Translator’, in The Translation Studies Reader, ed. Lawrence Venuti (London: Routledge, 2000), pp. 75–83; The Translator as Writer, ed. Susan Bassnett and Peter Bush (London: Continuum, 2006); Venuti, The Translator’s Invisibility. 62 This development can be seen in the series of volumes and connected conferences on medieval translation initiated by The Medieval Translator: The Theory and Practice of Translation in the Middle Ages, ed. Roger Ellis (Cambridge: D. S. Brewer, 1989); also in the series of panels at Kalamazoo, and associated publications such as Medieval Translators and their Craft, ed. Jeanette Beer, Studies in Medieval Culture 25 (Kalamazoo, MI: Western Michigan University, 1989). Rethinking Medieval Translation: Ethics, Politics, Theory, ed. Emma Campbell and Robert Mills (Cambridge: D. S. Brewer, 2012) offers a series of studies on medieval translation and modern translation theory. 63 Rita Copeland, Rhetoric, Hermeneutics and Translation in the Middle Ages: Academic Traditions and Vernacular Texts (Cambridge: Cambridge University Press, 1991).

Introduction

19

approach to individual texts.64 The same techniques and rhetorical practices which could be used in imitating and adapting classical models could be applied by the vernacular translator to his model, whether that model was Latin or another vernacular. Even in this brave new world of studies of medieval translation into the vernacular, the focus has been largely on translation from classical languages, particularly Latin: inter-vernacular translation is only rarely the main topic of study.65 As Copeland’s work has shown, vernacular translation of Latin texts was often part of an ambitious project of linguistic rehabilitation: the act of translation demonstrates the adequacy of the target language for the presentation of material hitherto confined to the source language, and, by implication, for the production of original material which might previously have been written only in the source language.66 Until these recent preoccupations with translation theory, a major focus of interest in medieval 64 Rita Copeland, ‘The Fortunes of “Nor verbum pro verbo”: or why Jerome is not a Ciceronian’, in The Medieval Translator, ed. Ellis (1989), pp. 15–35; Karen Pratt, ‘Medieval Attitudes to Translation and Adaptation: The Rhetorical Theory and the Poetic Practice’, in The Medieval Translator, ed. Roger Ellis (London: Westfield Medieval Publications, 1991), pp. 1–27. 65 Important exceptions include Rikhardsdóttir, Medieval Translations; Eriksen, Writing and Reading; and, of particular relevance here, Stephen H. A. Shepherd, ‘The Ashmole Sir Ferumbras: Translation in Holograph’, in The Medieval Translator, ed. Ellis (1989), pp. 103–21. See also Karen Pratt, ‘Direct Speech – A Key to the German Adaptor’s Art’, in Medieval Translators and their Craft, ed. Beer, pp. 213–46; Brenda Hosington, ‘The Englishing of the Comic Technique in Hue de Rotelande’s Ipomedon’, ibid., pp. 247–64; Earl Jeffrey Richards, ‘The Fiore and the Roman de la Rose’, ibid., pp. 265–84; Ana Pairet, ‘Intervernacular Translation in the Early Decades of Print: Chivalric Romance and the Marvelous in the Spanish Melusine’, in Translating the Middle Ages, ed. Karen L. Fresco and Charles D. Wright (London: Ashgate, 2012), pp. 135–46. The relatively small number of vernacular texts in the corpus of translations into French (90% are translations from Latin) is demonstrated in the output of the monumental transmédie project: Translations médiévales: Cinq siècles de traductions en français au moyen âge (XIe–XVe siècles): Étude et répertoire, ed. Claudio Galderisi with Vladimir Agrigoroaei, 3 vols (Turnhout: Brepols, 2011), in which see Michel Zink’s ‘Préface’, p. 10. 66 Copeland, Rhetoric, Hermeneutics and Translation, p. 3.

20

Charlemagne in Medieval England

translation was on biblical translation, as it was here that the approach of the translator was most likely to be articulated. However, there are important differences between literary texts like the Charlemagne romances and the translation of Scripture, sacred and inspired by God where, therefore, there was some argument for ad verbum translation.67 Jeanette Beer divides medieval translations into two main camps: the literal and the literary. The former groups together ‘functional’ texts and sacred Scripture: ‘While the inherited legal formularies, Scripture, and philosophic/scientific treatises compelled translators toward close lexical equivalence, literary translation necessitated no such devices.’ 68 She goes on to ascribe this practice of ‘unfaithful translation’ to ‘the shifting medieval attitudes toward authorship and authority’.69 However, medieval translation practice was far from uniform, and the fact that Roger Bacon thought it necessary to admonish translators not to add material of their own suggests that this was common practice, even though such freedom to edit was thought undesirable in translations of ‘functional’ or religious material.70 With the specific texts at the heart of our study, the Middle English and Anglo-Norman redactions of the Charlemagne material, an explanation for the particularly dynamic approach to translation should perhaps be sought in the conditions of manuscript culture and the nature of the source texts themselves.71 The chanson de geste as a genre was particularly subject to mouvance, an instability which went beyond minor formulaic changes and included substantial remaniement of the 67 Jeanette Beer, ‘Introduction’, in Medieval Translators and their Craft, ed. Beer, pp. 1–7 (p. 1). 68 Ibid., p. 4. 69 Ibid. Kelly distinguishes two authority structures: ‘personal’ and ‘positional’. The former allows the translator to take authority over the text; the latter is more text-centred. In medieval practice authoritative texts, in particular the Bible, would not be open to the same kind of editing and alteration by the translator as were historical and fictional texts (The True Interpreter, pp. 205–18). 70 Bacon, cited in J. D. Burnley, ‘Late Medieval English Translation: Types and Reflections’, in The Medieval Translator, ed. Ellis (1989), pp. 37–53 (p. 49). 71 Tim William Machan, ‘Manuscript Culture’, in The Oxford History of Literary Translation in English, I, ed. Ellis, pp. 29–44.

Introduction

21

source text.72 Scribes and remanieurs could adapt, expand and abbreviate according to the demands of individual patrons or the exigencies of performance, for this was a genre which, even as it moved away from its oral roots, continued to be performed. This freedom in adapting inherited material was part of the way the medieval scribe worked. The accepted instability of the medieval manuscript text allowed the scribal adaptor to make the changes he saw as necessary. The medieval translator took the work of the remanieur one stage further, since his ‘reworking’ of the text also translated it into another language. It is thus not surprising that, on a continuum with adaptation at one end and ad verbum translation at the other, the translations of the Old French verse texts into Middle English are at the adaptation end. This freedom to adapt which was part of manuscript culture and, as we have noted, particularly a feature of the dissemination of the chansons de geste, showed signs of changing in the later Middle Ages. Dynamic translation did not cease to exist, but texts which had previously been open to adaptation and appropriation became more fixed in their form and wording when, with the introduction of printing, the source text in the form of a printed book itself became less subject to constant remaniement. The later printed Charlemagne texts provide a useful comparison in consideration of translation practice. Charlemagne narratives were first printed in English by Caxton and translated by him, no doubt with some commercial imperative to complete them quickly.73 Caxton’s Four Sons of Aymon (c. 1490) and Lyf of Charles the Grete (1485) have been criticized in modern times for their ad verbum approach: In his translation, Caxton has followed his original so closely and even slavishly, that at times it is difficult, if not impossible, to understand his meaning without a reference to the language of 72 Paul Zumthor, Essai de poétique médiévale (Paris: Seuil, 1972), pp. 65–75; Zumthor, ‘Intertextualité et mouvance’, Littérature 41 (1981), 8–16. 73 Joerg Fichte, in ‘Caxton’s Concept of “Historical Romance” within the Context of the Crusades: Conviction, Rhetoric and Sales Strategy’, in Tradition and Transformation in Medieval Romance, ed. Field, pp. 101–13, considers the inter-relation of principle and commercial impulses with regard to Caxton’s Godeffroy of Boloyne, concluding that ‘if his appeal [to the common goal of the liberation of the holy places] was not heeded, Caxton had the consolation of having supported a good cause and having made good money in the process’ (p. 113).

22

Charlemagne in Medieval England the original. Frequently he has used the very words of the French author, and still more frequently he has merely given them an English dress.74

This is the judgement of the Early English Text Society’s editor of Charles the Grete, while the editor of Caxton’s translation of Les Quatre Fils Aymoun calls it ‘perhaps one of the most literal that has ever been produced in the English language’, so close that it can scarcely be considered an independent version of the story.75 Another nineteenthcentury editor comments that Caxton was ‘content to follow his author with almost plodding fidelity’. 76 Caxton himself frequently used the prologue to assert his fidelity to his source, making use of what have been described as ‘fairly standard late medieval translators’ topoi of fidelity and humility’.77 Such close translation was not unusual practice in the later Middle Ages and it is not unreasonable to connect this to the increasing stability of the text through the use of printing,78 though, as Lotte Hellinga points out, ‘appearing in print does not necessarily signal the stabilization of a text’. 79 Again, Caxton’s prologues give us insight 74 William Caxton, The Lyf of Charles the Grete, ed. Sidney Herrtage, EETS ES 36, 37 (London: 1880–1), p. vii. 75 William Caxton, The Four Sons of Aymon, ed. Octavia Richardson, EETS ES 44, 45 (London, 1884–5), p. viii. The extreme fidelity of Caxton’s translations makes them unprofitable for the purposes of our textual analyses; however, we examine the circumstances of their production in Chapter 3. For discussion of Caxton’s choice of these Matter of France romances, see Megan Leitch, ‘Thinking Twice about Treason in Caxton’s Prose Romances: Proper Chivalric Conduct and the English Printing Press’, Medium Ævum 81 (2012), 41–69. 76 The Game and Playe of the Chesse, ed. William Axon (London: Elliot Stock, 1883), p. xxxiv. 77 A. E. B. Coldiron, ‘William Caxton’, in The Oxford History of Literary Translation in English, I, ed. Ellis, pp. 160–9. 78 On fifteenth-century translation practice, see Blake, Caxton and his World, pp. 147–50; The Oxford History of Literary Translation in English, I, ed. Ellis. 79 Lotte Hellinga, ‘From Poggio to Caxton: Early Translations of Some of Poggio’s Latin Facetiæ’, in Makers and Users of Medieval Books: Essays in Honour of A. S. G. Edwards, ed. Carol M. Meale and Derek Pearsall (Cambridge: D. S. Brewer, 2014), pp. 89–104 (p. 91). For a useful

Introduction

23

into this tension between a degree of ongoing editing of a text and increased stability, for Caxton used the paratext to explain the changes he made to his sources;80 in manuscript culture, the licence a translator had to edit his source text, to add or subtract material, abbreviate, embellish and elaborate, did not need to be commented on. But even so, the increased stability of the text was neither total nor sudden, any more than the increasing monolingualism of England was immediate. Texts which appeared in print also circulated in manuscript. Early printers did continue to view their source texts as ‘malleable, revisable, interpretable works’.81 Nonetheless, a gradual, but inexorable, process of textual stabilization was taking place, which would not be seriously challenged until the era of the internet, wikis and blogging. All translators have a target readership in mind which will, in part, determine the principles of translation. A modern translator may seek what Beer terms ‘the modern ideal of audience response equivalence’. Aiming for a response from the reader of the target text which is equivalent to that of the reader of the source text permits radical changes, including changes of form, where the literary form of the source text, in our case the chanson de geste, has no equivalent in the literary tradition of the target language. Beer argues that medieval translation never had ‘audience response equivalence’ as its goal, and goes on to say that ‘appropriateness of form was determined from the predicted response of a particular target audience, never from an attempted match between presumed past response and presumed present one’. 82 However, in the case of the English Charlemagne romances, while the evidence of paratextual matter, particularly prologues, does not support any suggestion that medieval translators sought to replicate any ‘presumed past response’ (and such an ideal would seem to be contrary to the medieval tendency to ignore the historical context of the source text), the experimentation with different verse forms in the English verse texts summary of ‘revisionist histories of the book and print culture’ displacing earlier ideas of the ‘fixity’ of print, see Alexandra Gillespie, Print Culture and the Medieval Author: Chaucer, Lydgate, and their Books, 1473–1557 (Oxford: Oxford University Press, 2006), p. 9. 80 Coldiron, ‘William Caxton’, pp. 166–7. 81 Anne Coldiron, English Printing, Verse Translations, and the Battle of the Sexes, 1476–1557 (Farnham: Ashgate, 2009), p. 15. 82 Medieval Translators and their Craft, ed. Beer, p. 2.

24

Charlemagne in Medieval England

may indeed spring from a desire to find a literary form equivalent to that of the heroic gestes of France.83 In the treatment of content, rather than form, appropriation, rather than ‘audience response equivalence’, seems to have been the approach to the translated text, as the rest of this book will demonstrate; the target audience or readership was as important in a medieval context as in a modern one, and the production of the translation may equally have been driven by commercial imperatives, at least some of the time.84 The patron may have determined not only the matter to be translated, but also the form of the translation and the manuscript context in which the translation is found.85 A change in manuscript context, as in the case of the Anglo-Norman Fierenbras, could also mean a different way of reading a text, a different meaning.86 Indeed, the vexed question of why some French Charlemagne texts circulating in England were translated while others were not may be answered, at least in part, by the preferences of patrons. For example, the aristocratic and gentry patrons of translations would have had a vested interest in not extending beyond the francophone elite the textual community of narratives which seemed to justify rebellion against authority, such as Renaud de Montauban and Aspremont.87 83 See Chapter 1, pp. 86–97. 84 Coldiron, English Printing, returns at several points to the question of commercial motivation, e.g. pp. 4, 111. 85 See Roger Ellis, ‘Patronage and Sponsorship of Translation’, in The Oxford History of Literary Translation, I, ed. Ellis, pp. 98–115. For a comparison of the context of a fourteenth-century translator working under royal patronage and a sixteenth-century author translating his own work without patronage, see Jeanette Beer, ‘Patronage and the Translator: Raoul de Presles’ La Cité de Dieu and Calvin’s Institution de la religion Chrestienne and Institutio religionis Christianae’, in Translation and the Transmission of Culture between 1300 and 1600, ed. Jeanette Beer and Kenneth Lloyd-Jones (Kalamazoo, MI: Medieval Institute Publications, 1995), pp. 91–142. 86 See the discussion of British Library MS Egerton 3028, pp. 138–44. 87 See M. J. Ailes and Ad Putter, ‘French in Medieval England’, in La Francophonie européenne, ed. D. Offord and Vladislav Rjéoutski (Peter Lang: Bern, 2014), pp. 51–80. The insular situation differed from that on the continent in two respects: the presence of two vernaculars in England,

Introduction

25

The particular context in which our texts were translated was a multilingual one, and this must have affected the nature and purpose of translation. The multilingual nature of medieval England has been the subject of increasing attention in recent decades, in a period when modern Britain has also had to come to terms with the experience of various languages sharing the same geographical space.88 It is now recognized that ‘English–French bilingualism remained a central fact in the linguistic life of England well into the late medieval period’.89 Even after the Charlemagne texts had been translated into Middle English, they continued to be copied in French. The Pseudo-Turpin Chronicle circulated in England in three languages: Latin, French (continental and Anglo-Norman)90 and Middle English; in the wider context of the British Isles it was also translated into Welsh and Irish.91 At the same time it should be remembered that the source texts for all our translations, apart from the versions of the Pseudo-Turpin Chronicle, were and in France the existence of fiefdoms so vast as to more than rival the French king’s lands. 88 See, for example, Medieval Multilingualism: The Francophone World, ed. Christopher Kleinhenz and Keith Busby (Turnhout: Brepols, 2010); Multilingualism in Medieval Britain, ed. Judith A. Jefferson and Ad Putter (Turnhout: Brepols, 2013); Conceptualizing Multilingualism in Medieval England, c. 800–c. 1250, ed. Elizabeth M. Tyler (Turnhout: Brepols, 2011). An acceptance of the reality of multilingualism also lies behind the ‘Medieval Francophone Literature outside France’ project: http://www. medievalfrancophone.ac.uk/ [accessed 23/08/2016]. 89 Richard Ingham, ‘Anglo-Norman: New Themes, New Contexts’, in The Anglo-Norman Language and its Contexts, ed. Richard Ingham (York: York Medieval Press, 2010), pp. 1–7. 90 Traditionally, earlier insular French, which has very distinctive dialectal markers, has been called ‘Anglo-Norman’ and later insular French termed ‘Anglo-French’; more recently the terms ‘French of England’ and ‘insular French’ have been used to cover the whole period. We use the term ‘Anglo-Norman’ to distinguish works written in the dialect of French used in England, along with the more extensive term ‘insular French’. 91 Marianne Ailes and Suzanne Leedham, ‘Le Pseudo-Turpin en Angleterre’, Cahiers de recherche médiévales et humanistes 25 (2013), 495–514; Celtic material will be covered in the volume on Charlemagne in the Celtic and Scandinavian Worlds, ed. Helen Fulton and Sif Rikhardsdóttir, in the series ‘Charlemagne: A European Icon’.

26

Charlemagne in Medieval England

chansons de geste, a genre particularly associated with France (to the extent that some modern scholars have denied the recognition of the genre in an Anglo-Norman context).92 The distinction between ‘domestication’ and ‘foreignization’ used by Antoine Berman and Lawrence Venuti is much less clear in a context where the geographic area of production is the same and the cultural zones (to borrow a term from Stephen Greenblatt) of the source and target language overlap.93 Linguistic permeability is increasingly seen as part of the landscape of medieval England.94 The large number of lexical borrowings from French into English makes an assessment of our texts as translations particularly problematic. The Middle English Sir Ferumbras is an interesting case: its high proportion of romance words is clearly different from the ‘lazy’ taking over of French words into English in Caxton’s translations. To some extent this usage can be seen as an aspect of the fashionable enhancement of the text, but is it also a way of retaining a sense that this is and remains a ‘French’ text, an acknowledgment that the text is a ‘romance’ in the etymological sense of a text (originally) written in a romance language? Viewed or read thus, the technique might be seen in modern terms as deliberately 92 See Chapter 1, pp. 36, 49, 57–9; Luke Sunderland challenges this notion that the Matter of France belongs to France in ‘Bueve d’Hantone / Bovo d’Antona: Exile, Translation and the History of the chanson de geste’, in Rethinking Medieval Translation, ed. Campbell and Mills, pp. 226–42 (pp. 226–9). 93 Venuti, The Translator’s Invisibility; Antoine Berman, L’Épreuve de l’étranger (Paris: Gallimard, 1995); Antoine Berman, ‘Translation and the Trials of the Foreign’, in The Translation Studies Reader, ed. Venuti, pp. 276–90; Stephen Greenblatt, ‘Culture’, in Critical Terms for Literary Study, ed. Frank Lentricchia and Thomas McLaughlin, 2nd edn (London: University of Chicago Press, 1995), pp. 225–32. For a brief definition of ‘domestication’ and ‘foreignization’ see The Routledge Companion to Translation Studies, ed. Jeremy Munday (London: Routledge, 2009), pp. 183, 189. 94 Ingham, ‘Anglo-Norman: New Themes, New Contexts’, p. 2; Jocelyn Wogan-Browne, ‘General Introduction: What’s in a Name? The “French” of England’, in Language and Culture in Medieval Britain, ed. WoganBrowne et al., pp. 1–13; Emma Campbell and Robert Mills, ‘Introduction: Rethinking Medieval Translation’, in Rethinking Medieval Translation, ed. Campbell and Mills, pp. 1–20 (p. 11).

Introduction

27

‘foreignizing’. The chansons de geste belong equally to the culture of medieval France and to the francophone culture of medieval England and elsewhere.95 Although French was always a minority language in medieval England in purely numerical terms, it was never unfamiliar and was not therefore the vehicle for an alien culture.96 For the translation of our texts, the time in which they were redacted or copied, and the changes made to the texts to render them meaningful in a different context, may be as significant as the different language. Much work on medieval translation has made use of authors’ prologues to discuss approaches to translation, taking into consideration what the translators themselves say about both approach and purpose.97 The general assumption made is that ‘the point of translation is to make a text accessible to its readers who lack the requisite skills to read the original’. 98 Prefaces of medieval translations from Latin into medieval French which tell us the purpose of the translation support this. Thus, for example, the author of La Vie de Seint Clement makes it clear that his purpose in translating his Latin source text is to make it available to those with no Latin, so that they may learn from it.99 He is writing in the early thirteenth century and explicitly states that he expects everyone but the most vulgar to be able to read French. Our Anglo-Norman 95 Sunderland, ‘Bueve d’Hantone / Bovo d’Antona’; M. J. Ailes, ‘What’s in a Name? Anglo-Norman Romances or chansons de geste?’, in Medieval Romance, Medieval Contexts, ed. Rhiannon Purdie and Michael Cichon (Cambridge: D. S. Brewer, 2011), pp. 61–76. See also Chapter 1. 96 Ardis Butterfield writes of the ‘entanglements between peoples, cultures, histories and languages’, which are the subject of her study The Familiar Enemy, p. 393. 97 See, for example, The Idea of the Vernacular: An Anthology of Middle English Literary Theory, 1280–1520, ed. Jocelyn Wogan-Browne et al. (Exeter: University of Exeter Press, 1999); Elizabeth Dearnley, Translators on Translation: French–English Translation and the Middle English Translator’s Prologue, forthcoming. 98 Gloria Allaire, ‘Literary Evidence for Multilingualism: The Roman de Tristan in its Italian Incarnations’, in Medieval Multilingualism, ed. Kleinhenz and Busby, pp. 145–53 (p. 151). 99 La Vie de Seint Clement, vol. I: Text: 1–7006, ed. Daron Burrows (London: ANTS, 2007); the poet declares he is writing so that ‘plusurs genz pru en eussent’ [‘so that many people may have benefit from it’] 34.

28

Charlemagne in Medieval England

translation of the Pseudo-Turpin Chronicle asserts in a similar vein that it is written so that those who read or hear the tale may both learn and be entertained by it: Le mettray en romaunz ke ceus ke le orrunt i preynount essaumple e s’i delitunt a oyer les hauz feez et les hauz miracles. [I have put it into the romance language so that those who hear it may learn from the example and take pleasure in hearing of great deeds and mighty miracles.]100 The Pseudo-Turpin Chronicle stands out among our translated texts in the visibility of the translator; only Caxton’s later printed prose translations are similarly marked as translations. Anne Coldiron has nuanced Venuti’s comments regarding the invisibility of the translator in works dating from the eighteenth to the twentieth centuries, with an analysis of the ‘visibility’ of the translator in earlier times as ‘part of a complicated, powerful textual system designed to guarantee and to display a text’s auctoritas’.101 The Pseudo-Turpin is the one insular Charlemagne text which is given such a ‘hierarchy of authority’, beginning with St James himself telling Turpin to write the chronicle, and including the patrons of the translator, William de Briane. Where the Middle English verse Charlemagne romances have prologues, they are similar to those of the chansons de geste – exhortations to the listeners, with, sometimes, a reference to the content; they do not present themselves as translations.102 The translator is invisible; the translation functions as an independent text. It is probably safe to speculate that at least part of the purpose behind the fact of translation was indeed to ‘make a text accessible to its readers who lack the requisite skills’, but this is only part of the story. Some of the owners of the manuscripts would certainly have had a degree of competence in French; evidence suggests that levels of competence in French may even have improved during some periods of the Hundred 100 The Anglo-Norman Pseudo-Turpin Chronicle of William de Briane, ed. Ian Short, ANTS 25 (Oxford, 1973). 101 A. E. B. Coldiron, ‘Visibility Now: Historicizing Foreign Presences in Translation’, Translation Studies 5 (2002), 189–200 (p. 190). 102 Several of the Middle English Charlemagne romances are acephalous: see Chapter 3.

Introduction

29

Years’ War, due to increased contact with our neighbours across the Channel;103 indeed, writing of events in 1360, John M. Bowers notes the ‘francophilia, not to mention francophonia’ of the English court.104 Some patrons were aristocrats who owned French books, and may even have sometimes supplied the translator with their source text.105 The translations, however, broadened the textual community, making available to whole households the culture and narratives which were part of the (still) prestigious French-language culture. It is worth noting the tension, or even incompatibility, between translation as appropriation and translation as access.106 The verse Charlemagne romances, with their invisible translators, present as appropriations of the source texts, replacing (or displacing) them; the Pseudo-Turpin in its prologue is presented as giving access to the source text. It has been argued that the decision to write in English was a political act,107 but this act was not necessarily a simplistic one of promoting an ‘English’ identity over a ‘French’ one, as Bowers argues with regard to Chaucer’s output.108 Nicholas Watson claims that ‘writing in English raised questions about national/cultural identity and about

103 Anne Curry et al., ‘Languages in the Military Profession in Later Medieval England’, in The Anglo-Norman Language and its Contexts, ed. Ingham, pp. 74–93. 104 John M. Bowers, ‘Chaucer after Retters: The Wartime Origins of English Literature’, in Inscribing the Hundred Years’ War in French and English Cultures, ed. Denise N. Baker (New York: State University of New York Press, 2000), pp. 91–125 (p. 92). 105 This is likely to have been the case for William of Briane, translator of the Pseudo-Turpin Chronicle, see Short’s edition, p. 5. The Earl of Oxford supplied Caxton with a copy of Christine de Pizan’s Livre de Fait d’Armes to translate (Coldiron, ‘William Caxton’, p. 166). It is even possible that he had some role in providing Caxton’s source text for the Foure Sonnes of Aymon. 106 Copeland, Rhetoric, Hermeneutics and Translation, pp. 224–6. 107 Nicholas Watson, ‘The Politics of Middle English Writing’, in The Idea of the Vernacular, ed. Wogan-Browne et al., pp. 331–52; see also R. F. Yeager, ‘Politics and the French Language in England during the Hundred Years’ War’, in Inscribing the Hundred Years’ War, ed. Baker, pp. 126–57. 108 Bowers, ‘Chaucer after Retters’, passim.

30

Charlemagne in Medieval England

the consequences of the spread of literacy’.109 When the texts being translated are focused on a king of France, the emperor Charlemagne, it would be facile to see the translation as a rejection of French and francophone culture. Arguably, the use of English was as much about cultural identity as about national identity. In the early thirteenth century the clerical translator of the Vie de Seint Clement could assume that those who could read at all would be able to read French, but as literacy increased this was no longer true.110 Translation implies both contact and an acceptance of alterity. French was an integral part of insular literary culture; the fact of translation of these French texts into English both acknowledged the alterity of the French and suppressed that difference by appropriating the Matter of France for England. Moreover, translation or transference from one context to another does not always imply inter-lingual translation.111 The AngloNorman redactions of texts originally composed in another dialect of Old French are part of a continual process of adaptation and reworking of texts, a process of which inter-lingual translation is simply the most radical expression. The copying of the French texts into different manuscript contexts and their adaptation to different literary criteria is another aspect of the instability of the text which arises out of medieval manuscript culture.112 It is this very instability, of course, that produces the conditions for the adaptations and appropriations of the Charlemagne tradition in new insular contexts, in both vernacular languages, over a period of 300 years. In this book we read the story of the so-called ‘Matter of France’ in multilingual late-medieval Britain by piecing together a contextual 109 Nicholas Watson, ‘The Politics of Middle English Writing’, p. 331. 110 For discussion see Lotte Hellinga and J. B. Trapp, ‘Introduction’, and J. B. Trapp, ‘Literacy, Books and Readers’, both in The Cambridge History of the Book in Britain, vol. III: 1400–1557, ed. Lotte Hellinga and J. B. Trapp (Cambridge: Cambridge University Press, 1999), pp. 1–20 and pp. 31–43. 111 Paul Ricoeur, On Translation, trans. Eileen Brennan (London: Routledge, 2006) writes of translation between different registers of the same linguistic community (p. 10; also cited by Catherine Batt, ‘Introduction’, in Translating the Middle Ages, ed. Fresco and Wright, pp. 1–7, p. 1); in a medieval francophone context we would also want to include translation from one dialect to another. 112 Machan, ‘Manuscript Culture’, p. 31.

Introduction

31

understanding of the genesis and reception of Charlemagne texts in England, in both insular vernaculars, looking at the various ways in which the legends appear to have been used at different times. We explore several identifiable sites of particular interest in the insular Charlemagne tradition: the marked concentration on narratives portraying Christian and Saracen heroes in single combat; the choice of texts representing the advance of Saracen armies into Christian lands rather than vice-versa; the proliferation of Middle English versions of these stories in the fourteenth and fifteenth centuries. The book recognizes three significant stages in the textual history: first, the appearance of Charlemagne-related texts in Anglo-Norman (whether derived from continental French sources or Anglo-Norman compositions) in the thirteenth century (an era of crusades); second, the making of translations into Middle English, and new copies of Anglo-Norman texts, in the fourteenth century (coinciding with the outbreak of the Hundred Years’ War); and third, the production of further Middle English translations and adaptations, and new copies of older texts, in the fifteenth century (when there were fears of Turkish invasion in Europe). However, the structure of the book reflects the clear insular focus on a small corpus of narrative traditions: the stories of Roland, Fierabras and Otinel. Each individual text is closely examined as a revisioning of the tradition in its own specific moment, and the reception of the texts in these three different periods is investigated in light of contemporary cultural, political and religious concerns, drawing on the evidence provided in the manuscript (or print) context of each. Our aim is thus to engage with the major questions about these insular Matter of France narratives on their own terms.113 113 In this, we take a different approach from that adopted in most other critical analyses of the insular tradition, which have tended to be synthetic and thematic, discounting the integrity of each text and its role within the larger manuscript collection it may be in. Examples range from Metlitzki, The Matter of Araby, to Siobhain B. Calkin, ‘Saracens’, in Heroes and Anti-Heroes in Medieval Romance, ed. Neil Cartlidge (Cambridge: D. S. Brewer, 2012), pp. 185–200. A notable exception is Janet Cowen, ‘The English Charlemagne Romances’, in Roland and Charlemagne in Europe: Essays on the Reception and Transformation of a Legend, ed. Karen Pratt (London: King’s College Centre for Late Antique and Medieval Studies, 1996), pp. 149–68.

1 Acculturating Charlemagne: The Insular Literary Context The Circulation of French-Language Charlemagne Material in England Reception of the Chanson de geste

T

he Matter of France entered insular literary consciousness largely through two genres: the chanson de geste and the chronicle, or rather the pseudo-chronicle in the text of the Pseudo-Turpin Chronicle. If the latter apparently brought a respectable clerical imprimatur, the former might be expected to have had more popular appeal. While only a restricted group of texts became fully appropriated into Middle English popular culture through translation into Middle English, there is evidence of much wider dissemination of the legends of Charlemagne in French-language texts. The chansons de geste are more distinct by form than any other French narrative genre. After some early experimentation in versification, by the middle of the twelfth century most French narratives were written in rhyming couplets.1 The chansons de geste, however, retained the verse form of the earliest surviving examples of the genre, namely the laisse, a strophe of variable length united by assonance or rhyme and with a line length of ten or twelve syllables. The basic unit of meaning is the hemistich. The narrative is related using an episodic structure. With these formal features we find a characteristic discourse related to the oral origins and dissemination of the chanson de geste. This discourse can be epitomised as using different forms of repetition: formulaic lines and half-lines, different patterns of repetition to link the laisses (including

1 On the experimentation with versification in the twelfth-century lai of Piramus et Tisbé, see https://www.liverpool.ac.uk/media/livacuk/ modern-languages-and-cultures/liverpoolonline/piramus.pdf, pp. 18–21.

32

The Insular Literary Context

33

reprise, picking up some of the sense of one laisse in the following one), and different forms of parallelism.2 Some aspects of the reception of the chanson de geste in the insular context, and in particular in England, have long puzzled critics. All of the oldest chansons de geste survive only in Anglo-Norman manuscripts. The Chanson de Roland, the iconic text of French nationalism, and in particular of nineteenth-century French nationalism,3 survives in its oldest known form in an Anglo-Norman manuscript, as noted above; this version of the text is known as ‘the Oxford Roland’, because of its modern resting place, the Bodleian Library.4 Almost as venerable (but set in the reign of Charlemagne’s son Louis), the only complete text of the non-cyclical version of the Chanson de Guillaume is an AngloNorman manuscript (BL MS Add. 38663); a further fragment discovered in St Andrews University is also Anglo-Norman.5 Another Charlemagnerelated text surviving uniquely in an insular manuscript is the fragment of Gormont et Isembart.6 Probably also dating from the twelfth century, the 2 For an introduction to the genre, see Catherine M. Jones, An Introduction to the chansons de geste (Gainesville: University Press of Florida, 2014); François Suard, La Chanson de geste (Paris: Presses Universitaires de France, 1993); Dominique Boutet, La Chanson de geste: Forme et signification d’une écriture épique au moyen âge (Paris: Presses Universitaires de France, 1993). 3 For an analysis of the nineteenth-century exploitation of the Roland legend and nationalistic interpretations, see Isabel N. Di Vanna, ‘Politicizing National Literature: The Scholarly Debate around La Chanson de Roland in the Nineteenth Century’, Historical Research 84 (2011), 109–34. The most anglophobic attitudes were those of the first editor of the text, Francisque Michel, and the later commentator Léon Gautier. Di Vanna expresses the perspective of Michel succinctly: ‘it seems that Michel could not accept that the national poem had been created by the enemy’ (pp. 118–19). 4 Bodleian Library MS Digby 23. 5 See below, p. 40. On the Chanson de Guillaume as an Anglo-Norman text, see P. E. Bennett, ‘La Chanson de Guillaume, poème Anglo-Normand?’, in Au carrefour des routes d’Europe: La Chanson de geste, Xe Congrès International de la Société Rencesvals (Aix-en-Provence: Publications du CUER MA, 1987), pp. 259–81. 6 Brussels, Bibliothèque royale, portefeuille II. 181; Gormont et Isembart: Fragment de chanson de geste du XIIe siècle, ed. Alphonse Bayot, 3rd edn,

34

Charlemagne in Medieval England

Pèlerinage de Charlemagne now exists only in later transcripts and editions, the unique manuscript, again an insular copy, having disappeared from the British Library in 1879.7 The reason for the survival of these texts only in Anglo-Norman copies is unclear: is it, as Rosalind Field has suggested, because insular tastes were conservative, so these older texts were not rejected?8 Or was England, in the words of Dominica Legge, ‘in the van’ of the writing and copying of these texts, rather than lagging behind?9 Given that the earliest manuscripts are insular, the evidence points to what Melissa Furrow designates ‘a cutting-edge Anglo-Norman reading public’.10 In fact we find manuscripts containing chansons de geste in England throughout the Middle Ages.11 Most of the surviving manuscripts actually copied in England date from the thirteenth century, perhaps not surprisingly as this was a period of great invention and popularity of the genre in France also, and a time when the use and knowledge of French was possibly at its most widespread in England. However, some manuscripts are later: BL MS Egerton 3028, for example, was copied in the fourteenth century. Other manuscripts copied in CFMA 14 (Paris: Champion, 1931). 7 Le Pèlerinage de Charlemagne, ed. and trans. Glyn S. Burgess (Edinburgh: British Rencesvals Publications, 1998). On the peculiarities of the manuscript of the Pèlerinage de Charlemagne, see Jules Horrent, Le Pèlerinage de Charlemagne: Essai d’explication littéraire (Paris: BellesLettres, 1961), pp. 140–50. 8 ‘The prolonged life of the chansons de geste in Anglo-Norman England is intriguing and indicates a taste for a slightly archaic, morally concerned and heroically active type of narrative poetry’ (Field, ‘Romance in England, 1066–1400’, p. 154). William Calin also deems insular readers ‘a conservative public that esteemed texts gone out of fashion in the continent’ (The French Tradition, p. 136). 9 Dominica Legge, Anglo-Norman Literature and its Background (Oxford: Clarendon Press, 1963), p. 3; see also Dominica Legge, ‘Archaism and the Conquest’, Modern Language Review 51 (1956), 227–9. 10 Melissa Furrow, Expectations of Romance: The Reception of a Genre in Medieval England (Cambridge: D. S. Brewer, 2009), p. 105; Melissa Furrow, ‘Chanson de geste as Romance in England’, p. 67. 11 Keith Busby, Codex and Context: Reading Old French Verse Narrative in Manuscript, 2 vols (Amsterdam: Rodopi, 2002), pp. 488–513; Furrow, ‘Chanson de geste as Romance in England’, pp. 65–72.

The Insular Literary Context

35

France in the fifteenth century were for English patrons.12 All of this suggests an ongoing taste for the genre. In Gormont et Isembart the king concerned is Louis, son of Charlemagne in the French epic tradition. Some sections of the narrative concern England, and all the early evidence of the narrative is insular: the narrative is found, with variations, in Geoffrey of Monmouth’s Historia Regum Britanniae, Wace’s Roman de Brut and Gaimar’s Estoire des Engleis.13 The only manuscript dates from the thirteenth century and is a fragment of a mere 661 lines.14 The editor of the chanson de geste fragment considers the underlying language of the text to be that of central France, implying a continental original, though the fragment was copied in England. The narrative (which can be reconstructed with some confidence from internal allusions in the surviving fragment and from other versions) centres on a renegade Christian knight in the service of a pagan king, Gormont, who besieged Cirencester, later handing over the lands he conquered in England to the Saxons, at whose request he had invaded the land in the first place. The fragment of chanson de geste deals with the end of Gormont’s life, in a battle against Louis, and the death-bed repentance of Isembart. We find here the same concerns about Saracen–Christian conflict, particularly 12 For example, BL MS Royal 15 E vi, or BL MS Royal 16 G ii. 13 Geoffrey of Monmouth, Historia Regum Britanniae, The History of the Kings of Britain, ed. Michael D. Reeve, trans. Neil Wright (Woodbridge: Boydell Press, 2009), pp. 256–7, 264–5; Wace, Roman de Brut, ed. and trans. Judith Weiss (Exeter: University of Exeter Press, 1999), ll. 1194–98, 13379–13662; p. 30, pp. 337–43; Gaimar, Estoire des Engleis, ed. Alexander Bell, ANTS 14–16 (Oxford, 1960), ll. 3235 ff. Wace, as a Channel Islander, illustrates well the difficulty in trying to define too sharply what we mean by Anglo-Norman. 14 Gormont et Isembart, ed. Bayot, p. v. The legend was also known on the continent, as can be seen by later allusions in the chronicles of Philippe Mouskès and of the anonyme de Béthune, as well as in Guillaume le clerc’s Fergus, and much later in Loher und Maller, a German redaction of the fifteenth-century Lohier et Mallart (Bayot, pp. xii–xiii); see also Caterina Menichetti and Maria Teresa Rachetta, ‘The Gormond et Isembart Legend in the French Monastic Chronicles of the Thirteenth Century’, in The Proceedings of the Oxford Congress of the International Société Rencesvals, ed. Marianne J. Ailes, Anne E. Cobby and Philip E. Bennett (Edinburgh: British Rencesvals Publications, 2015), pp. 469–94.

36

Charlemagne in Medieval England

Saracen aggression, that we find in the other early chansons de geste and in all the chansons de geste copied in England, but in a peculiar narrative with a more complex moral framework; there is no clear ‘hero’ and the central protagonists are the Saracen and the renegade. The allusions in historical and pseudo-historical texts confer some authority on this disturbing narrative, but there is no Middle English romance based on it. These early chansons de geste may have some claim to being considered ‘insular’, but, in that they are probably based on continental models, they also demonstrate the difficulty of separating ‘insular’ and ‘continental’ at this early stage. At the very least, the existence of these early texts and, in some cases, relatively early manuscripts testify to a taste for the chanson de geste in England.15 While in the past it has generally been considered that no chanson de geste was actually composed in England, this depends on how narrowly the genre is defined. Given the distinctiveness of the form and discourse, we consider that, in general, texts written in chanson de geste form should be considered chansons de geste.16 Part of the uncertainty over this definition is caused by the implications of translating chanson de geste as ‘Old French epic’, a phrase which puts the focus on content at the expense of form. For example, W. R. J. Barron questions the validity of the convention by which critics ‘dignify the earliest treatment of the Matter of France as epic, distinguished by seriousness of purpose, solidarity, and worth of values from the fantasy and escapism of 15 On the taste for the genre, see Short, ‘Literary Culture at the Court of Henry II’, pp. 335–61; also Furrow, ‘Chanson de geste as Romance in England’, pp. 67–72; Furrow, Expectations, pp. 96–106; Legge, AngloNorman Literature, pp. 4–5. 16 Genre definition is, of course, not quite as mechanical as this. However, the form undoubtedly establishes the horizon of expectations. See below pp. 57–9; Legge, in Anglo-Norman Literature, states that there is ‘no trace of an Anglo-Norman chanson de geste’ (p. 3); on the Roman de Horn and Boeve de Haumtone as chansons de geste, see Ailes, ‘What’s in a Name?’; also Ailes, ‘The Anglo-Norman Boeve de Haumtone as a chanson de geste’, in Sir Bevis of Hampton in Literary Tradition, ed. J. Fellows and I. Djordjević (Cambridge: D. S. Brewer, 2008), pp. 9–24; Short, in ‘Literary Culture at the Court of Henry II’, considers the later section of the Destruction de Rome to be the only Anglo-Norman chanson de geste (p. 352). There are a few chronicles which use the laisse of the chanson de geste to give a heroic resonance to the narrative.

The Insular Literary Context

37

romance’, and coins the term ‘romanticized epic’ to describe the later chansons de geste such as Fierabras.17 A formulation such as ‘romanticized epic’ is unnecessary if the chanson de geste is understood in formal terms and without undue emphasis on the earliest (and atypical) example of the Chanson de Roland. A fundamental question to consider here is whether so few chansons de geste were translated into English because these were the only ones circulating in England. Was the wider chanson de geste tradition, beyond the Charlemagne material, also well-known in insular culture?18 To answer this question, it is necessary to look beyond the surviving manuscripts. The continental chanson de geste tradition is normally divided into four groups or cycles, in a classification partly derived from the division made by the poet Bertrand de Bar sur Aube in his chanson de geste, Girart de Vienne (c. 1170). Bertrand describes three gestes: Des rois de France est la plus seignorie, Et l’autre après, bien est droiz que jeu die Fu de Doon a la barbe florie, Cil de Maience qui molt ot baronnie … La tierce geste, qui molt fist a prisier, Fu de Garin de Monglenne au vis fier. (13–16, 46–47)19 [First in rank is that of the kings of France, And I should tell you that the next Is that of Doon of the white beard, [Doon] de Mayence who was most noble … The third geste, much to be admired, Was that of Garin de Monglane of the proud countenance.]

The fourth cycle, the cycle des croisades, was a later development. Of Bertrand’s three family groupings, that of Garin de Monglane (also known as the cycle de Guillaume) was to achieve the greatest

17 Barron, English Medieval Romance, pp. 106, 98–9. 18 Field, ‘Patterns of Availability and Demand’, pp. 77–84. 19 Girart de Vienne, ed. Wolfgang G. van Emden, SATF (Paris, 1978); translation our own.

38

Charlemagne in Medieval England

cohesion, with a number of cyclical manuscripts.20 As time went on, the cycle of Doon de Mayence brought together the traitors and those originally seen as justified rebels against the king;21 the poems of this cycle are found in one major cyclical manuscript. While Bertrand’s is a useful classification, however, it is not absolute. Bertrand uses it in order to place his own text in the literary tradition, and particularly in the moral framework suggested by the cycle de Monglane; but the process of cyclification was dynamic and Bertrand’s groupings reflect just a moment in that process. Moreover, the corpus cannot always be neatly divided up.22 The Mayence poems are also linked to the cycle du roi by the prominent role given in them to Charlemagne, and so, in their circulation in England, they form part of the narrative of the reception of Charlemagne in the insular context. Of these groups, it does seem it was the poems of the cycle du roi which achieved most success in the insular context; from a modern perspective this seems ironic, but as we have noted, the association of these narratives with French nationalism was an invention of the nineteenth century. Only the following chansons de geste are listed in Dean and Boulton’s inventory of surviving Anglo-Norman manuscripts:23 Aspremont (7 manuscripts and fragments); La Chanson de Roland (1 manuscript); La Chanson de Guillaume (1 manuscript); La 20 Cyclical manuscripts present the texts in narrative sequence with some texts altered to better fit with the others. On these cycles see Suard, La Chanson de geste, pp. 79–82; on cyclicity in Old French, see Cyclification: The Development of Narrative Cycles in the Chansons de geste and Arthurian Romances, ed. Bart Besamusca (Amsterdam: Royal Netherlands Academy of Arts and Sciences, 1994); Luke Sunderland, Old French Narrative Cycles: Heroism between Ethics and Morality (Cambridge: D. S. Brewer, 2010). 21 M. J. Ailes, ‘Some Observations on the Grouping of Epics of Revolt in Manuscripts and Compilations’, Reading Medieval Studies 10 (1984), 3–19; on the cycle of Doon de Mayence, see La Geste de Doon de Mayence dans ses manuscrits et dans ses versions, ed. Dominique Boutet (Paris: Champion, 2014). 22 Girart de Vienne itself, carefully attached to the family of Guillaume d’Orange (or Garin de Monglane), is thematically linked to that of Doon de Mayence through the group of poems dealing with rebellious barons. 23 Ruth Dean and Maureen Boulton, Anglo-Norman Literature: A Guide to Texts and Manuscripts (London: ANTS, 2000).

The Insular Literary Context

39

Destruction de Rome (2 manuscripts); Fierabras (2 surviving and 1 lost Anglo-Norman manuscripts, one containing the remanié Fierenbras); Gormont et Isembart (1 manuscript); Otinel (1 manuscript; 2 short fragments). All but the Chanson de Guillaume and Gormont et Isembart are texts of the geste du roi. The dates of the Anglo-Norman manuscripts of the Matter of France suggest a sustained and continuous interest in the material. Moreover, there is no significant chronological gap between texts’ being composed in continental France and appearing in England. We have mentioned the twelfth-century manuscript of the Chanson de Roland. The insular manuscripts of Aspremont all date from the thirteenth century. The parameters of the likely date of the insular Otinel manuscript (Cologny Geneva, Bodmer Library MS 168) are from the middle of the thirteenth to the early fourteenth century. The dating of the longer Otinel fragment (known as the Mende fragment) is no less uncertain than that of the Bodmer manuscript.24 The Hanover MS (Hanover Landesbibliothek IV 578), the earlier of two manuscripts containing the Destruction de Rome and Fierabras, is considered to date from the late thirteenth and early fourteenth centuries, the Destruction section being slightly earlier than the Fierabras (see Chapter 2). The Egerton MS (BL MS Egerton 3028), which also contains both the Destruction de Rome and a redaction of the Fierabras narrative, is from the fourteenth century. The sumptuous 24 Jean-Baptiste Camps suggests, on the basis of some characteristics of the hand, a possible twelfth-century date. Palaeographically this is possible though not certain, and such an early date would push back the date of composition of Otinel; see ‘Otinel et l’Europe: Éléments pour une histoire de la diffusion de la geste’, in The Proceedings of the Oxford Congress of the International Société Rencesvals, 2012, ed. Marianne Ailes, Anne E. Cobby and Philip E. Bennett (Edinburgh: British Rencesvals Publications, 2015), pp. 137–56. Otinel must post-date Fierabras, which can be dated quite closely to the end of the twelfth century or the very beginning of the thirteenth (Marianne Ailes, ‘The Date of Fierabras’, Olifant 19 (1994–5 [1999]), 25–71); Dean and Boulton (Anglo-Norman Literature, p. 53), suggest the middle of the thirteenth century for the Mende fragment and this seems more likely. All of this puts the Mende fragment as being copied quite soon after the composition of the poem. The poem itself, however, does not seem to be Anglo-Norman, with no indication in the rhymes of Anglo-Norman phonology; we are grateful to Professor Richard Ingham of the University of the City of Birmingham for confirming this analysis of the language.

40

Charlemagne in Medieval England

Shrewsbury Book (BL MS Royal 15 E vi), given to Margaret of Anjou to mark her wedding to Henry VI, can be dated to 1445 and though not Anglo-Norman is part of the story of the reception of the material in England. Further evidence that other chansons de geste were circulating and even being copied in England can be seen in the fragments of other texts which survive, not all of which are listed in Dean and Boulton’s catalogue, some having been discovered since its publication. The Chanson de Guillaume of BL Add. MS 38663 is a non-cyclical text. A fragment of another version of the same narrative has recently been discovered in St Andrews University Library along with a fragment of Fouques de Candie;25 the Chanson de Guillaume fragment dates from the thirteenth century and the Fouques from the early fourteenth; both exhibit characteristic Anglo-Norman orthography. The recent discovery of these fragments, providing new evidence that Guillaume texts were known and copied in England, underlines the need to be wary of arguing from lack of surviving texts. Fouques de Candie also survives in three other Anglo-Norman fragments which have been generally ignored in considerations of the chanson de geste in England.26 While insular manuscripts do not display the same tendency to full cyclification found in some continental chanson de geste manuscripts, ‘anthologization’ did occur (as discussed below, pp. 138–51). The Chanson de Guillaume of the St Andrews fragments has undergone some alteration so that it fits better with Fouques de Candie, which may suggest a pairing of texts rather like the pairing of the Destruction de Rome and Fierabras.27 Renaud de Montauban, the core text of the Mayence cycle, also known as Les Quatre Fils Aymon, is the subject of an unusual manuscript of insular origin: Bodleian Library MS Hatton 59 consists of three extensive fragments of this chanson de geste in at least three different 25 M. J. Ailes and P. E. Bennett, ‘Deux nouveaux fragments de poèmes sur Guillaume d’Orange’, in Proceedings of the Oxford Congress, ed. Ailes et al., pp. 87–105. 26 It does not figure in Dean and Boulton, Anglo-Norman Literature. 27 Ailes and Bennett, ‘Deux nouveaux fragments’; Marianne Ailes and Phillipa Hardman, ‘Texts in Conversation: Charlemagne Epics and Romances in Insular Plural-Text Codices’, in Insular Books: Vernacular Miscellanies in Late Medieval Britain, ed. Raluca Radulescu and Margaret Connolly (London: British Academy Publications, 2015), pp. 31–47.

The Insular Literary Context

41

hands, bound together at a later date.28 Comparison with the different manuscripts included in Antonella Negri’s edition of the Vaucouleur episode of Renaud de Montauban shows them likely to represent a longer version of the narrative.29 Another chanson de geste outside the Charlemagne material copied in England is a largely ignored text dealing with crusade material, which has never been fully edited and survives in two Anglo-Norman manuscripts.30 This text, known variously as the Siège, the Prise and the Estoire d’Antioche, with its narrative of the taking of Antioch during the First Crusade, may be a translation of Baudri de Bourgueil’s Historia

28 The existence and condition of these fragments may suggest that the text was circulating in England in unbound booklets of single episodes: M. J. Ailes, ‘Témoins fragmentaires de la geste de Maience et la reception du cycle en Angleterre’, and Antonella Negri, ‘In limine du manuscrit Hatton 59 (Oxford, Bodleian Library)’, both in La Geste de Doon de Mayence, ed. Boutet, pp. 115–36 and pp. 97–111, respectively; Busby, Codex and Context, p. 509. 29 Antonella Negri, L’episodio di Vaucouleurs nelle redazioni in versi del ‘Renaut de Montaubon’ (Bologna: Pàtron, 1996); see also: Franz Josef Mone, ‘Die Haimonskinder’, Anzeiger fur Kunde der deutschen Vorzeit 6 (1837), cols 189–205; J. C. Matthes, ‘Die Oxforder Renaushandschrift, MS Hatton 42, Bodl. 59, und ihre Bedeutung für die renaussage; nebst einem Worte über die übingen in England befindlichen Renausmss’, Jahrbuch für romanische und englische Sprache und Literatur, n.s. 3 (1876), 1–32; Busby, Codex and Context, pp. 377, 509, 582. 30 Bodleian Library MS Hatton 77; BL MS Add. 34114 (the Spalding manuscript); a ‘crowd’ transcription project led by Emma Goodwin can be found at dhcrowdscribe.com/crowdmap-the-crusades/transcribe/. Fragments of the text also exist in two other manuscripts: see The Historia Ierosolimitana of Baldric of Bourgueil, ed. Steven Biddlecombe (Woodbridge: Boydell Press, 2014), p. lxxii, citing Peter R. Grillo, ‘Vers une édition du texte français de l’Historia Jerosolimitana de Baudri de Dol’, in Autour de la Première Croisade: Actes du Colloque de la Society for the Study of the Crusades and the Latin East (Clermont-Ferrand, 22–5 June 1995), ed. M. Balard (Paris: Publications de la Sorbonne, 1996), pp. 6–16. Evidence that the crusade cycle itself was known in England is cited above in the Introduction, a copy of the gestes being borrowed for Eleanor of Castile from the Master of the Temple.

42

Charlemagne in Medieval England

Ierosolimitana, though this has still to be fully investigated.31 It stands outside the First Crusade cycle as such but shares with it a connection to real and contemporary (or near contemporary) events. Here chronicle meets chanson de geste as the adaptor draws upon Baudri’s material to create his own text in the form of a different genre. It seems likely the two extant manuscripts are insular copies of a continental text which has not otherwise survived.32 Of course, being copied in England is only part of the story. There was throughout the Middle Ages, even during times of enmity between the lands ruled by the king of France and those held of the king of England, considerable commerce between the two realms, and manuscripts of French origin were owned and bought by insular readers of French as well as by French natives living in England. Though only fragments of insular copies of Mayence texts survive, there is considerable evidence that this cycle was known in England. There is in particular a fair amount of evidence for the circulation of Renaud de Montauban. Indeed, an early reference to the text is found in De naturis rerum of the English scholar Alexander Neckham (d. 1217).33 Gérard Brault has also identified four coats of arms in the thirteenthcentury Roll of Arms known as the Heralds’ Roll as being those of the four sons.34 Yet no English version appears until Caxton’s translation (c. 1490) of the fifteenth-century mise-en-prose, Les Quatre Fils Aymon, apparently working with a copy of one of the two versions published in Lyon c. 1482–5.35 It seems likely, from the fragmentary witnesses and the evidence of manuscript catalogues (see pp. 44–52), that the French 31 For a partial edition of this text, based on MS Hatton 77, see La Chanson de la Première Croisade, ed. Gabel de Aguirre). Very little study has been made of this text, but see Meyer, ‘Un récit en vers français de la première croisade’; on the manuscripts see also Busby, Codex and Context, pp. 530, 766; The Historia Ierosolimitana, ed. Biddlecombe, p. lxxii; Jennifer Markey, ‘From History to Epic: The Siege of Antioch in chanson de geste Literature’ (PhD thesis in preparation, University of Bristol). 32 Ed. cit., pp. 70–1. 33 Les Quatre Fils Aymon, ou Renaud de Montauban, ed. and trans. Micheline de Combarieu du Grès and Jean Subrenat (Paris: Gallimard, 1983), p. 16. 34 Gérard J. Brault, ‘The FitzEdmund Arms (Heralds’ Roll 586–89) and the French Epic Renaut de Montauban’, The Coat of Arms 12 (1997), 2–6. 35 Hellinga, ‘From Poggio to Caxton’, p. 92.

The Insular Literary Context

43

chanson de geste was known in England, but Caxton, or John de Vere, Earl of Oxford, who commissioned the work, chose the more modern genre of the mise-en-prose as the source text, in line with Caxton’s other prose romance publications. Another fragment of a Mayence cycle text survives in a form which shows it to have been in England at least by the fifteenth century. The fragment of Maugis d’Aigremont in The National Archives at Kew was probably copied in the west of France, to judge by the dialectal characteristics.36 The single bifolium was conserved due to its use as a protective sleeve for some documents from the reigns of Henry VII and Henry VIII. A previous owner has inscribed his name, ‘Thomas Wodde’, at the foot of the first folio in a hand which can plausibly be dated to the fifteenth century. BL MS Royal 16 G ii, another fifteenth-century manuscript which may have been made for insular consumption, exhibits a number of peculiarities. The most obvious of these is its hybrid form: the text offers the mise-en-prose of Renaud de Montauban sandwiched between two sections of the verse chanson de geste. It was not copied in England, but at the same workshop in Rouen as the well-known fifteenth-century Shrewsbury Book (BL MS Royal 15 E vi). While coats of arms of both John Talbot, Earl of Shrewsbury, and Margaret of Anjou abound in that book, the related Royal 16 G ii manuscript has no evidence of patronage, yet is a rich and expensive volume, if not as sumptuous as the Shrewsbury Book. It seems surprising that such a manuscript bears no coat of arms or other mark of ownership or patronage, and it is only possible to speculate about whether an English patron had commissioned it;37 it certainly found its way to England, for it is listed in the Richmond inventory of 1535.38 The Shrewsbury Book itself is a 36 TNA SC12/37/34; edited in M. J. Ailes, ‘Deux fragments inconnus de Maugis d’Aigremont’, Romania 116 (1998), 415–30; on the linguistic traits, see pp. 416–19. 37 M. J. Ailes, ‘Deux manuscrits de la chanson de geste de l’automne du moyen âge: Un renouvellement du genre?’, in Mélanges Roussel, ed. Françoise Laurent (forthcoming) offers the speculative suggestion that it may have belonged to Cecily Neville, wife of the Duke of York. It may be that the decoration of the manuscript was never completed because of the English departure from Normandy. 38 http://www.bl.uk/catalogues/illuminatedmanuscripts/record. asp?MSID=8360&CollID=16&NStart=160702 [accessed 23/08/2016].

44

Charlemagne in Medieval England

key volume for the way Charlemagne material was used in England, to be discussed further.39 It contains a livre de Charlemagne which consists of Aspremont, Simon de Puille and Fierabras, but also has a version of La Chevalerie Ogier and a prose Renaud de Montauban, the same text as that found in the prose section of Royal 16 G ii. Behind surviving manuscripts and fragments lie the shadows of texts long since lost and now known only by references in library catalogues and inventories. The most complete library catalogues are those of religious institutions, many of which had some books in vernaculars.40 However, it is not always easy, or possible, to identify texts listed in catalogues, even where an incipit is given; sometimes it is not even possible to know in what language a text was written.41 With monastic and ecclesiastical institutions, titles are often given in Latin even when the text is in the vernacular, sometimes (but not always) with gallice 39 See Chapter 2, pp. 146–8. See also Jade Bailey, ‘Fierabras and the Livre de Charlemaine in British Library MS Royal 15 E vi: An Edition and Study’ (unpublished PhD thesis, University of Bristol, 2014), part of the AHRCfunded research project. 40 Busby points out that the ‘most significant evidence for ownership of vernacular narrative manuscripts in religious institutions … comes from England’ (Codex and Context, p. 749). 41 A typical example would be the reference to ‘all his books of romances’ in the will of Sir William Trussell in 1389: John Scattergood, ‘Literary Culture at the Court of Richard II’, in English Court Culture in the Later Middle Ages, ed. V. J. Scattergood and J. W. Sherborne (London: Duckworth, 1983), pp. 29–43 (p. 36). An inventory of the goods of Edward III’s youngest son, Thomas, Duke of Gloucester, was drawn up when they were forfeited to the Crown, the duke having been found guilty of treason. Among the eighty-four volumes of ‘livres de diverses rymances et Estories’ listed, the majority in French or Latin, no Charlemagne text is named. However, Thomas did own some untitled quires (presumably unbound) in French (‘divers veil quayers fraunceis saunz nouns’) and ‘un veił livre rumpus de Fraunceis de rymances’. Anthony Tuck, ‘Thomas, Duke of Gloucester (1355–1397)’, in Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004), also online at http://www. oxforddnb.com [accessed 23/08/2016]; Viscount Dillon, William St John Hope, ‘Inventory of the Goods and Chattels Belonging to Thomas Duke of Gloucester and Seized in his Castle of Pleshy, Co Essex, 21 Richard II (1397)’, Archaeological Journal 54 (1897), 275–308 (pp. 301–2).

The Insular Literary Context

45

or in gallicio noted in the inventory. Wills and inventories of property may be very unspecific and mention only ‘a boke in Frenshe’. 42 Some Benedictine monasteries owned a number of vernacular manuscripts, including some Charlemagne texts.43 Peterborough Abbey had a number of secular vernacular texts, including in one volume a version of the Tristan legend and ‘Amys et Amilion gallice’. 44 Another volume held the following: a) Fabule de animalibus et auibus moraliter gallice (identified by the editor as probably being the fables of Marie de France)45 b) Qualiter Sibilla regina posita sit in exilium extra Franciam et quomodo Makayre occidit Albricum de Mondisdene c) Versus de quodam claustrale facti The second item in this manuscript is the only reference in an insular context to the Chanson de la reine Sibille, or Macaire.46 The contents of another volume are itemized as Gui de Bourgogne and Otinel (Gesta Otuelis gallice).47 Both Otinel and Gui de Bourgogne are texts of the cycle 42 For example, note the references to ‘all his French books’ in the will of Roger de la Warr (1368), and to ‘all the French and Latin books’ of Lady Say (1369), in K. B. McFarlane, The Nobility of Later Medieval England (Oxford: Clarendon Press, 1973), p. 236. 43 On Latin texts see the discussion below of the Pseudo-Turpin Chronicle. 44 Peterborough Abbey, ed. Karsten Friis-Jensen and James M. W. Willoughby, Corpus of British Medieval Library Catalogues 8 (London: British Library in association with the British Academy, 2001), item BP 21 203, p. 130. De Mandach, Naissance et développement de la chanson de geste en Europe, 6 vols (Geneva: Droz, 1961–93), I, 268–74, seems to assume that manuscripts in Peterborough Abbey library were perforce copied there; he presents Peterborough as a ‘centre de diffusion des Gestes de Charlemagne et de Chansons de Roland’ (p. 268); this is mere speculation, as Busby points out (Codex and Context, p. 755); de Mandach speculates further about the possible identification of some of the manuscripts at Peterborough (Naissance et développement, pp. 272–4). 45 Peterborough Abbey, ed. Friis-Jensen and Willoughby, item BP 21 331, p. 171. 46 Correctly identified by the editor, Peterborough Abbey, ed. Friis-Jensen and Willoughby, pp. 171–2. 47 The editors of the medieval catalogue at Peterborough (Peterborough Abbey, ed. Friis-Jensen and Willoughby, item BP 21 338, pp. 173–4)

46

Charlemagne in Medieval England

du roi and, as such, might seem to belong together. According to the exceptionally detailed catalogue of Dover Priory library, compiled in 1389, its collection included a number of vernacular texts, some of which can be more securely identified than others.48 Le Romonse du roy Charlemagne is considered by the editor to be the Chanson d’Aspremont.49 Another volume includes in one codex texts in three languages: Latin, French and English. One item is La Romonse de Ferumbras; another, the Gesta Octouiani imperatoris in gallicis, can be identified as Octavian in French; while the Gesta Karoli magni in gallicis, with an incipit of Ore escutz segnouris, tentatively identified by the editor as ‘Renaud de Montauban or an unidentified text on the same subject’, could be any of a number of chansons de geste, or a French Pseudo-Turpin.50 David Bell, the editor of the relevant volume of the Corpus of speculate that the scribe made a mistake, writing ‘Guy de Burgoyne gallice’ when he meant to put ‘Gui de Warewic’ (Bodmer Library MS 168 contains these two texts in that order, though with Waldef preceding the Gui de Warewic: see Françoise Vielliard, Manuscrits français du moyen âge: Catalogue (Cologny-Genève: Fondation Martin Bodmer, 1975), pp. 93–9). This seems a strange mistake to have made, as Gui de Warewic was a popular insular text, but the Peterborough cataloguer does, according to the editors, have a ‘habit … of omitting the first item in a volume (p. 174); see Busby, Codex and Context, p. 755; de Mandach also speculates about this manuscript (Naissance et développement, I, pp. 267–8). 48 Although the second section of the catalogue includes incipits they do not always clearly identify texts. 49 Dover Priory, ed. William P. Stoneman, Corpus of British Medieval Library Catalogues 5 (London: British Library in association with the British Academy, 1999): different information about each volume is given in each of three sections, the first of which was for the precentor, the second for the brothers and the third for scholars wishing to find a particular volume; the text identified as Aspremont is listed as le Romonse du roy Charlemagne, item 364 (part 1, p. 41, part 2, p. 147, part 3, p. 226). 50 For item 170, the multilingual codex, see Dover Priory, ed. Stoneman, pp. 99–100; see also M. R. James, The Ancient Libraries of Canterbury and Dover (Cambridge: Cambridge University Press, 1903), pp. 430, 484, 460–1; Madeleine Blaess, ‘L’Abbaye de Bordesley et les livres de Guy de Beauchamp’, Romania 78 (1957), 511–18; Busby, Codex and Context, pp. 751–3.

The Insular Literary Context

47

British Medieval Library Catalogues series, noted of the library at the Premonstratensian Abbey at Titchfield in Hampshire that the ‘number and variety of the volumes in the Titchfield Library were remarkable’, as they had ‘a large and varied collection of materials in French’.51 Listed in their catalogue of 1400 is a vita Amici et Amilonis, though whether this referred to the romance or chanson de geste version of this semi-hagiographic narrative we cannot tell; its title suggests it is being perceived as a ‘vita’ or saint’s life, rather than a romance or secular text.52 The catalogue also lists in one volume: Gesta Karoli Francie in quaterno; Cato in ritimo; bella Karoli et Agulandi.53 Bell comments that the third item in this manuscript ‘obviously refers to a Romance of the Charlemagne cycle (Carolus is Charlemagne; Agulandus is a Saracen warrior) perhaps related to Raimbert de Paris’ Chevalerie Ogier de Danemarche … but the title is too imprecise to permit exact identification’. 54 We can, perhaps, be a little more precise than this: it is likely that the narrative here is either a Pseudo-Turpin Chronicle or a version of the Chanson d’Aspremont, both of which deal with Charlemagne’s battle against Agolandus and both of which had considerable success in England.55 Another chanson de geste, less clearly identifiable, is listed slightly earlier in the catalogue, in a miscellaneous manuscript, as ‘Belle Lodowyci regis filii Karoli, contra paganos’.56 This 51 David N. Bell, The Libraries of the Cistercians, Gilbertines and Premonstratensians, Corpus of British Medieval Library Catalogues 3 (London: British Library in association with the British Academy, 1992), pp. 180, 181; the catalogue survives as BL Add. MS 70507. 52 Item 207g in the catalogue; Bell, Libraries, p. 248. 53 Item 224 in the catalogue; ibid., p. 253. 54 Ibid. 55 Furrow, Expectations, p. 103, considers this to be a copy of the Chanson d’Aspremont. R. M. Wilson, ‘The Medieval Library of Titchfield Abbey’, Proceedings of the Leeds Philosophical and Literary Society (Literary and Historical Section) 5 (1940), 150–77, cited by Bell, draws attention to the fragment of an unknown chanson de geste (CUL MS Add. 3303) in which Agolant combats Ogier, edited by Paul Meyer, ‘Fragments de manuscrits français’, Romania 35 (1906), 22–67 (pp. 22–31). 56 Entry 217r; Bell, Libraries, p. 251. Bell notes: ‘this entry would seem to indicate some epic tale akin to the Chanson de Roland … but I have not been able to identify it precisely’.

48

Charlemagne in Medieval England

could be a Guillaume cycle text, as suggested by Furrow,57 or indeed a non-cyclical Guillaume text, as they frequently deal with Louis’ battles against the pagans; however, Louis is not often seen in action personally in Guillaume texts, so it could equally be a copy of a version of Gormont et Isembart, in which Louis indeed fights against the pagans.58 Sometimes legacies can say more about the reading habits of those in secular life than about monastic practice. Thomas Arnold, a monk of St Augustine’s in Canterbury, left a significant number of books to his abbey, ones which suggest the library of a literate secular.59 They comprise: Guy of Warwick, Yppomedon and other romances; Quatre Fils Aymon (Renaud de Montauban) in French; French books (no title); Launcelot; the Graal; Per le Galois; Devotions in French; a ‘French Book’; a codex of Guillaume d’Orange texts (previously thought to have been William the Marshal in French);60 Book of Hertu (Arthur?); another French book.61 Not all these books can be identified, but here is further evidence that Renaud de Montauban was known in England. St Augustine’s also owned two copies of Aspremont and a copy of Gui de Bourgogne as well as what may have been a Pseudo-Turpin.62 The books left by Guy de Beauchamp, Earl of Warwick, to Bordesley Abbey in 1305 also give some indication of the reading of an English 57 Furrow, Expectations, p. 103. 58 The library also held two copies of Gui de Warewic and one of Beues de Suthampton; see entries 216, 222, and 223. 59 James, The Ancient Libraries, pp. 294, 296, 297, 372, 373–4; on Thomas Arnold as a donor, see p. lxxiii; see also St Augustine’s, Canterbury, ed. B. C. Barker-Benfield, 3 vols, Corpus of British Medieval Library Catalogues 13 (London: British Library in association with the British Academy, 2008), II, 1431–2. 60 Now Bodleian Library MS Fr.e.32; St Augustine’s, ed. Barker-Benfield, entry BA 1533, pp. 1438–9. 61 James, The Ancient Libraries, item 1526 is ‘Katir ffitz Edmound in gallico’ (p. 373); James notes that while Christchurch was rich in Anglo-Saxon books, St Augustine’s was better off in French (and possessed one manuscript in German), and the priory at Dover had a mixture of French and English (pp. lxxxiv–lxxxv); see also St Augustine’s, ed. BarkerBenfield, entry BA1 1526, p. 1436. 62 St Augustine’s, ed. Barker-Benfield, entry BA1 1517, p. 1431; BA1 1519, now BL Add. MS 35289; BA1 1520, now Bodmer Library MS 11, p. 1433.

The Insular Literary Context

49

nobleman of the fourteenth century.63 The phrase which introduces this element of his donation refers to ‘touz les Rom[aunces] desouz nomes’, but here the word ‘romaunces’ almost certainly refers to the language in which they were written rather than the genre of the texts.64 They included a copy of Fierabras, apparently bound with a copy of the Estoire de Guillaume le Marechal;65 no codex containing both these texts has survived. Beauchamp’s library of chanson de geste material was quite extensive, with, in addition to the Fierabras, several texts in one volume: ‘Un volume qe parle des quatres principals gestes de Charles e de dooun de Meya[n]ce e de Girard de Vienne et de Emery de Nerbonne’. Of the individuals named, Aymeri de Narbonne and Girart de Vienne are eponymous heroes of texts of the Monglane, or Guillaume d’Orange, cycle,66 and Doon de Mayence is in texts of the Mayence cycle. Another volume in the donation contains the ‘romaunce [de] Emmond [e] d/e/ Ageland e deu roy Charles dooun de Nantoile e le Romaunce de Gwyoun de Nantoyl’.67 The first of these may well be another copy of Les Quatre Fils Aymon; the other items have been identified as Aspremont, Doon de Nanteuil,68 and Gui de Nanteuil; the 63 London, Lambeth Palace Library MS 577; Bell, Libraries, pp. 4–10 (p. 5); Blaess, ‘L’Abbaye de Bordesley’, commented that the list ‘porte la marque des goûts littéraires d’un très grand seigneur anglais du début du XIVe siècle’ (p. 518). 64 Furrow, in ‘Chansons de geste as Romance in England’, argues that the fact that both geste and romaunce are terms used by both chansons de geste and romances in England means that they are one genre; she cites as evidence the wording of the Bordesley donation. This would suggest a greater separation between continental and insular francophone literary culture than actually occurs. A French psalter in the donation is described as ‘un sauter de Romaunce’ (Bell, Libraries, p. 8) indicating that the word romaunce must refer to the language. 65 ‘Un volume del Romaunce des Marschaus et de Ferebras de Alixandre’ (Bell, Libraries, p. 9); there is no reason to assume, as Bell does, that this refers to a prose version of Fierabras. 66 Aimeri is one of Girart’s uncles. 67 Bell, Libraries, p. 6; Blaess, ‘L’Abbaye de Bordesley’, p. 513. 68 A lost chanson de geste linked to the cycle des barons revoltés; see M. J. Ailes, ‘Doon de Nanteuil and the Epic of Revolt’, Medium Ævum 52 (1983), 247–57.

50

Charlemagne in Medieval England

Nanteuil texts form a sub-cycle of the geste de Mayence.69 This is clearly a codex of various chansons de geste, so again we would take the word ‘romaunce’ as meaning a ‘narrative text in French’. Another volume contains the ‘romaunce de Gwy et de la reygne tut enterement’.70 This text is more difficult to identify as Gui is a common name in the chansons de geste; it could refer to Gui de Nanteuil again or Aye d’Avignon, or even to a version of Fierabras, as towards the end of the Fierabras narrative Gui de Bourgogne weds Floripas and they become king and queen of the former Saracen lands held by Floripas’s father.71 There are two manuscripts of Guillaume d’Orange: a Girart de Vienne and ‘Un volume del Romaunce de Willame de Orenges e de Tedbaud de Arable’, tentatively (and not very convincingly) identified by Madeleine Blaess as the Prise d’Orange; the Enfances Guillaume, where Tedbalt has a more significant role, is a more likely identification.72 There has been some discussion of this donation. Blaess surmises that the English nobleman might have been trying to get rid of books he no longer wanted.73 The wording of the donation is, however, clear. Members of the family who wish to read the books will have access to them. It seems that the reason for the donation is more probably to be found in the unstable times: Guy de Beauchamp has placed valuable books in the safekeeping of the monastery.74 In doing so he also made the books available to a wider readership, not just the monks but also those who borrowed books from them. As is well known, the royal court in England in the fourteenth century looked to what is often thought of as French chivalric literature, and it is important to remember that this consisted of more than

69 Doon de Nanteuil was one of the sons of Doon de Mayence in the thirteenth-century chanson de geste Gaufrey, ed. F. Guessard and P. Chabaille (Paris: APF, 1859), ll. 79–85. 70 Bell, Libraries, p. 8; Blaess, ‘L’Abbaye de Bordesley’, p. 513. 71 See Ailes and Bennett, ‘Deux nouveaux fragments’. 72 Ibid.; Tedbalt does not feature in the extant Prise d’Orange; a version of Fouques de Candie or Les Enfances Vivien seems more likely. 73 Blaess, ‘L’Abbaye de Bordesley’, p. 518. 74 Ailes and Bennett, ‘Deux nouveaux fragments’; Busby, Codex and Context, p. 749.

The Insular Literary Context

51

Arthurian romance:75 Charlemagne also had his place. Juliet Vale, in her study of the court of Edward III, describes a New Year’s gift given to the king by Queen Philippa in 1333 which included a ewer decorated with ‘ymaginubus Julii Cesar’, Jude Machabei, Regum Charleman’ et Arthuri, Rouland’, Oliueri, Galuan et Launcell’ de lacu’.76 On the death of Queen Isabella in 1352, a list of her books was drawn up:77 she owned a mixture of Latin and French books, many devotional. A list of secular books, ‘libri romanizati’, includes the following: Unus liber romanizatus de Duce de Basyns Unus liber cons’ de Emery et [sic] Nerbon Unus liber cons’ de Baudrous Unus liber cons’ de Oviel in gallic’ The book of ‘Duce de Basyns’ may refer to the chanson de geste Daurel et Beton, which survives only in an Occitan text, though Basin is not a particularly uncommon name in chansons de geste.78 Aimeri de Narbonne figures in several chansons de geste but the most likely would be the thirteenth-century eponymic Aymeri de Narbonne, also part of the Bordesley donation. Juliet Vale tentatively identifies ‘cons de Baudrous’ as Le Roman de Baudouin de Constantinople.79 The text of Oniel/Oviel 75 Furrow, ‘Chansons de geste as Romance in England’, points out that the fifteenth-century Shrewsbury Book contains no Arthurian material. 76 Juliet Vale, Edward III and Chivalry: Chivalric Society and its Context, 1270–1350 (Woodbridge: Boydell Press, 1982), p. 45; Vale points out the iconographic importance of this gift as an early visual representation clearly inspired by the concept of the Nine Worthies. Vale also notes a specific association of Edward with, among other heroes, Charlemagne, in the verse chronicle of the fourteenth-century Brabaçon writer Jan Boendale, who in his Van den derden Edewaert places Edward in the company of ‘heroes of the past: the Trojans, Cyrus of Persia, Alexander, Maccabeus, Charlemagne and Godefroy de Bouillon’ (p. 93); see Henry S. Lucas, ‘Edward III and the Poet-Chronicler John Boendale’, Speculum 12 (1937), 367–9. 77 TNA E101/393/4, fol. 8r; see Vale, Edward III, pp. 50, 170. 78 André Moisan, Répertoire des noms propres de personnes et de lieux cités dans les chansons de geste, 5 vols (Geneva: Droz, 1986), I, 213. There is a Duke Basin in the Fierabras tradition. 79 Vale, Edward III, p. 129, n. 125.

52

Charlemagne in Medieval England

remains unidentified but it could possibly be a misreading of Otinel, especially as the Anglo-Norman manuscript also uses the form Otuel in the first seventy lines of the text. All of this demonstrates that manuscripts of a wide range of chansons de geste were owned by individuals and institutions in England. These texts clearly formed part of the lighter matter owned by the rich and the learned. The evidence tells us nothing about whether the middle classes did or did not know these tales and legends. It remains the case, however, that, while texts from the Monglane and Mayence cycles were apparently available, the overwhelming majority of the chansons de geste known to have been circulating in England were from the cycle du roi. Texts in which rebellion is to some extent justified, or at least understood, and which were known in England are few: Aspremont, widely copied, but never translated into English,80 and Renaud de Montauban, which was belatedly translated but seems not to have been copied so extensively in England.

Chronicles and Pseudo-Chronicles Alongside the chansons de geste, and perhaps more obviously destined for consumption by the more educated in society, were the chronicles and pseudo-chronicles. A few copies of Einhard’s ninth-century chronicle life of Charlemagne, the Vita Karoli Magni, are found in libraries in Britain, particularly monastic libraries,81 but the very popular Pseudo-Turpin Chronicle, the Historia Karoli Magni et Rotholandi, was more important and influential. This text, composed in Latin around 1140,82 probably in 80 How this text was received in England will be discussed in more detail in the Anglo-Norman chapter of Charlemagne in Medieval Francophonia in the series ‘Charlemagne: A European Icon’. 81 English Benedictine Libraries: The Shorter Catalogues, ed. R. Sharpe et al., Corpus of British Medieval Library Catalogues 4 (London, British Library and the British Academy, 1996), B71, item 84, p. 433. One of the institutions which had a copy of Einhard was Reading Abbey: Alan Coates, English Medieval Books: The Reading Abbey Collections from Foundation to Dispersal (Oxford: Clarendon Press, 1999), pp. 43, 149; Coates notes ‘the interest of members of the Reading community in history and chronicles’ (p. 70). 82 The Anglo-Norman Pseudo-Turpin Chronicle, ed. Short, p. 1; Walpole’s edition gives c. 1145 as the date of composition of the Latin text (The Old

The Insular Literary Context

53

France, drew some of its material from the chanson de geste tradition. It purported to be an account of Charlemagne’s Spanish war, written by Archbishop Turpin of Rheims, who, contrary to the account given in La Chanson de Roland, had survived the Battle of Roncevaux. Though to the modern reader it is, as Short succinctly expresses it, ‘an uneasy marriage of the epic and the homiletic, the clumsy handiwork of pious propagandists’,83 it seems to have been accepted as a genuine historical account,84 knew astonishing success, and survives in an extraordinary number of manuscripts and versions: there are six different French translations from between 1200 and 1230 alone.85 In the second decade of the thirteenth century an Anglo-Norman clerk, William de Briane, translated the Latin into Anglo-Norman, and it is this insular text which is of particular interest here, though it was not the only French version of the Pseudo-Turpin to be circulating in England. It is evident that the slightly earlier text, sometimes referred to as the ‘Johannes’ translation, commissioned by Renaud, Count of Boulogne, in 1206, also circulated in England, as this was the source of Roland and Vernagu.86 French Johannes Translation, p. xi). C. Meredith-Jones concludes in his edition that the narrative was compiled between 1116 and 1145 (Historia Karoli Magni et Rotholandi, ou Chronique du Pseudo-Turpin (Paris: Droz, 1936), p. 74). 83 The Anglo-Norman Pseudo-Turpin Chronicle, ed. Short, p. 1. 84 Jean-Claude Vallecalle, ‘La Réception de la Chronique du Pseudo-Turpin en Europe’, Cahiers de recherches médiévales et humanistes 25 (2013), 465–9. 85 On the Pseudo-Turpin Chronicle across Europe, see Le Livre de SaintJacques et la tradition du Pseudo-Turpin, ed. Jean-Claude Vallecalle (Lyon: Presses Universitaires de Lyon, 2011); on the French tradition see The Old French Johannes Translation of the Pseudo-Turpin Chronicle, ed. Walpole, and La Traduction du Pseudo-Turpin du manuscript Regina 624, ed. Claude Buridant (Geneva: Droz, 1976), and the review of these editions by Ian Short, Medium Ævum 47 (1978), 123–30; Gabrielle M. Spiegel, in ‘PseudoTurpin: The Crisis of the Aristocracy and the Beginnings of Vernacular Historiography in France’, Journal of Medieval History 12 (1986), 207–23, excludes the Anglo-Norman version from her figure of six, counting only those versions composed within the confines of the French realm; see also The Anglo-Norman Pseudo-Turpin Chronicle, ed. Short, p. 2. 86 Ronald N. Walpole, ‘The Source MS of Charlemagne and Roland and the Auchinleck Bookshop’, Modern Language Notes 60 (1945), 22–6;

54

Charlemagne in Medieval England

England was a site of activity at the beginning of the composition of historical texts in the vernacular, specifically in French. It may be that the existence of the Anglo-Saxon Chronicle, and therefore familiarity with the idea that this kind of ‘true’ material could be written in the vernacular, was one reason why England was innovative in the writing of chronicles in French. All the earliest French-language chronicles have insular connections, at least if the Channel Islands, and therefore the production of Wace, can be included under this definition.87 These texts are all in verse, the normal form for narrative in French in the twelfth century. The bringing together of the insular tradition of writing verse chronicles in the vernacular and the new prose PseudoTurpin Chronicle gave an imprimatur of authority to material already circulating in the chansons de geste. For Gabrielle Spiegel, the significance of the translation into the vernacular is that the ‘vernacular prose history emerges as a literature of fact, integrating on a literary level the historical experience and expressive language proper to the aristocracy’, and is part of a response to specific concerns of the Northern French aristocracy;88 the translation in an insular context, under the patronage of the prominent Anglo-Norman courtier Warin Fitzgerold, underlines the continuity between continental and insular French. Although the Anglo-Norman version may not have been widely disseminated, the Pseudo-Turpin was circulating in England in both Latin and French, and copies are found in a number of library catalogues. Crowland Abbey owned a copy of Turpini Historia, probably a Latin Pseudo-Turpin.89 Rochester seems to have had at least two copies of the Pseudo-Turpin, one in a volume with William of Jumièges’ Historia Normannorum ducum and Fulcher of Chartres’ Historia on Walpole’s idea that there once existed a composite Charlemagne and Roland, see below pp. 76–9. 87 Dean and Boulton describe Wace’s Brut as belonging ‘to Anglo-Norman literature by its content, its influence, and the number of its AngloNorman manuscripts’ (Anglo-Norman Literature, p. 2). 88 Spiegel, ‘Pseudo-Turpin’, p. 207 (abstract). Spiegel argues that the aristocracy of France was instrumental in the promotion of the new idea of writing prose chronicles in French, one of the earliest being the Pseudo-Turpin. 89 English Benedictine Libraries: The Shorter Catalogues, ed. Sharpe et al., B25, item, 5, p. 126.

The Insular Literary Context

55

Hierosolimitana,90 the other, entitled Miracula sancti Iacobi apostoli cum jstoriade Runcievallo, a version of the Liber Sancti Jacobi (which includes the Pseudo-Turpin).91 Dover priory also had two copies, one of which was in French.92 It may have had a third copy, if, as suggested above (p. 46), the gesta Karoli magni in gallicis was a French Pseudo-Turpin.93 It seems that Peterborough Abbey also held several copies of the PseudoTurpin: one volume in the late fourteenth-century catalogue contained a Gesta Karoli secundum Turpinum episcopum quomodo adquisiuit Hispaniam as well as Liber de Gestis Normannorum (identified as ‘possibly the chronicle of William of Jumièges’);94 another volume in the same catalogue contained (among other items) ‘quedam de gestis Karoli magni’, possibly ‘excerpts from the Pseudo-Turpinus’ .95 In another volume, what is probably Geoffrey of Monmouth’s Historia Regum Britanniae (item a) is followed by the Pseudo-Turpin Chronicle divided into episodes (items b–e):96 a) Historia britonum b) Gesta Karoli Regis magni in Hispania quomodo liberavit uiam iacobitanam a potestate paganorum c) Bellum contra Eygolandum d) Bellum contra Ferraentum e) Bellum Runcievallis All of these manuscripts seem to have been in Latin, but another copy of what may be identified as the Pseudo-Turpin Chronicle was in French, found with two other texts in French: a) tractatus de [pro + /con]fessione gallice 90 Ibid., item 64, p. 485. 91 Listed in ibid., p. 512. 92 Dover Priory, ed. Stoneman, item 202d, p. 105, and item 369, p. 148; the latter gives an incipit in French. 93 Ibid., item 170k, pp. 99–100. 94 Peterborough Abbey, ed. Friis-Jensen and Willoughby, BP21, item 93, p. 90. 95 Ibid., BP21, item 176, pp. 120–2. 96 Ibid., BP21, item 322, p. 168. There is no basis for de Mandach’s assumption that this manuscript was in French (Naissance et développement, I, 272); compare Busby, Codex and Context, p. 755.

56

Charlemagne in Medieval England

b) De .vii. mortalibus peccatis gallice c) Quomodo Karolus adquisiuit coronam domini gallice d) De bello Vallis Runcie cum aliis gallice The editors suggest that items c and d in this manuscript are ‘excerpts from the Pseudo-Turpinus’. 97 While any (or all) of these three manuscripts may have contained a complete text of the Pseudo-Turpin, they are in each case entered in the catalogue as episodes.98 The presence of the Pseudo-Turpin Chronicle in such large numbers in monastic libraries demonstrates the status it was given as a history of Charlemagne. Indeed, evidence that the Pseudo-Turpin is being presented as ‘history’ can be seen as much in the company it keeps as in the text itself. It often sits in the same codex as histories of Britain or next to them in the catalogues. In these monastic manuscripts it was not infrequently copied with other Latin chronicles, in particular the Historia Normannorum ducum of William of Jumièges (Rochester, item 64; Peterborough, item 93) and Geoffrey of Monmouth’s Historia Regum Britanniae (BL MS Arundel 220; Peterborough, item 322).99 In other surviving manuscripts it is found with Higden’s Polychronicon.100 The history of Charlemagne is thus implicitly linked with that of the English monarchs. This is in keeping with the observations made above about the context in which some of the surviving Charlemagne texts are found. Many of the manuscripts listed in the catalogues of monastic libraries have not been identified, so it is difficult to come to even an approximate idea of how many copies of the Pseudo-Turpin were in England, and in what languages. All that we can say with certainty is that the fact of there 97 Peterborough Abbey, ed. Friis-Jensen and Willoughby, BP21, item 309, p. 165. 98 This is particularly interesting in view of the subdivisions in the Middle English Otuel and Roland, based on the Pseudo-Turpin Chronicle, which present the narrative as a series of battle episodes (see below, pp. 371–4). It is also possible that the chansons de geste may have circulated in episodic form in England; see Ailes, ‘Témoins fragmentaires de la geste de Maience’. 99 Peterborough Abbey, ed. Friis-Jensen and Willoughby, p. 168. 100 BL MS Royal 13 D I; CUL MS Dd.i.17; Stephen H. A. Shepherd, ‘The Middle English Pseudo-Turpin Chronicle’, Medium Ævum 65 (1996), 19–34 (19–20).

The Insular Literary Context

57

being only one surviving manuscript of the Anglo-Norman version cannot be taken to mean that Pseudo-Turpin tradition was not significant in England. It evidently was.

Insular Developments of the French-Language Charlemagne Material The Insular Chanson de geste The English Channel in the Middle Ages was as much a route for communication as a barrier. Between England and the rest of the francophone world there existed not only family bonds but also trade links and cultural exchange, not just between courts but also between towns.101 Yet the literary traditions of different parts of medieval Francophonia did not all develop in the same way. The legends of Charlemagne not only circulated in England, but were also subject to textual mouvance in that insular context; in addition, there were also some peculiarly insular developments of the chanson de geste tradition. It is this difference in the way the genre developed that has led some critics to deny the validity of the term chanson de geste for Anglo-Norman compositions. Dominica Legge, in her seminal study of Anglo-Norman literature, was the first to articulate the idea that there is ‘no trace of an AngloNorman chanson de geste’, though there are ‘romances in chanson de geste form’.102 This distinction is, as suggested above, open to challenge. It is certainly true that there is no completely new insular chanson de geste of any of the main cycles. Two of our core texts, however, though derived from a pre-existing continental tradition, should be treated as newly redacted insular chansons de geste. La Destruction de Rome is a prologue to the Fierabras narrative which is not found anywhere without a version of Fierabras, but it is a uniquely insular ‘prequel’ text.103 Fierenbras is entirely derived from the continental Fierabras but is so thoroughly reworked that it is difficult to deny it the status of a new chanson de 101 See, for example, Butterfield, The Familiar Enemy, esp. pp. 201–32; Bennett, ‘France in England’. 102 Legge, Anglo-Norman Literature, p. 3. 103 For further discussion, see pp. 59–62.

58

Charlemagne in Medieval England

geste.104 These two texts are distinct from any continental text, and the fact that they have been created from a pre-existing tradition should not detract from their importance: this was, after all, a common method of composition in the Middle Ages. Both of these texts are part of the cycle du roi and represent a substantial contribution to the development of the Fierabras tradition. Insular versions of Otinel and Aspremont are not radically reworked and as such should be considered Anglo-Norman copies rather than new Anglo-Norman texts. The older chansons de geste which survive only in Anglo-Norman copies are, as noted above, single-manuscript texts, and any consideration of the degree to which these texts have been reworked in an insular context and to suit insular tastes can therefore only be speculative. The critical assumption with the Chanson de Roland tends to be that it is the later versions which are remaniements, though as all the stemma which have been proposed suggest a lost original, it is possible to surmise that some remaniement may have taken place between the first written text and the Oxford version.105 Furrow specifically contrasts the ‘extraordinary formal variation’ of the insular Chanson de Roland, Chanson de Guillaume, Gormont et Isembart and the Pèlerinage de Charlemagne with the ‘more formally homogeneous genre that was to develop on the continent’. 106 She notes the fact that ‘no two have the same form’, pointing to their differences in line length and their sporadic use or otherwise of a refrain; but, given the early date of these texts, this is not surprising – the chanson de geste was, indeed, a ‘genre in creation, a genre not yet fixed’. 107 No genre, however, is entirely fixed; 104 Keith Sinclair, ‘Fierabras in Anglo-Norman: Some Cultural Perspectives’, in Anglo-Norman Anniversary Essays, ed. Ian Short (London: ANTS, 1993), pp. 361–77, argues for this text to be treated as an Anglo-Norman composition as it is thoroughly reworked; M. J. Ailes, ‘Fierabras and AngloNorman Developments of the chanson de geste’, in Epic Studies: Acts of the Seventeenth International Congress of the Société Rencesvals for the Study of Romance Epic, Storrs, CT (July 2006), ed. Leslie Zarker Morgan and Anne Berthelot = Olifant, special issue, n.s. 25.1–2 (2006 [2009]) 97–110. 105 For a summary of the stemma see Duggan, ‘General Introduction’ to La Chanson de Roland: The French Corpus, ed. Duggan et al., I, 1–38. On the remaniement of the Oxford Roland, see Chapter 4. 106 Furrow, ‘Chanson de geste as Romance in England’, p. 65. 107 Ibid.

The Insular Literary Context

59

the genre of chanson de geste remains dynamic and flexible, but with the irregular laisse as a marker. Although, as noted above, the language of these texts suggests a continental source, with no indication of Anglo-Norman phonology in the rhyme schemes, it may be that these texts were written down at an early date when the distinction was not yet significant,108 and that some of them took their first written form in Anglo-Norman rather than continental territory.109 Whatever the geographic location where those texts were first written down, England certainly provided fertile ground for their development, and, as Furrow states, was a ‘scriptorium for [the] dissemination [of the genre] and [a] library for its reception’. 110 The clearest contribution made by Anglo-Norman writers to the Charlemagne tradition is in the development of the Destruction de Rome material. In the extant Fierabras, the story of the destruction of Rome by attacking pagans under the leadership of the Saracen king Fierabras is briefly related as the background, the history behind the narrative of this late twelfth-century chanson de geste. Fierabras had attacked Rome, stolen precious relics of the Passion, killed the pope and sacked the city. The whole narrative of Fierabras follows on from this as Charlemagne seeks to regain the stolen relics, but in the extant poem the preliminary history is told initially in just a few lines when Fierabras himself is introduced:111 Et si wuloit par forche desus Ronme regner, Et tous cheuz de la terre a servage torner.

108 Ian Short, ‘Introduction’ to his edition of the Oxford text, in The French Corpus, I, p. 39. 109 This is at least the implication of Furrow’s analysis in Expectations, p. 106 and p. 108, where she refers to the ‘earliest phase’ of the genre as ‘the Anglo-Norman phase’. Clear dialectal evidence for this is absent, though the possibility should be borne in mind that the dialects were not entirely separate when the texts were first written down, which would also suggest that the cultures could not be separated. Bennett, in ‘La Chanson de Guillaume, poème Anglo-Normand?’, stops short of positing an Anglo-Norman original, though it is clear that the surviving manuscript represents an Anglo-Norman redaction. 110 Furrow, Expectations, p. 107. 111 For a full summary of the narrative, see Chapter 5.

60

Charlemagne in Medieval England Mais chil dedenz nel woudrent soufrir ne greanter; Por tant les fist destruire et Seint Pierre gaster. Mort i a l’apostoille et fist a duel finer, Et nonnainz et mongnie et mostiers vïoler; S’emporta la corronne qui mout fait a loer. (ed. Le Person, 54–60) [He wished to reign over Rome through force And take into slavery those of the land. But those in [Rome] did not wish to endure or accept this So he destroyed them and laid waste St Peter’s He killed the Pope and brought great sorrow He ravaged nuns and monks and great churches And took the crown which is most worthy of praise.]

A number of studies have demonstrated that this brief account refers to what may have been a well-known tale, an older version of the poem, narrating the sack of Rome, the stealing of the relics and the arrival of Charlemagne and the peers to avenge and rescue the holy city, culminating in a combat in which Fierabras is defeated by Oliver. Vestiges of a narrative of a battle outside Rome remain in the extant Fierabras; these correspond closely to a summary of the narrative as found in Philippe Mouskès’ Chronique rimée.112 To this older version of the chanson de geste was added what has been described as the queue postiche, a narrative of a beautiful Saracen princess and peers imprisoned within her father’s castle, while the narrative of the destruction of Rome was cut to produce the extant chanson de geste. The truncated opening left much unexplained, and at some point it seems that a poet, considering the story inadequate, added a new ‘prequel’ filling in the details of the attack on Rome, but with Floripas, the Saracen princess, and Balan, father to both Fierabras and Floripas, now given 112 Philippe Mouskès, Chronique rimée des rois de France, ed. Baron F. de Reiffenberg, Ac. roy. de Belgique, 2 vols (Brussels, 1836–8), ll. 4664–716; see also http://www.narrative-sources.be/naso_link_en.php?link=1136. For a more detailed discussion of the development of the legend, see Marianne Ailes, ‘A Comparative Study of the Medieval French and Middle English Verse Texts of the Fierabras Legend’ (unpublished PhD thesis, University of Reading, 1989), pp. 15–61.

The Insular Literary Context

61

important narrative roles. This new Destruction de Rome survives only in two Anglo-Norman manuscript versions. The more difficult question to answer is whether it only ever existed in Anglo-Norman. Was the Destruction de Rome an Anglo-Norman text?113 The two manuscript versions of the Destruction de Rome are very different. The shorter text in BL MS Egerton 3028, accompanying the Anglo-Norman Fierenbras, is undoubtedly an Anglo-Norman redaction. The longer text in Hanover Niedersächsische Landesbibliothek MS IV 578 is more problematic.114 Gustav Gröber, the first editor of the slightly older Hanover manuscript, mutilated the text to restore it to what he considered to be its ‘original picardism’. 115 His conviction that this had been the dialect of the poem was determined as much by his belief that the Destruction and Fierabras were by the same author as by his linguistic analysis of the poem. His analysis is actually very thorough, though his conclusions are unsatisfactory. Having described the text, with the prejudices of his time, as ‘cruellement depravé par de nombreux anglicismes’ (which he lists), he goes on to point out some Norman forms and a few Picard forms, some of which (such as the use of le for the feminine article) are also found in Anglo-Norman.116 The idea 113 An extract of the Destruction de Rome is included in Vernacular Literary Theory from the French of Medieval England, ed. Jocelyn Wogan-Browne, Thelma Fenster and Delbert W. Russell (Cambridge: D. S. Brewer, 2016); this book appeared too late for us to take into account. 114 The manuscript is described in the catalogue, Handschriften der Niedersächsischen Landesbibliothek Hannover, ed. Helmar Härtel and Felix Ekowski, 3 vols (Wiesbaden: Otto Harasowitz, 1982), II, 161–4; the catalogue is available online at http://www.manuscripta-mediaevalia. de/?AREA_2#|16. 115 ‘La Destruction de Rome: Première branche de la chanson de geste de Fierabras’, ed. Gustav Gröber, Romania 2 (1873), 1–48. 116 Ian Short, Manual of Anglo-Norman, 2nd edn (Oxford: ANTS, 2013), p. 51; Mildred K. Pope, From Latin to Modern French (Manchester: Manchester University Press, 1934), ¶1252 (iii), p. 465. Gröber has been much criticized, most recently in Luciano Formisano’s edition of the text: La Destructioun de Rome, ed. L. Formisano (Florence: Sansoni, 1981), pp. 6–8; Formisano, ‘Alle origini del lachmannismo romanzo. Gustav Gröber e la redazione occitanica del Fierabras’, Annali della Scuola Normale Superiore di Pisa. Classe de Lettere e Flosofia, 3rd series, 9 (Pisa, 1979), pp. 247–302. See also

62

Charlemagne in Medieval England

that the text might be considered in two parts, having, as it were, two stages of development, was put forward by A. Stimming and developed by Formisano, whose analysis of the language is the most thorough.117 Formisano argues for ‘une anglo-normandisation progressive’, dividing the text after line 698; both Formisano and Stimming see the latter part of the text as indubitably Anglo-Norman.118 While the dialectal characteristics of the latter part of the text are more marked, Formisano concedes that there is difficulty in identifying any other dialect in the first 698 lines. He lists a few details with a generally northern flavour,119 but all of these characteristics, even though they may primarily be associated with other dialects, can also be found in AngloNorman. Formisano proposes two possible explanations for the greater preponderance of Anglo-Norman features after l. 698: either the text was composed on the continent and partially reworked, or the original author was indeed Anglo-Norman but made an effort to conform to a continental scripta.120 The second possibility is just as likely as the first. Formisano’s edition in the ANTS ‘plain text’ series: La Destructioun de Rome, ed. L. Formisano, ANTS Plain Texts 8 (London: ANTS, 1990). 117 A. Stimming, ‘Die Entwicklungsgeschichte der Destruction de Rome’, Zeitschrift für romanische Philologie 40 (1919–20), 550–88; La Destructioun de Rome, ed. Formisano (1981), p. 48. 118 Stimming, ‘Die Entwicklungsgeschichte’, p. 585; La Destructioun de Rome, ed. Formisano (1981), p. 48. Short considers the second section of the Destruction as the only Anglo-Norman chanson de geste (‘Literary Culture at the Court of Henry II’, p. 352). 119 La Destructioun de Rome, ed. Formisano (1981), pp. 55–6. These are: the reduction of iee > e, a characteristic of northern French and AngloNorman (Short, Manual, p. 69; Pope, From Latin to Modern French, p. 479); a distinction between -ant and -ent in the rhyme scheme (laisse VII is in -ent, but is, as Formisano himself points out, too short to allow us to conclude definitively that there is a phonetic opposition between the sounds, a characteristic of Picard and Walloon, but also of Anglo-Norman (Short, Manual p. 49)); the use of the feminine personal pronoun el, which was most common in western texts, but which was also found alongside ele in Anglo-Norman (Short, Manual, pp. 131, 149; Pope, From Latin to Modern French, p. 464; for a more detailed discussion of the dialectal characteristics, see Ailes, ‘A Comparative Study’, pp. 114–19; for the discourse, see Ailes, ‘What’s in a Name?’, pp. 65–9). 120 La Destructioun de Rome, ed. Formisano (1981), p. 55.

The Insular Literary Context

63

One can further speculate that either there were two poets involved, or the Anglo-Norman poet progressively reverted to his more familiar orthography. It is also worth considering what contribution to the insular Charlemagne tradition was made by the Anglo-Norman Pèlerinage de Charlemagne, in which the emperor and his peers travel first to Jerusalem and then to Constantinople in response to the queen’s taunt that the emperor of Constantinople is reputed to be even more magnificent than her own husband; in the course of his travels Charlemagne collects various relics of the Passion. This text is unique in the degree to which it mixes the comic and the heroic, even to the extent of mocking its own heroes, though humorous elements are found in many texts of the genre.121 As with the Oxford Roland and La Destruction de Rome, it is difficult to know how much this text has been remanié in an insular context, for we have no continental manuscript with which to compare the Anglo-Norman. Anne Latowsky assumes that it was ‘in the cosmopolitan world of the Anglo-Angevin court that the imperial-themed Voyage of Charlemagne to Jerusalem and Constantinople originated’. 122 Theodor Heinermann, on the other hand, suggests a continental origin, with Eleanor of Aquitaine as Louis VII’s queen as a model for Charlemagne’s queen in the Pèlerinage.123 Linguistically this redaction is clearly Anglo-Norman, but not necessarily from an 121 On the ambiguity of tone, see Le Pèlerinage, ed. Burgess, pp. xlii–xliv; see also Anne E. Cobby, Ambivalent Conventions: Formula and Parody in Old French (Amsterdam: Rodopi, 1995), pp. 82–158; Anne Cobby, ‘Religious Elements in Le Voyage de Charlemagne à Jérusalem et à Constantinople’, in Au Carrefour des routes d’Europe, I, 367–82; J. L. G. Picherit, The Journey of Charlemagne to Jerusalem and Constantinople (Birmingham, AL: Summa, 1984) refuses the word ‘parodic’ (p. iv); this is rebutted by Margaret Burrell, ‘The Voyage of Charlemagne: Cultural Transmission or Cultural Transgression?’, Parergon, n.s. 7 (1989), 47–54. 122 Anne A. Latowsky, Emperor of the World: Charlemagne and the Construction of Imperial Authority, 800–1229 (Ithaca, NY: Cornell University Press, 2013), p. 227. 123 Theodor Heinermann, ‘Zeit und Sinn der Karlsreise’, Zeitschrift für romanische Philologie 56 (1936), 497–562; Burgess considers this plausible, see Le Pèlerinage, ed. Burgess, p. xi; Horrent, on the other hand, is critical of all attempts to particularize the satire, Le Pèlerinage de Charlemagne, pp. 116–22.

64

Charlemagne in Medieval England

Anglo-Norman original; there was certainly an older tradition relating Charlemagne’s supposed journey to the East.124 There has been much discussion of the genre of the Pèlerinage, largely because it does not fit the expected ‘epic’ tone.125 Arguably, the issue here is not whether it is or is not a chanson de geste, but our questionable assumption that all chansons de geste were epic. Insular versions of chansons de geste include the very heroic and serious Chanson de Roland, the Chanson de Guillaume with its admixture of comic elements, and the ‘romanticized epic’ Fierabras.126 Without a continental comparator we cannot know whether the parodic elements of the Pèlerinage were an insular addition to the tradition or were already present in an older redaction. Jules Horrent described the text as we have it as a ‘remaniement original’, raising the whole question of when a reworked poem becomes a different poem.127 What we can say is that the insular manuscript ensured the survival to modern times of a unique text, one in which, as Anne Cobby has described it, ‘Charles and the Franks fall short of their epic ideals and make a poor showing against the romance characters; yet the epic tradition itself … is vindicated through its religious dimension, and this gives them victory over the romance world’. 128 Other insular developments of the genre of chanson de geste include its appropriation for narrative concerned with native heroes. It is this deviation from the subject matter of continental chansons de geste, in particular that of the Carolingian epic, that has led to the classification of

124 See Gabriele, An Empire of Memory, pp. 41–70. 125 John L. Grigsby, ‘Le Voyage de Charlemagne, pèlerinage ou parodie?’, in Au carrefour des routes d’Europe, I, 567–81, discusses the different labels assigned to this text by its early editors and commentators: ‘ein Gedicht’ (Koschwitz), ‘Heldengedicht’ (Morf, Koschwitz) and ‘chanson’ (Gaston Paris), concluding that the term ‘chanson’ then set up certain horizons of expectation (p. 567). 126 Barron, English Medieval Romance, p. 99; for further discussion, see pp. 36–7. 127 Horrent, Le Pèlerinage de Charlemagne, p. 122; he is arguably giving the surviving Pèlerinage text the same status as is given to the Anglo-Norman Fierenbras, a text so thoroughly remanié that it can be treated as a text in its own right (see Chapter 5). 128 Cobby, ‘Religious Elements’, p. 379.

The Insular Literary Context

65

Anglo-Norman chansons de geste as romance or, at best, hybrid texts.129 These are Legge’s ‘romances in chanson de geste form’: the Roman de Horn and the chanson of Boeve de Haumtone. One of these heroes, Boeve was actually exported to continental Francophonia and beyond, and there is a continental as well as an Anglo-Norman version of this chanson de geste.130 Rather than reject these texts from the canon because of their departure from the traditional subject matter, we would rather see them as insular developments of the genre, evidence of the dynamism of the chanson de geste in England. The form of the genre was further appropriated by a few AngloNorman chroniclers. It seems that Jordan Fantosme (d. c. 1180) and Peter of Langtoft (d. c. 1305), both churchmen, adopted the laisse of the chanson de geste as the structural base for their chronicles in order to suggest that their protagonists, Henry I and Edward I respectively, belonged in the same world as the heroes of the chanson de geste.131 Matthew Paris, similarly, used a laisse form for his Vie de Saint Auban.132 One other chanson de geste which may have an insular origin and which is almost a hybrid between chronicle and chanson de geste is the translation of Baudri de Bourgeuil’s Historia Ierosolimitana, mentioned above. The poet adapting Baudri’s narrative may, like Fantosme, Langtoft, and Matthew Paris, have adopted the form of the Charlemagne texts in order to convey that he is writing 129 Furrow, ‘Chanson de geste as Romance in England’; Judith Weiss, ‘Insular Beginnings: Anglo-Norman Romance’, in A Companion to Romance from Classical to Contemporary, ed. Corinne Saunders, Blackwell Companions to Literature and Culture (Oxford: Blackwell, 2004), pp. 26–44 (p. 30). 130 Albert Stimming, Der festländische Bueve de Hantone (Dresden: Gedruckt für die Gesellschaft für romanische Literatur, 1911–20). 131 Pierre de Langtoft, La Règne d’Édouard Ier, ed. Jean-Claude Thiolier (Paris: Université de Paris XII, 1989); the use of laisses in this text implies that as late as the early fourteenth century Langtoft recognized this as the appropriate form for an epic tale. On Fantosme, see Philip E. Bennett, ‘L’Épique dans l’historiographie anglo-normande: Gaimar, Wace, Jordan Fantosme’, in Aspects de l’épopée romane, ed. van Dijk and Noomen, pp. 321–30; on Langtoft, see M. Dominica Legge, Anglo-Norman in the Cloisters (Edinburgh: Edinburgh University Press, 1950), pp. 70–4. 132 Matthew Paris, La Vie de Saint Auban, ed. Arthur Robert Harden (Oxford: ANTS, 1968).

66

Charlemagne in Medieval England

about the actions of heroes. It is largely Anglo-Norman writers who appropriated this form of versification for the writing of texts which might be described as chronicles or hybrids, part chronicle and part chanson de geste. These Anglo-Norman writers working in the chanson de geste tradition seem to have belonged to a different cultural context from those extolling the deeds of Charlemagne and the peers, or the treacheries of the Mayence family, on the continent. In England, while French undoubtedly had a higher status than the other vernaculars spoken, it is simplistic to consider French the language of the court and the elite, English the language of ordinary lay people, and Latin that of churchmen, as though the three did not even meet, let alone intersect. French was widely used in the twelfth and thirteenth centuries, and even into the fourteenth – so much so that in the twelfth or early thirteenth century Denis Piramus could argue for its utility across different groups: Que en franceis le poent entendre E li grant et li mendre. [For in French it can be understood By both the great and the least.]133

While the chroniclers mentioned above all wrote in French, they would also have functioned in Latin: binary oppositions of popular and learned, or oral and written, are not especially helpful models. It is more appropriate to think of a spectrum of language use, in which the French used in England, though it may have covered a large part of this spectrum, as Piramus suggests, was still of higher status than English. Yet the chansons de geste composed on the continent signalled their oral origins and continued to be in some sense ‘popular’ literature, to be listened to rather than read. This complex intersection of the learned and the popular gave rise to the texts using the chanson de geste form in England, whether Charlemagne chansons de geste, narratives of insular heroes, or chronicles, and produced some of their particular literary characteristics. 133 Denis Piramus, La Vie Seint Edmund, ed. Florence L. Ravenel, Bryn Mawr College Monographs 5 (Bryn Mawr, PA: Bryn Mawr College, 1906). The same argument would be used in the thirteenth century by the author of the Vie de Seint Clement, and in the fourteenth by William of Nassyngton, in his Speculum vitae, for writing in English.

The Insular Literary Context

67

Perhaps the most obvious characteristic of these insular texts, including the redactions of the Charlemagne material, is that they are all relatively short and economically narrated. This is also a characteristic of the early exemplars of other chansons de geste, the texts listed above as surviving only in Anglo-Norman. It may be that the relatively short length and tight structure of texts such as the Chanson de Roland and the Chanson de Guillaume is due to their antiquity; they may have survived in England because of an insular taste for fast-paced narratives; or it could be that the surviving versions had already undergone some adaptation for insular taste. Both the Roman de Horn and Boeve de Haumtone are relatively early, from the twelfth century, but the AngloNorman Fierenbras is considerably later: the only manuscript dates from the fourteenth century, by which time most continental chansons de geste were much longer. The Destruction de Rome is even shorter and brisker in its narration. This suggests a long-standing insular preference. It seems that continental chansons de geste developed in one direction, becoming longer texts with more episodes, and the insular chansons de geste developed in another, retaining the shorter narrative shape of the earliest exemplars. This tendency to relative brevity and tight structure seems of a piece with greater narrative coherence than is normally found in a continental chanson de geste. Taking as the clearest example the two Fierabras narratives, where there is a fair idea of the stemma and the relationship between the Anglo-Norman and continental texts, the link between brevity and narrative cohesion is evident. The Egerton Fierenbras is considerably shorter than the so-called ‘Vulgate’ Fierabras (1775 lines compared to 6408 in Le Person’s edition). Much of this is achieved by simple omission of repetitions, of description and of formulaic material, but there is also some ‘streamlining’ of the narrative with, for example, a conflation of episodes when Richard de Normandie makes his escape form the besieged castle during another sortie, instead of a special sortie being made to allow him to make his escape.134 This fits with an insular taste for shorter narratives (see pp. 85–6). Another feature of Anglo-Norman chansons de geste is the way they combine traditional chanson de geste discourse with a more clerical rhetoric. While rhetorical devices are not absent from continental

134 See Chapter 5.

68

Charlemagne in Medieval England

chansons de geste,135 they are found with greater frequency in the Anglo-Norman material, with instances of chiasmus and word play not otherwise typical of the genre.136 The forms of discourse characteristic of the genre are not lost, but they are adapted. This will be looked at in detail when we examine the separate traditions, but it is worth noting now that, for example, the Anglo-Norman Otinel, in abbreviating the text, loses a number of vers d’intonation; on the other hand, Otinel does not add rhétorique scolaire.137 The Destruction de Rome retains the formulaic discourse of the chanson de geste, but has no examples of the kinds of parallelism known as laisses similaires and laisses parallèles.138 Fierenbras, despite its brevity, makes extensive use of epic parallelism; it also introduces alliterative word-play (such as ‘donez moi le doun’ (115), ‘un jouste jousté (185), ‘La ceinture … enceinter’ (1067–68), ‘ceinture … ceinte’ (1109)), and uses rhetorical structure to emphasize the chanson de geste topos of the prière du plus grand péril.139 All of this is entirely in keeping with seeing these texts as coming out of a milieu which may be courtly or ecclesiastic, produced by poets with a certain level of education.

The Corpus of Anglo-Norman Charlemagne Texts What we see in this ‘history’ of the Charlemagne material in England is a gradual narrowing of the corpus. There is clearly a great deal of chanson de geste material available in England throughout the Middle Ages. While the majority of surviving manuscripts are cycle du roi texts, the narratives of some of the Mayence texts (also, of course, part of the wider Charlemagne tradition), particularly Renaud de Montauban, were known, and several Guillaume texts were copied and possibly 135 Sarah Kay, ‘The Nature of Rhetoric in the chanson de geste’, Zeitschrift für romanische Philologie 94 (1978), 305–20. 136 Ailes, ‘Fierabras and Anglo-Norman Developments’, p. 104; chiasmus is rare in the chanson de geste but is found in Boeve de Haumtone. 137 Ailes, ‘What’s in a Name?’, pp. 63–5. 138 Ibid., pp. 65–9. The standard terms used for the patterns of repetition and parallelism of the genre come from Jean Rychner, La Chanson de geste: Essai sur l’art épique des jongleurs (Geneva: Droz, 1955), pp. 68–125; see also Jones, An Introduction, pp. 12–16. 139 Ailes, ‘Fierabras and Anglo-Norman Developments’, pp. 104–6.

The Insular Literary Context

69

underwent revisions in English scriptoria. Alongside the chansons de geste, the narratives of the Matter of France were being read and disseminated, perhaps in a different context, through the Pseudo-Turpin Chronicle. The presence of copies of this text in monastic libraries is noted above, and the Latin text was clearly well known in England, as it was across Europe.140 The codicological evidence, the fact that the Pseudo-Turpin Chronicle is so often found with ‘genuine’ Latin chronicles, suggests that this text was taken to be, and was received as, history (this is not an insular phenomenon but a European one). The reduction of the corpus is not, moreover, a constant process. Manuscripts and texts continued to be imported into England throughout the medieval period, right up to the fifteenth century when the period of the dual monarchy led to influential and wealthy English patrons, such as John Talbot, Earl of Shrewsbury, sometimes spending extended periods in France and bringing back with them books and manuscripts. However, for all the rich potential of available material, only the traditions of Fierabras, Otinel and the narratives relating to Roland (the chanson de geste and the Pseudo-Turpin Chronicle) would be translated into Middle English. Possible reasons for this will be considered in the next chapter as we look at how this material was treated and appropriated in the insular context. Rosalind Field has argued that the reason only certain texts are found in Middle English versions is because the choice was already made and only these texts, Fierabras, Otinel and the Chanson de Roland, were available in Anglo-Norman. However, as noted above, the texts of the Mayence cycle do seem to have been known, and in some parts of the country the tales of Guillaume may also have circulated, though on the evidence available it is difficult to know how widespread was the diffusion of these texts, and no doubt it varied across both time and place. Nevertheless, it is apparent that even if these other narratives were read and listened to, those most thoroughly appropriated, through translation or remaniement, and even those most frequently copied, were those chansons de geste of the Charlemagne cycle concerned with major conflict, that is, the religious conflict between Saracen and Christian. We would concur with the observation made by both Melissa Furrow and Rosalind Field of a strong thematic interest: the texts copied all 140 Marianne Ailes, ‘Pseudo-Turpin Chronicle’, in Encyclopedia of the Medieval Chronicle, ed. Graeme Dunphy (Leiden: Brill, 2010), pp. 1454–5.

70

Charlemagne in Medieval England

deal with the Saracen Other.141 However, this is not the whole story. The surviving Guillaume text and fragments and the extensive AngloNorman copies of Aspremont point to widespread knowledge of texts which deal with this very theme, but yet were not translated into Middle English. We discuss this question below, and further in relation to the evidence of more precise political engagement in individual texts and manuscripts which will be analysed in the following chapter. It is striking that all the texts which would be translated into Middle English were already undergoing some changes in insular French, yet these adapted texts are not always the sources of Middle English translations. At the same time, although the Pseudo-Turpin Chronicle lies behind several Middle English texts, none of these comes directly from a Latin source. The key to the selection of texts to be translated does seem to be the prior existence of the Anglo-Norman redactions and narratives.

The Matter of Charlemagne in English: Developing a Tradition The Choice of Matter of France Texts in Middle English It has long been noted that the choice of texts in the Middle English Matter of France corpus is far more limited than the range offered by the French tradition and, as discussed above, is more limited even than the body of insular French-language texts available. However, it is worth reiterating the other side of this observation: that there is no extant Middle English text adapted from French that does not derive from a source already available in some form in an insular AngloNorman version. The Middle English romances often show traces of affiliation to more than one branch of their French source’s tradition: for example, the Song of Roland, while following the model of the Chanson de Roland as represented in the insular Oxford version, also includes material from the continental rhymed Roland tradition; the Ashmole Sir Ferumbras, among the copious alterations made in the translator’s holograph manuscript, shows evidence of drawing on alternative

141 Field, ‘Patterns of Availability and Demand’, p. 81; Furrow, ‘Chansons de geste as Romance in England’, pp. 63–46, 71; see Chapter 2.

The Insular Literary Context

71

textual versions;142 and all three of the Middle English versions of Otinel display a complex pattern of affiliation across both insular and continental traditions.143 While it is impossible to prove that all these instances of textual hybridity occurred in the process of translation into English rather than in some now lost French-language antecedent text, the characteristic freedom with which the English writers tended to treat the inherited narrative tradition, as compared with the relative fidelity of French-language versions, makes it a likely hypothesis. Although every cycle du roi text translated into Middle English was not only already circulating in England but had already been adapted in an Anglo-Norman version (even if this was not the version used for the English translation), the fact that continental versions of the texts were sometimes used suggests that the Middle English translators were not constrained in their choice of material: it is not the case that they were unable to access these texts unless an Anglo-Norman version was available, but rather that insular scribes/adaptors and translators alike were choosing to appropriate the same few texts. Even the two romances that have no known antecedent, The Sege of Melayne and Rauf Coilyear, are likely to have originated as a ‘prequel’ or response to one of the insular favourites, Otinel.144 It thus seems clear that the pre-selection of this small body of Charlemagne texts, initially appropriated for an insular context by their adaptation into Anglo-Norman versions, was endorsed by the English writers in their repeated translations and rewritings of the same few texts; it is therefore justifiable to consider the insular Matter of France texts in both languages as a coherent group of connected narratives with a particular widespread and enduring appeal to late-medieval insular audiences. As discussed above, the apparent neglect by the Middle English translators of the insular-adopted chansons de geste Renaud de 142 For example, in line 3597 (ed. Herrtage): ‘An heȝ þan wente þus barouns stout, And at þe wyndowe loked out’, the word ‘wyndowe’ is altered from the original word ‘kernels’ (meaning battlements). The alteration accords with the Vulgate version as edited by Marc le Person: ‘fenestres’ (4182), whereas ‘kernels’ accords with the tradition represented by Bagnyon’s French prose version, as translated by Caxton: ‘bataylement’ (ed. Herrtage, p. 149/21). 143 See pp. 348–50. 144 See pp. 381–2.

72

Charlemagne in Medieval England

Montauban and Aspremont – and also, it seems, the other two chansons de geste known from fragments of Anglo-Norman manuscript copies, the Chanson de Guillaume and Fouques de Candie – demands some explanation. The possibility of a lost Middle English version of Renaud de Montauban is discussed below (pp. 79–81). For the others, all three deal with conflict between Christians and Saracens and so might be thought to appeal to the same taste as that which appreciated Fierabras and Otinel; however, their stories (unlike these) do not hinge upon the issue of religious conversion that seems to have held such a compelling interest for insular audiences. Conversion would have been an especially potent concern in the period after the Fourth Lateran Council of 1215, with its strong emphasis on pastoral issues of religious education and spiritual renewal for both clergy and laity, particularly through the practice of sacramental confession and penance; and narratives of conversion such as those of Fierabras and Otinel would provide highly dramatic examples of repentance and amendment. Besides providing a narrative parallel to the topical issue of penitential conversion for Christian readers, though, stories of the spectacular conversion of Saracen super-heroes offered fantasy fiction responses to a real fear in late-medieval Europe: the fear of Saracen invasion. Robert Rouse argues that all the Middle English crusading romances, including the Charlemagne romances, produced in the almost two centuries from Richard Coeur de Lion (c. 1300) to Caxton’s translation of Godfrey de Bouillon (1481), figure the ‘cultural trauma’ of the loss of the Holy Land, when the gains of the First Crusade were lost in the disaster of the fall of Acre in 1291: they are ‘recovery’ romances.145 This is a persuasive claim in respect of narratives of the crusade to the Holy Land, but it raises some puzzling questions. First, why were none of the crusade cycle epics translated into English? Copies of some crusading 145 Robert Allen Rouse, ‘Crusaders’, in Heroes and Anti-Heroes in Medieval Romance, ed. Cartlidge, pp. 173–83 (p. 174). Rouse is building on Christopher Tyerman’s discussion of ‘recovery literature’ (God’s War: A New History of the Crusades (London: Lane, 2006), p. 827). This idea of Middle English ‘recovery’ romances is further developed by Lee Manion in Narrating the Crusades (Cambridge: Cambridge University Press, 2014), where it is used to define the category ‘crusading romance’ as featuring (among other things, such as pilgrimage to the Holy Land) ‘crucially, a narrative pattern of loss and recovery’ (p. 65).

The Insular Literary Context

73

chansons de geste were available in England, as noted in the Introduction (pp. 9–10). Second, why did Middle English writers instead choose Fierabras and Otinel, which make no reference to the Holy Land? Rouse’s claim may in fact be more relevant to the insular production and copying of French-language Matter of France epics in the thirteenth and early fourteenth centuries, in the traumatic context of the fall of Jerusalem (1187) and the loss of Acre and mainland Outremer (1291). Christopher Tyerman points out that the period between Edward I’s crusade to the Holy Land (1270–2) and the beginning of the Hundred Years’ War (1337) was marked by the repeated imposition of papal taxes on the English Church ‘solely for crusading’ (in 1274, 1291, 1312, 1333), ‘but none thereafter’,146 and this concentration of resources on efforts to recover the Holy Land may be reflected in the contemporary insular interest in Matter of France chansons de geste featuring Charlemagne’s victories over Saracens and the recovery of relics of the Passion, especially in the context of thirteenth-century Anglo-Norman texts such as William de Briane’s translation of the Pseudo-Turpin Chronicle and the anonymous First Crusade epic based on the chronicle of Baudri de Bourgueil. There is thus a case for reading the insular French texts, in particular La Destruction de Rome and Fierabras, as ‘recovery’ narratives in line with Rouse’s argument; but the change of focus in the Middle English rewritings in the fourteenth and fifteenth centuries is pointedly shown by the much diminished and altered significance of the ‘recovery of the relics’ theme in the three Fierabras-related texts. On the other hand, the Middle English reworkings of the Fierabras and Otinel traditions consistently magnify another theme in the chansons de geste: the threat to European nations and to Christian society posed by hostile Saracen powers. These are less obviously ‘recovery’ romances: they reflect an entirely different anxiety, that of invasion. As we shall argue, one reason why the focus of fourteenth- and fifteenth-century Middle English translation and rewriting is on these two texts may be that they address in fictional form the real and present danger of Turkish victory over, and possession of, European Christian homelands.

146 Tyerman, England and the Crusades, p. 229.

74

Charlemagne in Medieval England

Evidence for Lost Middle English Charlemagne Romances In The Lost Literature of Medieval England, R. M. Wilson cites one lost Middle English romance of the Matter of France: the text named Olive and Landres and known only from its late thirteenth-century translation into Old Norse prose in the Karlamagnús saga.147 Smyser gives an account of this earliest Charlemagne romance: it concerned the falsely accused Olive, who is eventually vindicated by her son Landres, nephew of Charlemagne. As Smyser observes, it has very little in common with any of the extant texts in the Charlemagne tradition.148 Wilson concludes from the evidence of wills, inventories and catalogues that while there are numerous references to Matter of France texts, with one exception ‘they are invariably in French or Latin, and provide no evidence for any lost romances in English’.149 There is similarly unproductive witness in those lists of romance subjects found as a topos in Middle English writings,150 which regularly include the names Charlemagne, Roland and Oliver, but usually without indication of any specific narrative texts, or indeed of the language of narration. Rather than pointing to any more extensive corpus of Middle English translations, they may simply be referencing the widespread knowledge of these subjects in insular culture. However, there are a few more precise references to possibly lost English Charlemagne narratives. The prologue to Otuel and Roland in the Fillingham MS (BL MS Add. 37492),151 and the probable duplicate prologue in the now lost opening lines of Roland and Vernagu in

147 R. M. Wilson, The Lost Literature of Medieval England, 2nd edn (London: Methuen, 1970), pp. 109–10. 148 Smyser, ‘Charlemagne Legends’, p. 81. 149 Wilson, Lost Literature, p. 111. The one exception is the reference to a ‘gesta Guydonis de Burgundia in patria lingua’ at St Augustine’s, Canterbury, which Wilson takes to indicate a lost text in English; this more probably refers to a copy of Gui de Bourgogne in French. 150 The lists typically appear in the prologues to didactic works (such as Cursor Mundi) seeking to distance their subject matter from frivolous fiction, or in other romances placing themselves in a recognizable tradition. 151 As is evident in the stanzas reproduced below, the text of Otuel and Roland has suffered in transmission.

The Insular Literary Context

75

the Auchinleck MS (NLS MS Adv. 19.2.1),152 have provoked some discussion of possible lost romances: Herkenyth lordynges, & ȝeuyth lyst In the worchype of ihesu cryst Off a conquerour that was yhote syr Charlemayne, howe he wan Galys of spayne with ful grete honour, And howe that he ouer cam [line missing] with full grete vygour, And howe Rowland & othyr knytys to Aȝeyn four knytys fouȝtyn tho, and euer was a grete warryour. & the kyng Ebryan helde werre aȝeynes ham, and greued hem ful sore, And howe Rowland slowe vernagu thorugh the myȝt of ihesu, that leued in false lore, And the caytyf Emoun helde werre aȝens charlyoun Thrytty wyntyr & more, and magre hym and al hys, thorugh the myȝt of Ihesus, In the mount awbane thay wore. And ther Rouland the gode knyght Ouercom Otuell in fyȝt, Nowe ye schulle yhuyre, and was cristenyd withoute fayle

152 The opening leaf of Roland and Vernagu in the Auchinleck MS has been excised, but a few letters from the ends of lines of text remain and match the final letters of the equivalent lines at the opening of Otuel and Roland. See the digitized images of the manuscript in The Auchinleck Manuscript, ed. David Burnley and Alison Wiggins (NLS, 5 July 2003), version 1.1, http://auchinleck.nls.uk/editorial/history.html.

76

Charlemagne in Medieval England and helpe Charlys in many a batayle and was hym lef & dere. And all the dussyperys with hym Bothe Gwynes and syr turpyn that was stoute in fyȝt. And howe Charlys aftyrthan all hys fomen ouercam thorowe the grace of god almyȝt. with swerdys dynt withouten les, hys lond he held in grythe & pes, withoute warre and fyȝt. the Ercheboshope syr Turpyn All those dedys wrote in latyn whiles he hem sauȝ in syȝt. Off charlys that was so grym, nowe y wolle begynne of hym, to tell ȝou aryȝt. (Otuel and Roland, 1–44)

Discussion of the two romances Roland and Vernagu and Otuel and Roland in the twentieth century was largely focused upon the problematic question of their evident textual relation and the likelihood of their representing two separated fragments of a once continuous cyclic Middle English tail-rhyme romance of the Matter of France, based on the Pseudo-Turpin tradition combined with the chanson de geste Otinel. The issue was complicated by the presence of a different English version of Otinel, the couplet romance Otuel, as the sequel to Roland and Vernagu in the Auchinleck MS, a text which also shows textual relation to Otuel and Roland. The interrelated and supporting hypotheses of a lost Middle English cyclic text, given the title ‘Charlemagne and Roland’,153 of the circumstances of its production in a putative London bookshop,154 and of the dependence of the lost text on sources in a 153 Gaston Paris, Histoire poétique de Charlemagne (Paris: Librairie Bouillon, 1905), p. 156; H. M. Smyser, ‘Charlemagne and Roland and the Auchinleck MS’, Speculum 21 (1946), 275–88. 154 Laura Hibbard Loomis, ‘The Auchinleck MS and a Possible London Bookshop of 1330–1340’, PMLA 57 (1942), 595–627.

The Insular Literary Context

77

French manuscript miscellany available to the scribe/translators in the bookshop,155 are briefly summarized by Smyser in his influential survey of the Matter of France romances,156 and the problems inherent in the theories are noted in counter-arguments by Frederick Porcheddu157 and Judith Weiss.158 A fresh interpretation of the evidence is offered in Rhiannon Purdie’s study of the tail-rhyme tradition, to produce a theory that is, as she acknowledges, ‘far less tidy’ than the Loomis-Walpole-Smyser hypotheses, but is perhaps all the more convincing for just this reason, and which also ‘correspond[s] rather better to the histories of some other known medieval romances’. 159 Purdie argues that the likeliest explanation for the textual correspondences between Roland and Vernagu, Otuel and Roland, and Otuel is as follows: Otuel, a pre-existing Middle English adaptation in couplets of Otinel, was used as a crib by the author of the tail-rhyme Otuel story which forms the first part of Otuel and Roland, when composing his own closer version of the French chanson de geste; this would explain the notable verbal echoes between the two English texts.160 To this Otuel story was attached a tail-rhyme adaptation of episodes from the Pseudo-Turpin Chronicle (either by the same author or by another, writing in a similar dialect) to form the composite romance from which the extant Fillingham Otuel and Roland derives. This poem then inspired the author of Roland and Vernagu to return to the Pseudo-Turpin and to compose his own romance ‘as a kind of prequel’, incorporating the four-stanza prologue and the three-stanza description of Charlemagne (taken from the Pseudo-Turpin chapter 20, ‘De persona et fortitudine Karoli’, and used in both romances as a 155 Walpole, ‘The Source MS’. 156 Smyser, ‘Charlemagne Legends’, pp. 80–100. 157 Frederick Porcheddu, ‘Edited Text and Medieval Artifact: The Auchinleck Bookshop and “Charlemagne and Roland” Theories, Fifty Years Later’, Philological Quarterly 80 (2001), 463–500. 158 Judith Weiss, ‘The Auchinleck MS and the Edwardes MSS’, Notes and Queries 16.12 (1969), 444–6. 159 Rhiannon Purdie, Anglicising Romance: Tail-Rhyme and Genre in Medieval English Literature (Cambridge: D. S. Brewer, 2008), p. 123. 160 Purdie shows that the evidence better supports this hypothesis than the ‘standard explanation … that both derive from a common lost Middle English source’ (p. 120).

78

Charlemagne in Medieval England

mid-way marker) from the earlier poem,161 and concluding the final stanza with a three-line ‘cue’ to the connected story of Otinel: To otuel also ȝern Þat was a sarrazin stern, Ful sone þis word sprong. (878–80)

Purdie’s argument on the origins of these related texts is thoroughly convincing, but it is less persuasive when responding to the theory of the lost composite ‘Charlemagne and Roland’ romance. Instead of an original single-author cyclic romance, she proposes a lost ‘cycle’ manuscript compilation, a ‘book of Charlemagne’ made by collecting several ‘relevant narratives’ from different sources,162 on the grounds that the Otuel and Roland prologue ‘was clearly intended to represent a much fuller Charlemagne cycle than that represented by Otuel and Roland and Roland and Vernagu put together’ (p. 119).163 However, as 161 For discussion of the function of these descriptive passages as marking a new beginning in the narrative, see Phillipa Hardman, ‘Knight, King, Emperor, Saint: Portraying Charlemagne in Middle English Romance’, Reading Medieval Studies 38 (2012), 43–58 (p. 50). 162 Purdie, Anglicising Romance, p. 119. This is precisely what exists in the Shrewsbury Book (BL MS Royal 15 E vi), where diverse French Charlemagne texts are presented in similar format to produce a ‘liure de charlemaine’, but there is no evidence of manuscript production on this lavish scale in relation to Middle English romance texts. The mention of ‘A boke of Charlman’ in the late fifteenth-century list of English books found on a flyleaf in the Welles Apocalypse (c. 1310, BL MS Royal 15 D ii, fol. 211) could possibly point towards such a collection of narratives, but it could equally well indicate a single narrative text; Mary Hamel suggests it may be Caxton’s Charles the Grete (which itself closely translates Jehan Bagnyon’s composite ‘book of Charlemagne’, formed of Pseudo-Turpin material and Fierabras), or alternatively one of the extant English romances based on the Pseudo-Turpin Chronicle; see Mary Hamel, ‘Arthurian Romance in Fifteenth-Century Lindsey: The Books of the Lords Welles’, Modern Language Quarterly 51 (1990), 341–61 (p. 352, n. 43). 163 Frederick Porcheddu interrogates the lure of the mega-narrative theory for critics in relation to the Icelandic cycle, ‘The Cloning of the Karlamagnús Saga in Anglo-French Textual Criticism’, in Sagas and Societies, Proceedings of an International Conference at Borgarnes,

The Insular Literary Context

79

evidence of the range of other stories that might have been available, the garbled Otuel and Roland prologue is not promising. Apart from the story of Otinel and the episodes from the Pseudo-Turpin Chronicle already appropriated in these two romances (that is to say, Charlemagne’s conquest of Spain, his defeat of Ebrahim, and Roland’s defeat of Vernagu), the only reference to an identifiable story among the general celebration of Charles’s achievements is the mention of ‘Emoun’ and ‘mount awbane’. 164 Moreover, it seems from the evidence of the surviving manuscripts containing English Charlemagne romances that putting two compatible texts together to form a sort of narrative diptych was precisely what appealed to the compilers of the manuscripts.165 An alternative explanation is that which Purdie offers in the similar case of the prologue shared between Sir Orfeo and Lay le Freine: ‘The Orfeo– Freine prologue’s all-purpose description of the “Breton lay” makes it suitable for attachment to any such poem’ (p. 118), and this seems just as appropriate for the Charlemagne prologue: it was used simply as a formula that could introduce any story or collection of episodes from the Charlemagne tradition.166 To return to the prologue’s mention of ‘Emoun’ and ‘mount awbane’ (18–23): this briefly sketched story is the first surviving version in English of the chanson de geste Renaud de Montauban or Les Quatre Fils Aymon. As discussed above, the chanson de geste seems to have had Iceland, 5–9 September 2002, ed. Stefanie Würth, Tõnno Jonuks and Axel Kristinsson (https://publikationen.uni-tuebingen.de/xmlui/ bitstream/handle/10900/46205/pdf/13_fre~1.pdf;sequence=1 [accessed 24/10/2016]). 164 The story mentioned in lines 9–10 (‘howe Rowland & othyr knytys to | Aȝeyn four knytys fouȝtyn tho’) refers to the episode of Otinel where Roland, Olivier and Ogier fight four Saracen kings; in Otuel and Roland this episode is introduced by an extensive descriptive rubric (662–73); see pp. 371–3. The mention of ‘Gwynes’ (31) possibly refers to Ganelon’s role in the Battle of Roncevaux and its aftermath, told in the final part of Otuel and Roland. 165 See p. 352, and Ailes and Hardman, ‘Texts in Conversation’. 166 This argument in relation to both prologues was made by Derek Pearsall in his introduction to the facsimile edition, The Auchinleck Manuscript: National Library of Scotland Advocates’ MS 19.2.1, ed. Derek Pearsall and I. C. Cunningham (London: Scolar Press, 1977), p. x.

80

Charlemagne in Medieval England

some currency in England; nevertheless, Smyser finds its mention here surprising, ‘for there is nowhere else evidence that any version of the Foure Sonnes of Aymon ever appeared in English before Caxton’ (p. 89). But if this mention is indeed a hint of a lost Middle English romance, rather than testimony to widespread knowledge of the French chanson de geste, it suggests a rather different treatment of the tradition from the story as known in the chanson de geste. Here, Aymon is the central character (his sons are not named, nor is the famous horse Bayard) and he is presented in a bad light: a ‘caytyf ’ who opposed Charlemagne for thirty years. Indeed, this very simplified account of the original story may indicate a reason why this particular French chanson de geste was not translated into English romance: while Renaud de Montauban, with its scenario of semi-independent noble vassals resisting the authority of the monarchy, reflected the contemporary political situation in France when the chanson de geste was composed, it did not as obviously fit the scene of more centralized royal power in fourteenth-century England. The author of this brief English sketch ‘rewrote’ the story to produce a clearly moralized model of ‘wicked rebel’ versus ‘good king’ that would fit with the uniformly positive portrayal of Charlemagne and monarchy in the Middle English romances, but it is hard to imagine that it refers to an actual full-scale rewriting of the chanson de geste in these terms, and it is thus unlikely to imply the existence of a lost Middle English version. A somewhat more accurate sketch is given in Skelton’s Philip Sparrow: Of Quater Fylz Amund, And how they were sommonde To Rome, to Charlemayne, Upon a great payne, And how they rode eche one On Bayarde Mountalbon; Men se hym now and than In the forest of Arden. (651–8)167

By this time (c. 1505), English readers were acquainted with the story in Caxton’s translation of the French prose romance Les Quatre Fils Aymoun, published c. 1490 as The Right Pleasant and Goodly Historie of 167 John Skelton, The Complete English Poems, ed. John Scattergood (New Haven, CT: Yale University Press, 1983), p. 88.

The Insular Literary Context

81

the Foure Sonnes of Aymon (though Skelton’s language suggests he may be referring to the French text rather than Caxton’s). Two sixteenthcentury plays, The Four Sons of Aymon (c. 1581) and The Twelve Peers of France (before 1586), both now lost, were probably dramatic adaptations of Caxton’s prose romances The Foure Sonnes of Aymon and Charles the Grete rather than representing lost Middle English romances.168 The title of Caxton’s Foure Sonnes of Aymon is reproduced in two sixteenthcentury lists of popular fiction,169 though its presence alongside tales and ballads suggests the possibility of lost chapbook adaptations. A further intriguing hint occurs in The Complaynt of Scotland (1549), in which the list of currently well-known stories includes, besides ‘the tayl of the four sonnis of aymon’, three other Matter of France subjects: ‘the tail of the brig of the mantribil’ (presumably a version of Fierabras, but raising the interesting possibility that it represents a shorter narrative excerpted from the tradition and focusing on the capture of the bridge of Mautrible by Charlemagne and the peers), ‘rauf collzear’ (presumably a version of the Scots text printed in 1572 by Robert Lekpreuik),170 and ‘the seige of millan’ (presumably a version of the Middle English romance The Sege of Melayne). While none of these points to unknown or lost romances, they do indicate a widespread and enduring interest in the Matter of France in Britain.

The Question of Genre and the English Charlemagne Romances The question of genre is a particularly thorny one in relation to the insular Matter of France tradition, as discussed above in relation to the Anglo-Norman texts. Traditionally, the French-language poems, with their epic laisse form, have been classed as chansons de geste, and the later Middle English verse texts as romances, but these traditional 168 Helen Cooper, The English Romance in Time: Transforming Motifs from Geoffrey of Monmouth to the Death of Shakespeare (Oxford: Oxford University Press, 2004), pp. 414, 417. 169 ‘The foour suns of Aymon’ is named among Captain Cox’s stories in Robert Laneham, A Letter (1575), facsimile reprint (Menston: Scolar Press, 1969), p. 34, and ‘the tayl of the four sonnis of aymon’ is listed among the stories told by shepherds in ‘Ane monolog of the actor’, in The Complaynt of Scotland, ed. A. M. Stewart, Scottish Text Society, 4th series, 11 (Edinburgh, 1979), chapter 6. 170 See pp. 214–20.

82

Charlemagne in Medieval England

classifications have not gone unchallenged. Furrow, in a radical approach, dissolves the distinction with the claim that ‘rather than being in generic opposition to romances, chansons de geste and their Middle English descendants were romances in England’.171 As argued above, we take the view that the insular chansons de geste can be identified as such predominantly by their formal epic features rather than by a focus on either their ‘epic spirit’172 or ‘key subject-matter’, or on the ‘selfidentifying’ use of the terms ‘geste’/’romaunce’ in medieval texts.173 On the other hand, the frequent description of the category ‘romance’ as ‘capacious’ signals the difficulty of identifying romances by any easily recognized set of characteristics, formal or otherwise.174 We shall use the customary term ‘romance’ for all the Middle English Charlemagne texts, verse or prose, but with the recognition that the remarkable variation within the group in terms of translation practices, narrative style, and verse forms may be seen as witnessing to a continuing process of negotiation between the tradition of the chanson de geste and the expectations of Middle English romance. In fact, while the classification of Middle English romances as a whole is a problematic venture that has generated prolonged critical dispute,175 one of the most stable categories has been that which describes the romances under discussion here, usually referred to as the English Charlemagne romances, or as romances of the Matter of 171 Furrow, ‘Chanson de geste as Romance in England’, p. 72. 172 Barron, English Medieval Romance, p. 106. 173 Furrow, ‘Chanson de geste as Romance in England’, pp. 72, 67. 174 For an analysis of selected romances to demonstrate ‘generic homogeneity’ transmitted through ‘a single set of intertextual markers’, see Carol Fewster, Traditionality and Genre in Middle English Romance (Cambridge: D. S. Brewer, 1987), pp. 150–1. 175 See Dieter Mehl, The Middle English Romances of the Thirteenth and Fourteenth Centuries (London: Routledge, 1967), pp. 30–8; John Finlayson, ‘Definitions of Middle English Romance’, Chaucer Review 15 (1980/81), 44–63, 168–81; Paul Strohm, ‘The Origin and Meaning of Middle English Romaunce’, Genre 10 (1977), 1–28; Andrea Hopkins, The Sinful Knights: A Study of Middle English Penitential Romance (Oxford: Clarendon Press, 1990), pp. 1–20; Yin Liu, ‘Middle English Romance as Prototype Genre’, Chaucer Review 40 (2006), 335–53; Cooper, The English Romance in Time, pp. 1–22; Furrow, Expectations, passim.

The Insular Literary Context

83

France. The category is here taken to include the texts published by the Early English Text Society between 1879 and 1887 under the group title ‘The English Charlemagne Romances’, 176 together with three further texts edited from manuscripts that came to light subsequently.177 Yet the stability of the grouping may not imply critical consensus so much as a lack of interest in these texts, for they are often named only in order to be dismissed. Derek Pearsall, for instance, in a relatively recent discussion of medieval English popular romance, offers a brief characterization of what he terms ‘the typical romance’ that fits the Charlemagne texts as well as any other candidates, but he simply excludes from the list ‘the Matters of France and Antiquity’;178 John Finlayson refines his definition of Middle English romance as a genre by removing from consideration the whole group of ‘Charlemagne romances’ as insufficiently romantic and better described as ‘largely heroic works’. 179 Dieter Mehl, although briefly distinguishing between the verse texts in terms of their different lengths and style, refers frequently, without discussion, to ‘the Charlemagne romances’ as a homogenous group in a way that he never does in relation to, for instance, the Arthurian romances;180 while Rosalind Field, surveying the output of English romance in the fourteenth century, dismisses

176 The last text of this series, Huon of Burdeux, is excluded on the grounds that Lord Berners’s translation (c. 1515) is the first evidence of insular interest in this romance, though it was apparently popular in the sixteenth century. 177 Firumbras and Otuel and Roland in the Fillingham MS, acquired by the British Museum in 1907 and edited for EETS in 1935, and the Middle English prose Pseudo-Turpin, identified in San Marino, CA, Huntington Library MS HM 28,561 in the 1980s and edited for EETS as Turpines Story in 2004. See the Appendix for a full list of texts and editions. 178 Derek Pearsall, ‘The Pleasure of Popular Romance’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 9–18 (p. 12). 179 Finlayson, ‘Definitions of Middle English Romance’, p. 446. Calin counters this argument in The French Tradition, p. 434, and Furrow argues for a conflation of chanson de geste and romance in these and similar insular texts (Expectations, pp. 115–16; ‘Chanson de geste as Romance in England’, pp. 66–7). 180 Mehl, The Middle English Romances, pp. 18, 118, 122, etc.

84

Charlemagne in Medieval England

the whole corpus of Middle English Charlemagne romances without distinction.181 Studies of individual texts, however, have helped to recuperate the reputation of the romances as worthy of critical attention and to promote the recognition that while the thirteen texts that together constitute the Middle English Charlemagne romances are linked by common elements of subject matter (the basic material shared by all texts in this category, with the partial exception of Rauf Coilyear, consists of the court of Charlemagne, the fellowship of Roland and Oliver, the twelve peers, the treacherous character of Ganelon, and the defence of Christendom against incursion by the Saracens),182 they cannot be considered as an undifferentiated group. A 2009 discussion of the Sege of Melayne is prefaced by the observation that while crusade-style conflict between Christendom and Islam is indeed the characteristic feature of the Matter of France, … this does not mean that the holy-war convention guarantees any one particular kind of poem. … It is perhaps the very constancy of the holy-war meme that gives the Middle English Charlemagne-poet license to change things.183 This is a helpful approach to a body of romances containing multiple independent versions of the same source texts, where a ‘license to change things’, like the freedom of translators to adapt their material, may have empowered writers to use the Matter of France in different ways at different times, in response to different contexts. Crusade-style holy war as a narrative meme is not, of course, exclusive to Charlemagne romances.184 Lee Manion offers a recategorization of the genre of ‘crusade romance’ which focuses on 181 Field, ‘Romance in England, 1066–1400’, p. 172; see also Introduction, above. 182 This material was also part of late-medieval culture: the reputation of Charlemagne as one of the three Christian Worthies, the friendship of Roland and Oliver, and the treachery of Ganelon all had universal exemplary force. 183 Crofts and Rouse, ‘Middle English Popular Romance and National Identity’, pp. 87–8. 184 For discussion of the term ‘meme’ in relation to romance, see Cooper, The English Romance in Time, pp. 3, 21.

The Insular Literary Context

85

non-cyclic romances of knights ‘performing an alternative practice of crusading’, alongside the more traditional idea of ‘the English crusading romance’ in texts such as the ‘semi-historical’ Richard Coeur de Lion and fictions of Charlemagne.185 The crusading status of Guy of Warwick’s battles in the Holy Land has been much debated.186 Equally, the markedly pious character of the Middle English Charlemagne poems, which has often been noted as a distinguishing feature by comparison with the chansons de geste from which they derive, is shared with a wide range of texts that have variously been termed ‘homiletic’, ‘penitential’, or ‘hagiographic’ romances.187 It is important, therefore, to recognize the likely extent to which English rewritings of the Matter of France material will have been predetermined by the horizon of expectations already embedded among the writers and readership of Middle English popular romances.188 Common characteristics would include: a marked tendency to create separate adventure stories rather than story cycles; the frequent abbreviation of lengthy texts to produce shorter adaptations; the prioritizing of narration of action and direct speech; a corresponding 185 Lee Manion, ‘The Loss of the Holy Land and Sir Isumbras: Literary Contributions to Fourteenth-Century Crusade Discourse’, Speculum 85 (2010), 65–90 (pp. 84–5, 68–9). These ideas are developed in Manion, Narrating the Crusades (2014). 186 See, for example, Judith Weiss, ‘The Exploitation of Ideas of Pilgrimage and Sainthood in Gui de Warewic’, in The Exploitations of Medieval Romance, ed. Ashe et al., pp. 43–56; Robert Allen Rouse, ‘An Exemplary Life: Guy of Warwick as Medieval Culture-Hero’, in Guy of Warwick: Icon and Ancestor, ed. Alison Wiggins and Rosalind Field (Cambridge: D. S. Brewer, 2007), pp. 94–109 (pp. 98–101); Rebecca Wilcox, ‘Romancing the East: Greeks and Saracens in Guy of Warwick’, in Pulp Fictions of Medieval England: Essays in Popular Romance, ed. Nicola McDonald (Manchester: Manchester University Press, 2004), pp. 217–40. 187 It should be noted that insular versions of chansons de geste may also show an increase in religious awareness: see pp. 116–19, below. 188 For extensive discussion of the category ‘popular romance’, see Pearsall, ‘The Pleasure of Popular Romance’, pp. 9–18, and essays in The Spirit of Medieval English Popular Romance, ed. Ad Putter and Jane Gilbert (Harlow: Longman, 2000); Pulp Fictions, ed. McDonald (2004); Companion to Medieval Popular Romance, ed. Radulescu and Rushton (2009).

86

Charlemagne in Medieval England

bias towards fast-paced narratives, frequently focusing on the exploits of an individual hero; the tendency to present a simplified contrast between virtuous heroes and vicious enemies; a taste for touches of vivid realism and sardonic humour. Indeed, these characteristics, as likely to occur in the English Charlemagne texts as they are in any other Middle English popular romances, seem to some extent to represent insular late-medieval taste in fiction of this kind, whichever language is in use, despite the marked differences in formal characteristics between chansons de geste and romances (see above, p. 67). An exception must be made for Caxton’s two prose translations Charles the Grete and The Four Sons of Aymon, both of which translate a fifteenth-century French prose rewriting of an earlier chanson de geste with extraordinary fidelity, but the anonymous Middle English prose Pseudo-Turpin shows rather more freedom in its translation, exhibiting many of the same tendencies towards abbreviation, action, focus on individual heroes, and touches of humour, thus participating in the same dynamic approach to translation found in the other insular manuscript texts, whether Anglo-Norman or English.

Verse Forms and Experimentation in English Metrical Romances of Charlemagne In the Middle English tradition, Pearsall firmly connects popular romance with the flowering of one particular verse form, following the findings of Rhiannon Purdie’s study: ‘The development of the twelveline tail-rhyme is the triumph of form in the popular romance.’ 189 For the Charlemagne romances in particular, Purdie’s argument offers a second compelling association with tail-rhyme in the suggestion that ‘tail-rhyme may, in fact, have been perceived as an appropriate English equivalent to the laisses used in Old French or Anglo-Norman chansons de geste, a genre similarly defined by a combination of subject matter and poetic form and which, significantly, is primarily associated with the celebration of Charlemagne’.190 A count of the ten verse romances reveals that nearly half – four and a half of the texts – are cast in one 189 Pearsall, ‘The Pleasure of Popular Romance’, p. 17, citing Purdie, Anglicising Romance. 190 Purdie, Anglicising Romance, p. 9. See also Calin, The French Tradition: ‘The laisse in a chanson de geste, like the tail-rhyme stanza and the tail-line

The Insular Literary Context

87

version or another of the twelve-line tail-rhyme stanza, and that these span the whole period from which the extant texts date, from Roland and Vernagu (c. 1330–40) to Otuel and Roland (1475–1500). This testifies (if Purdie’s suggestion is accepted) to a longstanding perception among English adaptors that this was a suitable verse form for reflecting the use of the epic laisse as a structural unit. The other five and a half verse texts, however, are not in tail-rhyme stanzas, and these present a wide range of formal variation: couplets, quatrains, septenaries, alexandrines, and thirteen-line rhymed alliterative stanzas. The list below indicates the verse form or forms as usually described for each of the Middle English texts, grouped under the main source texts. 1) From Fierabras (Old French and Anglo-Norman chanson de geste in alexandrines): Firumbras (BL MS Add. 37492, the Fillingham MS), 6-stress couplets The Sowdone of Babylone (Princeton University Library MS Garrett 140) [derived from AN La Destruction de Rome + Fierabras; cf. BL MS Egerton 3028], 4-stress cross-rhymed quatrains, with variant 4/3-stress stanzas Sir Ferumbras (Bodleian Library MS Ashmole 33), rhymed septenaries + 6-line (4/3-stress) tail-rhyme stanzas 2) From Otinel (Old French and Anglo-Norman chanson de geste in decasyllabics): Otuel a Kniȝt / Otuel (NLS MS Adv. 19.2.1, the Auchinleck MS), 4-stress couplets Duke Rowland and Sir Otuell of Spayne / Roland and Otuel (BL MS Add. 31042, the London Thornton MS), 12-line (4/3stress) tail-rhyme stanzas Otuel and Roland (Fillingham MS) [derived from Otinel + OFr prose Johannes Pseudo-Turpin], 12-line (4/3-stress; var. 3/3-stress) tail-rhyme stanzas, with two rhyme-schemes

itself, functions as a narrative and formal unit with linking devices’ (p. 441).

88

Charlemagne in Medieval England

3) From other sources: Roland and Vernagu (Auchinleck MS) [derived from OFr prose Johannes Pseudo-Turpin], 12-line (4/3-stress; var. 3/3-stress) tail-rhyme stanzas The Sege of Melayne (London Thornton MS) [no source known], 12-line (4/3-stress) tail-rhyme stanzas The Song of Roland (BL MS Lansdowne 388) [derived from La Chanson de Roland], 4-stress couplets with alliteration The Taill of Rauf Coilyear (print 1572) [no source known], alliterative 13-line rhymed stanzas This remarkable variety of verse forms is especially interesting in light of Furrow’s account of the line and laisse structures in early Anglo-Norman chansons de geste, which we discuss on pp. 58–9. While, as we argue there, the variation in these early chansons de geste is not quite as remarkable as Furrow’s account suggests, it nevertheless indicates a characteristic insular willingness to experiment. It may be that the heterogeneity of form in the Middle English Charlemagne romances similarly signals a site of uncertainty and experimentation. While the various types of tail-rhyme stanza employed in half the texts would indicate a widespread acceptance of this form’s general appropriateness for translating the laisse of chanson de geste into English verse, the widely differing solutions to the challenge attempted in the other texts perhaps also demonstrate a conscious engagement with their epic heritage. Rosalind Field notes parallel developments in both later Anglo-Norman and Middle English versification in response to the ‘fashion for stanzaic verse’, in that ‘several of the Anglo-Norman works [are] in regular monorhyme stanzas rather than true laisses, just as, in Middle English verse, alliteration is combined, sometimes very successfully, with stanza forms’;191 and although she uses the language of literary degeneration, speaking of these as ‘contaminated’ forms, her discussion implies instead development, experimentation and conscious choice. It is thus possible to see the marked formal variation that 191 Rosalind Field, ‘The Anglo-Norman Background to Alliterative Romance’, in Middle English Alliterative Poetry and its Literary Background: Seven Essays, ed. David Lawton (Cambridge: Brewer, 1982), pp. 54–69 (p. 61).

The Insular Literary Context

89

characterizes the body of insular Charlemagne texts as contributing to the ongoing process of adaptation and re-presentation of the traditional matière in light of evolving and changing cultural expectations. In some cases, the variations may be witnessing to a complex history of adaptation and borrowing from multiple sources, such as Purdie describes in relation to the tail-rhyme Matter of France romances,192 but in these and other cases, as we have argued elsewhere, the variant forms of the English Charlemagne romances may also be taken to represent a state of continuous engagement and experimentation with the problem of producing appropriate English equivalents for the vers and laisses of the French chanson de geste tradition.193 The Fierabras group presents at least four different verse forms, with the Fillingham Firumbras using what seems the closest approximation to the alexandrines of the original chanson de geste: six-stress lines with a medial caesura (though its use of rhyming couplets is closer to French romance norms). The Sowdone of Babylone begins by using a fairly regular four-stress quatrain, but through most of the text these quatrains alternate irregularly with what Smyser calls the Sowdone’s ‘ballad stanza’, a metrical scheme which is indistinguishable, apart from its layout in the manuscript, from the rhymed septenaries found in the first part of the Ashmole Sir Ferumbras.194 However, alongside this evidence of difference and experimentation there is one observable constant: all four verse forms can be understood as providing approximate equivalents for French alexandrines: units of six or seven stressed syllables that (no matter how they were displayed in manuscript texts) would be heard by readers and listeners as rhymed long lines with a medial caesura. The three Middle English texts in the Otinel group display none of these patterns: the Auchinleck Otuel uses four-stress couplets arranged in verse paragraphs of between six and twenty-eight lines, while three different kinds of twelve-line tail-rhyme stanza are used in the Thornton Roland and Otuel and the Fillingham Otuel and Roland. The two other texts usually attached to the Otuel group, that is, the Auchinleck Roland and Vernagu and the Thornton Sege of Melayne, each of which in different ways can 192 Purdie, Anglicising Romance, pp. 115–25. 193 Ailes and Hardman, ‘How English are the English Charlemagne Romances?’, pp. 45–7. 194 The second part of the Ashmole Sir Ferumbras uses a six-line tail-rhyme stanza.

90

Charlemagne in Medieval England

be thought of as a prequel to the Otinel story, are both also structured in twelve-line tail-rhyme stanzas. Following Purdie’s findings, all four can be perceived as adopting the tail-rhyme stanza as an ‘anglicized’ equivalent for the decasyllabic laisses of Otinel. The sense of experimentation is especially marked where there is not only variation between independent versions of the same source narrative, but also a change of verse form within a single text. The clearest example of what looks like formal experimentation within a text is presented by the Ashmole Sir Ferumbras. This manuscript text is assumed to be the translator’s holograph because of the fortuitous preservation of the partial draft in what was originally the parchment binding of the manuscript.195 The draft relates to about one-seventh of the first part of the poem, the half written in septenary couplets with internal rhymes, and the changes between the draft and the revised copy are interestingly inconsistent. While it would be satisfying to find, as Shepherd implies, that the alterations are all improvements, producing a more ‘eloquent’ and more metrically ‘finished’ text, this is not necessarily the case. Sometimes the extra syllables added in the revision serve to congest the verse rather than to improve the metre. For instance, when Oliver sets out to encounter Fyrumbras, the draft reads: ‘& Olyuer was fre to fond & forþ he gan to fare’, using the alliteration to mark the stressed syllables in the line, while the revised version reads: ‘& Oliuer þat was wel fre to fond ; tok is leue to fare’ (341). The reviser seems to have wanted to improve the simple paratactic syntax of the draft, and also to include the mention of leave-taking, which is a major issue in the French text at this point; these must have been more pressing concerns than maintaining a smooth metrical line. On the other hand, as pointed out long ago by Walther Fischer,196 a passage of nearly twenty lines in the draft is written in alexandrines, imitating the metre of Fierabras, and these have been revised in the second copy to produce regular septenary 195 Sir Ferumbras, ed. S. J. Herrtage, EETS ES 34 (London, 1879), pp. xiv–xvi; Shepherd, ‘The Ashmole Sir Ferumbras’; Phillipa Hardman, ‘Bodleian Library, MS Ashmole 33: Thoughts on Reading a Work in Progress’, in Middle English Texts in Transition, ed. Simon Horobin and Linne Mooney (York: York Medieval Press, 2014), pp. 88–103. 196 Walther Fischer, ‘Zur me. Romanze “Sir Ferumbras”: das Verhältnis des ersten Entwurfs. (V. 331–759) zur Reinschrift’, Archiv für das Studium der neueren Sprachen und Literaturen 142 (1921), 25–54.

The Insular Literary Context

91

lines. The interesting question is, why did the translator choose in the first place to use alexandrines for this passage? It is a passage of particular significance, in which Ferumbras describes to Oliver the miraculous powers of the healing balm, a relic of Christ’s burial, that he stole from Rome, and the contrast in metre draws attention to it, as well as perhaps making an implicit statement about the authenticity of this translated account of the relic. Unfortunately, we cannot know whether this device of using alexandrines was originally adopted in relation to other relics passages later in the text, as no draft survives.197 Shepherd demonstrates the sometimes marked fidelity of the Ashmole Sir Ferumbras as a translation of the French text; moreover, it seems that the translator was experimenting with formal as well as verbal imitation, for we find quite frequent groups of couplets sharing the same rhyme, and even internal rhyme, as if to reproduce something of the effect of the French monorhymed laisses. The experimental nature of the project is indicated by alterations between the draft and revised versions in order to change the rhyme words of couplets and achieve these more elaborate extended rhymes. In this passage, the story continues from Oliver’s leave-taking as he rides out to meet the Saracen Fyrumbras; the draft reads fluently, preserves the detail from the original French that Fyrumbras is under a tree, and creates an effective little narrative of anticipation building up to the moment of meeting: Oliuer torneþ him þanne wiþ an hardi chere, Toward þat heþen manne he rideþ a softe amblere; Til he cam þer þat he was him þoȝte ech stap was þre, At þe laste he fyndeþ Fyrumbras liggyng vnder a tre. (draft)

In the revised version much of this is sacrificed to produce the extended -as / -ere rhyme: Dvc Oliuer him rideþ out of þat plas ; in a softe amblere, Ne made he non oþer pas ; til þey wern met yfere: And wan he cam þer as he was ; þyderward he caste ys chere,

197 There is an interesting parallel in the Anglo-Norman text of Otinel, where two laisses (lines 17–46) are clearly written in alexandrines, unlike the rest of the text and the whole of the OFr version, all of which are in decasyllabics.

92

Charlemagne in Medieval England & fyndeþ þer sir Fyrumbras ; liggyng on þe erþe there. (344–7)

Another example, where the draft already has the four identical end-rhymes, shows the reviser refining his extended rhyme-scheme by altering the internal rhymes, at some cost to narrative cogency and metrical regularity. Fyrumbras is enquiring whether Oliver is already wounded: He askede þanne of Olyuer, þat houede þer him to byde: ‘Me þenkþ þou ert yhert ful sere & berest a wonde wyde By þat blod þat renneþ doun by þyn oþer side.’ Olyuer saide, ‘þat is noȝt so,’ & turnde him hit to hide. (draft) & þan he askede of Olyuere ; þat houede þer him t’abide, If þat he any wonde bere ; in ys body þat tyde: ‘Me þynkþ þou hast a wonde þere,’ ; said he, ‘in þy syde.’ ‘þerof ’, quaþ Olyuer, ‘ne haue þou no fere,’ ; & turnd him þat sor to hyde. (500–3)

While experimentation may not have improved the versification here, it does show a deliberate, continuing engagement with the defining formal characteristics of French epic discourse. The other question is why the verse form changes half-way through the text from septenary couplets to six-line tail rhyme stanzas. One reason for the change may be to do with the observed preference in Middle English romance for subdividing longer works into shorter narrative units, such as is seen in the well-known example of the Auchinleck Guy of Warwick, where the content of the Anglo-Norman romance is repackaged as three consecutive but separate Middle English romances, and their different verse forms serve to enforce their discrete status.198 A further possible explanation for the choice of tail-rhyme seems to be that the basic three-line unit provides a more spacious and flexible equivalent to the French alexandrine for the

198 For further discussion of this hypothesis in MS Ashmole 33, see Ailes and Hardman, ‘Texts in Conversation’, p. 40. For discussion of the possibility of a change of source at this point, see Chapter 5, pp. 272–3.

The Insular Literary Context

93

somewhat wordy style of the Ashmole translator.199 The change occurs at a point where the Ashmole text is following the form of the French poem very carefully, placing a large capital (3351) to mark the section corresponding to laisse 101, and another (3411) to mark the next section, now in tail-rhyme, corresponding to laisse 102.200 The French peers are besieged with Floripas in the tower, discussing whether or not to send to Charlemagne to rescue them. If the speech of Richard of Normandy is compared in the French and English texts, it seems that the Ashmole poet is treating the tail rhyme unit as more or less equivalent to the long line of the chanson de geste: Premier en apela Richart de Normendie: ‘Seignors, ce dist li dus, ne vos mentirai mie, Nos sonmes chi enclos en ceste tor antie; Je sai bien que a longues n’i durrerons nos mie. Ker mandon a Karlon et secors et aie.’ (Fierabras, ed. Le Person, 4039–43) Þanne spak Richard of Normaundy | To þe barons þat stode hym by: | ‘Herknyaþ for ȝour honour, Wel ȝe wyteþ we buþ her enclos | Hard byseged wyþ our foes | & with strengþe & gret vygour; Of o þyng lordes beo ous sure, | Her mowe we noȝt longe dure, | Bot ous come socour: Sende we þerfor to þe Emperer, | þat he come with his power | & delyuery ous of the tour.’ (3419–26)

This is one of the very few cases in the text where four tail lines rhyme together instead of only two, unusually producing a twelve-line stanza. It perhaps signals another experiment by the Ashmole poet in translating the formal features of the French laisse into English verse. 199 Shepherd gives an example of ‘direct borrowings’ from Fierabras and ‘a parallel of lineation (in this new scheme it is a strophe-for-line parallel)’ in Sir Ferumbras, 4899–904 and Fierabras (ed. Le Person), 5290–2a (Shepherd, ‘The Ashmole Sir Ferumbras’, pp. 107–8, n. 5). 200 Laisse 101 has sixty-nine lines and the corresponding section of the Middle English translation consists of sixty septenaries; laisse 102, with twenty-four lines, corresponds to sixteen tail-rhyme strophes.

94

Charlemagne in Medieval England

A very large proportion of the Sowdone of Babylone is written in flexible rhyming quatrains of four or three stresses, frequently falling into apparent ballad stanzas, as noted above, where the syntax supports a reading in septenary couplets with internal rhymes like those of the Ashmole Sir Ferumbras. Also like that text, there are occasional tours de force where the poet extends the rhyme, as here: Tho spake Savaryȝ with wordes on hye | And saide ‘My felowes alle, This daie prove you men worthy | And faire you al shal befalle. Thenke yat Criste is more mighty | Than here fals goddis alle, And he shal geve vs the victorie | And foule shal hem this day bifalle.’ (192–9)

Much of the Sowdone is a free reworking of the Destruction de Rome + Fierabras material as found in the Egerton version, but there are passages where the English poet’s rhymed lines can be seen to translate the French alexandrines quite closely: A itant vint Estragot, qi port un mace de fer. Grant coupe s’en voit sur la porte doner. Romeins i kurent le port colice sus sachier. (La Destruction de Rome, ed. Brandin, 581–3) Tho the grete gloton Estagote | With his mighty mace sware On the gatis of Rome he smote | And brake hem alle on thre thare. In he entrid at the gate, | The porte-colis on him thai lete falle. (427–32)

By contrast with this flexible narrative style rendering the translator’s adaptation of the Anglo-Norman poems, the opening lines of the Sowdone are written in stiffly regular tetrameters. The poet seems to have had a clear idea of appropriate verse styles: lofty language and stately syntax for the inserted introductory passages; simple phraseology in flexible couplets for translating the long lines of French epic narrative. The use of six-stress lines with a strong medial caesura in the Fillingham Firumbras is unusual among Middle English romances;201

201 Ad Putter overlooks Firumbras when noting the rare use of alexandrines in Sir Ferumbras: ‘The Metres and Stanza Forms of Popular Romance’, in

The Insular Literary Context

95

but the English adaptor is not without technical skill in exploiting the opportunity for rhetorical patterning offered by the form, as here: Rowland wex ful pale // & so dud Olyuer, And eythyr made hys mone // that routhe was to huyre. (209–10)

The six-stress couplet form is not rigidly adhered to: the flexible verse varies from four stresses to seven in a line, but six is the standard number.202 Since four-stress lines and septenaries or ballad metre are so much more common (and one suspects more comfortable) in Middle English narrative verse,203 the persistent attempt to sustain these alexandrine-equivalent six-stress lines throughout the Fillingham Firumbras seems to indicate a specific desire to appropriate some feature of the form of Fierabras, although no attempt is made to reproduce the characteristic laisse structure of the chanson de geste: the couplets are grouped into just ten divisions in the text as against the eighty-three assonanced laisses found in the equivalent section of Fierabras. However, as Purdie has shown, twelve-line tail-rhyme, with its extended single rhyme, can be seen to reproduce something of the effect of monorhymed laisses. This is particularly marked in the more demanding forms of the tail-rhyme stanza used in the Thornton Roland and Otuel and the first half of the Fillingham Otuel and Roland, where couplets rhyme as well as tail lines. A stanza sometimes directly matches a short laisse (for example, Otuel and Roland, 75–86, parallels Otinel, 63–76), but usually a group of stanzas parallels a laisse (for example, Otuel and Roland, 87–113 parallel Otinel, 77–136). A different system of textual divisions that may equally reflect the structure of laisses in Otinel occurs in the Auchinleck MS, where the couplet Otuel is set out in the format used for almost every text in the manuscript: short lines in two A Companion to Medieval Popular Romance, ed. Radulescu and Rushton, pp. 111–31 (p. 114, n. 12). 202 Derek Pearsall points to the unstable metre of the romances of Gamelyn and Beryn, both in couplets, which alternate irregularly between the six-stress line and the perhaps more familiar seven-stress line (‘The Metre of The Tale of Gamelyn’, a paper given at the Romance in Medieval Britain Conference, University of Bristol, 2014). 203 Compare the Ashmole Sir Ferumbras and The Sowdone of Babylone.

96

Charlemagne in Medieval England

columns, with paraphs at frequent intervals. The tail-rhyme stanzas of its companion romance, Roland and Vernagu, are indicated by a paraph at the first line of each, whereas the paraphs in Otuel occur at less regular intervals, and organize the couplets into verse paragraphs. These units of discourse correspond to some extent to the laisses of the Anglo-Norman Otinel, in that the start of almost every laisse in the first half of the chanson de geste (fewer in the second half) is matched by the beginning of a verse paragraph in Otuel, though there are further subdivisions in many of these paragraphs. As a further reflection of French epic practice, Otuel and Roland has a unique system of text divisions: six-line narratorial interpolations which present the narrative as a sequence of battles. The first few correspond exactly to the beginnings of laisses in Otinel. The style of these passages is modelled on the conventional call for attention in chansons de geste, and the formulaic repetition in all of them (‘Here beginneth a battle snelle / strong / fell’) reproduces the characteristic effect of Old French epic discourse (for example, in Fierabras numerous laisses begin: ‘Mout fu fort la bataille’). As the English poem progresses, the occurrence of this structural device shows increasing independence from the French source, but the adapted form is faithfully maintained throughout as a system of textual divisions that self-consciously alludes to the rhetorical practice of the chansons de geste.204 The Song of Roland and The Taill of Rauf Coilyear are the only Charlemagne romances to draw extensively on the insular tradition of alliterative verse, and, despite their obvious dissimilarities, there is an interesting aspect to the versification of each that can perhaps be seen as experimentation in response to French epic models. The Song of Roland has been criticized for its ‘feebly alliterative’ form,205 and compared unfavourably with the true ‘epic spirit’ of the alliterative tradition embodied in the alliterative Morte Arthure;206 but this is to assume that the poem represents an attempt to employ alliteration in accord with 204 The Auchinleck Otuel introduces a similar epic device in its call for attention halfway through the romance (669–672): see pp. 367–8. 205 Derek Pearsall, ‘The Alliterative Revival: Origins and Social Backgrounds’, in Middle English Alliterative Poetry and its Literary Background, ed. Lawton, pp. 34–53 (p. 36). 206 Barron, English Medieval Romance, p. 92. Herrtage states: ‘The metre of the poem is alliterative, but the rules are far from being strictly observed’

The Insular Literary Context

97

the same tradition, whereas the evidence suggests a less structured use of the alliterative heritage. The text is composed in four-stress rhyming couplets, a form shared with several other Middle English romances dealing with legendary historical subjects, such as Arthour and Merlin, Kyng Alisaunder, and The Seege of Troye. For long stretches the narrative moves fluently with little or no alliteration. However, passages of heightened drama and emotion or (as Barron notes) of violent action are marked with a range of alliterative effects, from coupled words (‘wonderfull wais’ (14)) to lines resembling traditional alliterative verse: ‘Helmes and hedes he hewithe of stout’ (747).207 These alliterative features provide the poem with an element of English heroic style that can parallel the formal features of epic discourse in the chanson de geste (though it should be noted that the imperfect unique manuscript copy preserves some traces to suggest that the process of combining rhyme and alliteration might still have been incomplete when the exemplar was produced).208 The poet of The Taill of Rauf Coilyear also selected a verse form that combines alliteration and rhyme, but in this case with the added sophistication of a demanding (but not uncommon) thirteen-line stanza form. Thus, despite the broad humour that characterizes this late addition to the corpus of Charlemagne texts, the technical refinement of its versification testifies to the seriousness of its engagement with the French epic tradition, by adopting a formal structure in the thirteen-line alliterative rhymed stanza that parallels the discipline of the assonanced or rhymed laisse of the chansons de geste.

(‘The Sege off Melayne … with a fragment of ‘The Song of Roland’, ed. Herrtage, EETS ES 35 (1880), p. xxii). 207 Field describes a comparable instance in the case of Ipomadoun A, where the Middle English translator uses heavy alliteration, ‘the heroic style of his own linguistic tradition’, to render the effects of the Anglo-Norman poet’s exploiting ‘the chanson style to embellish scenes of battle and heroic action’ (‘The Anglo-Norman Background’, p. 62). 208 For discussion of this point, see Hardman, ‘Bodleian Library, MS Ashmole 33’, pp. 100–2.

98

Charlemagne in Medieval England

The Matter of France and National Identity in the Middle English Romances Introducing his descriptive commentary on the English Charlemagne legends in the Romances volume of A Manual of the Writings in Middle English, 1050–1500, Smyser gives an account of the Old French cycle du roi as celebrating ‘the predominant national hero of the French’, and assumes that these chansons de geste, being ‘lastingly infused with the spirit of patriotism’, could never be ‘at home’ in other European national cultures in the way that the Arthurian cycle was (p. 80).209 ‘The Charlemagne legends’, he claims, ‘lost an essential element of patriotic appeal when exported’, which he takes to explain the ‘undistinguished’ quality of the Middle English romances.210 Whatever the merits of this claim in relation to the Matter of France when translated into other European languages,211 its inadequacy to explain the case in the complex linguistic, cultural and political situation of late-medieval England is obvious, and we have indicated some more nuanced modern critical approaches to these questions of language, nation and identity as they relate to the translation of vernacular texts in the Introduction. However, it is important also to interrogate the Middle English texts: what evidence do they provide of attitudes or anxieties in relation to these issues? The Auchinleck MS (c. 1340), which contains the two earliest copies of Middle English romances of Charlemagne (Roland and Vernagu and Otuel), has been much discussed in terms of national identity.212 In an 209 Smyser adds: ‘It is not odd that Chrétien de Troyes should treat of Arthurian themes, and the Grail story is as much at home in Germany as Wales’ (‘Charlemagne Legends’, p. 80). 210 Smyser explains the relative literary success of the Taill of Rauf Coilyear by its being ‘only nominally a Charlemagne romance’ (ibid.); but as we argue (pp. 104, 396–400), it shows thorough assimilation of the Matter of France tradition. 211 For discussion of these texts, see further volumes in the series ‘Charlemagne: A European Icon’. 212 Thorlac Turville-Petre uses the Auchinleck MS as a case study on the cultural construction of Englishness in his seminal book, England the Nation: Language, Literature, and National Identity, 1290–1340 (Oxford: Oxford University Press, 1996); Siobhain Bly Calkin concentrates on the same manuscript in Saracens and the Making of English Identity: The

The Insular Literary Context

99

essay on ‘Middle English Popular Romance and National Identity’, Thomas Crofts and Robert Rouse take Auchinleck as their prime example for assessing the ‘complex Englishness(es)’ to be found in Middle English romance,213 and while they do not discuss its two Charlemagne texts in any detail, they offer a robust corrective to hopes of finding sharply defined English versus French identities: Any expectation that these romances will show anti-French sentiment is soon disappointed: … Points of contact between the Matter of France and the reading material of the English suggest that either a) it was a non-issue for romance, which was very busy doing other things, or that b) appropriation, rather than polarization, was the chief register of national difference. (p. 86) These alternative suggestions are both borne out in an examination of the two Auchinleck Charlemagne romances, which differ sharply in their representation of national identity. Roland and Vernagu makes no explicit reference to national identity at all. Charlemagne is identified as the emperor of Rome, where he is found in his palace; France is named, together with England and Denmark, as among his territorial possessions, and these different locations are reflected in the varied origins of his most celebrated peers: Oger, called the Dane (43), and Roland, who names France as his place of birth (668). The mention of England here is not surprising,214 as it is found in the source, the Johannes redaction of the Old French Pseudo-Turpin Chronicle, and also in the best-known chanson de geste, the Chanson de Roland. It does, Auchinleck Manuscript (New York: Routledge, 2005), and reads its two Charlemagne texts, Roland and Vernagu and Otuel, focused on the idea of simultaneous difference and sameness between Christian knights and Saracens, as an opportunity for exploring contemporary relations between English and French knights. See also Libbon, ‘The Invention of King Richard’, and her forthcoming book, Richard I and the Idea of England. 213 In A Companion to Medieval Popular Romance, ed. Purdie and Cichon (2009), pp. 79–95. 214 Pace Calkin, who claims it is an ‘addition’ in Roland and Vernagu (Saracens, p. 17).

100

Charlemagne in Medieval England

however, point to an early opportunity afforded by the Charlemagne tradition for appropriation by writers and readers in England, an opportunity that was strikingly seized by the Anglo-Norman reviser of the Oxford Chanson de Roland, to claim not only that England was among Charlemagne’s conquests, but also that he chose to make it his own personal domain.215 But no such claim is made in Roland and Vernagu; it seems that here the question of national identity is not a site of contest, or even of particular interest. In Otuel, by contrast, references to Charlemagne’s knights as ‘the French’ are retained, and even added to. In a striking incident, Roland, stung by Otuel’s scorn, promises that if they should meet in battle, Otuel, ‘þat heþene kniȝt’ as he is twice named (279, 284), should never again ‘despice freinchs man’ (290), posing a direct opposition between the two identities. It seems that in Otuel, ‘Frenchness’ as opposed to Saracen ‘Otherness’ has been appropriated and internalized by the Middle English adaptor as an acceptable, inclusive Christian identity that insular readers and listeners can embrace in their narrative hero, and one that can equally represent Otuel himself after his conversion when he fights alongside Roland and Oliver: ‘Þo þe freinche kniȝtes seien | Þe sarasins fallen wiþ hare eien’ (1099–1100). The case of the Ashmole Sir Ferumbras (c. 1380) is similar: alongside the heightened presentation of religious difference between Christians and Saracens (as discussed pp. 273–4), the text retains frequent references to the identity of the Christian knights as ‘þe frenschemen’, while at the same time invoking the reader’s identification with them as ‘oure cristen men’. Fifteenth-century copies of Charlemagne romances provide similar evidence of both erasure and appropriation of French identity. As we discuss in Chapter 6, the two romances in the Thornton MS (c. 1440) represent the identity of the Christian knights differently, although both make extensive use of the first-person plural pronoun to include the reader and narrator with the knights in a common identity: in Roland and Otuel this is almost exclusively figured as a shared Christian identity with no mention of nation or language, whereas in the Sege of Melayne the explicitly French identity of ‘oure folke’ seems as thoroughly appropriated as it does in Otuel. In the Fillingham MS (1475–1500), Otuel and Roland follows the precedent of Auchinleck Otuel in retaining the references from the French-language source to Charlemagne’s 215 See Introduction, pp. 3–4 above, and Chapter 4, pp. 227–32.

The Insular Literary Context

101

knights as ‘the frenche’ in opposition to the ‘hethyn’ Other;216 while in the second half of the text, based on the Johannes Pseudo-Turpin, the opposed identities are constructed as ‘the cristen’ versus ‘the hethen’, ‘the paynemes’ or ‘the sarisins’, who are uniquely further characterized as ‘godes wytherlynges’ (2296, 2645) and ‘the deueles lemes’ (2184) or ‘the fendys lemes’ (2648), to accord with the text’s heavily pointed account of Charlemagne’s actions as part of Providential history, carrying out God’s will. A similar perspective is created in Firumbras in the same manuscript, where Charlemagne’s knights are presented as ‘Oure cristenmen’ and the narrator frequently exhorts the reader to pray on their behalf in their battles against ‘the sarisins’ (see p. 297). France is named as Charles’s kingdom but, unlike the Ashmole Sir Ferumbras, all mentions of the knights as ‘French’ in the inherited Fierabras tradition are here erased. The third Middle English romance based on the Fierabras tradition, The Sowdone of Babylone, takes an intermediate approach, sometimes reproducing references to ‘the Frenshmen’ as given in the related Egerton texts, and sometimes erasing them or replacing them with ‘the Cristen’. Most tellingly, perhaps, the Middle English Song of Roland (rewriting the text perceived in modern times as the national epic of France) replaces every narratorial reference in the chanson de geste to Charlemagne’s men as ‘Franceis’ with ‘the cristyn’, or simply ‘our folk’, ‘our men’, where ‘the cristyn’ are opposed to ‘the hethyn’, and the reader is incorporated into support for ‘oure cristyn men’ in the same way as in the Fillingham Firumbras. However, in speeches attributed to the Saracens, the Christians are referred to as ‘the frenche’ (739, 898, 1006), and the different perception of identity perhaps helps to distance the Saracen viewpoint from the shared Christian identity constructed among the ‘fellowship’ of knights, narrator and reader.217 Whether by erasing national identity or by appropriating ‘Frenchness’ as an acceptable insular identity, all these English romances subsume the question of national identity within the larger issue of Christian identity in opposition to a non-Christian ‘Other’. 216 This fits with Rhiannon Purdie’s argument that a version of the couplet Otuel was used as a crib by the author of the tail-rhyme translation of Otinel that lies behind Otuel and Roland (see above, pp. 77–8). 217 For the importance of ‘fellowship’ as a theme in this version of the tradition, see Chapter 4, pp. 244–5.

102

Charlemagne in Medieval England

The Coherence of the Middle English Charlemagne Romances The Matter of France chansons de geste exhibit a high degree of intertextual allusion that serves to construct and consolidate the identity of the literary and cultural tradition to which they belong;218 the Middle English romances continue this trait, connecting separate poems by added cross-references (as in the closing lines of Roland and Vernagu quoted above, referring forwards to the story of Otinel), and alluding by imitation to incidents in other Charlemagne texts. For example, in the Sowdone of Babylone Ferumbras claims that he has long sought revenge on Oliver, who slew his uncle Sir Persagyne (1259–62), imitating Otuel’s seeking revenge on Roland for killing his uncle Sir Vernagu (Roland and Otuel, 313–18); in a combat scene between Oliver and Ferumbras, the Ashmole Sir Ferumbras makes several narrative additions including the detail that Oliver struck Ferumbras a blow that shaved his beard (612–18), which seems to allude to the scene in the Otinel tradition when Otuel shaves Clarel to the teeth (Roland and Otuel, 1318–20). All this contributes to a strong sense that the Middle English Charlemagne romances themselves constitute a consciously coherent body of texts within the insular Matter of France tradition. Confirmation may be provided by intertextual allusions to the Charlemagne material found in romances outside the core tradition, obviously drawing on recognizable motifs that would enrich their different context: a striking example can be seen in Octovian Imperator, the variant Southern version of the Middle English romance Octavian.219 The Northern version of the romance, closer to the French source, introduces the central scene (in which the young Florent reveals his chivalric prowess in single combat against a Saracen giant) by briefly sketching a background of war between the unidentified Sowdon’s marauding army and the helpless people of France, whose five bold

218 See Marianne Ailes, ‘Fierabras and the Chanson de Roland: An Intertextual Diptych’, Reading Medieval Studies 28 (2002), 3–22; Marianne Ailes, ‘Chivalry and Conversion: The Chivalrous Saracen in the Old French Epics Fierabras and Otinel’, Al-Masaq: Studia AraboIslamica Mediterranea 9 (1996–7), 1–21; Ailes, ‘Otinel: An Epic in Dialogue with the Tradition’. 219 Octovian Imperator, ed. Frances McSparran, Middle English Texts 11 (Heidelberg: Winter, 1979).

The Insular Literary Context

103

knights are summarily despatched by the giant.220 In Octovian Imperator considerably more detail is added, both to the history of the Sowdon and to the combat between the giant and the Christian knights, which relates this episode to celebrated archetypes in the Matter of France tradition. A whole stanza provides the title and territorial conquests of the Sowdon: Of Babylonye þe hegh sowdan Werrede vpon Crystene men; Boþe into hys hond he wan Gales & Spayne, Lumbardie & ek Tuskan Rome & Allemeyne. (907–12)221

His Saracen faith is made explicit: he ‘left yn Teruagaunt’ (919), and his message to the king of France declares his ultimate aim: ‘Crystendom schall adoun | Fram eueryche man’ (945–6).222 The hideous appearance and strength of the giant ‘of Egypte’ are fully described, with details of his boar-like tusks and bristles (920–36), while the champions who ride out against him one by one are identified as ‘þe .xii. dusepers of Fraunce’ (980). When all are slain, the narrative comments: ‘Þys was vnhap & hard chaunce | To alle Crystendome’ (983–4). The agenda here is clear: echoes of similar hostile challenges and Saracen giants in the Otinel and particularly the Fierabras traditions create a parallel between the threat to the king of France in this romance and Charlemagne’s conflicts with Balan/Laban and Garcy, and prepare the way for Florent to outshine the legendary heroes Roland and Oliver and the other peers. Later, in another unique addition, when Florent fights the Saracen he kills 220 Octavian, ed. Harriet Hudson, in Four Middle English Romances, 2nd edn (Kalamazoo, MI: Medieval Institute Publications, 2006). Octavian is one of the texts identified by Lee Manion as a non-cyclic crusading romance. 221 The English Charlemagne romances typically increase the geographical specificity of Saracen lands and conquests. 222 In the Northern Octavian, the Sowdon expresses anti-Christian vengeance only after he thinks a Christian knight has attempted to abduct his daughter (1112–35). Octovian reflects the heightened Saracen aggression typical of English Charlemagne texts.

104

Charlemagne in Medieval England

his horse, at which the Saracen complains (1097–9), an incident that seems to be modelled on one in the Otinel story,223 inviting the reader or listener to recognize the parallel between this romance and that archetypal narrative of Christian–Saracen combat. A similar but more substantial appropriation of the Matter of France occurs in The Taill of Rauf Coilyear, the latest of the EETS designated ‘English Charlemagne Romances’;224 it is set in Paris at the court of King Charles, but only in the last quarter of the poem is any capital made out of the choice of Charlemagne for the ‘Unrecognized King’ story that occupies most of the narrative. As critics have noted, despite the humorous, even slapstick, elements of the narration, the text exhibits a high level of technical sophistication, and behind its imitative reuse of the Saracen versus Christian knight combat and conversion scene, prominent in both the Fierabras and Otinel traditions, is evidence of serious engagement with those issues of cultural and religious confrontation that are central to the Matter of France texts.225 This kind of creative reimagining of iconic episodes from the insular Charlemagne legend, with its concentration on Fierabras and Otinel, in romances that otherwise belong outside the core tradition, points to a familiarity with the Charlemagne texts that writers were able to assume as an integral part of the horizon of expectations they shared with the readers of Middle English popular romance. It also exemplifies the adaptability of the tradition, the ‘license to change things’ noted by Crofts and Rouse, that can help to explain the persistent and repeated engagements by English poets with the same few texts over one and a half centuries. Some things, however, are resistant to change. For example, trace evidence of insular familiarity with the paired heroes of the Charlemagne legends persists in records of brothers christened Roland and Oliver,226 in proverbs, and in diverse literary and documentary

223 Otinel: Chanson de geste, ed. F. Guessard and H. Michelant, APF 1 (Paris: Vieweg, 1859), ll. 450–3; Otuel, in ‘The Taill of Rauf Coilyear’ , with the Fragments of ‘Roland and Vernagu’ and ‘Otuel’, ed. S. J. H. Herrtage, EETS ES 39 (1882), ll. 469–74. 224 Edited by Herrtage (together with Roland and Vernagu and Otuel) as Part VI in the series. 225 See pp. 396–400. 226 Short, ‘Literary Culture at the Court of Henry II’, p. 357.

The Insular Literary Context

105

references alluding to the duo.227 It is thus not surprising to find that the close relationship of Roland and Oliver is often expressed in the Charlemagne texts in symmetrical verbal structures that can sometimes be traced throughout the whole tradition across languages and centuries, as one striking instance may serve to demonstrate. In Fierabras, when the knights launch a surprise attack on the Saracens, the Emir Balan jumps out of the window to safety, escaping his pursuer: ‘Compaigns, dist Olivier, est vos il escapé? – Oïl, ce dist Rollant, tant sui gen plus irés.’ (Fierabras, ed. Le Person, 3105–6) [‘Companion’, said Oliver, ‘Has he got away from you?’ ‘Yes’, said Roland, ‘I am very angry about this’.] ‘Oliver’, dist il [Rollant], ‘vous est il eschapee? – Veir’, dit Oliver, ‘pur ceo su mult irré’. (Fierenbras, ed. Brandin, 943–4) [‘Oliver’, said Roland, ‘has he got away from you?’ ‘Indeed’, said Oliver, ‘I am very angry because of this’.] ‘Felawe,’ saide sir Olyuere: ‘ys he ous now ascapid?’ ‘Ȝe, forsoþ,’ saide he [Rolond] ther: ‘ac oþerweys y hadde yschape hit’ (Sir Ferumbras, 2317–18) ‘Felowe,’ seyd Oliuer, ‘ys he alyue ous fro?’ ‘Ȝe, for god,’ sey Rouland, ‘therfore me ys who.’ (Firumbras, 42–3) Roulande than came rennynge And axed where was Laban. Olyuere answerede moornynge And saide howe he was agoon. (The Sowdone of Babylone, 2047–50)

227 Turpines Story: A Middle English Translation of the Pseudo-Turpin Chronicle, ed. Stephen H. A. Shepherd, EETS OS 322 (Oxford, 2004), pp. xli, 84.

106

Charlemagne in Medieval England

‘Compaing, dist Olivier a Roland, l’admiral vous est il eschappé? Ouÿ, certes, dist Roland, ‘dont je suis mal content.’ (Bagnyon’s Fierabras, p. 100 [56v]) [‘Companion’, said Oliver to Roland, ‘Did the emir get away from you?’ ‘Yes, indeed’, said Roland, ‘I am most unhappy about it’. ] ‘Brother,’ sayd Olyuer, ‘the admiral is escaped fro you.’ ‘Certeyn,’ ansuerd Rolland, ‘therof am I euyl contente.’ (Caxton’s Charles the Grete, p. 121, ll. 4–6)

Despite their prose form, both Jean Bagnyon’s Fierabras (derived from the verse Fierabras) and Caxton’s close translation faithfully preserve the shape of the exchange. Three of the four English texts follow the same model in having Oliver question Roland, addressing him as ‘fellow’ or ‘brother’ after the French ‘compaigns’, while The Sowdone, closely related to the Anglo-Norman Egerton text, shares its different attribution of speeches (Roland questions Oliver) but increases the symmetry of the phrasing, despite recasting the passage in indirect speech, perhaps as part of a project to create a more formal, ‘literary’ text (see pp. 343–5). Two of the English verse texts, moreover, indicate the shared nature of the comrades’ loss by replacing the French ‘vos’ with ‘ous’, 228 thus reinforcing the effect of the symmetrical phrasing as a marker of the fellowship between them. The persistence of such verbal structures in the Otinel/Otuel textual tradition is not quite as striking, as there are fewer and less varied witnesses; nevertheless, some interesting and long-lived parallels can be seen. For example, the central proposition of this chanson de geste is that the newly baptized Saracen Otinel is incorporated into the brotherly fellowship of Roland and Oliver – the iconic pair of knights is transformed into a trio. To underline this, when Otinel’s conversion has finally been made clear to the Emir Garcy, another Saracen champion, a Turk, emerges to challenge all three of Charlemagne’s pre-eminent peers: Roland, Oliver – and Otinel. Au roi Garsile a demandé le gant,

228 This change could be the result of a misreading of the minims in French ‘uos’, but even if accidental it contributes to the different emphasis in the Middle English texts.

The Insular Literary Context

107

Du premier cop por ocirre Rollant Ou Olivier ou Otinel le grant. (Otinel, ed. Guessard and Michelant, 1629–31) [He demanded the glove from King Garsile To kill Roland with the first blow, Or Oliver, or the huge Otinel.] Al rei garsie a demande le guant Del primer coup pur occire Rollant V oliuer v ote le vaillant. (AN Otinel, Bodmer MS 168, fol. 220ra) [He demanded the glove from King Garsile To kill Roland with the first blow, Or Oliver, or the valiant Otinel.] he askede leue at Sir Garcy there To Iuste with Rowlande and Olyuere, theire bothere dede to dyghte. […] he called firste on Rowlande, One Otuell stalworthe for to stande, And sythen One Olyuere. (Roland and Otuel, 1366–8, 1372–4)

The corresponding passage is missing from Otuel and Roland,229 but the same point is made a few lines earlier when the target trio of peers is named as Garcy’s army prepares for hostilities: Alle þretyng charlys the kyng, And Both Otuel and Olyuer, And also eche dussypere, And Rouland, hys gode derlyng. (Otuel and Roland, 1579–82)

229 The Fillingham MS is imperfect at this point. Seven folios are missing after the episode breaks off: ‘A torkeyes was prykyng out before’ (1613).

108

Charlemagne in Medieval England

In an interesting variation in the Auchinleck Otuel, the Turkish champion engages Roland (1380), another ‘stout sarazin’ takes on Oliver (1399), while a third ‘noble sarazin’ is faced with Otuel (1417), in three passages whose parallel structure recalls the patterned laisses of chansons de geste. At the end of the episode there is a general mêlée, in which the three heroes are the only knights named on the French side: Roulond & oliuer hulpen wel, & þe douȝty otuwel. (Otuel, 1457–8)

Earlier in the narrative, Roland had offered the promise of this knightly fellowship as an incentive to Otuel to convert to Christianity: Et moi et toi soromes compaignon, Si conquerrons et chautiax230 et dongon Ja plus de toi ne quier .I. esperon. (Continental Otinel, 518–20) [And you and I will be companions, And we will conquer castles and strongholds I will not seek to spur against you again.] E io e tu serrum tut dis mes compaingnun E oliuer a nus auisterun Si conquerrum chastel e dunchun Cite ne Marche ne bon chastel gascun Ja plus de tei ne quier un esperun. (AN Otinel, fol. 213va–b) [And you and I will be aways companions And we will keep Oliver with us And we will conquer castles and strongholds citadels and marchlands and good Gascon castles I will not seek to spur against you again.]

230 The editors emend this MS reading to ‘chastel’ in line with the Bodmer MS. We agree that ‘chautiax’ (‘chattels’) is most likely a scribal error, so have translated it as ‘castles’.

The Insular Literary Context

109

And thou and I & Oliuer Mowen wende togedyr in fere In-to batayle and in-to fyȝt. Ne schulle we fynde in no londe None that schall vs withstonde Neyther kyng ne knyght. (Otuel and Roland, 515–18) And thow and I and Olyuer, We schall be felawes all in fere, & trauell nyghte & daye: We schall ryde bothe ferre & nere, Wyn Citees & townnes dere, & gode horses at assaye. (Roland and Otuel, 523–8)

Notable here is the more fully inclusive formula shared by the AngloNorman and Middle English texts (‘E io e tu … E oliuer’; ‘And thow and I and Olyuer’), reinforcing the idea that the new group of three will take the place of the former pair of peers. The more expansive images of the future conquests they will achieve together are also striking. But the crucial promise of fellowship is common to all these versions: to be ‘compaignon’, ‘felawes’, ‘togedyr in fere’. 231 As these emblematic details indicate, the insular body of texts, while faithfully preserving the core ideology of knightly fellowship in the Matter of France, can also engage creatively with the tradition; and creative engagement in the Anglo-Norman chansons de geste and the Middle English romances is the focus of our discussion in the following chapters. The few texts that were translated into Middle English were closely connected, both in their narratives and in their themes. These connections, and the appropriation of the narrative traditions to address specific insular concerns, will be a major strand in our analysis of the Charlemagne texts.

231 In the Auchinleck Otuel Roland omits this promise and offers only the hand of Charlemagne’s daughter, his cousin Belecent.

2 Charlemagne ‘Translated’: The Anglo-Norman Tradition

T

he insular tradition of Charlemagne is concentrated in a small number of related core texts, repeatedly reworked in both languages. The theory that so few chansons de geste were translated into English during the Middle Ages because only certain texts were circulating in England in French is difficult to sustain, as noted in Chapter 1, for while fewer manuscript witnesses to the genre survive in insular than continental manuscripts, it is clear that other texts were known that have not survived or survive only in fragmentary form. At the very least we must admit to uncertainty; we cannot argue from an absence of surviving texts that a particular narrative was not known in England. Yet it remains striking that we have Middle English translations of only three chansons de geste: Fierabras, Otinel and the Chanson de Roland.1 This chapter is largely concerned with the engagement of the insular French versions of these narratives in their particular cultural and political milieux, including the manuscript contexts in which they are found. Before turning to the major thematic concerns shared by the insular texts it is worth noting that there are also particularly strong thematic and narrative links connecting the original Old French texts which lie behind the insular narratives: Fierabras, Otinel, and both the chanson de geste and chronicle versions of the Roland narrative. Clear intertextual allusions invite us to read the Fierabras and Otinel material together, and both in conjunction with the matter of Roncevaux.2 Intertextual dialogue works in a complex way in the chanson de geste, as these 1 The Middle English Song of Roland, while most closely related to the Oxford Roland, also draws on the continental rhymed Roland tradition: see Chapter 4. 2 For more detail on the relationship between the French texts, see Ailes, ‘Chivalry and Conversion’; Ailes, ‘Otinel: An Epic in Dialogue’. The term ‘matter of Roncevaux’ is used here to include both the different redactions of the chanson de geste and the Pseudo-Turpin Chronicle.

110

The Anglo-Norman Tradition

111

narratives continue to be disseminated, at least in part, through oral performance as well as in manuscript; the written texts would continue to be received aurally. The texts are also particularly unstable and subject to remaniement, as noted in the Introduction. This allows ongoing debate and reconsideration of the themes raised by the narratives. As the texts are adapted and reinterpreted, the connections between them become more complex. In the continental tradition, the Pseudo-Turpin Chronicle draws on the narrative found in the Chanson de Roland, while Otinel is best understood as a response to Fierabras. Fierabras almost certainly exploits the Pseudo-Turpin as well as the rhymed version of the Roland,3 and is situated in fictional time in relation to the treachery of Ganelon as recounted in the Chanson de Roland; this treachery will lead to the Battle of Roncevaux in which Roland is killed (Fierabras, ed. Le Person, 6396–402). Similarly, in the later poem Otinel, we are reminded that the twelve peers loved each other so much that they would not leave one another ‘until the day when they died at Roncevaux’.4 This small group of texts, all responding to one another, provide the core material later to be appropriated as part of English-language culture in the English Charlemagne romances. The narrative of Otinel has been described as a calque of Fierabras; it can be seen rather more positively as a response to Fierabras.5 The two narratives are in some aspects mirror images of each other. They show different responses to the same concerns around baptism and conversion, conflict with the Saracen Other, and the roles of female figures in the epic tradition. For example, Fierabras is converted largely as a consequence of his defeat by Oliver, his defeat proving that ‘right is might’, and the Christian God is demonstrated to be the one true, powerful God. The Holy Spirit does have a role in ‘enlightening’ 3 M. J. Ailes, ‘From Epic to Chronicle and Back: The Pseudo-Turpin Chronicle and the Chanson de geste Fierabras’, in Thirty Years of Medieval Studies at the University of Reading, 1965–95: A Celebration, ed. Anne Curry (Reading: University of Reading, 1995), pp. 17–24; on the use Fierabras makes of the Rhymed Roland, see also Ailes, ‘Fierabras and the Chanson de Roland’. 4 ‘De ci au jor que il furent morant | En Roincevaux’ (Otinel, ed. Guessard and Michelant, 8–9). 5 Gautier, Les Épopées françaises, III, 398; for a detailed response see Ailes, Otinel: An Epic in Dialogue’.

112

Charlemagne in Medieval England

Fierabras, but comes into play only after Oliver has defeated him. The Saracen Otinel, by contrast, is converted through the action of the Holy Spirit alone, and that when he is on the point of defeating Roland. Marc Le Person presents the conversions of Fierabras and Otinel as parallels, but in fact they contrast: Otinel’s is theologically orthodox, in stressing the action of the Holy Spirit, while Fierabras’s change of faith is brought about in a very epic manner, in that the superiority of the Christian God is first demonstrated through the might of the Christian warrior.6 There is a similar interplay in the depiction of the most dominant female role in each text. The Saracen princess Floripas defies Western behavioural norms of the time, but is acceptable because her violent interventions serve to help the peers; at the same time her behaviour corresponds to (and is part of the development of) the generic convention of the belle sarrasine.7 This convention was, by the early thirteenth century, sufficiently established for the author of Otinel to offer a response to it: the Christian princess Belissant, while not unwilling to be Otinel’s bride, behaves with circumspection. She is bestowed upon him by her father, Charlemagne; this is how in Fierabras Gui de Bourgogne says he would wish to receive his bride, but he is forced to accept Floripas under the threat that she will hand the peers over to the emir if he refuses. The texts are, as it were, in dialogue, responding to the same concerns and conventions but in different ways. It is also worth noting that the only continental manuscript of Otinel, Vatican Library MS Reg. lat. 1616, also contains Fierabras. While this is a fifteenth-century compilation of two manuscripts which had previously been separate, it nonetheless suggests that on the continent, as in England, these two texts were read together at least some of the time.8 6 Fierabras: Traduction en français moderne, ed. Marc Le Person (Paris: Champion, 2012), p. 348. 7 On Floripas and the development of the genre, see Philip E. Bennett, ‘The Storming of the Otherworld, 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 in Celebration of his Seventieth Birthday, ed. Philip E. Bennett and Wolfgang G. van Emden (Reading: Société Rencesvals British Branch, 1984), pp. 1–14. 8 The two texts, by different scribes and slightly different in date, were bound together in the fifteenth century (Fierabras, ed. Le Person, p. 50).

The Anglo-Norman Tradition

113

Themes and Concerns The genre of chanson de geste as a whole is highly politically engaged and this remains true of the texts circulating in England. Reactions and responses differ from text to text, but major themes found in both continental and insular material are embedded within the narratives themselves. Indeed, there may be some indication in the narratives and the issues explored in them as to why these particular texts were so popular and so thoroughly appropriated in England. We will look at three major concerns which are explored in several texts: the Saracen– Christian conflict, the significance of relics, and the role or depiction of the monarchy.

Crusades and the Religious Other Religion – worship, prayer, conversion and crusading – lies at the heart of our Old French texts, and the observation made by both Field and Furrow, that the texts which were adapted as the Middle English Charlemagne romances had common concerns, about religious warfare in particular, is an important one.9 The Saracen–Christian conflict is inherent in all the narratives, the general shapes of which do not, on the whole, change dramatically in the Anglo-Norman versions; however, there are some specific developments. The narrative of the Chanson de Roland in all its versions has two central strands: Charlemagne’s wars against the Saracens in Spain, and the treachery of one of his own men, Ganelon. The Christian–Saracen conflict, largely a matter of physical warfare in the chanson de geste (appropriately for a ‘song of deeds’), is articulated in a more theological way in the Pseudo-Turpin Chronicle in all three languages in which it circulated in medieval England (Latin, continental and insular French, and Middle English). In the Chanson de Roland, the Pseudo-Turpin Chronicle, Fierabras and Otinel the general conflict is often particularized and represented by a kind of judicial duel which will demonstrate the superiority of the Christian God: in the Roland Charlemagne defeats and kills Baligant; in the Pseudo-Turpin Roland first combats the giant Vernagu; in Fierabras Oliver defeats Fierabras; and in Otinel 9 Field, ‘Patterns of Availability and Demand’, p. 81; Furrow, ‘Chanson de geste as Romance in England’, pp. 63–4, 71; see also Busby, Codex and Context, p. 502.

114

Charlemagne in Medieval England

Roland fights with, and is almost killed by, Otinel, who in turn, after his conversion, combats the unconverted Clarel. In comparison to the continental and Latin versions, the AngloNorman Pseudo-Turpin Chronicle by William de Briane develops the importance of the religious material and its overt didacticism.10 While the conflict between Christians and Saracens is a major focus of all versions of the Chanson de Roland, it is in the Pseudo-Turpin tradition, rather than the Chanson de Roland itself, that conversion becomes a major theme, with lengthy theological discussion between Roland and Vernagu interrupting their physical combat. However, the theme is not absent from the Chanson de Roland: the conversion of the Saracen queen Bramimonde and mass conversions at the beginning and end of the Chanson de Roland (at Cordres and Saragossa, laisses 8 and 266), when all those who will not convert are killed, raise the issue of conversion, though in a way that is uncomfortable for the modern reader.11 The role of Charlemagne himself is central to our understanding of the way in which both insular and continental texts deal with the crusades. The Oxford version of the Chanson de Roland describes Charlemagne at the outset as ‘nostre emperere magnes’. In the wider French tradition Charlemagne serves as a model for Christian unity, bringing together many peoples under one rule and purpose. The representation of Charlemagne’s men in the chanson de geste tradition from the earliest texts encompasses geographical areas which were not historically under Charlemagne’s direct rule: among his inner circle, alongside Naimes de Baviere and Richard de Normandie, we find Guillermez l’Escot, and Ogier le Danois.12 This sense that the emperor 10 The Anglo-Norman Pseudo-Turpin Chronicle, ed. Short, p. 9; Ailes and Leedham, ‘Le Pseudo-Turpin en Angleterre’; see Chapter 4. 11 For discussion of forced conversion see Marianne Ailes, The Song of Roland: On Absolutes and Relative Values, Studies in Mediaeval Literature (Lampeter: Mellen, 2002), p. 144; Wolfgang G. van Emden, La Chanson de Roland (London: Grant and Cutler, 1995), p. 91. 12 This presentation of the great emperor as the supreme secular Christian ruler in Christianity’s against the infidel is used to effect by the Norman chronicler of the Third Crusade, Ambroise, who travelled with Richard the Lionheart to the Holy Land at the end of the twelfth century. He contrasts the (supposedly) unified spirit of Charlemagne’s campaign

The Anglo-Norman Tradition

115

unites Christendom in its opposition to the Saracen threat, and is ‘ours’, is crucial to the representation of Charlemagne’s role. Similarly, in the complete Anglo-Norman redaction of Otinel, Charlemagne remains nostre emperere, with an extra reference to ‘our’ emperor where, instead of ‘Desus le pont estoit Karles le ber’ (‘The noble Charles was on the bridge’; ed. Guessard and Michelant, 742), it reads ‘Sur le pont est nostre emperere ber’ (‘Our noble emperor is on the bridge’; Bodmer MS, fol. 215ra). This text, like others in the tradition, presents Charlemagne as the leader of Christendom, just as Charlemagne is ‘our Emperor’ in the Anglo-Norman Roland manuscript because he is the secular Christian leader. The struggle between Christians and Saracens in Spain as recounted in the Roland material, both in the chanson de geste and in the Pseudo-Turpin Chronicle, can thus imaginatively stand in for the crusades in the Holy Land during the twelfth and thirteenth centuries, when England as well as France was actively involved in ventures that drew the forces of Christendom together against the common enemy.13 The narrative of Fierabras, in all its versions, continental and AngloNorman, is focused on the recovery of some relics stolen by the Saracen emir Balan from Rome when his army sacked the city. The possession of Passion relics by the Saracens is of key symbolic importance, as was acknowledged when, in crusading reality, Saladin took possession of the True Cross in 1187 at the Battle of Hattin.14 To leave such holy items in Saracen hands would be unthinkable. The crusading conflict is epitomized early in the text in a lengthy episode of single combat, taking up about one quarter of the whole text, between Oliver and Fierabras, Balan’s son. This contains echoes of Roland and Vernagu’s conflict in the Pseudo-Turpin chronicle, and of Charlemagne and Baligant’s fight in the Chanson de Roland, representing the wider battle between Saracen and Christian; Oliver defeats Fierabras, who is converted to Christianity.15 Such ‘trials by combat’ were something of a rhetorical set piece in the chanson de geste. The combat is echoed towards the end of against the Moors with that of the crusaders in his own passagium, where there was division and disloyalty between the various nationalities (The History of the Holy War, ed. Ailes and Barber, 8459–87). 13 See Tyerman, England and the Crusades. 14 Phillips, Holy Warriors, p. 130. 15 Ailes, ‘From Epic to Chronicle and Back’, pp. 17–24. It is probably at this point that the older, lost chanson de geste ended.

116

Charlemagne in Medieval England

the extant chanson de geste when Charlemagne fights with and defeats Balan; the ideological assumption in both is that ‘right is might’, so the representative of God will always win. The Anglo-Norman Fierenbras, as noted in Chapter 1, radically recasts the narrative, with considerable abbreviation, and this includes the abridgement of some of the long, sometimes formulaic, prayers. Charlemagne’s extensive intercession for Oliver during the opening combat, for example, is shortened in all the abbreviating texts. It follows a structure similar to that of a credo épique, listing God’s deeds in history, from creation to salvation, preceding the actual petition. In the Vulgate (continental) narrative of Fierabras it runs for sixty-eight lines (ed. Le Person, 1221–87).16 The Anglo-Norman Hanover MS (containing the Vulgate narrative) gives only twenty-eight lines to the prayer, and the Anglo-Norman Fierenbras has seventeen; this, however, does not indicate a decreased interest in the religious theme but is simply part of the pattern of abbreviation, where formulaic material is sacrificed to give greater narrative momentum. In the Vulgate Fierabras the long prayer is preceded by one which is more surprising, as Charlemagne threatens to attack the church, abandon the stolen Passion relics, and give up his crown if God does not spare Oliver: ‘Damledex’, che dist Karles, ‘se vos cest plaist soufrés Ke Olivier soit veincus ne mordriz ne matés, Chertes toute iert destruite seinte crestïentés; Je ne larrai en Franche ne mostiers ne autés Qu’il ne soit abatus et jus acraventés, Et larrai la coronne dont ge sui mout penés Et les dignes reliques por cui sui mout irés.’ (Fierabras, ed. Le Person, 1213–19) [‘Lord God’, said Charles, ‘If you permit That Oliver should be vanquished or defeated The whole of holy Christianity will certainly be destroyed. I will not leave in France a single church nor altar That will not be destroyed or razed to the ground,

16 Even among the continental manuscripts there is some variation. An earlier edition based on MS A has a sixty-five-line prayer (Fierabras, ed. A. Kroeber and G. Servois, APF (Paris: 1860), 1169–1233).

The Anglo-Norman Tradition

117

And I will give up my crown, which pains me so much And the worthy relics which I am so angry about.’ ]

While this motif is not unique in the chanson de geste tradition,17 it is more common to find Saracens threatening their gods, as Balan does in Fierabras, than to find Christians doing so. This prayer is completely absent in the Anglo-Norman Fierenbras. Even if this was just part of the abbreviating process, the effect of taking it out is to reduce controversiality in the presentation of religious practice while stressing the religious rather than the personal aspect of the combat. In the Vulgate text Charlemagne is, in a very human way, desiring to protect those he cares about; in Fierenbras the focus remains firmly on the need to defeat the Saracen. The Destruction de Rome in both its redactions deals with a crusading theme by bringing it to the heart of Christendom as the Saracens attack Rome.18 Morally, the stark opposition we expect between Christian and Saracen may be undermined by the fact that the Christians are the initial aggressors, as Balan attacks Rome in retaliation for an attack on one of his merchant ships. If a just war is a defensive one, then it would seem that Balan has some justice on his side. On the other hand, when he takes the battle into Christian territory and attacks Rome itself he becomes the aggressor and it becomes the duty of Christians to go to the aid of Rome. In the Egerton version, the Roman leader Savari harangues his troops, emphasizing the defensive nature of their cause and the expectation that God will give them the necessary strength (St Peter is also invoked because the city to be defended is Rome): ‘Seignurs baruns’, fait il, ‘francs chivalers, Vous savéz bien qe paiens ont la terre gastés, Penséz de vos femmes et de vos heritéz. Gardéz qe nule cowardie en vous seit trovéz.

17 Compare the pope himself railing against God in the Couronnement Louis, ed. E. Langlois, 2nd edn (Paris: Champion, 1984), 1086–9; here, however, the pope’s ‘threat’ could be read as a fear that he will not be able celebrate mass in St Peter’s church. 18 For the shorter Egerton text, see ‘La Destruction de Rome et Fierabras, MS Egerton 3028 de Musée Britannique’, ed. L. Brandin, Romania 64 (1938), 18–100. For the longer Hanover text, see La Destructioun de Rome, ed. Formisano.

118

Charlemagne in Medieval England

Chescuns se tienge od l’autre, ja n’i seit fauséz. Les paiens assaldroms od nos brancs asceréz. Gardéz que vilanie ne seit de nous parléz. Dieu et sainte Pier nous durra poestéz.’ (La Destruction de Rome, ed. Brandin, 243–50) [‘My lord barons’, he said, ‘good knights, You know well that the pagans have laid waste the land Think of your wives and your inheritance. Ensure that no cowardice is found in you. Let each stay with another, let no one falter, We will attack the pagans with our steel swords. Let nothing unworthy be said about us. God and St Peter will give us strength.’]

There is no doubt that this is a war in which defence of one’s own is closely associated with a religious conflict, where the enemy are identified as pagans and the Christians will rely on God’s strength. The narrative of Otinel is arguably even more centred on the religious conflict than that of Fierabras:19 in Otinel, the combat between the Saracen Otinel and the Christian Roland is later paralleled by a second conflict between the converted Otinel and Clarel, the pagan who refuses to convert and acts as foil to Otinel. The exchange of challenges between Otinel and Clarel is cast with different emphasis in the two manuscript texts, Anglo-Norman and continental. In the continental version Otinel has more to say, but in the Anglo-Norman one, while Clarel is given more space with which to condemn himself by his insults to Christianity, he is also credited with a rather more precise understanding of Christian practices. In the continental version he suggests that Christianity is a polytheistic religion, constructing his understanding of it as a mirror image of his own religion (itself constructed in the chanson de geste as a distorted mirror image of Christianity): Que cele loi que tu a recoillie Envers al nostre ne vaut pas une alie; Toi ne tes Diex ne valent une pie. (Otinel, ed. Guessard and Michelant, 1292–4)

19 For a summary of the narrative see Chapter 6.

The Anglo-Norman Tradition

119

[For this religion which you have acquired Is not worth a clove of garlic next to ours; You and your god are not worth a straw.]

In the Anglo-Norman manuscript, on the other hand, his assertions of the inferiority of the Christian religion are more precise and more accurate: Ke tis baptesmes ne la cristienie Ne cele messe ne prestre sacrefie Vers nostre lei ne valt un alie Melz valt Mahun ke fait le fiz Marie. (Bodmer MS, fol. 218rb) [For neither your baptism, nor Christianity Nor this mass, nor the consecrated priest Is worth a clove of garlic compared to our law. Mohammed is worth more than the son of Mary.]

A specific concern for the fate of Eastern Christendom may be perceived in the Hanover Destruction de Rome, which includes Constantinople among the territories under Saracen rule (ll. 79, 130, 207). The manuscript appears to have been copied more than 150 years before Constantinople would fall to the Turks, but possibly close to the date when the city, in the hands of Latins after the conquest of 1204, reverted to the Byzantines in 1261.20 Indeed, the loss of the city of Constantinople to the Byzantines may have seemed little better than losing it to the Saracens, so suspicious were the Latins of their Greek co-religionists; as schismatics the Greeks were arguably worse than the pagans.21 Considering the time-frame in which all these insular 20 Michael McLagan, The City of Constantinople (London: Thames and Hudson, 1968), p. 114. 21 This distrust of the Greeks is evident in texts such as the Fourth Crusade chronicles by Robert de Clari and Villehardouin. The situation did not improve as time went on. The fourteenth-century writer William Adam ‘claimed it was more important to attack the Greeks than the Moslems’, on the grounds that they had once known the truth (cited in Antony Leopold, How to Recover the Holy Land: The Crusade Proposals of the Late Thirteenth and Early Fourteenth Centuries (Aldershot: Ashgate,

120

Charlemagne in Medieval England

texts were composed and copied, it is self-evident that responses to and considerations of the themes of crusading and conversion could not be stable as the texts were recopied over a period of several centuries. The Oxford text of the Chanson de Roland is usually dated around 1100; the earliest Latin version of the Pseudo-Turpin Chronicle dates from the early twelfth century, while the Anglo-Norman version dates from the early thirteenth century, by which time copies in Latin and the continental French translation were already circulating in England.22 The continental Fierabras dates from around 1200; Otinel was composed at some time in the early thirteenth century.23 The AngloNorman remaniement of Fierabras, the text of BL MS Egerton 3028, may be as late as the 1260s (see below), while the Anglo-Norman Hanover MS of the continental text of Fierabras dates from the fourteenth century.24 Although modern historians have, until recently, focused on crusading in the twelfth and thirteenth centuries, it remained a live issue with projects being planned, if not carried out, until the fall of

2000), p. 99); for a letter from Gregory IX directly comparing Muslims, heretics and ‘schismatics’ (i.e. Orthodox Greeks), see Rebecca Rist, The Papacy and Crusading in Europe, 1198–1245 (London: Continuum, 2009), pp. 143–4. 22 Ailes and Leedham, ‘Le Pseudo-Turpin en Angleterre’. 23 The dating of Otinel has never been satisfactorily addressed. Paul Aebischer suggests the very early date of the first half of the twelfth century for an early version of the text, largely on the evidence of the names Ottonellus and Ottinelus being found in Italy in the 1170s and 1180s: Paul Aebischer, Études sur Otinel (Berne: Editions Francke, 1960), pp. 146–8. Even if we accept this insubstantial evidence, it would only point to the existence of a narrative about a character Otinel, not necessarily the chanson de geste. The earliest surviving manuscript, the Mende fragment of 293 lines in the Bibliothèque nationale de France (BnF MS NAF. 5094; http://gallica.bnf.fr/ark:/12148/btv1b6000371s) has been variously dated: Ernest Langlois proposed the early thirteenth century in his ‘Deux fragments épiques: Otinel, Aspremont’, Romania 12 (1883), 433–57 (p. 434); it is described as mid-thirteenth century by Dean and Boulton, Anglo-Norman Literature, p. 53. 24 L. Brandin, ‘Le Manuscrit de Hanovre de la Destruction de Rome et de Fierabras’, Romania 28 (1899), 489–507; La Destructioun de Rome, ed. Formisano (1981), p. 13.

The Anglo-Norman Tradition

121

Constantinople to the Turks in 1453.25 The movement would change substantially in the later medieval period. Sylvia Schein’s detailed analysis of the projects of the fourteenth century concludes that since the loss of the Holy Land in 1291 was at first considered only temporary, it initially led to a new strategy rather than to ideological changes. Indeed, Jonathan Riley-Smith commented on ‘a revival of fervour’ following the fall of Acre.26 However, according to Schein, in the longer term a new military strategy led to changes in the concept itself, as the approach necessitated an army of professional warriors; increasingly, the way ordinary pious Christians could contribute to the crusade (and thereby receive a crusading indulgence) was through financial support.27 She goes on to point out that although there was no larger-scale crusade, ‘neither aristocratic nor popular enthusiasm for the crusade faded’. 28 The central narrative of Fierabras, Otinel and the Chanson de Roland is, in each case, concerned with the struggle against the Saracens, but it seems largely the Muslims of Spain and Italy rather than those of the Holy Land.29 Here the threat is on the borders of, or even within, 25 On a rash of projected proposals and treatises in the late thirteenth and early fourteenth centuries, see Leopold, How to Recover the Holy Land; on the later crusades, see Norman Housley, The Later Crusades, 1274– 1580: From Lyons to Alcazar (Oxford: Oxford University Press, 1992); Documents on the Later Crusades, 1274–1580, ed. and trans. Norman Housley (Basingstoke: Macmillan, 1996); Sylvia Schein, Fidelis Crucis: The Papacy, the West, and the Recovery of the Holy Land, 1274–1314 (Oxford: Clarendon Press, 1991); Riley-Smith, The Crusades, pp. 245–97. 26 Ibid., p. 246. 27 Schein, Fidelis Crucis, pp. 258–68; Riley-Smith notes that ‘the successful recruitment of peasant crusaders in the fifteenth century demonstrated that popular feeling was still finding expression’, but also that the ‘more conventional crusading armies seem to have become more professional … through the employment of mercenaries and the use of contracts for service’ (Riley-Smith, The Crusades, p. 246). 28 Schein, Fidelis Crucis, p. 266. 29 On crusading in Spain, see Joseph F. O’Callaghan, Reconquest and Crusade in Medieval Spain (Philadelphia: University of Pennsylvania Press, 2003); R. A. Fletcher, ‘Reconquest and Crusade in Spain, c. 1050–1150’, in The Crusades: Essential Readings, ed. Thomas F. Madden (Oxford: Blackwell, 2002), pp. 51–67; Robert I. Burns, Moors and Crusades: Collected Studies (London: Variorum, 1978); Bernard F. Reilly,

122

Charlemagne in Medieval England

Christendom, as Rome itself is attacked and the Muslims seem to be from southern Europe (though Balan is described in the Destruction de Rome as reigning as far as the Red Sea, ed. Formisano, l. 75). It is clear in all the insular texts and insular copies of continental texts that concern about the Saracens, and fear of the threat on the very borders of Christianity, were potent issues throughout the later Middle Ages. However, the political engagement of these insular texts was not limited to this topic.

Relics and Sacred Places Crusading as an expression of religious commitment is integral to the narratives of all the insular Charlemagne texts, but it was not an activity open to all. The cult of relics, on the other hand, was a fundamental part of the expression of spirituality in the Middle Ages.30 Relics have also been associated with the reception of the chanson de geste since the birth of modern criticism, particularly in the work of the philologist Joseph Bédier.31 In some chansons de geste the relics of the heroes themselves are presented as objects of veneration. In the Oxford Roland the bodies of the heroes are returned to Blaye, but, we are told, pilgrims may see Roland’s Olifant in Bordeaux, where Charlemagne has placed it on the altar of Saint-Seurin. It is not surprising to find relics in the hilt of Roland’s sword, epitomizing the interconnection between piety and war. The Pseudo-Turpin tradition was linked from the beginning to the The Contest of Christian and Muslim Spain, 1030–1157 (Oxford: Blackwell, 1992); on the (somewhat ambivalent) reception of Charlemagne in Spain, see Charlemagne and his Legend in Early Spanish Literature and Historiography, ed. Matthew Bailey and Ryan D. Giles, Bristol Studies in Medieval Cultures (Cambridge: D. S. Brewer, 2016). 30 See Charles Freeman, Holy Bones, Holy Dust: How Relics Shaped the History of Medieval Europe (New Haven, CT: Yale University Press, 2011). 31 Joseph Bédier, Les Légendes épiques: Recherches sur la formation des chanson de geste, 4 vols (Paris, 1908–13); Bédier advocated an individualist view of the origins of the genre, arguing that the historical narratives on which many of the older chansons de geste were based were picked up in monastic libraries. On the debates about the origins of the genre, see Wolfgang G. van Emden, ‘ “ La Bataille est aduree endementres”: Traditionalism and Individualism in chanson de geste Studies’, Nottingham Medieval Studies 13 (1969), 3–26; Emden, La Chanson de Roland, pp. 11–13.

The Anglo-Norman Tradition

123

pilgrim’s visits to Santiago de Compostela and to the cult of St James, under whose instructions Pseudo-Turpin writes his account. The historic Charlemagne was known to be a collector of relics,32 and two of the chansons de geste which circulated in England had relics acquired by the emperor at the core of their narrative: indeed, some commentators have described them as a ‘relics cycle’.33 They offer different narratives of the translation of major Passion relics from the East to the West. The earlier Pèlerinage de Charlemagne (never, as noted above, translated into Middle English) presents a narrative in which Charlemagne travels to Jerusalem and Constantinople, where he is presented with gifts of relics by the emperor. In this chanson de geste the gift of relics is combined with a less than heroic episode when the peers, drunk, make various boasts, which the emperor of Constantinople then requires them to fulfil.34 It is difficult to reconcile the narrative of Le Pèlerinage with that of Fierabras: in both texts some of the relics end up in Paris, but in the Pèlerinage they are taken straight from Jerusalem, while in Fierabras they were first in Rome. It is not unknown for mutually incompatible traditions to circulate,35 but the Fierabras narrative, which presented a more consistently heroic origin for the Passion relics, had such a widespread reception across Europe that its success may account for the relative lack of success of the Pèlerinage.36 Although the myth of Charlemagne’s journey to the 32 See Freeman, Holy Bones, Holy Dust, pp. 69–77. 33 Le Person, Fierabras: Traduction, pp. 36–61. 34 For a discussion of this scene, see Le Pèlerinage, ed. Burgess, pp. xxx– xxxv; Cobby, Ambivalent Conventions, pp. 91–5, 143–5; Cobby, ‘Religious Elements’, pp. 368–71. 35 Neither narrative is compatible with the Latin tradition of the Descriptio, where Charlemagne travels to the East in response to a divinely inspired message from the emperor of Byzantium to evict pagans from the Holy Land; see Le Pèlerinage, ed. Burgess, pp. xlvi–xlvii; Latowsky, Emperor of the World, pp. 2–3. 36 The Swiss prosateur Jehan Bagnyon did bring the two narratives together. In his Histoire de Charlemagne, book 1, part 3, he describes how on a visit to Constantinople Charlemagne is given thorns from the Crown of Thorns, one of the nails which had pierced Christ, a piece of the True Cross, and the sudary (ed. Hans-Erich Keller (Geneva: Droz, 1992), pp. 24–5); it does not seem to concern him that there is a strong overlap

124

Charlemagne in Medieval England

East was well known in its Latin account, the chanson de geste seems to have had a more limited reception, surviving in the one Anglo-Norman manuscript.37 The Fierabras tradition was associated with relics of the Passion from its beginning. The earliest surviving version, the continental ‘Vulgate’ text, has a specific group of Passion relics at the core of its narrative, and all the evidence suggests that the older, lost, version of the tale was centred even more clearly on the theft and recovery of relics, taken by the Saracens from Rome and sought by Charlemagne. In the texts which survive, the recovery of the relics is closely linked to the rescue of the peers taken prisoner by Balan, the Saracen emir. Charlemagne’s power resides in both: the warrior heroes are a source of power, but possession of the relics, witnesses of God’s favour and of his special relationship with Charlemagne, is of equal importance. The continental chanson de geste as it now stands takes some of the focus away from Rome; the evidence suggests that the lost text was much more focused on the aid Charlemagne gave to Rome – indeed, the title invented for this text by Bédier, Rome perdue et reconquise, is probably a more apt description of the narrative than the title Balan used by Gaston Paris.38 The text of around 1200, however, which contains the familiar story of Floripas and her lover as well as the regaining of the relics, has a different city at its heart: Paris. Paris was by this time very firmly the main seat and power base of the king of France, Philip Augustus, and the Fierabras narrative is framed by the presence of the relics at Saint-Denis and Notre Dame in Paris. The text opens with a reference to the annual Lendit fair in Paris, as the commonplace convention of Saint-Denis as the place where a roll containing the narrative was found is linked to a more specific and more topical allusion to the fair:

with the relics rescued by Charlemagne in the Fierabras narrative, which occupies the bulk of his book (ed. Hans-Erich Keller, pp. 27–173), or that the relics are tested in a similar way in the two narratives. 37 On the Latin tradition, see Chapter 1. 38 Joseph Bédier, ‘La Composition de la chanson Fierabras’, Romania 17 (1888), 22–51 (p. 24); he also suggested the title Rome sauvée, see Bédier, Les Légendes épiques, IV, 160; Paris, Histoire poétique de Charlemagne, pp. 251–3. It is likely that Balan did not feature in the lost chanson de geste: see Ailes, ‘A Comparative Study’, pp. 26–38.

The Anglo-Norman Tradition

125

A Seint Denis en Franche fu le roules trouvez; Bien chent et chi[n]quante ans i a l’estoire estez. Or en orrez le voir, s’entendre le woulez, Si con Karles de Franche, qui tant fu redoutez, Recomquist la corronne, dont Dex fu corronez, Et les seintismes cloz, et le signe honorez, Et les autres reliques dont ill i out assez. A Seint Denis en Franche fu le tresor portez; Au perron a l’Endit fu partis et donnez Por les saintes eglises dont vous aprés orrez: Por chen est il encore li Lendis apelez. (Fierabras, ed. Le Person, 5–15) [The roll was found in Saint-Denis in France The story had been there for a good 150 years If you choose to listen you will now hear the truth How Charles of France, who was so feared Regained the crown with which God was crowned And the most holy nails and the honoured sudary And the other relics of which there were plenty. The relics were carried to Saint-Denis in France; They were distributed on the steps, at the Lendit [fair] To the holy churches about which you shall hear; For this reason it is still called the Lendit.]

The text ends with Charlemagne distributing the relics regained from the pagans, specifically to Saint-Denis and Notre Dame in Paris (ed. Le Person, 6384–94). There is no question, it seems, of returning them to Rome. Here the sacred objects are linked with the specific locus which is also the locus of power of the French kings, namely the abbey church at Saint-Denis. The relics in the narrative are not treated in a theologically sophisticated way. The first relic to affect the progress of the narrative is the balm with which Christ was embalmed; the Saracen champion Fierabras carries the containers of balm, which has healing properties, into his combat with Oliver. Fierabras, in his chivalry, offers it to the wounded Oliver, who refuses to use it unless he wins it by conquest, which, predictably, he does. He then uses it to heal himself of his

126

Charlemagne in Medieval England

wounds before disposing of it summarily by throwing it into the river.39 The indiscriminate way in which the balm works, not dependent at all on the faith of the consumer, and the irreverence with which Oliver then treats it, throwing it into the Tiber, suggest that this was not originally a religious object at all. It is treated more like Floripas’s magic girdle, which will keep the wearer from starving (and which also ends up in water when the thief wearing it is toppled into the moat). It seems to be an attempt to appropriate and ‘Christianize’ an element of non-religious magic.40 The Fierabras poet has also Christianized the motif of the white hart, which in this poem appears to lead Richard to safety across an expanse of water in response to his prière du plus grand péril. What has happened to the balm seems to be of the same order. It is integrated more securely into the narrative of the longer version of the Destruction, where it is explicitly one of the relics taken by Fierabras (ed. Formisano, 1291–5). The other relics, the Passion relics specifically associated with the Lendit fair in Paris, are treated with more reverence. They are in the safekeeping of Floripas, who brings them out for the peers to adore. The relics actively aid the peers when they are besieged in the former Saracen stronghold of Aigremore: when Naimes holds them up before the Saracens, they all fall back dazzled and blinded. There is an implicit contrast to the Saracen idols and the Saracen treasure, which are held up and then thrown out of the castle of Aigremore, the treasure causing the pagans to scrabble around to rescue it. The pagan Saracens can recognize only the false gods of worldliness and idol worship. The dissociation of the sacred objects from the specific locus with which they were associated enables further changes. The relics having been translated from Rome to Paris, the texts are now translated into a different context in England, where their connection to sites of pilgrimage in Paris is of less immediate appeal. However, the fact that the relics continue to be important in the insular versions of the narrative is evident both from the narrative versions themselves and 39 The text is ambiguous about whether the balm should be applied externally or swallowed. See Le Person, Fierabras: Traduction, p. 48. 40 Such magic in chansons de geste can be neutral, associated with neither Saracen necromancy nor Christian faith. See M. J. Ailes, ‘Faith in Fierabras’, in Charlemagne in the North, ed. Bennett et al. (London: Grant and Cutler, 1993), pp. 126–33.

The Anglo-Norman Tradition

127

Fig. 1. Fierabras takes the Passion relics from St Peter’s

from the manuscript treatments. The two surviving Anglo-Norman manuscripts (the Egerton and Hanover MSS) renew the more specific association with Rome by the presence alongside Fierabras of the ‘prequel’ narrative La Destruction de Rome. The sacking of Rome by the Saracens, recalled in a few lines at the opening of the Vulgate Fierabras, is re-elaborated. The linking of the two texts alters the textual focus, by separating the narrative of the relics from that of the peers and by moving the place of action back to Rome, which is itself a sacred place, and one with resonance across the whole of Europe. While Fierabras makes clear that the relics had been in Rome and had been stolen thence by the Saracens, the Destruction provides the narrative which leads up to that theft, but for most of the time this text’s concern is with the holy

128

Charlemagne in Medieval England

Fig. 2. Laban flees Rome with the relics

city of Rome itself. Not that the relics are unimportant: rather, they assume their importance as the narrative progresses but are not its main focus. Both manuscripts also have a thoroughly worked out pictorial scheme in which the relics have a significant place.41 Among the scenes selected for illumination in both manuscripts is the taking of the relics in the Destruction de Rome (Egerton, fol. 80r; Hanover, fol. 21r, see Fig. 1). The relics are also prominent in the depiction of the pagans returning by sea from Rome in the Egerton MS (fol. 81r, Fig. 2), and in the parallel closing image, which shows Charlemagne returning by sea, with the Cross displayed prominently (Egerton, fol. 118r, Fig. 3). The importance of the relics can also be seen in the Fierabras section of the Hanover 41 Both manuscripts have an extraordinary number of illuminations: the Egerton MS has 118 in a codex of 118 folios, including two full page miniatures, with 55 of these images in the Destruction de Rome and Fierenbras sections; the Hanover MS, which contains only the two texts, has 103 illuminations in 100 folios.

The Anglo-Norman Tradition

129

Fig. 3. Charlemagne sailing home with the relics

MS, which illustrates the scene when Naimes holds up the relics and the attacking Saracens fall back (fol. 88r; ed. Le Person, 5435; ed. Hilka, 5253). Near the end of the Hanover Fierabras there is a further depiction of Charlemagne and the peers adoring the relics (fol. 98v). The dissociation of the relics from Paris allows more flexibility in the lists of relics mentioned. The cult of saints was both local and international, and this is reflected in the cult of relics. The relics named in the continental Vulgate text are the major relics of the Passion associated with Saint-Denis and Notre Dame de Paris: the Crown of Thorns, the shroud and the nails. The exact dates when Saint-Denis

130

Charlemagne in Medieval England

acquired its various relics of the Passion are difficult to determine with certainty, though it appears to have had a nail in the twelfth century. This would fit with Fierabras and also with the Pèlerinage de Charlemagne, which similarly privileges the relics of Saint-Denis: the nail and crown in the Pèlerinage go to Saint-Denis, and the remaining relics are distributed to anonymous sites throughout the kingdom.42 Saint-Denis acquired a piece of the Cross in 1205, according to Rigord, although the relic of the Holy Cross owned by Notre Dame was already shown at the Lendit fair.43 The continental manuscripts of Fierabras make no reference to the Cross, but this is perhaps not surprising: Notre Dame was in dispute with the abbey, which claimed precedence in the Lendit procession, arguing that the origins of the fair lay with the abbey’s relics;44 the continental text would seem to support the abbey’s claim, and therefore ignores the cathedral’s most important relic. The AngloNorman manuscripts, on the other hand, include the Cross, at least in the images.45 Separating the text from the local dispute allows this important Passion relic a place in the images, at least. The dissociation of place and relics also allows the insular remanieurs to change the final resting places of the relics when they are distributed: in the Egerton Fierenbras Saint-Denis receives only the Crown of Thorns, while the 42 Anatole Frolow, La Relique de la vraie croix (Paris: Institut Français d’Etudes Byzantine, 1961), pp. 328–9, 387, 488; Bédier, ‘La Composition de la chanson de Fierabras’, p. 26; Le Person, Fierabras: Traduction, p. 55. 43 Oeuvres de Rigord et de Guillaume le Breton, Historiens de Philippe Auguste, ed. H. François Delaborde, Société de l’Histoire de France (Paris, 1882), pp. 162–3; Bédier, Les Légendes épiques, IV, 137–41. 44 Saint-Denis probably did not acquire a piece of the Cross until 1205, shortly after the composition of the extant Fierabras, but had possessed a piece of the Crown of Thorns since the ninth century; see Rohault de Fleury, Mémoire sur les instruments de la Passion (Paris: Lesort, 1870), pp. 116–17; The Encyclopedia of Medieval Pilgrimage, ed. Larissa J. Taylor et al. (Leiden: Brill, 2010), pp. 6–9, available online at http://www.academia. edu/2611937/. 45 The image of the Saracens’ return from Rome (Egerton, fol. 81r) features the Cross and lance in the boat bearing Balan; the image of Fierabras taking the relics (Egerton, fol. 80r) shows the Cross, three nails, the lance and the Crown of Thorns. The Hanover MS shows Floripas holding the Cross when the relics are held up before the pagans (fol. 88r).

The Anglo-Norman Tradition

131

nails go to ‘Nostre Dame de Boloigne’ (ed. Brandin, 1767) and the shroud goes to Notre Dame in Paris.46 Perhaps more significant is the inclusion of an additional relic in the Destruction de Rome and the Egerton Fierenbras, namely the lance. (The lance is also among the relics listed in the Fierabras narrative listened to by Robert the Bruce in the account by John Barbour.)47 Rudolf Mehnert considered this to be merely a result of misreading, since the list of the relics may be attached to an account of the Crucifixion, as at the beginning of the Hanover Destruction where the lance is mentioned in conjunction with the other instruments of the Passion (ed. Formisano, 49–52);48 nothing could be more natural than for it to be listed among the relics, particularly when the text is separated from the Lendit fair. Laura Hibbard Loomis suggests there may be a more deliberate local appropriation of the narrative, at least in the Middle English versions.49 A tradition, dating back possibly to the ninth century, tells of a gift of Passion relics from Charlemagne to the English king Athelstan. Loomis argues that this tradition lay behind the inclusion of the lance among the relics in the Middle English texts. If the lance was already known in England as a Passion relic which had been owned by Charlemagne and given to the king of England, it would not have been surprising for an insular remanieur to have added this relic to the list of those stolen from Rome in the Destruction. The Egerton Fierenbras only mentions the

46 Among the continental manuscripts MS B (BnF f. fr. 1499) abbreviates the end and lacks the distribution of the relics. The third relic is called the ‘signe’ in the Vulgate, translated as a shroud, but this meaning is rarely attested. 47 The fourteenth-century Scottish poet John Barbour summarizes the Fierabras narrative as listened to by the Scottish hero: Barbour’s Bruce, ed. Matthew P. McDiarmid and James A. C. Stevenson, 2 vols, STS, 4th series, 15 (Edinburgh, 1985), II, book 3, 436–62); the inclusion of the lance among the relics and the form of the Saracen emir’s name – Lawayne, similar to Laban – suggest that the version known to Barbour may have been close to that of the Egerton text. 48 R. Mehnert, ‘Alt und Neue Fierabras-Fragen’, Zeitschrift für romanische Philologie 60 (1940), 49–62 (p. 52). 49 Laura Hibbard Loomis, ‘The Athelstan Gift Story: Its Influence on English Chronicles and Carolingian Romances’, PMLA 67 (1952), 521–37.

132

Charlemagne in Medieval England

lance once, in the prologue, where it may have been included as part of the process of uniting the Destruction de Rome and the Fierenbras text: Ceo est del roi Charls, ke Fraunce ad a bailer, Cum il recunquist les relikes au verrai justiser, Les clowes et la corone, qe tant font a preiser Et la lance et l’ensigne, dunt sun corps fist pener. (ed. Brandin, 4–6) [It is about the king Charles who rules France About how he reconquered the relics of the True Judge: The nails and the crown, which caused so much pain, And the lance and the shroud, which so hurt his body.]

What these alterations demonstrate is a process of appropriation and integration. On transferring the tradition to England, the changes to the list of relics provide a relatively easy way to make the text more ‘at home’. These insular French-language texts do not change the treatment of the relics, but simply dissociate relic and place; the later Middle English texts would introduce more substantial changes.

Monarchy Crusade-oriented religious concerns such as those discussed above were also part of those Charlemagne narratives circulating in England that were not translated into Middle English. What distinguishes these texts from those which were translated is not the crusade, which was central to the narratives, but the attitude to monarchy and the representation of Charlemagne himself. Renaud de Montauban and other texts of the Mayence cycle portray a king who took bad advice, who acts unjustly and whose major vassals may be pushed to a rebellion which is at least partly justified. It is not that the rebellious barons had nothing to say to an English aristocracy during the later medieval period, but perhaps their specific appeal to the political elite explains why they circulated in French and not in English.50 Aspremont (the other chanson de geste that was circulating widely in England), like the Mayence cycle texts 50 It is possible that the copy of Renaud de Montauban in BL MS Royal 16 G ii was owned by Cecily Neville, wife of Richard, Duke of York, who would certainly have appreciated the political issues of the chanson de

The Anglo-Norman Tradition

133

features a noble who refuses to pay homage to the king. The refusal is not justified in the way it is in the Mayence poems, and though Girart de Fraite later contributes greatly to the defeat of the Saracens, he also responds violently to Charlemagne’s messenger, Turpin, one of his kinsmen. Girart represents the independence of Burgundy from Charlemagne. The relevance for the situation of the English, negotiating with Burgundy against France during the Hundred Years’ War, is evident and may be one reason for Aspremont’s popularity among the French-literate aristocracy in England. In the end, its message is similar to that of our core texts, that only in Christian unity can the Saracens be defeated. It may therefore be pure chance that Aspremont was not translated into Middle English. The Pèlerinage de Charlemagne, on the other hand, paints an idiosyncratic picture of Charlemagne which could at best be described as ambiguous. Here Charlemagne and his peers are seen wasting their talents, forced to fulfil outrageous boasts made when drunk. Yet at the end of the narrative Charlemagne does return to Paris triumphant, his power affirmed by the gift of relics he brings back from Constantinople, and his status as the most powerful ruler in Christendom assured.51 Charlemagne in the Oxford Roland, indeed in the wider Roland tradition, has been much studied, and the depiction of the king-emperor in this text is almost certainly behind his portrait in both Otinel and Fierabras. It is quite a complex image: Charlemagne takes counsel with his barons, as a medieval king would be expected to; he is also a warrior king who can defeat the Saracen leader yet is unable to prevent the tragedy of his nephew’s loss. Further, he is a ruler with a special relationship with God, not only seen praying to God, but with God communicating with him in dreams and directly enabling him to defeat Baligant in single combat; yet, at the end, he is very human in his sorrow and weariness when God calls him to another scene of action. geste; see Ailes, ‘Deux manuscrits de la chanson de geste de l’automne du moyen âge’, forthcoming. 51 Cobby, Ambivalent Conventions, p. 156, concludes that it is not so much Charles and the peers who are vindicated as the epic tradition itself. An extension of this reasoning would conclude that, while Charlemagne can be the butt of jokes in this text, his dignity, not as a man but as the emperor par excellence of the epic tradition, must be finally restored.

134

Charlemagne in Medieval England

In the continental Fierabras tradition Charlemagne is not always wise, but nor is he the capricious tyrant of the Mayence cycle texts: the opening argument between Charlemagne and Roland reveals a king of violent and hasty temper, who exercises little tact in dealing with his headstrong nephew; indeed, the quarrel leads to Charlemagne striking Roland, who has to be restrained from fighting back. The quarrel is provoked by the way Charlemagne had rated the prowess of his older knights above that of the young men led by Roland: all the Vulgate witnesses except the Hanover MS ascribe this to the influence of too much wine. Charlemagne is inclined to accept the advice of the wrong counsellors, being almost on the point of returning to France and leaving the imprisoned peers to their fate, when Richard de Normandie arrives just in time with a message from the peers asking for help. Charlemagne’s court is somewhat unruly, as Ganelon’s proposal to return to France without the peers provokes a fight (ed. Le Person, 4635–49), though this is brought to an end by the intervention of Fierabras and of Charlemagne himself. Perhaps most striking of all is his threat of turning away from Christianity and destroying the churches of France if Oliver is defeated in combat with Fierabras (quoted pp. 116–17). On the other hand, he is shown leading his men in battle, fighting well himself at the taking of the city and bridge of Mautrible, and finally defeating Balan, the Saracen emir, though to do so he needs the help of some of his barons (ed. le Person, 6027–32); this is, however, presented as God’s intervention. In the Anglo-Norman Fierabras tradition we find subtle but significant changes to this complex and nuanced picture of Charlemagne. In the Hanover MS his lack of tact in handling Roland is not ascribed to drunkenness; this is also the case in the Egerton Fierenbras, where the whole episode is considerably abbreviated.52 More 52 The loss of temper which leads Charlemagne to quarrel with Roland is the only ambiguous note in the presentation of Charlemagne which remains in the insular versions of Fierabras; see also Chapter 5. On this quarrel in the continental tradition, see Wolfgang G. van Emden, ‘The Reception of Roland in Some Old French Epics’, in Roland and Charlemagne in Europe, ed. Pratt, pp. 1–30 (pp. 13–15); van Emden, ‘La Réception du personnage de Roland dans quelques oeuvres plus ou moins épiques des 12e, 13e et 14e siècles’, in Aspects de l’épopée romane, ed. van Dijk and Noomen, pp. 353–62 (p. 354).

The Anglo-Norman Tradition

135

importantly, in the Egerton Fierenbras Charlemagne needs no help from Richard, Roland and Oliver to defeat Balan (ed. Brandin, 1650–63). This abbreviated text also lacks the outbreak of fighting in Charlemagne’s camp. The overall tendency is to simplify and idealize the image of the king. Otinel offers a view of Charlemagne which may again be a response to the continental Fierabras. It opens with praise of Charlemagne, who is presented as the best of a distinguished line, a positive exemplum of kingship: S’orra la flor de la geste vaillant Du fiz Pepin, le riche roi poissant. (ed. Guessard and Michelant, 3–4) [You will hear of the flower of the valiant line Of the son of Pepin, the great, powerful king.]

There is a fracas at his court, but it is caused directly and entirely by the Saracen messenger Otinel. As in Fierabras, Charlemagne observes the religious practices expected of a king, but while this is in part subverted by his startling threat of apostasy in Fierabras, in Otinel it is rather supported by his generous offering on the eve of Roland’s combat with Otinel (ed. Guessard and Michelant, 267–70). He personally stands as godfather to Otinel at his baptism, before betrothing him to his daughter.53 Charlemagne himself is less frequently seen engaged in combat; he is rather the man in charge, making decisions and allowing his barons to gain the victory, before rewarding them with lands at the end. The Anglo-Norman copy changes little of this. The core insular texts are as concerned with the depiction of the king as with the conduct of the war against the infidel, but the two themes are clearly linked. In Fierabras, Otinel and the Chanson de Roland Charlemagne as king must seek advice from his barons (though he does not always heed it), and disaster threatens when bad advice would have Charlemagne abandon both peers and relics in Fierabras. Wise counsellors do not seek peace but do seek the conversion and submission of the Saracen: thus in the Chanson de Roland, Naimes 53 Strictly speaking this constituted spiritual incest, but the poet does not seem to have been concerned by this.

136

Charlemagne in Medieval England

advises Charlemagne to accept the submission of the Saracens (though strategically, as the audience knows, it is the wrong advice). In all these narrative traditions, success over the religious Other also means more power accrues to Charlemagne. The fact that the texts which were most successful in England present a portrait of Charlemagne as a positive example of contemporary kingship suggests a conservative political reading of the texts, and this is borne out by the political exploitation of the texts indicated by the codicological context of some of the narratives in England.

The Manuscripts and their Contexts If the narratives themselves demonstrate ideological concerns, the particular redactions and their manuscript contexts suggest political appropriation of the Charlemagne material. The insular Frenchlanguage manuscripts range in date and quality from the plain twelfthcentury Bodleian Library MS Digby 23 of the Chanson de Roland54 to the sumptuous fifteenth-century BL MS Royal 15 E vi.55 Most have been subject to considerable palaeographical study, but what is known about the owners varies considerably, in part because it is as yet not possible to map Anglo-Norman manuscripts on a linguistic basis.56 The discussion of each manuscript which follows will consider codicological and other evidence for the ownership of the various manuscripts, and for their ideological or political uses.

Bodleian Library MS Digby 23 MS Digby 23 is probably the best-known manuscript in medieval French studies. It is a twelfth-century manuscript, but the precise date has 54 http://image.ox.ac.uk/show?collection=bodleian&manuscript=msdigby 23b [accessed 23/08/2016]. 55 http://www.bl.uk/catalogues/illuminatedmanuscripts/record. asp?MSID=18385&CollID=16&NStart=150506 [accessed 23/08/2016]. 56 The possibility that this could be done has been put forward by JeanPascal Pouzet, ‘Mapping Insular French Texts? Ideas for Localisation and Correlated Dialectology in Manuscript Materials of Medieval England’, in The Anglo-Norman Language and its Contexts, ed. Richard Ingham (York: York Medieval Press, 2010), pp. 102–29.

The Anglo-Norman Tradition

137

been much discussed.57 This, one of the earliest manuscripts in French literary history and one of the most important, is also one of the most modest. Short has described the parchment as ‘mediocre, sometimes poor quality, thick and somewhat coarse’, and the hand as ‘modest, informal and sometimes clumsy’, 58 a contrast to the later and relatively high-status manuscripts of the other texts. It was at some point bound with a Latin version of Plato’s Timaeus which in the thirteenth century was bequeathed to the Augustinian Abbey of Osney, Oxford, by Master Henry de Langley (a prebendary first at Henry III’s chapel in Bridgnorth Castle, who later held the prebend of Momerfield).59 Whether the Roland was already bound with the Timaeus when it was at Osney is not known for sure, but copies of chansons de geste were not uncommon in medieval insular monastic libraries (see Chapter 1). The significance of such early witnesses to the chanson de geste tradition surviving only in insular copies has been discussed in some detail above. It has been suggested that there is a fruitful analysis in reading the two texts together, both texts exemplifying ‘the important place myth or narrative occupies in the search for, and articulation of, truth’. 60 This may offer an interesting post-medieval approach, challenging modern readers to find new ways of looking at both texts. It is less instructive about medieval readings of either text. Though the manuscripts may have been together at Osney, the current binding is post-medieval and their binding together may well have been accidental.

57 The general consensus is that it probably dates to sometime before the middle of the century, but later dates have also been proposed; see Short/ Duggan, The French Corpus, I/19. 58 Ibid., I/15. 59 Ibid., I/17; the library catalogue, which lists the two parts of the manuscript separately, gives the place of origin of the Timaeus as France but it could also be England: http://manuscripts.cmrs.ucla.edu/ Manuscripts_list.php?a=return; but see Short/Duggan, The French Corpus, I/17. 60 Sarah-Jane K. Murray, ‘Plato’s Timaeus and the Song of Roland: Remarks on Oxford Bodleian MS Digby 23’, Philological Quarterly 83 (2004), 115–26.

138

Charlemagne in Medieval England

British Library MS Egerton 3028 MS Egerton 3028 brings together an abbreviated version of Wace’s Brut with the Anglo-Norman Fierabras texts, La Destruction de Rome and Fierabras. The compilation of texts, with its associated programme of illumination, suggests some connection between the rulers of England and the great emperor (rather like the juxtaposition of versions of the Pseudo-Turpin Chronicle with the chronicles of Geoffrey of Monmouth or William of Jumièges, discussed in Chapter 1). The Egerton MS is rare among early insular manuscripts of historical writing in being highly illuminated. Alison Stones describes it as ‘belonging to a special category of densely illustrated secular manuscripts made between 1250 and 1350 in England for patrons, mostly anonymous, who were particularly interested in historical, hagiographical and literary works in Latin and French’.61 A date of 1338–40 is proposed by Vernon Underwood, since the Brut is extended to the reign of Edward III and refers to Edward’s beginning his attack on France, but this is a date for the poem, for this version of the Brut,62 and no more than a terminus post quem for the manuscript. For Stones the sequence of texts – Brut, La Destruction de Rome, Fierabras – links ‘the history of England and its kings and the beginning of the Hundred Years’ War and France with the triumph of Right – Christianity over Islam – on the Continent under Charlemagne’.63 She hypothesizes further that the ‘division of Spain between Fierabras and Gui de Bourgogne may have been chosen to represent a parallel to the

61 Alison Stones, ‘The Egerton Brut and its Illustrations’, in Master Wace: A Celebration, ed. Glyn S. Burgess and Judith Weiss ( Jersey: Société jersiaise, 2006), pp. 167–76. 62 Vernon Philip Underwood, ‘An Anglo-Norman Metrical Brut of the XIVth Century’ (unpublished PhD thesis, University of London, 1937), pp. 5–6. Lines 3259–61 of the text refer to Edward the noble conqueror ‘qui are est’ [‘who now is’] and to his great war ‘encuntre le Roi de France en sa terre’ (3265); as Underwood points out (p. 5), Philip crossed the French frontier late in 1339. The final lines of the poem ask God to give him victory, suggesting that the poem was written while Edward was still on campaign. Underwood also suggests that this might be an autograph manuscript, but this is no more than speculation. 63 Stones, ‘The Egerton Brut’, p. 169.

The Anglo-Norman Tradition

139

dividing of France between the kings of France and England’;64 this came into force as a consequence of the Treaty of Brétigny in 1360 so could suggest a slightly later date for the manuscript. There are certainly some interesting parallels with contemporary politics, but perhaps the idea of a crusade in Spain, integral to the Fierabras narrative, is key. In 1329 Edward paid homage to King Philip VI of France and immediately negotiations began for a joint French–English crusade in Spain.65 Any direct link can only be speculative, but for this manuscript to link the kings of England with Charlemagne in a narrative of crusade against Spain must have had contemporary resonances at the very least. Our texts were not speaking into a vacuum, but were responding to, or rather were expressions of, a religious and ideological reality. Far from ‘ignoring the deadly rivalry between [France and England]’ in their presentation of the idealized past, as Robert Warm suggests,66 they are, in fact suggesting a solution. Charlemagne as a representation of Christian unity may well be one factor in the production of the Egerton manuscript. However, the manuscript goes beyond promoting an ideal of a Christendom united under one ruler; the association of the history of England with that of Charlemagne also promotes a consideration of the king of England as the one who will unite Christians. Looking for clues as to the possible patron of this manuscript, Stones considers some of the heraldic shields which are found in many of its 118 miniatures. It is indeed a remarkable manuscript for the extent of its heraldic decoration. A full-page miniature of Charlemagne shows him bearing the arms of France ancient: azure semé of fleurs de lis or. Above him hang two smaller shields: gules a lion rampant or and gules a lion statant guardant or – or a badly drawn lion passant (leopard?) or. Lion 64 Ibid. 65 Timothy Guard, Chivalry, Kingship and Crusade: The English Experience in the Fourteenth Century (Woodbridge: Boydell Press, 2013), p. 52, describes this as a ‘face-saving show of crusade enthusiasm’, but also acknowledges that ‘Edward II and Edward III were both sensitive to the prestige of the Iberian reconquista’ (p. 51). See also Tyerman, England and the Crusades, pp. 246–8; Tyerman considers that while ‘there is little sign that he was ever particularly eager to crusade’ (p. 246), Edward III was aware of the international prestige which would have accrued to him from supporting crusading initiatives. 66 Warm, ‘Identity, Narrative and Participation’, p. 87.

140

Charlemagne in Medieval England

coats were relatively common and a number of nobles in fourteenthcentury England bore gules a lion or, though the lion statant on a shield was rare in medieval England. Both these coats can be found in the later Middle Ages associated with mythical kings of England. The gules a lion rampant or coat was attributed to several early mythical kings in the Kings of Britain Roll (temp. Henry VI);67 by the beginning of the sixteenth century a lion statant guardant, though with colours reversed, is found attributed to Brutus himself.68 Although our manuscript is earlier than these heraldic sources, the allusion seems likely given the contents of the codex. The lion statant/passant could equally (or also) be an allusion to the arms traditionally attributed to Aquitaine: a single lion or crowned leopard, which is first depicted on coins (thus giving no evidence about the colours).69 The Black Prince was to be made ‘Prince of Aquitaine’ in 1362: this suggests a possible royal connection. Stones is no doubt correct to say that the patron seems to have an interest in French territory and the ‘desire for the retention of English holdings 67 The roll survives only in a late sixteenth-century copy, College of Arms MS Vincent 170; see Dictionary of British Arms: Medieval Ordinary, ed. Thomas Woodcock et al., 4 vols (London: Society of Antiquaries, 1992– 2014), vol. I, ed. D. H. B. Chesshyre and Thomas Woodcock (1992), p. 131. As Stones points out, the arms are also those of the earls of Arundel (Stones, ‘The Egerton Brut’, pp.  175–6); however, in the context the connection to the early kings seems more likely. 68 In Prince Arthur’s Book we find or a lion statant guardant gules quartered with Arthur’s three crowns as the arms of Brutus; see Lord Howard de Walden, Banners, Standards and Badges from a Tudor Manuscript in the College of Arms (The de Walden Library, 1904), p. 30; See Dictionary of British Arms, vol. I, ed. Chesshyre and Woodcock, p. 195; Stones (writing before the publication of the Dictionary of British Arms) tentatively links the lion passant coat to Llewelyn ap Griffyd, ‘whose arms quarterly or and gules a lion passant countercharged themselves made reference to the arms of England, gules three lions passant guardant or’ (‘The Egerton Brut’, pp. 175–6). 69 The ‘crowned leopard’ was first struck in 1355, during the reign of Edward  III; see Armorial de Gelre, ed. Michel Popoff and Michel Pastoureau (Paris: Léopard d’or, 2012), pp. 198–9; Age of Chivalry, ed. Alexander and Binski, p. 477; see also Caroline Shenton, ‘Edward III and the Symbol of the Leopard’, in Heraldry, Pageantry and Social Display, ed. Coss and Keen, pp. 69–81 (p. 76).

The Anglo-Norman Tradition

141

there in the wars with which the Egerton’s Brut continuations end’. 70 However, the heraldic evidence can be taken further to conclude that the compiler of the codex is deliberately making a link between Charlemagne and the early, legendary kings of England. If the arms used throughout the whole manuscript are considered more closely, an even greater political engagement with the English claim to the French throne may be discerned. In the Brut section, the early rulers of England, Hengist, Arthur and Edward I, bear the gules three lions passant guardant or, which was by then the established royal arms of England;71 Frolle, the Roman ruler of France, carries azure three fleurs de lis or (Fig. 4), the arms that would be formally adopted by Charles V in about 136572 (though as earlier examples of the three fleurs de lis can be found, this does not necessarily suggest a later date for MS Egerton 3028).73 Charlemagne, in the two chansons de geste in the manuscript, is depicted carrying the traditional French coat of arms, often referred to now as France ancient: azure semé of fleurs de lis or (Fig. 5). Depending on the precise date of the manuscript, this could be significant. Edward III formally adopted the arms of England quartering France in 1340, the earliest possible date for the manuscript, though recent research suggests he may have been using the two arms quartered since shortly after his accession in 1327.74 Edward used the traditional azure semé of fleurs de lis or and it may indeed have been for this reason that Charles V of France was later to adopt the modified coat of arms.75 Given that Egerton 3028 was copied around this time, it may 70 Stones, ‘The Egerton Brut’, p. 175. 71 See folios 18r, 27r, 41r, 53r, 63r. On the early history of the Royal Arms of England, see Adrian Ailes, Royal Arms of England (Reading: University of Reading, 1982). 72 Adrian Ailes, ‘Heraldry in Medieval England: Symbols of Politics and Propaganda’, in Heraldry, Pageantry and Social Display, ed. Coss and Keen, pp. 83–104 (p. 91). See W. M. Hinkle, The Fleurs de Lis of the Kings of France, 1205–1488 (Carbondale and Edwardsville: Southern Illinois University Press, 1991), pp. 42–6. 73 The heraldist William Hinkle, following the work of Max Prinet, suggests that the three fleurs de lis was common from 1228 (Hinkle, The Fleurs de Lis, p. 42). 74 Ailes, ‘Heraldry in Medieval England’, p. 89. 75 Ibid.

142

Charlemagne in Medieval England

Fig. 4. Arthur bears the three lions passant guardant of England; Frolle bears France modern: three fleurs de lis

be that giving Charlemagne the traditional French royal coat of arms would be more likely to associate him with Edward III than with the king of France; while the king of France is instead associated with Frolle, the French ruler, depicted with the modified version of the French arms, who is killed by Arthur, bearing the arms of England, in the Brut section of the manuscript. The visual link between Arthur and Charlemagne, and King Edward of England, would have been striking, particularly as Arthur was normally depicted carrying a coat of azure three crowns or.76

76 See, for example, Michel Pastoureau, Les Chevaliers de la Table Ronde (Lathuile: Editions de Gui, 2006), pp. 29–35.

The Anglo-Norman Tradition

Fig. 5. Charlemagne bearing France ancient: semé of fleurs de lis

143

144

Charlemagne in Medieval England

Moreover, Arthur not only defeats (and kills) Frolle: he then wins the whole of France which he claims as his own.77 It seems that the codex as a whole is being used to promote the claim of Edward to the throne of France by associating the enemy of the Brut section with the form of arms which would be adopted by the French kings, while Charlemagne himself, at the beginning of the paired Destruction de Rome and Fierabras, bears the arms that were being used by Edward, thus appropriating the Matter of France on the side of the English. Even at this date, the choice of French-language texts seems appropriate for a codex supporting the claim of the English king to the French crown.78 This is in keeping with the association, noted in Chapter 1, of the Pseudo-Turpin Chronicle with chronicles of the kings of England. Charlemagne, as a figure of Christian unity, thus becomes associated with Edward III, who was at this time calling for a joint crusade with France.

Hanover Niedersächsische Landesbibliothek MS IV 578 Hanover Landesbibliothek IV 578 is a composite manuscript containing two texts: Fierabras and La Destruction de Rome. The two parts of the manuscript are written in different hands and are generally considered to be of slightly different date, the Destruction dating from the end of the thirteenth century and the Fierabras from the beginning of the fourteenth century. However, the bringing together of the two parts must have been carried out at an early date, as there is a complete and coherent programme of illumination, seemingly carried out at the 77 Fierabras’s own arms, azure four lions or, Stones describes as a ‘puzzle … not solved by recourse to standard heraldic sources’ (Stones, ‘The Egerton Brut’, p. 175). The illuminator is taking his cue from the text: in the Vulgate version of the text Fierabras bears a shield of ‘Quatre lïoncheauz d’or’ (four little golden lions) with an image of Apollo ‘delés le bouglés’ (near the boss), possibly in pretence (Fierabras, ed. Le Person, 694–5); in the Egerton text Fierabras’s arms are described thus: ‘Puis ad pris l’escu od qatre lions dorréz’ (‘Then he took [his] shield with four golden lions’; La Destruction de Rome et Fierabras, ed. Brandin, 163). 78 Susan Crane comments on the appropriateness of writing in French for Chandos Herald’s Vie du Prince Noir, composed c. 1386; see Susan Crane, ‘Anglo-Norman Cultures in England, 1066–1460’, in The Cambridge History of Medieval English Literature, ed. Wallace, pp. 35–60 (p. 53).

The Anglo-Norman Tradition

145

same time throughout the whole codex; it is clearly related to the programme in the Egerton MS, but without its political import. That the source manuscripts of the Hanover MS and the Egerton MS were related is sufficient to explain the degree of parenté in the texts, and may also explain the similarities in the programme of illuminations.79 The Hanover MS again shows an image of Charlemagne bearing a coat of arms which is semé of fleurs de lis, as are his banner and tent.80 As this manuscript is slightly earlier than the Egerton MS, these arms of France are what we would expect regardless of the place of production or political bias. The difference between the two manuscripts highlights the heraldic propaganda of the Egerton MS.81 The illuminations of the Egerton MS are similar in technique to those of the Hanover Destruction, with the illuminations in plain red frames, though walls, banners and other features may spill out over the frame. Although Stones points out a major difference in the Fierabras section of the Hanover MS, in that there is no solid colour in the illuminations of this section, she is unconvinced that there is actually a change of artist, as was proposed by

79 Ailes, ‘A Comparative study’, pp. 270–83. 80 Folios 26r, 27r, 28v, 81r, 80r, 83r, 84r, 94r, 99r. Here it is Richard de Normandie who bears a coat which refers to the royal coat of arms: a chevron cotised between three fleurs de lis; this demonstrates that the illuminator understands how heraldry functions, but does not appear to reflect any medieval English coat of arms. See Dictionary of British Arms, II, 495; Gui de Bourgogne’s coat also includes fleurs de lis: on a fess three etoiles, between fleurs de lis; Balan’s coat is a parody of Richard’s: a chevron between three lions’ heads. For a description of the Hanover MS, see Härtel and Ekowski, Handschriften, II, 160–4. 81 Fierabras in the texts of the Hanover MS bears the same coat as in the Egerton MS – four lions – though the inhabited initial S at the beginning of the Destruction (fol. 1r.) portrays a knight holding a shield with an ugly Saracen head depicted on it, a possible reference to the Apollo’s head in the description of the coat of arms in the Vulgate text: ‘A son cole ad pendu son fort escu … | Quatre lions d’or i avoit paynturez: | L’image dAppolyn estoit sure le bouclez’ (Chanson de Fierabras, ed. Alfons Hilka and André de Mandach (Neuchâtel: Pré-publication de Neuchâtel, 1981), ll. 666–8); compare the Hanover Destructioun, ed. Formisano: ‘En son escu avoit .iiii. lions panturee’ (1230).

146

Charlemagne in Medieval England

Louis Brandin, though there is a change of scribe.82 Both manuscripts use a line drawing and colour wash technique. Stones describes the artistic quality of the manuscripts as ‘not very high’, and suggests this is why the manuscripts have received little attention from art historians.83 Yet the Hanover MS in particular has a verve and energy in the illuminations, with a strong sense of movement conveyed by the swing of the horse caparisons and angles of human bodies.84

British Library MS Royal 15 E vi (The Shrewsbury Book) At least a century later, heraldry can also be seen to be part of the propaganda of the Shrewsbury Book; its celebrated genealogy of the royal couple and extensive fleurs de lis decoration was overtly political in its promotion of the dual monarchy of France and England. The Book of Charlemagne had a major place in the manuscript, and the iconography of Charlemagne as a dual monarch, both French and Germanic, Holy Roman Emperor and king of France, again fitted that political aim. While it might seem natural to present a French queen of England with a codex which included tales of the great emperor, as discussed above, Charlemagne was also associated with the English monarchy, so it is also not surprising that it is the emperor with European reach who has his place in this manuscript, rather than the other king of medieval legend, Arthur of the Britons. The Shrewsbury Book was presented to Queen Margaret, bride of King Henry VI, by John Talbot, Earl of Shrewsbury. His political concern was to promote the dual monarchy, the king of England as 82 L. Brandin, ‘Le Manuscrit de Hanovre’, p. 493; Stones, ‘The Egerton Brut’, p. 170. 83 They are omitted from the otherwise comprehensive survey of manuscripts illuminated in the British Isles: Lucy Freeman Sandler, Gothic Manuscripts, 1285–1385, Survey of Manuscripts Illuminated in the British Isles, 2 vols (London and Oxford: Harvey Miller and Oxford University Press, 1986); Stones points out this omission in her review, Speculum 66 (1991), 624–8. 84 The term romance in the explicit: ‘Ici est le finement del romance de Fierenbrace d’Alisandre’, refers to the language in which the text was written, not to its genre, but a post-medieval reader, perhaps in the seventeenth century, replaced the word ‘romance’ by ‘L’estoire’; Le Person notes the correction in his edition, Fierabras, p. 39.

The Anglo-Norman Tradition

147

monarch of both England and France. The genealogical tree at the beginning of the codex is dominated by a background of fleurs de lis and the codex consistently presents Charlemagne as both king of France and emperor; he is attired in a blue robe scattered with fleurs de lis (fols 25r, 43r, 70r, 155r) and at the same time wears an imperial closed crown (fols 25r, 43r, 70r, 155r, 176v, 181v, 182r). This dual role is vividly brought home by the heraldry of his shield, which impales the fleurs de lis of France with the double-headed imperial eagle (fol. 25r).85 The iconography of Charlemagne thus presents him both as king of France and as emperor, at once belonging to the French and, by association, to the English claiming France, and to a larger polity, the empire. This representation of Charlemagne bearing both arms is found in a number of manuscripts and can be traced back to the late thirteenth century.86 85 Bailey, ‘Fierabras and the Livre de Charlemaine’, discusses the implications of this impaled coat in the context of Talbot’s political views; this was also discussed by Bailey in ‘The Portrayal of Charlemagne in Text and Image in British Library MS Royal 15 E vi’, a paper given at Leeds International Medieval Congress, July 2012. 86 It can be traced to the Enfances Ogier, particularly as found in MS 3142 of the Arsenal Library, an illustrated manuscript of the works attributed to the late thirteenth-century poet Adenet le roi; see Louis Carolus-Barré and Paul Adam-Even, ‘Les Armes de Charlemagne dans l’héraldique et l’iconographie médiévales’, in Mémorial d’un voyage en Rhenanie de la Société nationale des antiquaires de France (Paris, 1953), pp. 289–308; Gérard Brault, ‘Adenet le roi et l’héraldique médiévale’, Olifant 25 (2006), 141–9. Both text and image give Charlemagne dimidiated arms: ‘Armes parties d’or et d’azure portoit, | Dedenz l’azur fleurs de lis d’or avoit | Et demi aigle noire sur l’or seoit’ (‘he carried arms divided, on the one side azure and on the other or, with golden fleurs de lis on the azure and half a black eagle on the or’). See Les Enfances Ogier, ed. Albert Henry, Les Oeuvres d’Adenet le roi, III (Bruges: Rijksuniversiteit te Gent, 1956), 5004–6, quoted in Gérard J. Brault, ‘Adenet le Roi and the Coats of Arms of Charlemagne and his Twelve Peers in a Medieval Roll’, in ‘Contez me tout’: Mélanges de langue et littérature médiévales offerts à Herman Braet, ed. Catherine Bel, Pascale Dumont and Frank Willaert, La république des lettres 28 (Louvain: Peeters, 2006), pp. 115–28 (p. 117). For examples of this dimidiated arms, see Brussels, Royal Library MS 9243, the Chronique de Hainault, c. 1438; see fol. 214r, reproduced in Rita Lejeune and Jacques Stiennon, La Légende de Roland dans l’art du moyen âge, 2 vols (Brussels:

148

Charlemagne in Medieval England

While this double coat of arms was not thereafter universally attributed to the king-emperor, French manuscripts often did use the impaled coat of arms, though some show him with only the fleurs de lis.87 Indeed, although MS Royal 15 E vi shows him with the impaled coat of arms in the one illumination in which we see him with a shield, the predominant heraldic symbol of the manuscript is the fleurs de lis. MS Royal 15 E vi was compiled for a particular political context and for use at the English royal court. It would seem that the dual role of the emperor reflects the king of England’s claim to a dual monarchy, and that John Talbot is strongly asserting his continued belief in that dual monarchy in the codex presented to the queen of England.88

Arcade, 1966), p. 315. Folio 125 has an image of the Nine Worthies with Charlemagne bearing an impaled shield: the image is reproduced by Carolus-Barré and Adam-Even, ‘Les Armes de Charlemagne’, p. 290. When the double-headed eagle and the fleurs de lis of France are found together on the same shield they are variously dimidiated, impaled or quarterly, though the last is less common; Carolus-Barré and Adam-Even give three images based on fifteenth-century armorials: Armorial Van der Ersten shows the two coats of arms dimidiated; Armorial MontjoieChandon shows them impaled; Armorial normand shows them quarterly. In BL MS Royal 15 E vi they are impaled. The manuscript was probably copied some fifty to a hundred years after the composition of the chanson de geste (Carolus-Barré and Adam-Even, ‘Les Armes de Charlemagne’, p. 291). 87 Some manuscripts depict him with the fleurs de lis alone; e.g. Jean Mansel’s La Fleur des histoires, Brussels, Royal Library MS 9232, also Copenhagen, Royal Library Thott 568; see Lejeune and Stiennon, La Légende de Roland, I, 328–30, II, plates 387 and 388; the fifteenth-century Berne Burgerbibliothek manuscript of the same text shows him carrying the fleurs de lis and the double-headed eagle. The most famous depiction which shows Roland with the eagle is the statue of Bremen; on the statues, see Lejeune and Stiennon, La Légende de Roland, I, 354–63, plate 58. 88 Bailey, ‘Fierabras and the Livre de Charlemaine’; André de Mandach, ‘A Royal Wedding-Present in the Making: Talbot’s Chivalric Anthology (Royal 15 E vi) for Queen Margaret of Anjou, and the Laval-Middleton Anthology of Nottingham’, Nottingham Mediaeval Studies 18 (1974), pp. 56–76.

The Anglo-Norman Tradition

149

Louvain University MS G 171 (The Didot MS) The Didot MS, which was destroyed in a fire in 1940, dated from the thirteenth century, and was described as being apparently of no better than medium quality,89 with a mise-en-page typical of a thirteenthcentury manuscript: single column, the first letter of each line set aside, and initials alternating in red and blue at the beginning of each laisse.90 It seems that the manuscript was acephalous and that the text of Fierabras (fols 53–130) followed a copy of Boeve de Haumtone (fols 1–52). This coupling of two texts in chanson de geste form accords with the insular tradition, which seems to have preferred manuscripts of paired texts to the cyclical manuscripts that developed on the continent. Cologny Geneva, Bodmer Library MS 168 Otinel, which as noted above can be considered as a text in dialogue with Fierabras,91 is found in company with Fierabras in its only complete continental manuscript, Vatican MS Reg. lat. 1616; the manuscript as it survives is a composite codex, in which a copy of Fierabras, dated 1317 and now incomplete, was bound together with a fourteenth-century copy of Otinel.92 In the insular tradition, however, Otinel is found in different company. There is one more or less complete Anglo-Norman manuscript, MS Bodmer 168, and two fragments.93 In the Bodmer 89 It was described in the Didot sale catalogue as a ‘jongleur’ manuscript (Fierabras, ed. Le Person, p. 27). 90 Busby, Codex and Context, p. 24, suggests that double column is more common in the thirteenth century. 91 Ailes, ‘Otinel: An Epic in Dialogue’. 92 Le Person accounts for the compilation as follows: ‘En fait ces deux poèmes ont été ajoutés au XVe siècle à un ancien recueil qui groupait des œuvres en latin’ (Fierabras, ed. Le Person, p. 50). He points out that the two chanson de geste manuscripts were misbound, with a section of Fierabras appearing in the middle of Otinel (p. 51); the mistake was presumably not noticed because, as he observes, both texts have the same mise-en-page. Busby notes that Otinel was copied in two fourteenthcentury hands similar to that of Fierabras, localized to Brittany and discussed alongside Anglo-Norman manuscripts (Codex and Context, p. 508). 93 Dean and Boulton, Anglo-Norman Literature, p. 53.

150

Charlemagne in Medieval England

MS, which dates from the late thirteenth century or early fourteenth century,94 Otinel is accompanied by two insular romances: Waldef and Gui de Warewic. It is rare in continental manuscripts to find chansons de geste with romance texts, but these are romances with an epic flavour. The manuscript is copied by one hand. There are no narrative links between the three texts, but there are thematic links, as all three texts include scenes of conflict between Christian and Saracen, though this theme is more central to Otinel than to the other two texts: the crusading theme is significant only in the second part of Gui, and in Waldef could be seen as just another source of adventure, and in particular of challenge and troubles.95 Possibly of more significance, given the links seen in the other manuscripts between the continental Charlemagne and the royal house of England, is the implicit association of insular heroes with the heroes of the Charlemagne texts. The narratives of English heroes, a group 94 The date of this manuscript has been much discussed, with earlier scholars suggesting an early (if vague) date of the twelfth century: C. Sachs, Beiträge zur Kunde altfranzösischer, englischer und provenzalischer Literatur aus französischen und englischen Bibliotheken (Berlin, 1875), p. 50; Hermann Sachier, ‘Beschreibung der Cheltenhamer Handschrift 8345’, in Miscellanea di studi critici in onore di Vicenzo Crescini (Cividale: Stagni, 1927), pp. 315–25. More recently scholars have supported a late thirteenthor early fourteenth-century date: Vielliard, Manuscrits français du moyen âge, pp. 93–9 (a photograph of Otinel, fol. 211r, is reproduced in plate 8); Busby, Codex and Context, pp. 502–3; Dean and Boulton, Catalogue of Anglo-Norman Texts, p. 53. These dates have been reviewed recently by Jean-Baptiste Camps, who noted that it is rare to find text writing on the top line of manuscript, as is the case here, after the middle of the thirteenth century, though the hand does suggest the later date (Vielliard, Manuscrits français du moyen âge, p. 93). 95 M. J. Ailes, ‘Gui de Warewic in its Manuscript Context’, in Guy of Warwick, ed. Wiggins and Field, pp. 12–26; Busby also sees the ‘presence of the epic Christian–Saracen opposition’ in the romances as authorizing ‘their codicological conjuncture’ (Codex and Context, p. 502; an image of the manuscript is found on p. 503); Camps has suggested that Otinel might be something of an after-thought, as it does not begin on a new gathering but follows on from the end of Gui de Warewic and is, especially at the end, an abbreviated text (‘Otinel et l’Europe’, n. 36), but we are not convinced by this argument.

The Anglo-Norman Tradition

151

which modern criticism has labelled ‘ancestral romances’ and which includes both Waldef and Gui de Warewic, would appear to be more concerned with baronial pride than anything else, but here perhaps the modern label is misleading.96 The ‘ancestral’ link with aristocratic families of England may have developed in the reception of these texts, in their appropriation by families such as the Earls of Warwick, rather than springing from the original patrons. They also deal with broader matters, and bringing all three texts together in one manuscript associates the insular heroes with the crusading ethos of the chanson de geste97 while setting the story of Otinel in the context of two stories featuring ‘historical’ English heroes: as Rosalind Field argues, Waldef shows ‘the extensive, pervasive influence of Wace’s Brut’, and is itself the model for Gui de Warewic.98 Thus the manuscript context in the Bodmer MS creates something similar to the combination of Charlemagne texts and English history in the Egerton MS. This does not seem to be a random collection of narratives. The manuscript bears signatures of owners in three fourteenthcentury hands: fol. 6 ‘Jane Grey’; fol. 207v ‘Anne Elhy’; and fol. 206v ‘Anne Rey’. Vielliard suggests this links the manuscript to the Gray family, an English family of bibliophiles.99 What is certainly clear is that the readers of this tale of crusading violence included women: indeed, English women in particular seem to have been major patrons of Frenchlanguage literature, including chansons de geste.100 96 Susan Dannenbaum, ‘Anglo-Norman Romances of English Heroes: “ancestral romances”?’, Romance Philology 35 (1981/2), 601–8. 97 A similar association is effected by the use of the chanson de geste form for the non-Charlemagne texts the Roman de Horn and Boeve de Haumtone. 98 Rosalind Field, ‘Waldef and the Matter of/with England’, in Medieval Insular Romance: Translation and Innovation, ed. Judith Elizabeth Weiss, Jennifer Fellows and Morgan Dickson (Cambridge: D. S. Brewer, 2000), pp. 25–39 (p. 29). Field notes the aspect of ‘penitential romance’ in both Waldef and Gui (p. 38), and this may be linked to the conversion of Otinel as a connecting thread between all three texts. 99 Vielliard, Manuscrits français du moyen âge, p. 94; Busby (Codex and Context, p. 734) suggests that the signatures are slightly later than the fourteenth-century date ascribed them by Vielliard. 100 On women as patrons, book owners and readers, see Jocelyn WoganBrowne, ‘“Ceste livre lisez … chescyun jour”: Women and Reading,

152

Charlemagne in Medieval England

The Otinel Fragments There is little evidence regarding the original manuscript context or owners of the Otinel fragments. The shorter of the two fragments presents as an after-thought, a jotting down of a point of interest at the end of a manuscript. It consists of only four lines at the top of a blank column at the end of BnF MS fr. 25408, a manuscript of didactic texts, formerly in the library of Notre Dame de Paris. The manuscript is dated 1267 and the hand of the lines of Otinel is judged to be of a similar date.101 We can only surmise that these few lines of Roland’s défi to the pagan Otinel caught the interest of an owner. The defiance of the Christian to the pagan may have been deemed suitably didactic. That Otinel is found in company with another chanson de geste, Aspremont, in the longer Mende fragment, more certainly of Anglo-Norman provenance, is not surprising, as we have noted the insular preference for pairing epic texts and both these texts were popular in England. British Library, MS Arundel 220 MS Arundel 220 is described in the British Library catalogue as a ‘miscellany of mostly historical texts’. 102 This is not a chanson de geste manuscript but one in which a Charlemagne text which purports to be a prose chronicle is found in a manuscript with more miscellaneous material. The contents include a number of historiographical texts in Latin, including Geoffrey of Monmouth’s Historia Britonum, c. 1230–1430’, in Language and Culture in Medieval Britain, ed. WoganBrowne et al., pp. 239–53; Carol Meale, ‘“… alle the bokes that I haue of latyn, English and frensch”: Laywomen and their Books in Late Medieval England’, in Women and Literature in Britain, 1150–1500, ed. Carol M. Meale, 2nd edn (Cambridge: Cambridge University Press, 1996), pp. 128–58. Priscilla Bawcutt demonstrates the difficulties in making sense of the (sometimes) scanty evidence in ‘“ My bright buke”: Women and their Books in Medieval and Renaissance Scotland’, in Medieval Women: Texts and Contexts in Late Medieval Britain, ed. Jocelyn Wogan-Browne et al. (Turnhout: Brepols, 2000), pp. 17–34. 101 Camps, ‘Otinel et l’Europe’, n. 5; for a description of the manuscript, see http://gallica.bnf.fr/Search?ArianeWireIndex=index&p=1&lang=EN&f_ typedoc=manuscrits&q=fr+25408&x=0&y=0. 102 http://www.bl.uk/catalogues/illuminatedmanuscripts/record. asp?MSID=1697&CollID=20&NStart=220 [accessed 23/08/2016].

The Anglo-Norman Tradition

153

Trivet’s Annales regni Regis Edwardi I, and John Pike’s Historia Regum Britannorum, Saxonum, Danorum, et Normannorum.103 The vernacular texts are: the Prologue to the Prophecies of Merlin, in French (fols 4, 5); the Anglo-Norman Pseudo-Turpin Chronicle (fols 284–98v); Walter de Bibbesworth’s Tretiz de langage (fols 299–305v); satires and verses in French verse (fols 305v–307). Most of the Latin texts were copied by one scribe; the scribe who was responsible for the Anglo-Norman texts also copied a Latin account of the claims of the English kings to the Scottish crown (fols 278–82) and Latin marginal notes added to the Easter table (fols 315–28); according to Short, these were intercalated ‘before and after the already existing Latin texts’, so the manuscript developed in at least two stages.104 Despite the range here there is some specific association: most of the texts in the manuscript are Latin chronicles, and the Pseudo-Turpin is added as a vernacular chronicle with a text that supports claims of the English king to another crown. Again, a text which holds up the Emperor Charlemagne as an example is associated with historical texts that relate to the English monarchy. In its final incarnation this is a bilingual manuscript, though it was not originally conceived as such.105 Short deduces from the entries to the Easter Table that ‘the second part of our manuscript could have been copied by a monk attached to the Cathedral Priory of Norwich during the third decade of the fourteenth century’, and further points out that the date of the copying would be consistent with ‘the period of active restocking of the cathedral library which followed its destruction by fire in 1272’.106 A link between the family of the original patron, Warin Fitzgerold, and the Despensers may account for the text’s being in Norwich. Henry Despenser was bishop of Norwich between 1308 and

103 Catalogue of Manuscripts in the British Museum, n.s., vol. I, part 1: The Arundel Manuscripts (London, 1834), pp. 62–3. 104 The Anglo-Norman Pseudo-Turpin Chronicle, ed. Short, p. 10. 105 There is also a small amount of English in the codex, but it is essentially French and Latin. 106 The Anglo-Norman Pseudo-Turpin Chronicle, ed. Short, p. 11; see N. R. Ker, ‘Medieval Manuscripts from Norwich Cathedral Priory’, Transactions of the Cambridge Bibliographical Society 1 (1953), 1–28 (pp. 5–6); Ker notes that the monks hired scribes to copy books during the period 1272–1325 (p. 8).

154

Charlemagne in Medieval England

1319.107 Moreover, Henry himself led a military expedition to Flanders, probably at the instigation of the pope, which could therefore be considered a ‘crusade’.108 This prelate’s interest in a text which valorizes the archetypal warrior-bishop Turpin would not be surprising, so the bishop may well have owned a copy of William de Briane’s translation. The assumed date of copying, on the other hand, in the second quarter of the fourteenth century does not fit with Henry as the owner of the manuscript from which it was copied. If, however, we assume that the vernacular texts may have been copied later, as Short suggests, then these texts could well have come from the household of the worldly prelate. One might wonder at the taste that could include light-hearted verse such as the lyric ‘nul ne peot a touz plere’, one of the assorted French poems (fol. 305), along with, indeed immediately following, Walter of Bibbesworth’s Tretiz de langage (fols 299a–305c), and in conjunction with the pious, crusading Pseudo-Turpin Chronicle. The short vernacular poems which end the manuscript seem to be partly a matter of filling the pages with, literally, miscellaneous matter. The joining together of the Latin and vernacular texts would not create a more bizarre juxtaposition than was already found among the French texts. Rather, the linking of Charlemagne with historical texts relating insular history is only what we have already observed in monastic libraries (see Chapter 1) and in Egerton 3028. It seems that Charlemagne was not considered at all out of place in association with the kings of England in the sense of identity constructed by insular manuscripts.

Conclusions It is clear that in the thirteenth century texts moved freely between England and France, as we would expect. The manuscripts copied or 107 The Anglo-Norman Pseudo-Turpin Chronicle, ed. Short, p. 11. Hugh Despenser had the guardianship of a descendant of Warin Fitzgerold’s nephew sometime between 1308 and 1318. Hugh’s grandson was bishop of Norwich from 1370 to 1404; see R. G. Davies, ‘Despenser, Henry (d. 1406)’, Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004), also online at http://www.oxforddnb.com [accessed 23/08/2016]. 108 N. Housley, ‘The Bishop of Norwich’s Crusade, May 1383’, History Today 33/5 (1983), 15–20.

The Anglo-Norman Tradition

155

used in England varied from the minimally decorated and relatively inexpensive, to the most sumptuous. While we may only be able to speculate about the purpose of the Anglo-Norman adaptations and copies, the major themes – matters of religion, the unity of Christendom, fear of the religious Other, and a positive exemplum of kingship – are important in all of these texts. Moreover, we should not underestimate the power of narrative, and the narratives of Fierabras, Otinel, and the mythical events of Roncevaux, related in both the Chanson de Roland and The Pseudo-Turpin Chronicle, made good stories that were popular across Europe. With violence and religion and a strong element of a love story, all the ingredients of a successful tale were present. We can also see signs of appropriation for a new context in the treatment of some of these matters, particularly the dissociation of the relics from the Lendit fair in Paris, a renewed focus on Rome, and an awareness of vulnerability to attack. While the purpose for the ‘translation’ to the west of the Matter of France may be a matter for speculation, the political and ideological uses of some of the manuscripts seem clear. Given that the genre is politically engaged from the time of its first written texts, these uses demonstrate both continuity with the continental tradition and an insular appropriation of it.

3 Charlemagne ‘Appropriated’: The Middle English Tradition Adapting the Anglo-Norman Inheritance: Themes and Concerns

T

he enhanced emphasis on religious warfare discerned in the insular French-language tradition is decidedly to the fore in the English Charlemagne romances, which often take the opportunity to sharpen the confrontation between two opposed faiths. In the Middle English Song of Roland, for example, Roland’s first encounter on the battlefield, where he kills Amaris,1 is clearly figured as a stereotypical combat between Christian and heathen, by contrast with the French tradition, where the issue at this point is exclusively the honour of France and Charlemagne. In the English version, Amaris challenges Roland as representative both of Charlemagne and of Christianity: ‘wher art thou, Roulond, leder of charles? | thy lay is fals, and also thy lordes’ (655–56). Having felled ‘that fals kinge’, Roland responds: ‘thy soule … to satanas I beteche! | thou shalt neuer greve man þat to god will seche’ (663–64). Similarly, whereas the Egerton Destruction de Rome has the Roman leader Savari encourage his troops with conventional thoughts of their wives and heirs and the fear of shame (245–9), in the Sowdone of Babylone Duke Savaryz’s speech focuses instead on the battle as a proxy war between the Christian and Saracen gods: Thenke yat Criste is more myghty Than here fals goddis alle; And he shal geve vs the victorie, And foule shal hem this day bifalle. (196–9)

The earliest of the Middle English adaptations of Otinel, the

1 Amaris’s role corresponds to that of Aelroth at this point in the French chanson de geste (Oxford version, laisse 93).

156

The Middle English Tradition

157

Auchinleck Otuel, specifically represents the climactic single combat between the converted Otuel and the Saracen Clarel as a trial of strength between their respective deities (1265–70). In fact, the prologues in all three Otuel texts (Otuel, Otuel and Roland, and Roland and Otuel) stress the conflict between heathen and Christian as their central concern, whereas in the French Otinel, in both manuscript versions, the prologue ends with the promise of a hitherto untold adventure of renowned Charlemagne.2 The English poems promise ‘bolde batailles … Þat was sumtime bitwene | Cristine men & sarrazins kene’ (Otuel, 4–6), tales of the doughty douzeperes ‘Þat wele couthe feghte with a Saraȝene’ (Roland and Otuel, 17), and stories of Roland’s achievements defeating infidel opponents (Otuel and Roland, 15–17, 24–5). Indeed, the series of added ‘chapter headings’ that subdivide the text in this last romance explicitly introduce battles between Christians and Saracens. The texts tend to emphasize the aggression of the Saracen sultan and attribute to him genocidal intentions towards all Christians, providing an unequivocal rationale for Charles’s wars as a defence of Christendom, and not specifically, as in Otinel, of ‘douce France’ (153). In the Ashmole Sir Ferumbras, unique additions serve to exaggerate the anti-Christian hostility of the Saracen champion and to highlight the explicitly Saracen and Christian identities of the two opponents, Fyrumbras and Oliver (see pp. 273, 278). Similarly, in the Fillingham Firumbras, the peers are identified as Christian knights rather than as French barons, and the reader or listener is conscripted into supporting them by a network of added or elaborated meta-narrative material, commenting on God’s guidance and grace towards the Christian forces, and exhorting readers and listeners to become personally involved in the story (see p. 297). In all the Middle English texts there is strong evidence of the insular preference for representing Saracen– Christian conflict through the narrative motif of single combat between individuals championing the opposed faiths: instances are preserved

2 ‘Qu’avint à Kalle, que Dex parama tant | Qu’il fist miracles por lui en son vivant’ (18–19; ‘that happened to Charles, whom God loved so much that He did miracles for him in his lifetime’), Otinel, ed. Guessard and Michelant; variant in Bodmer MS 168, fol. 211r: ‘k avint a charlemaine si subitement’.

158

Charlemagne in Medieval England

from the source narratives and in some cases elaborated,3 and episodes on the same model are created in the ‘new’ Matter of France romances of the Sege of Melayne and Rauf Coilyear (see pp. 385–6 and pp. 397–9). The interest in the topic of Saracen conversion, evident in all the insular Charlemagne romances, goes beyond the mere fact of selecting the stories of Fierabras and Otinel. In the Sowdone of Babylone, for example, the conversion of Ferumbras is given added significance: the claim that Ferumbras was known after his death as St Florence of Rome, not present in the closely related AN Egerton MS, is imported from the Vulgate Fierabras,4 here elaborated with a miniature hagiographic life: God for him many miracles shewed, So holy a man he bycame; That witnessith both lerned and lewde, The fame of him so ranne. (1487–90)

The story of the hostile persecutor of Christians who experiences sudden conversion and becomes a saintly champion of Christ is of course a familiar one, echoing the narrative of the Conversion of St Paul, told in Acts. It may be that the Middle English adaptations of the Fierabras and Otinel stories exaggerated the anti-Christian hostility of the Saracen challengers at least in part to create a more emphatic and easily recognizable instance of the well-known model of Pauline conversion, presented in the Golden Legend as an encouragement to repentance, for ‘no sinner, no matter how grievous his sin, can despair of pardon when he sees that Paul, whose fault was so great, afterwards became so much greater in grace.’5 Each of the Charlemagne romances in the Fillingham MS uniquely 3 See, for instance, the added marginal lines in the Ashmole MS (pp. 274–5). 4 Rome is substituted for the place named in the French text, which has ‘seint Florens de Roie’ (Fierabras, ed. Le Person, 1948–9). The reference to ‘Florent’ as Fyrumbras’s baptismal Christian name occurs in the Ashmole Sir Ferumbras, but his future sainthood is not mentioned. 5 Jacobus de Voragine, The Golden Legend: Readings on the Saints, trans. William Granger Ryan, 2 vols (Princeton, NJ: Princeton University Press, 1993), I, 119.

The Middle English Tradition

159

adds a conversion at the end of the story that echoes the central episode of the Saracen hero’s conversion: in Firumbras, all Floripas’s maidens are christened, adding to the impact of the scene in which the heroine is herself baptized; in Otuel and Roland, the Saracen king Garcy (unlike his counterpart in all other versions, French or English) opts to convert and is given instruction and baptism by Bishop Turpin. This extra emphasis on Christian conversion is equally, though differently, conveyed in the Ashmole Sir Ferumbras, when Charlemagne offers baptism to the captured Saracen emir, Balan. Here there is extended and more explicit religious instruction than in the Old French Fierabras, taking Balan article by article through a paraphrase of the Apostles’ Creed and exhorting him to believe.6 Ac arst þou most forsake Mahone And belyue on gode sone þat in marye y-kened was; & suþþe of hure body y-bore, Wyþoute wem & wyþoute hore, As sunne goþ þorȝ þe glas. Þow most also belyue thus: þat he suffrede deþ for vs, To kepe ous fram helle pyne, & þat he aros þe þridde day And to helle tok þo þe way, And delyuerede þar is hyne; And þat he þerafter to heuene steȝ & syt on ys fader riȝt hond an heȝ, & ys in trynytee: And suþþe sente þe holy gost To ys decyples he louede most, And het men y-fulled bee. Þou most byleue on holychurche, After hure lawe forto wurche, And on forȝyft of synne;

6 It is notable that while the Middle English romances extend credal material in narrative instances of this kind, they typically omit the ‘epic credo’ uttered in different circumstances by the heroes in chansons de geste.

160

Charlemagne in Medieval England

And on þe dredful domesday Wan ech man schal rysen on such aray As he dayeþ ynne, And come before god present, And fonge ther ys iuggyment, to ioye oþer pyne to wende, After þat þat he doþ here, Body & saule to dwelle yfere, Euere wyþouten ende. Bylyf þou as y haue þe tauȝt. (5723–43)

Nevertheless, Balan remains obdurate and in the end Charlemagne still has him beheaded. Similarly, in the episode in Otuel and Roland when the now converted Otuel begs Charlemagne to let him take the fight against the Saracen Clarel, Otuel reports the dispute between himself and Clarel the previous day, and whereas the French account is limited to the usual insults against each other’s religion,7 in Otuel and Roland Otuel responds to Clarel’s attack on Christian beliefs with a corrective summary of the articles of the faith: y schall ȝow telle euery word, How it began, ende and ord, The stryf between ous to. He sayd that oure god was nought worth a tord, And that he wold proue with dynt of swerd, To whom that it wolde do; And sayde that we were thourȝ hym y-lore, That of a woman was y-bore, And schent for euermore; For hys lesyng and for hys sawe Vppon a cros he was y-drawe: Alle thus sayde he me to thare.

7 In the Anglo-Norman version, Clarel shows more precise knowledge of the Christian faith, see Chapter 1.

The Middle English Tradition

161

y answered and sayde nay, That He Was Bore of a may, To saue al menkynde, And ros & to helle toke þe way – þat byfyl vppon þe þrydde day – And satan brouȝt in bonde, And toke þerout ous and Adam, And all with hym þo god nam, And syþe þe holy gost sende, and after hys rysyng, vp sty To hys fadyr vp an hy. Þys we hauen in mynde. (1319–42)

Despite Otuel’s attempted instruction, Clarel, like Balan, shows no sign of conversion. This concern with religious instruction is a persistent feature of the English Matter of France texts. Conventional episodes on the same pattern are created in the two ‘new’ romances, the Sege of Melayne and Rauf Coilyear, while the romance of Roland and Vernagu is fashioned from chapters in the Pseudo-Turpin Chronicle in order to focus on Roland’s religious disputation with the Saracen giant. In the Sege of Melayne, the Sowdan Arabas taunts the captured Roland with the provocative claim that ‘The false lawes of Fraunce sall downn, | The Rewme sall leue one Seynt Mahownn’ (394–5). Roland rejects the idea that France would ever ‘lose oure crysten lawe’ (408), and counters the Sowdan’s argument with a succinct outline of the Christian ‘lawe’: For sothe, þou Sowdane, trowe þou moste One þe Fader and þe Sone and þe holy goste, thire thre are alle in one: Þat Borne was of Marye free, Sythen for vs dyede one a tree, In other trowe we none. (409–14)

However, the Sowdan merely laughs and ridicules the Christians’ belief at great length, with echoes of the jeers offered to Christ (in the Synoptic Gospels) by the bystanders and soldiers at the Crucifixion

162

Charlemagne in Medieval England

(415–56). The Saracen Magog in Rauf Coilyear, offered conversion by Roland in the middle of his combat with Rauf, is at first equally scornful of Roland’s arguments for the spiritual benefits of Christian belief versus the false promises of his own Saracen faith (895). There are several other occasions in the inherited narratives when Christian characters try to persuade infidels to convert, and however much the Middle English adaptations elaborate the theme of religious instruction, it is notable that the attempted conversion is always unsuccessful. For example, in Roland and Vernagu, where Christian hero and Saracen giant conduct a lengthy question-and-answer session on the articles of the Creed before Vernagu abruptly proposes that they fight to determine which is the better faith, Roland prays for help and an angel uniquely reassures him that there is no alternative to killing the giant, for conversion is impossible: þei alle prechours aliue To cristen wald him schriue, Gode nold he neuer be. (812–14)

The Middle English prose Pseudo-Turpin, in its version of the same material, similarly adds a small detail that emphasizes the lost potential for the Saracen’s conversion when, at the end of their dispute, Ferakutte states: ‘I can noȝt se it ne belyue it’ (899), adding the sense that he might have converted to Christian belief to the simple idea of intellectual knowledge in the source.8 The only Saracens who are converted in the end – Fyrumbras/Ferumbras and Otuel, and Magog in Rauf Coilyear – are not persuaded by preaching or argument. As in the French traditions, Otuel experiences conversion as a miraculous gift of divine grace, after Charles’s prayer on behalf of the near-defeated Roland, while Fyrumbras feels the power of the Christian God through Oliver’s success in overcoming him in combat,9 as does Ferumbras in the Sowdone of Babylone (1353–62) after Charles’s prayer (answered by an angel) for 8 The standard Latin reads ‘prorsus ignoro’. 9 The details of Fierabras’s looking up to heaven at the moment of his wounding and being enlightened by the Holy Spirit (Fierabras, ed. Le Person, 1570–2) are replaced in the Ashmole text by his falling to his knees and thanking God (751).

The Middle English Tradition

163

Oliver’s victory.10 As noted above, Magog rejects Roland’s persuasions both spiritual and material (in the form of a wife and rich dowry), but he voluntarily converts because he has seen the power of God when Christians have called on him in great affliction (944–6), a grace for which Roland immediately thanks Christ ‘that the that grace send’ (948). Since they do not lead up to triumphant conversion moments, why do the texts include these sometimes much elaborated episodes focusing on the articles of the Creed? One logical answer in terms of the narrative situations is implied by the angel’s message in Roland and Vernagu: that the obdurate unbeliever who refuses the offered grace of conversion has unequivocally damned himself. But it seems likely that the texts are also capitalizing on the opportunities provided in the narratives to address a specific contemporary concern: to guard against the danger of heresy, in this case by reinforcing orthodox vernacular teaching of the basic elements of the faith.11 Despite the denunciations of romances (including those of Charlemagne) as unprofitable fables by writers from the author of Cursor Mundi to the sixteenth century, there is ample evidence from the manuscript contexts in which romances were compiled with other texts that they were being read as edifying works capable of imparting moral lessons.12 The programme of basic religious instruction of the faithful that was recommended by Archbishop Pecham in his canons of the Council of Lambeth, Ignorantia sacerdotum (1281), responding to the demands of the Lateran Council of 1215, and that was reiterated by Archbishop Thoresby in his injunction, translated into the vernacular as the Lay Folks’ Catechism (1357), would be all the more earnestly promoted in the fifteenth century, in line with the focus on religious orthodoxy in the Constitutions of Archbishop Arundel (1409). The pastoral initiatives were intended to ensure a minimum universal knowledge of essential Christian belief, 10 This passage closely resembles that in the Egerton Fierenbras (275–302, 332–71). 11 This initiative may be seen in the light of much larger-scale efforts to provide orthodox religious resources in the vernacular, such as Nicholas Love’s Mirror of the Blessed Life of Jesus Christ. See Andrew Cole, Literature and Heresy in the Age of Chaucer (Cambridge: Cambridge University Press, 2008), p. 187. 12 Phillipa Hardman, ‘Learning Lessons in Middle English Romance’, Reading Medieval Studies 37 (2011), 1–13.

164

Charlemagne in Medieval England

and vernacular didactic texts aimed towards this end often occur in the same manuscript contexts as popular romances.13 These ‘conversion attempt’ episodes in the romances could thus offer the reader or listener some recapitulation of their own instruction in the articles of belief, while simultaneously helping to engage their support for their fellow Christians within the narrative. A similarly lively concern with the essentials of Christian doctrine is more widely reflected in the Middle English Pseudo-Turpin. Following the inherited tradition, Roulonde summarizes the Creed in answer to Ferakutte’s initial question, ‘what is þicke “Criste” in whome þou beleuyste?’, saying: ‘Þe Sonne of God þe Fadur of heuene, þe wiche was bore of a mayde; he dyede on a crosse; he was beride in a sepulcre; he aros þe þredde day; he styede into heuene and syȝtte in þe riȝtte syde of his fadur’ (804–8). Prior to this episode, however, the translated chronicle opens with a unique addition to the inherited narrative, setting its account of St James’s apostolic mission and martyrdom in the context of Christ’s death and resurrection, so that it now spells out exactly what was the content of the apostles’ preaching: After oure Lord Ihesu Criste had sufferid deþe and paide þe rawnsome for synfulle man by his peynfulle passioun and rose fro deþe to lyue and at þe laste styede to heuene, þen þe holy appostlus, after þey had reseyued þe Holy Gost, þey departid hem into diuerse parties of the worlde to preche oure Lord Cristis lawis. (80–4) The echo of the Creed here anticipates the reiteration of Christian belief that Roulonde will later provide to Ferakutte, and emphasizes the importance of the themes of religious instruction and conversion throughout the whole first half of the narrative: not only in the Roulonde/Ferakutte debate but also in the history of Charlemagne’s reconversion of Galicia, and the account of his prolonged conflict and disputation with Aigolonde for the religious identity of Spain. While the particular religious significance of the theme of Passion relics in relation to cult pilgrimage sites in France is to some extent 13 For example, see the Auchinleck MS (described below, pp. 178–83), and CUL MS Ff. ii. 28.

The Middle English Tradition

165

moderated in the insular French-language Fierabras texts, the Middle English versions treat the topic with much greater freedom, either by removing references to the specific objects and naming them simply ‘the relics’, or by adding new items to the list. The Fillingham Firumbras seems to reflect a range of devotional cults in its unique list of relics. Where the continental Vulgate Fierabras names ‘la corronne et li trois clou … et le signe Jesu’ (ed. Le Person, 5415–16), Fillingham not only adds ‘the spere that was in goddys syde’ (1415), as found in the insular Destruction tradition,14 but also (perhaps responding to a copy that used the Anglo-Norman synonym ‘sudaré’) combines the mention of the shroud with a reference to the popular relic of the vernicle, the napkin or kerchief with which Veronica wiped Christ’s face and on which his image was imprinted: ‘the voluper & the sudary’ (1416).15 Further, in addition to the three Fierabras-related romances, the two Middle English romances created as ‘prequel’ responses to the Otinel tradition, Roland and Vernagu and the Sege of Melayne, both include material that specifically or by analogy speaks to a concern with the relics. Roland and Vernagu, based on the Johannes Pseudo-Turpin Chronicle, begins with the story of Charlemagne’s intervention to aid the emperor of Constantinople and the patriarch of Jerusalem against the hostile Saracen king Ebrahim, for which, at Charles’s request, the emperor rewards him with the gift of Passion relics: the Crown of Thorns, a fragment of the Cross in a reliquary, the spear, and one of the nails (together with other holy relics such as Our Lady’s smock, the rod of Aaron, and St Simon’s arm). As Smyser notes, the focus of the incident is clearly upon the relics, for ‘the English versifier quite forgets to mention Charlemagne’s battles against Ebrahim’,16 with the result that the reader simply takes for granted the angel’s prophecy that Charles 14 The spear is also mentioned in lines 595–6, 1769–70. 15 Middle English ‘voluper’ (kerchief) and ‘sudary’ (sweat-cloth) can both denote the cloth used by Veronica; ‘sudary’ is also used for the burial cloth wrapping the head according to biblical custom, and hence for a shroud. 16 Smyser, ‘Charlemagne Legends’, p. 90; it is possible that the translator overlooked the brief account of the battle in the French text as a result of eye-skip, as almost identical phrases are used for Charlemagne’s two entries into Constantinople, before and after delivering the Holy Land (Johannes, chapter III, lines 13, 18).

166

Charlemagne in Medieval England

would ‘sle þe sarrazin wiþouten fail’ (60) and follows the trajectory of the narrative towards the donation of the relics, the evidence of their power when 300 sick people are cured by their holy odour (107–9), and the miraculous proof of their validity (125–36). The interest this episode might have for a fourteenth-century English audience seems to inhere precisely in its difference from the relics theme in the Fierabras tradition: here there is no association of the relics with any particular place, but the miscellaneity of this collection of holy objects, together with the accurate depiction of the relic of a piece of the Cross ‘þat in a cristal was don in clos’ (114), invites comparison with the various relic collections that were to be seen in the reliquaries of many abbeys and churches, and which readers might themselves have venerated, taking comfort from stories of miraculous cures, like that of the 300 sick people, attached to them.17 The Sege of Melayne takes a different approach to the representation of powerful sacred memorial objects: here it is not an actual relic of the Passion that plays a part in the narrative, but an image that functions with the same kind of divine force. The Sowdan Arabas expresses his scorn for the Christian faith by laughing at his easy destruction of its gods, which is how he understands the carved wooden crucifixes found in every church (415–20), and sends for another: Goo, feche one of theire goddes In And if he in this fire will byrne Alle oþer sett att noghte. (421–3)

When the ‘faire rode’ is brought from a nearby church, the narration stresses the truthfulness of its representation of Christ in his Passion: ‘Fourmede ewenn als he gan blede’ (427), and the orthodox purpose of such images is immediately made clear as the Christian knights ‘by-gane þaire crede’ (428) and Roland prays to God, calling on the merit of Christ’s sacrifice: ‘Schewe þou, lorde, thi meracle this day, | Þat with thi blode vs boghte’ (431–2). The image of Christ crucified directs their thoughts to the God to whom they pray, and when a miracle does occur, 17 See Freeman, Holy Bones, Holy Dust. Freeman notes the tradition that Charlemagne wore ‘a necklace with two crystal amulets containing pieces of the True Cross and the hair of the Virgin Mary’ (p. 69).

The Middle English Tradition

167

they repeatedly attribute it to God’s grace.18 The Sowdan’s contrary assumption that the crucifix is a ‘god’, an idol, like the ‘Mawmettes’ of his own religion which he has set up for worship in churches throughout the land (28–30), is developed with grim humour as the cross withstands the Saracens’ repeated assaults with brands, brimstone, pitch, tar and torches, and the Sowdan concludes that the Christians must have tricked him: they ‘wethede þaire goddis þat þai may not byrn, | I wote wele it es soo’ (455–6). The humour of this episode builds on reminiscences of the theme in the Fierabras tradition that opposes genuine Christian relics and inert Saracen idols. For instance, the Sowdan mocks the impotence of this wooden ‘god’ who can help neither himself or anyone else to escape (439–44) in terms that recall the mockery and abuse directed towards the unresponsive Saracen idols by both Christian knights and the disappointed Saracen Balan in Fierabras. Similarly, the moment when the cross, in response to the knights’ prayers, miraculously shoots forth a fire that strikes the Saracens in the eyes (469–71) echoes the scene in Fierabras in which Naimes uses the power of the relics to blind a thousand attacking Saracens: ‘Les reliques lor mostre et il sont awelglé’ (5435, ed. Le Person). However, the effect on the Saracens in the Sege of Melayne, who are rendered blind, deaf, unable to move and insensible, is to convert them into the same condition as their own idols (472–80), and in this state they are treated much as the Christian knights deal with Balan’s ‘gods’ in Fierabras, being broken and thrown on the fire (484–9). In this insular text, the conventional Matter of France concern with relics has given rise to a set of scenes that, with almost programmatic effect, enact the parallel orthodox doctrine on the use of holy images, embellished with incidents like those often found in miracle stories.19 18 ‘This Meracle es schewede thorowe goddis grace’ (478; 494, 586). A similarly deliberate explanation of the orthodox doctrine of intercession is provided in relation to the miraculous white horses that bring the knights to Paris: it was by ‘þe myghte of god’ (518) and ‘thorowe þe prayere of seynt Denys’ (521). 19 For instance, several Miracles of the Virgin tell of her jealous concern for the reverence due to her image, often punishishing blasphemers with death. Examples include: ‘The Virgin’s Image Insulted’; ‘The Bleeding Child-Christ’; ‘Scoffer at Mary-Image Beaten’; ‘Saracens Cannot Deface Mary-Image’. See H. D. L. Ward, and J. A. Herbert, Catalogue of

168

Charlemagne in Medieval England

In the Ashmole Sir Ferumbras, the role of the Passion relics is reinterpreted by their being consistently identified as Charlemagne’s misappropriated personal property (see pp. 273, 276 below). After the initial identification of the stolen relics as ‘þe croune & þe nayles three’ (65, 135), very little mention is made of the specific sacred items: apart from the incident when the peers place the Crown of Thorns upon their helmets before setting out to rescue Guy (2959–64; cf. Fierabras, ed. Le Person, 3656–60), they are undifferentiated, simply referred to as ‘þe relyques’. Nevertheless, the relics feature with the same frequency as in the inherited tradition: indeed, there is even an added reference to the relics when Richard of Normandy is required to swear an oath to complete his mission even to the point of death, and uniquely in Ashmole the relics are fetched for the purpose (3507–8). Conversely, in the Sowdone of Babylone the theme of the relics is radically abbreviated in line with new and different emphases in this adaptation of the narrative (see pp. 334–5 below). Even in the climactic relics scene, where the Egerton text has Floripas present the crown, nails and shroud to Charlemagne for his adoration (1677–81), in the Sowdone they are simply ‘Relikes of grete honour’ (3138). In the Fillingham Firumbras, the treatment of the Passion relics is adapted to serve two specific new functions: first, the incidents involving them are reimagined to fit with an unequivocally orthodox attitude towards the veneration of relics (see pp. 297–300 below); secondly, the relics are used to elevate the status of Charlemagne himself as holy in the concluding scenes of the romance. Uniquely in this version of the story, the miraculous proving of the relics’ authenticity is brought to a climax with Charles explicitly praying for a sign that he is ‘worthy to haue thys holy thyng’ (1817) and God’s performing a ‘fayre myracle for Charlys the kyng’ (1824, 1821). No mention is made of distributing the relics to churches: the point of the relics in this version is simply to demonstrate by his winning ‘the Relykys of goddys passyoun’ that Charlemagne is ‘that gode, holy kyng’ (1833), as the text concludes. Charlemagne was in fact canonized in 1165, though, as an act of the anti-Pope Pascal III, it was not entered in the Roman Breviary; despite this, the tradition of his sanctity was evidently current

Romances in the Department of Manuscripts in the British Museum, 3 vols (London, 1883–1910), II, 586–740.

The Middle English Tradition

169

in England, for in the Life of St Barbara found in some manuscripts of the Gilte Legende there is a specific reference to the ‘emperour Karolus Magnus, whiche for his holynesse by specyalle priuylege ys cleped Sanctus Karolus’. 20 In the light of the Fillingham Firumbras’ concern with Charlemagne’s sanctity, it is tempting to think that there may even be an echo of Christ’s five wounds when Balam wounds Charles in five places on his body (1620), rather than fifteen places as in the French texts (6000, ed. Le Person). A similar focus on Charlemagne’s saintlike role is evident in the first half of Roland and Vernagu where, as Janet Cowen observes, ‘it is as if the numinous power of the relics had become vested in the person of Charles himself ’, and the narration of the miracles that bring about the conquest of Spain reads ‘like a pilgrimage account in which Charles has assumed the role of peripatetic saint, conveying miracles, rather than visiting them’. 21 It is worth noting that while Roland and Vernagu is quite closely modelled on the Pseudo-Turpin Chronicle, it adds another ‘fair miracle’ which God performs for Charles, not present in the source, after the series of miraculous victories, when an overnight harvest of ripe grapes appears on newly planted vines as a quasi-Eucharistic sign of divine favour (305–16). The idea of Charlemagne’s exemplary holiness is only one adaptation of the inherited concern in the Matter of France tradition with his role as monarch. The centrality of his role in the Middle English romances is signalled by the titles attributed to some of them in scribal paratexts: the Thornton Roland and Otuel announces itself ‘Off Cherlles of Fraunce’, 22 and the Fillingham Firumbras concludes ‘Explicit Kynge Charlys’. 23 These scribal titles acknowledge the enhanced part played by 20 This reference is not in the Latin source for the Life. See Supplementary Lives in some Manuscripts of the Gilte Legende, ed. R. Hamer and V. Russell, EETS OS 315 (Oxford, 2006). 21 Cowen, ‘The English Charlemagne Romances’, p. 153. 22 This text, unusually, has both a scribal title and colophon, naming it ‘Þe Romance of Duke Rowlande and Sir Ottuell of Spayne’ at beginning and end; but it also has a double attribution to Charlemagne. The title is followed by an additional title: ‘Off Cherlls of Fraunce’, and the colophon is bracketed with the word ‘Charlles’ added alongside. As well as all this surplus, there is a final explicit: ‘Explycit sir Otuell’. 23 Following normal practice, the compiler of the British Library catalogue entry for this text took the explicit to give the contemporary title of

170

Charlemagne in Medieval England

Charlemagne in the Middle English versions of the inherited narrative traditions, whether in added emphasis on his role and status, in more active intervention credited to him in battle scenes, or in new material invented to create a starring role for him. Following and building on the presentation of Charlemagne in insular French-language texts as supreme secular Christian ruler, the two earliest of the Middle English romances, Roland and Vernagu and Otuel, both adapt material from the opening of their source text to emphasize Charlemagne’s status as representative head of all Christendom. In Roland and Vernagu the original account of the hero’s conquests, as derived from the Pseudo-Turpin Chronicle via the thirteenth-century Old French Johannes redaction, is rearranged to construct a picture of his achievement of power and control throughout Europe to become finally, in a new addition, ‘lord of al christendome’ (15). Otuel omits most of Otinel’s chanson de geste prologue and substitutes an introductory paragraph in which King Charles of France is described as a ‘douȝty’ hero who overpowered Saracen aggression, and guaranteed the security of Christendom.24 A similar view of Charlemagne is inserted into the introductory or concluding material in most of the other Middle English romances; while the Ashmole Sir Ferumbras, which lacks both its beginning and ending, contains a lengthy additional passage that highlights Charlemagne’s inspirational role as leader of the Christian forces in the final climactic battle against the Saracens, facing apparently overpowering odds with unshakeable faith (see pp. 274–6 below). In the Middle English Pseudo-Turpin, the predominant emphasis from the outset is on Charlemagne’s status as emperor. The prologue announces its subject as in the standard Latin text: ‘how þat oure famose emperoure Charlis þe Grete delyuered þe lond of Spayne and

the work, and duly named the text ‘Kynge Charlys’. The same normal practice led the editor of the Catalogue of Romances in the Department of Manuscripts in the British Museum (1883) to take the final explicit in the Thornton Roland and Otuel as the title: ‘Sir Otuel’ (I, 954). It is not clear why the editors of the two romances for EETS did not follow the same practice. 24 He ‘made sarazins ful tame’ (8–9) and ‘meintenede cristendom ariȝt’ (14).

The Middle English Tradition

171

Galice fro the powere of þe Sarzyns by many a victorious dede’ (6–8);25 the unique added titulus names ‘the Storye of the Bataille of Rouncivale of Grete Charles the Emperoure’ (16–17); the chapter describing Charlemagne is here entitled ‘Off þe persone and of the strenkethe of Emperoure Charlis’ (52–3), explicitly recalling his imperial status.26 The climax of this chapter is the account of Charlemagne’s appearing on great Feast days with crown and sceptre and a drawn sword held before him. The standard Latin text states that this accords with imperial custom (‘more imperiali’), and the otherwise close English translation enhances this phrase with visual proof of Charlemagne’s imperial status: ‘he satte at mette in his trone imperialle’ (1079).27 Like the focus on his role as ruler of all Christendom, this emphasis on Charlemagne’s title as emperor speaks particularly to an interest in his supra-national significance. The two Charlemagne romances that include a version of the same descriptive chapter from the Pseudo-Turpin tradition, Roland and Vernagu and Otuel and Roland, are also alike in their portrayal of Charlemagne not only as the defender of Christendom from Saracen aggression, but specifically as a conqueror of other lands. In Roland and Vernagu, this derives directly from the Pseudo-Turpin material, where Charles is credited with the conquest and conversion to Christianity of England, France and many other parts of Europe; the romance adopts this element of Charlemagne’s career to create a descriptive epithet, ‘Charles þe conquerour’ (57), and supports it with repeated references to how he ‘wan’ cities, lands and treasure ‘Wiþ dint of swerd’ (270), and had all these territories ‘in [his] pouwer’ (178). Otuel and Roland begins with exactly this portrayal of Charlemagne, offering to tell ‘Off a conquerour | that was y-hote syr Charlemayne’ (3–4), who won Spain and overcame all his enemies (34). The phrase ‘Charlemayn the conquerour’ is repeated throughout the Otinel part 25 In his EETS edition Turpines Story, Shepherd notes that a later reference to Charlemagne in the phrase ‘imperatori nostro’ is replaced by ‘Kynge Charlis’ (937), without a possessive, in the ME translation and in all other C-text non-Francophone versions (p. 77). 26 The standard Latin text reads ‘Karoli’. 27 An earlier line similarly emphasizes the material display of Charlemagne’s power: ‘he satte in his trone of astate’ (535, translating the word ‘tribunal’ in the standard Latin text at this point).

172

Charlemagne in Medieval England

of the text, functioning as Charles’s accepted cognomen (92, 276, 408, 1041, 1064), and leading to the triumphal ending of the story, in which he is said to have ‘lyued in warre many ȝerys’ and won every battle that he undertook (1686–90). This simplified figure of ‘Charlemagne the conqueror’ has perhaps been fashioned in these romance representations to fit a stereotype that seems to have held a powerful fascination for contemporary insular readers, as shown for example in Robert Thornton’s collection of texts, which includes the ‘lyf of gret Alexander conqueror of all þe worlde’; the alliterative epic Morte Arthure, celebrating Arthur as conqueror and ‘Ouerlynge of all þat on the erthe lengez’ (3211); the ‘Romance of Kyng Richerd the þe Conqueroure’ who ‘wonne the holy lande’; as well as the two Matter of France romances, Roland and Otuel and the Sege of Melayne, featuring Charlemagne as conqueror of ‘many a lande’. 28 Conquest figures prominently in Charlemagne’s achievements in the Fierabras tradition, or rather ‘Recomquist’ (9, ed. Le Person), as the prologue indicates, for the epic closes with Charles’s having reconquered Spain – an issue that would still have been of topical interest in the late Middle Ages, when the three Middle English versions of Fierabras were composed and copied.29 However, the two fifteenthcentury romances show typical independence in their adaptation of Charlemagne’s reconquest of Spain. The Fillingham Firumbras takes a radical approach: Spain is not mentioned at all, at least, not in the surviving part of the text. At the end, Charles gives to Guy ‘that kingdom Ryche that balam helde’ (1757), Guy establishes his control of the land, and no more is heard of it; as discussed above, the focus on Charlemagne at the end of this version of the tradition is more hagiographic than political. At the end of Sowdone of Babylone, on the other hand, additional references to the ‘londe of Spayne’ are introduced, first as Charles bestows Laban’s possessions on Ferumbras and Guy, and second in a new summary of his conquests to conclude the poem, without any hint of further conflict as found in the French texts: Thus Charles conquered Laban, The Sowdon of Babyloyne

28 The Sege off Melayne, ed. Herrtage, 1089. 29 See Tyerman, England and the Crusades, pp. 276–8; Phillips, Holy Warriors, pp. 307–9.

The Middle English Tradition

173

That riche Rome stroyed and wan, And alle the brode londe of Spayn. (3259–62)

Here Charles is portrayed rather as rex pacificus: he bids farewell to Guy and Ferumbras, foreseeing for them an entirely peaceful future in Spain assured by their mutual support and strength (3215–18), and so instead of promising to come to their aid if needed with 40,000 troops, as in the Anglo-Norman text, here he invites them to visit him in France for their pleasure and to comfort his old age (3219–21). For the Sowdone poet, it seems it was not the image of Charlemagne as perpetual conqueror that held the greatest appeal, but Charlemagne portrayed as promoter of peace; it is not difficult to imagine that this could have been an attractive ideal in a century that saw the effects of the Hundred Years’ War and the Wars of the Roses. A similar preference may be discerned in some unique features of the representation of Charlemagne in the contemporary Middle English Song of Roland, where the English version greatly elaborates his response to Ganelon’s (false) report that the Saracen Soudan is willing to cease hostilities, convert to Christianity, and become Charles’s vassal. In the French Chanson de Roland tradition, Charlemagne briefly thanks God and Ganelon and declares the war over; here, in a long speech, he imagines a ceremonial welcome with gifts and a fortnight of feasting, to mark a future of peaceful coexistence between himself and the Soudan (40–51). The focus on Charlemagne’s reputation as invincible conqueror is supported in some of the Middle English texts by alterations or additions to the inherited narratives. For example, the Fillingham Firumbras rewrites the episode when Charles and his barons come to rescue the beleaguered Richard of Normandy, so as to give greater evidence of Charlemagne’s personal role as an action hero. Richard is at the mercy of the Saracen giant Golfagor, keeper of the bridge of Mautryble, and here, uniquely, it is Charles, not Ganelon, who first reaches the bridge when Richard sounds his horn. The significance of the change is indicated by an added comment: ‘Well was hym þat to mautryble come fyrst of alle!’ (1313). Shortly afterwards the narration states: ‘Charles was þe fyrst: er euer wolde he blynne, | he wan þe brygge hym byfore & entred with-inne’ (1316–7). In line with the tendency towards simplification in the insular Charlemagne tradition, the potentially troubling idea of the notorious traitor Ganelon’s

174

Charlemagne in Medieval England

spear-heading the victory is suppressed, and the scene is reinvented as an opportunity to assert the pre-eminence of the king. Charlemagne’s heroic status is maintained in the following action: he hyede hym swyþe, nold he nouȝte abyde, And sey Golfagor stray on hym wyde. ‘Euyl þryst on þey heued, with þy foule vice! þu hast ben a moche schrew in thyn offyce.’ And al hol as he was, the foule tyraunt, Charlys kest hym ouer the brygge with his on hand. (1318–23)

In the narrative tradition, when Charles arrives the Saracen giant is lying wounded but still alive and fighting; Charles, seeing him, laments the loss of his knights, at which point the giant is finally killed by the French and thrown into the river. The Fillingham version, however, presents Charles in a much more heroic role, as he confronts the unharmed Golfagor, denounces him, and then throws the giant bodily off the bridge ‘with hys on hand’ (1323).30 The text conscripts the reader and audience into loyal admiration for Charlemagne’s heroic martial values by referring to him repeatedly as ‘Oure noble kyng charlys, ful coraious in yre’ (1621, 1589). The same concern to enhance the actively heroic element of Charlemagne’s role can be seen in the Sowdone of Babylone. The extensive adaptation of material from La Destruction de Rome and Fierenbras to form this single narrative, particularly around the join between the two originally distinct poems, involved enlarging the roles of both Charlemagne and the Sowdone. The first part ends with a battle between them and, as in the source, Charlemagne himself takes the field while the Sowdone remains in his stronghold and sends his son Ferumbras out against Charlemagne; but whereas the chanson de geste briefly represents the battle by an encounter between Roland and Fierenbras and a general mêlée, the Middle English romance creates a new extended scene that showcases Charlemagne’s personal feats of arms. 30 The Ashmole Sir Ferumbras gives an intermediate version of the scene, in that Charles himself attacks the fallen Agolafre and eventually kills him before the giant is thrown into the river by others (4599–618).

The Middle English Tradition

175

Kinge Charles saugh Ferumbras; To him fast he rode And [h]it on the helme with his mace. That stroke sadlye abode. Ferumbras was woode for woo, He myght for prees come him to For no worldis thinge that myght be tho. King Charles anoon Ioye oute-drowe, And with his owen honde XXXti Sarseynys ther he slowe, That laie dede vppone the sonde; Many of hem therfore made joy inowe. Sir Lucafere of Baldas, He presed to Charles sone, And saide, ‘Sir, with harde grace, What hastowe here to done? I behight Laban to bringe the to him And the xij peris alle; Now shaltowe come from al thy kyn Into the Sowdans halle. Yelde the to me,’ he saide; ‘Thy life shalle I safe.’ A stroke on him than Charles layde; He made the paynym to rafe. He smote him on the helme With mownjoye, his gode bronde. Ne hadde he be reskued than, He hade slayn him with his honde. (843–70)

Here Charlemagne is shown at the centre of the battlefield action, swiftly delivering a mace-blow against Ferumbras, and, before Ferumbras can respond, drawing his sword and killing thirty Saracens. The next incident shows the Middle English poet’s careful attention to the plot, as the Saracen king Lukafer, who has vowed to capture

176

Charlemagne in Medieval England

Charles and the twelve peers in order to gain the hand of the Sowdone’s daughter Floripas, uniquely attempts to make good his promise by urging Charles to yield to him. His foolish presumption is answered by Charlemagne’s mighty sword-stroke. Like the scene on the bridge in the Fillingham Firumbras, this scene underlines Charles’s personal heroism by repeated use of the same conventional phrase: Charles overcomes his enemies ‘with his owen honde’ (851, 870). This aspect of Charlemagne’s individual prowess is highlighted in the Ashmole Sir Ferumbras, where new emphasis is laid on Charlemagne’s knightliness in a striking accolade put into the mouth of Fyrumbras at the moment of conversion. He yields himself to ‘Charlis kyng, þe beste knyȝt y-core | Þat is owar now lyuyng, oþer euere was her-before’ (766– 7). Compared with the stock description in the French text at the same point – ‘le fort roi corronné’ (1575, ed. Le Person) – Charlemagne is here portrayed as an idealized knight, a timeless embodiment of chivalry, and an exemplar of the values embraced by the converted Saracen Fyrumbras. This finds an interesting insular parallel in the Middle English prose translation of the Pseudo-Turpin Chronicle, where another translator has also highlighted Charlemagne’s knightly credentials in the chapter entitled ‘Off þe persone and of the strenkethe of Emperoure Charlis’, by extending the brief account of his soldierly expertise in the standard Latin text to claim him ‘moste douȝttyeste in dedis of armys, moste of knyȝttehode and chyualrye’ (1061–2). Knighthood is also a prominent concern of the comic Matter of France adaptation in Rauf Coilyear, where Charlemagne is shown in the act of making Rauf a knight, and expounding the rules and expectations that define the order of knighthood, particularly as they concern the reciprocal duties of knight and king (743–77). In their different adaptations of the French texts’ portrayal of Charlemagne, the Middle English romances, while properly responsive to their source material, show considerable creative independence. Each new text reimagines the portrait of Charlemagne in line with the particular priorities of its own reading of the legend, to construct Charles variously as powerful emperor, defender of Christendom, unbeaten conqueror, pattern of knighthood, warrior-hero, peacemaker, or saint; but they all seek to enhance or ‘foreground’ the figure of Charlemagne within the text. The same creative independence may be seen in the Middle English romances’ developments of the other characteristic insular concerns discussed above: even when the relics in

The Middle English Tradition

177

the Fierabras tradition, for example, are completely divorced from their cultural origins, each romance adapts the inherited material to produce a new construction of the relics to serve its own particular agenda. These different priorities and agendas are often most evident when the romances are read each in their own unique manuscript context.

The Manuscripts and their Audiences Each of the English Charlemagne texts is unique, and investigating each in its specific manuscript context will provide evidence for an understanding of its individual place in the evolving engagement of writers and readers with the Charlemagne legend. Additionally, three of the manuscripts, each of which contains two separate Charlemagne romances placed together, suggest that the texts were perceived in relation to each other: detailed analysis indicates ways in which these poems have been selected or adapted to produce connections, creating pairs of texts.31 Their common concerns may be shared with other materials within their manuscript context, and investigating these concerns can throw light on the remarkable persistence of the Charlemagne material in late-medieval English popular romance. The manuscripts containing the Middle English Charlemagne romances are not prestigious productions like some of the Frenchlanguage manuscripts for insular consumption described in the previous chapter. With one exception,32 they provide no evidence of intent to illustrate the texts with pictorial images, and few of them are decorated with anything more elaborate than coloured and sometimes flourished initial letters. Only one includes armorial designs to proclaim the identity of an owner or commissioner;33 and in only one other case is there any evidence to identify a medieval owner of the manuscript. 31 Ailes and Hardman, ‘Texts in Conversation’, pp. 39–47. 32 See the description of the Auchinleck MS below, which once had a modest scheme of pictorial illuminations. 33 The late-fifteenth-century Middle English prose translation of the Pseudo-Turpin Chronicle in Huntington Library MS HM 28,561 is bound up with a copy of Trevisa’s Polychronicon whose decorative programme contains coats of arms identifying the family of Mull in Gloucestershire (Turpines Story, ed. Shepherd, p. xviii; for a suggested connection with the Shrewsbury Book, BL Royal MS 15 E vi, see p. xxvi).

178

Charlemagne in Medieval England

Middle English Charlemagne romances survive in a range of different types of manuscript: small volumes containing a single text; large multi-text compilations designed to serve the various needs of a reading household; medium-sized collections of disparate texts copied by one or more scribes, or possibly assembled into a single volume by a contemporary owner. The manuscripts’ provenance is similarly diverse, ranging from London to Exeter, and from different areas of the Midlands to Yorkshire. All this indicates a widespread interest in the romances and, in some cases, in their potential for interaction with other texts. The manuscripts, dating from the mid-fourteenth to the late fifteenth centuries, are discussed here in approximately chronological order.

National Library of Scotland, Advocates MS 19.2.1 (c. 1330–40): Roland and Vernagu; Otuel The Auchinleck Manuscript has not lacked critical attention in modern times: as one of the earliest and most extensive collections of Middle English romances, it looms large in studies both of romance and of vernacular so-called miscellany manuscripts, yet after more than a century’s scrutiny, the evidence for its date, provenance and ownership remains circumstantial. Alison Wiggins outlines the scholarly consensus: Palaeography, style of illumination and internal references indicate that Auchinleck was most likely to have been produced between 1331 and 1340 but the identity of the earliest readers and owners remains unknown. Dialect and the apparently commercial and collaborative nature of this manuscript’s production imply that it was most likely to have been produced in London. As this is a large (and therefore costly) manuscript, professionally produced and with a carefully executed design scheme, it would seem that it must have been produced on commission for a specific purchaser.34 The script, layout, and decorative scheme show marked similarities to Anglo-Norman vernacular manuscripts of the late thirteenth and early 34 The Auchinleck Manuscript, ed. Burnley and Wiggins, version 1.1, http:// auchinleck.nls.uk/editorial/history.html. A full description of the manuscript is available here.

The Middle English Tradition

179

fourteenth centuries, such as, for example, MS Bodmer 168, described in the previous chapter,35 while the few remaining small and damaged miniatures36 have been compared to the work of the illuminator of the Queen Mary Psalter (BL MS Royal 2 B vii), produced in London c. 1310–20.37 This indicates that despite the difference in the choice of insular vernacular, there is more continuity than divergence between Auchinleck and the cultural context of Anglo-Norman manuscript tradition.38 A number of hypotheses have been advanced as to the social status and gender of Auchinleck’s original owner(s), from an ‘aspirant middle-class citizen, perhaps a wealthy merchant’, 39 to a woman of the nobility,40 to a very rich noble family of unrefined tastes,41 to a 35 Alison Wiggins, ‘The Manuscripts and Texts of the Middle English Guy of Warwick’, in Guy of Warwick, ed. Wiggins and Field, pp. 61–80 (p. 68); see also Ralph Hanna III, ‘Reconsidering the Auchinleck Manuscript’, in New Directions in Later Medieval Manuscript Studies, ed. Derek Pearsall (York: York Medieval Press, 2000), pp. 91–102 (pp. 97–8). 36 Otinel begins with a large flourished initial but no miniature; Roland and Vernagu probably began with a small miniature on the excised first leaf of the text (before fol. 263). 37 Timothy A. Shonk, ‘A Study of the Auchinleck Manuscript: Bookmen and Bookmaking in the Early Fourteenth Century’, Speculum 60 (1985), 71–91; see also Lynda Dennison, ‘ “ Liber Horn”, “Liber Custumarum” and other Manuscripts of the Queen Mary Psalter Workshops’, in Medieval Art, Architecture and Archaeology in London, ed. Lindy Grant, British Archaeological Association Conference Transactions 10 (London: Maney, 1990), pp. 118–34. 38 On the transition from Anglo-Norman to Middle English book production, see Andrew Taylor, ‘Manual to Miscellany: Stages in the Commercial Copying of Vernacular Literature in England’, The Yearbook of English Studies 33, ed. Phillipa Hardman, Medieval and Early Modern Anthologies and Miscellanies (2003), 1–17. 39 The Auchinleck Manuscript, intro. Pearsall and Cunningham, p. viii. 40 Felicity Riddy, ‘The Auchinleck Manuscript: A Woman’s Book?’, paper given at the Romance in Medieval England conference at Bristol in 1992. Riddy suggested Katerine de la Poole, whose family name occurs in the so-called Battle Abbey Roll (fols 105v–07). 41 Turville-Petre, England the Nation, p. 138. Turville-Petre proposes a family with a crusading tradition whose name appears in the Battle Abbey Roll

180

Charlemagne in Medieval England

sophisticated member of the London civic elite.42 Clearly, the purchaser had considerable wealth, but equally significant here is the suggestion that this was a book for the use of a whole family or household.43 The collection of texts, almost all in English verse,44 includes eighteen romances of various kinds, six female saints’ lives and religious narratives, five works of religious or moral instruction catering for different stages of education, four pious legends, three moral or religious debates, two moral tales,45 two political and moral complaints, a penitential prayer, a chronicle of England, and a poem in praise of women for the sake of the Virgin Mary. The whole book would function for the household as a repository of traditional wisdom and cultural ideals, both in worldly terms of chivalry, power, nation, and family, and in spiritual terms of contempt of the world and Christian observance (sets of ideals whose inherent tensions are directly confronted in romances such as Guy of Warwick, present in the Auchinleck MS). The two Matter of France romances, like several others in the collection, address both chivalric and religious concerns, here focused on the idea of conversion: in each, Roland engages in single combat ‘such as the Beauchamps and the Percies’ (p. 136). 42 Ralph Hanna III, London Literature, 1300–1380 (Cambridge: Cambridge University Press, 2005), chapter 3. 43 For some discussion of the manuscript in this context, see Turville-Petre, England the Nation, pp. 135–6; Phillipa Hardman, ‘Domestic Learning and Teaching: Investigating Evidence for the Role of “Household Miscellanies” in Late-Medieval England’, in Women and Writing, c. 1340–c. 1650: The Domestication of Print Culture, ed. Anne Lawrence-Mathers and Phillipa Hardman (York: York Medieval Press, 2010), pp. 15–33 (pp. 19–21). 44 The only exceptions are ten lines of Anglo-Norman verse alternating macaronically with English in the complaint known as ‘The Sayings of the Four Philosophers’ (fol. 105r), and the Latin quotations and verses that punctuate Alcuin’s sermon in the ‘Speculum Gy de Warewyke’ (fols 39–48) and the translation of Psalm 50 in ‘Dauid þe king’ (fol. 280). Minor as these are, they indicate a comfortable level of multilingualism among the readers of the texts. 45 One of these, ‘Þe wenche þat loued a king’ (fol. 256), is so badly mutilated that one can only guess from its companion text, ‘A penniworþ of witte’ (fols 256–9), that it was most likely a similarly entertaining tale with a socially conservative moral.

The Middle English Tradition

181

with a Saracen and also attempts to convert him to Christianity. They are presented as a sequential pair, intended to be read together, for although they occur in different booklets,46 copied by different scribes,47 one stands at the end of a booklet (quire 37) and the other at the beginning of what must always have been the next (quire 38): besides the catchword added at the foot of the last folio of quire 37, the first of the pair, Roland and Vernagu, ends with three lines designed to lead directly to the following story of Otuel (Vernagu’s nephew): ‘To Otuel also ȝern, | Þat was a sarrazin stern, | Ful sone þis word sprong’ (878–80). The story of Roland’s unsuccessful attempt to convert the unregenerate Saracen Vernagu is completed by the story of his combat with Otuel, which results in the conversion of the ‘sarrazin stern’ and his assumption of a new identity: ‘Otuel is my cristine name’ (1154). The placing of the two romances one after the other, and the manifestation of common concerns in both narratives, seem to indicate a programme of reinforcing the values they represent. The two romances share a number of marked characteristics in those passages that diverge from their source texts: each has an introduction presenting Charlemagne as chief lord of Christendom, opposed by a powerful heathen king; both lay new stress on the Saracens’ violent hostility to Christians; both increase the role of miracles in the narrative; and both give added emphasis to the display of chivalrous behaviour. For example, the Saracen giant Vernagu, in the middle of his single combat with Roland, desires to sleep, and Roland gives him leave with an elaborate assurance that he would under no circumstances slay a sleeping knight (620–5).48 Similarly, Vernagu is uniquely presented as responsive to the values of chivalry: he appreciates the knightly generosity of 46 The manuscript consists of self-contained booklets of varying size, arranged by the main scribe (Scribe 1). 47 Alison Wiggins summarizes the considerable debate on this question, ‘Are Auchinleck Manuscript Scribes 1 and 6 the Same Scribe? WholeData Analysis and the Advantages of Electronic Texts’, Medium Ævum 73 (2004), 10–26. 48 There is an exemplary parallel in Malory’s account of Sir Pelleas’s selfrestraint: ‘ “ Though this knyght be never so false, I woll never sle hym slepynge, for I woll never dystroy the hyghe Ordir of Knyghthode” ’ (The Works of Sir Thomas Malory, ed. Eugene Vinaver, 2nd edn, 3 vols (Oxford: Oxford University Press, 1977), p. 103).

182

Charlemagne in Medieval England

Roland’s action in placing a stone pillow beneath his sleeping head, and reciprocates by offering Roland friendship (641–9). There are parallel instances in Otuel, such as the moment in the one-to-one combat between Otuel and Roland when each knight accidentally kills the other’s horse: in this version alone, the incident is turned into a demonstration of battle-field courtesy, as Roland chivalrously waits for Otuel to recover from his fall and Otuel, in turn, consciously imitates Roland’s noble behaviour (465–8, 489–94).49 All this indicates that processes of adaptation in each of the Matter of France romances of Auchinleck have increased the existing similarities between the texts, both by the enhanced polarization of Christian and Saracen rulers, and by the construction of a pattern of knightly behaviour, conveyed through the exemplary conduct of the hero confronting his enemy in one-to-one combat. Both issues can be observed as shaping concerns in the manuscript as a whole: Siobhain Bly Calkin has highlighted the pervasive presence of the Saracen as Other in Auchinleck’s textual construction of English identity;50 and the manuscript provides for any young readers in the household what amounts to an informal programme of education in the Christian and knightly values of their elders.51 The Charlemagne material has been recontextualized within the cultural agenda of the Auchinleck MS 49 Malory provides a similarly educational instance when Gawain rebukes Marhaus for not dismounting when Gawain is unhorsed: ‘ “Gramercy”, sayde sir Marhaus, “of your jentylnesse! Ye teche me curtesy, for hit is nat commendable one knyght to be on horsebak and the other on foote” ’ (The Works of Sir Thomas Malory, ed. Vinaver, p. 96). 50 Calkin, Saracens and the Making of English Identity; Calkin discusses the role of the noble Saracen in both these texts in Chapter 1, but beyond noting the reference to Otuel at the end of Roland and Vernagu, does not consider the relation between the two narratives. 51 Hardman, ‘Domestic Learning’, pp. 20–1. For Auchinleck romances as texts for the young, see Nicole Clifton, ‘The Seven Sages of Rome, Children’s Literature, and the Auchinleck Manuscript’, in Childhood in the Middle Ages and the Renaissance: The Results of a Paradigm Shift in the History of Mentality, ed. Albrecht Classen (Berlin: De Gruyter, 2005), pp. 185–201; Phillipa Hardman, ‘Popular Romances and Young Readers’, in A Companion to Medieval Popular Romance, ed. Radulescu and Rushton, pp. 150–64 (pp. 156–60).

The Middle English Tradition

183

and plays its part in the preservation and reinforcement of the owning family’s idea of itself and its Christian, chivalric traditions.52

Bodleian Library MS Ashmole 33 (c. 1380–1400): Sir Ferumbras MS Ashmole 33 is a late fourteenth-century paper manuscript containing the single, unique text, Sir Ferumbras.53 Remarkably, it is accompanied by its original parchment wrapper, on which is preserved a draft version of over 400 lines of the text, and both the draft and the manuscript book exhibit scribal corrections or revisions which in some passages are very frequent and sometimes repeatedly address the same line. The manuscript and wrapper have been described by Stephen Shepherd,54 and images of both are reproduced in The Romance of the Middle Ages, which gives a succinct account of the known facts: The draft copy of the romance of Sir Firumbras [sic] is written on the back of documents relating to the dioceses of Exeter and Sherborne, dated 1357 and 1377. These were folded to form a protective envelope within which the fair copy of the romance was then stored.55 Richard Beadle has dated the unusual watermark of the paper stock used in Ashmole 33 to the last decades of the fourteenth century.56 The language of the text is a fit with the geographical area to which 52 There may be a connection (either real or imagined) between the Browne family whose names are inscribed at the end of the Battle Abbey Roll and the name ‘Brun’ that appears in the list (Hardman, ‘Domestic Learning’, pp. 19–20). 53 This is the title of the modern edition. However, the spelling of the Saracen king’s name throughout the text is Fyrumbras (or occasionally Firumbras). 54 Shepherd, ‘The Ashmole Sir Ferumbras’; Stephen H. A Shepherd, ‘Uninscribed Meaning: VRML Applications in Electronic editions of medieval Texts’, at http://myweb.lmu.edu/sshepherd/vrml33.htm [accessed 25/10/2016]. 55 Nicholas Perkins and Alison Wiggins, The Romance of the Middle Ages (Oxford: Bodleian Library, 2012), p. 62. 56 Richard Beadle, ‘Set Pieces: Medieval English Literary Autographs’, in ‘Aspects of Late-Medieval English Autograph Writings’ (The Lyell Lectures), given at Merton College, Oxford, 30 April 2013.

184

Charlemagne in Medieval England

the original papal documents relate,57 and as it is unlikely that anyone unconnected with the ecclesiastical milieu in which they circulated would have had access to the documents in order to reuse them in this way, it has generally been accepted that the scribal author of Sir Ferumbras was a member of the clergy in the vicinity of Exeter. Indeed, it is not impossible that a library in that area was the source of the Frenchlanguage text(s) on which this translation was based: it is known that the library of Exeter cathedral, for example, contained (unspecified) manuscripts in French in the fourteenth century,58 and, as made clear in Chapter 1, copies of chansons de geste were by no means uncommon in the libraries of religious houses.59 The question of the intended recipient of the new Middle English version is harder to answer. The two texts of Ashmole 33, on the parchment wrapper and in the paper book, are usually referred to as the ‘rough draft’ and the ‘fair copy’, yet a comparison of the two reveals that this is not the case, either in terms of their textual relation, or in terms of their relative levels of finish.60 There is copious evidence of the author’s returning again and again to the text to make changes and alternative versions of phrases, lines, or longer passages; and in the lines where both ‘draft’ and ‘copy’ versions are available, it is clear that each is a different state of the text. Both states include features of ordinatio – paragraph marks, initial capitals, and decorative Lombardic initials – and both show the scribe carefully marking the text with conventional signs and pointing fingers to indicate where additional or substitute text should be inserted. However, all this care produces a difficult, at times unreadable, copy, which can never have been intended for a recipient to own. In relation to the texts of Auchinleck, Derek Pearsall surmises that ‘Behind the extant copies 57 See Sir Ferumbras, ed. Herrtage, pp. xviii–xxvii; Purdie, Anglicising Romance, pp. 186–7. Richard Beadle points out that besides the language, palaeographical details also place the scribe in the south west (‘Set Pieces’). 58 The extant inventory of the cathedral’s library in 1327 includes: ‘Multi alii libri vetustate consumpti Gallice, Anglice, et Latine scripti, qui non appreciantur, que nullius valoris reputantur’. See George Oliver, Lives of the Bishops of Exeter and a History of the Cathedral, 2 vols (Exeter: William Roberts, 1861), I, 301–10 (p. 309). 59 See pp. 44–50; Blaess, ‘L’Abbaye de Bordesley’, pp. 511–18. 60 See Hardman, ‘Bodleian Library, MS Ashmole 33’.

The Middle English Tradition

185

presumably lie rough working drafts provided by the versifiers.’ 61 This is very likely what Ashmole 33 represents – a working draft in several states, prepared on waste parchment and coarse paper quires, marked up for transcription to the final ‘presentation’ copy. In the absence of such a copy, any indication of the possible identity of the intended owner must be sought in the details of the Middle English text itself. Although it has been described as ‘a full and close translation’ of the French source,62 Sir Ferumbras does in fact contain a coherent and sustained pattern of small changes that perhaps reveal something of the interests of its author and expected readers. Not surprisingly, given the clerical status of the assumed author, one set of changes concerns religious matters, as the English version represents Charlemagne as a lover of Holy Church and a generous almsgiver (1896–7, 2558), makes more frequent and precise reference to the rites of baptism,63 inserts the words of the marriage service between Guy and Floripas,64 perhaps echoes a biblical text in Fyrumbras’s expressed wish to convert his father,65 and adds instances throughout the text in which the peers show their proper religious disposition by giving thanks to God. Another set of additions speaks to a readership interested in matters of chivalry, introducing explicit references to courteous behaviour and especially courteous speech, often between knights and ladies,66 to questions of honour or worship, whether among Christians or Saracens,67 and repeatedly characterizing the knights as the flower

61 The Auchinleck Manuscript, introd. Pearsall and Cunningham, p. ix. 62 Smyser, ‘Charlemagne Legends’, p. 84. 63 See Siobhain Bly Calkin, ‘Romance Baptisms and Theological Contexts in The King of Tars and Sir Ferumbras’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 105–19. See also ll. 396–8, 548–9, 828–9. 64 ‘To wham y pliȝte trouþe ȝore | To haue & holde for euere more | On wedlak fre’, lower margin, fol. 77v. 65 He hopes to ‘Brynge him ȝut of his errour | Into þe betere waye’ (5761–2); cf. ‘Scire debet quoniam qui converti fecerit peccatorem ab errore viae suae, salvabit animam ejus a morte’ ( James 5:20, Vulgate text). 66 See lines 1833, 1844, 2038, 2055, 2058, 2070, 2093, 2806. 67 See lines 563, 1645, 1837, 2644, 2837, 3018, 3286, 4955, 5146, 5782.

186

Charlemagne in Medieval England

of their nation or ‘þe flour of Chyualarye’ (1107, 2341).68 There are also enhanced references to the material context of chivalry, such as richly decorated armour and luxurious furnishings,69 and to the details of siege warfare.70 An implied cultural context of this kind, together with the extreme care expended on translating and improving the text, suggests that the presumed fair copy for which the rough draft in Ashmole 33 was prepared might have been a manuscript of some expense and refinement.71 The regional language of the text implies not only that the text was produced in Devon, but that the manuscript would also circulate in the South-West; here a comparison might be made with Sir Gawain and the Green Knight, a contemporary romance in a similarly geographically defined dialect, whose literary sophistication has led to speculative association with a number of great magnates or gentry families with connections in the Cheshire area as potential patrons or recipients.72 While Sir Ferumbras is not in the same league, it might be argued that the combination of its enhanced focus on courtesy and chivalry, with the author’s evident concern to perfect the form of the poem, could equally point to an intended recipient among the noble or gentry families of the South-West.73 As with Sir Gawain and the Green Knight, 68 See also lines 125, 5173, 5537. 69 See lines 553, 623, 701–2, 1090–1, 1339–41, 3028, 3071. 70 See lines 3179ff., 5007–10, 5155–6. 71 Ad Putter makes a similar suggestion in relation to BL MS Cotton Nero A.x: ‘It may well be a clumsy copy of an original de luxe manuscript with illuminations, either a presentation copy or a repertory book commissioned by a magnate of wealth’ (An Introduction to the GawainPoet (London: Longman, 1996), p. 23). 72 See, for example, Edward Wilson, ‘Sir Gawain and the Green Knight and the Stanley Family of Stanley, Storeton, and Hooton’, Review of English Studies, n. s. 30 (1979), 308–16; Putter, An Introduction to the Gawain-Poet, p. 37; W. G. Cooke and D’A . J. D. Boulton, ‘Sir Gawain and the Green Knight: A Poem for Henry of Grosmont?’, Medium Ævum 68 (1999), 42–54; Francis Ingledew, ‘Sir Gawain and the Green Knight’ and the Order of the Garter (Notre Dame, IN: University of Notre Dame Press, 2006). 73 As a further sign of sophistication, Shepherd notes the ‘clever and playful sense of humour’ evident both in Ashmole 33’s modifications of the received narrative and in the occasional inventive drolleries incorporated

The Middle English Tradition

187

the production of a prestigious chivalric narrative in a regional English language (rather than French) might be understood as signalling local pride, a desire to proclaim ancestral connections with the place and tongue of a particular region.

British Library MS Add. 31042 (c. 1440): The Sege of Melayne; Roland and Otuel The two volumes produced by the North Yorkshire gentleman Robert Thornton of East Newton, known as the ‘Lincoln’ and ‘London’ Thornton manuscripts, may be thought of as a personal library of texts catering to the reader’s literary, religious and medical reading needs, which was compiled by the single scribe-owner over a long period.74 Lincoln Thornton is the more famous, on account of its large collection of Middle English romances,75 but London Thornton is perhaps more interesting in the way its arrangement of texts may be seen as serving a particular purpose. John Thompson argues that an analysis of Thornton’s compilation procedures reveals patterns of both thematic and practical concerns behind the ‘disorderly and unsettled nature of the material’ in this volume, and shows how the first booklet was apparently built up from an original ‘core’ pairing of The Northern Passion and The Siege of Jerusalem, with an abbreviated copy of Cursor Mundi added at the start and The Sege of Melayne followed by Roland and Otuel at the end, to produce ‘a recognizable historical sequence’ of texts connected by elaborate scribal colophons and incipits.76 The period thus covered

into initial letters and marginal symbols (‘The Ashmole Sir Ferumbras’, p. 121, n. 30). 74 Ralph Hanna III, ‘The Growth of Robert Thornton’s Books’, Studies in Bibliography 40 (1987), 51–61; John J. Thompson, Robert Thornton and the London Thornton Manuscript (Cambridge: D. S. Brewer, 1987). Robert Thornton’s name appears in scribal explicits etc. in both manuscripts. Thompson gives a full description of the London MS. 75 The texts in Lincoln Thornton are arranged in three large booklets, roughly by genre: romance, religious, medical. 76 Thompson, Robert Thornton and the London Thornton Manuscript, pp. 35, 48. For evidence confirming Thornton’s agency in the compilation of this sequence, see Michael Johnston, ‘Constantinian Christianity in the London Manuscript: The Codicological and Linguistic Evidence of

188

Charlemagne in Medieval England

extends from the Creation to the imagined time of Charlemagne,77 with Christ’s Passion as the central event in human history.78 The texts in the other booklets show some signs of organization, perhaps inherited (as Thompson suggests) from Thornton’s exemplars, but nothing as sustained as this.79 The interpretive agenda behind the sequence is apparently indicated in the wording of the introduction to The Siege of Jerusalem: ‘Hic Incepit Distruccio Ierarusalem [sic] Quomodo Titus & vaspasianus Obsederunt & distruxerunt Ierusalem et vi[n]dicarunt mortem domini Ihesu Christi’ (fol. 50r).80 The poem begins with an account of the Crucifixion and introduces the story of the destruction of Jerusalem as Christ’s reluctant vengeance on those who caused his death: For alle those harmys þat he hade he hastede hym noghte One thaym that velanye to venge þat his veynys braste

Thornton’s Intentions’, in Robert Thornton and his Books, ed. Fein and Johnston, pp. 177–204 (pp. 180–90). 77 For evidence of the remains of quire signatures that imply Thornton’s copy of Cursor Mundi originally began at the beginning, with the Creation, see George R. Keiser, ‘Robert Thornton: Gentleman, Reader and Scribe’, in Robert Thornton and his Books, ed. Fein and Johnston, pp. 67–108 (p. 93). 78 For discussion of this themed sequence, see Phillipa Hardman, ‘Reading the Spaces: Pictorial Intentions in the Thornton MSS: Lincoln Cathedral MS 91 and BL MS Add. 31042’, Medium Ævum 63 (1994), 250–74 (pp. 263–9). 79 Keiser argues for a much more extensive programme, including all the texts as far as the last quire to form ‘a Spiritual History’ in which ‘the Passion is the central matter of most of these works’ (‘Robert Thornton: Gentleman, Reader and Scribe’, pp. 104–5). See also John Finlayson, ‘The Context of the Crusading Romances in the London Thornton Romances’, Anglia 130 (2012), 240–63, for a reading of the compilation as a three-level Christian allegory of national, spiritual and personal life (p. 248). 80 Titles or colophons in other copies of this text, derived from the Vindicta salvatoris, mention only the siege and/or destruction of Jerusalem, but one copy of the couplet version known as Titus and Vespasian is named þe Vengaunce of Godys deþe (BL MS Add. 36983: this manuscript (dated 1442) shares many texts with London Thornton).

The Middle English Tradition

189

Bot aye taryede he the tym if thay torne wolde and lent thaym space þat hym spilt þofe it spede littill. Fourty wynter als I fynde and na faere ȝeris Or he oghte put at that prynce of þat pepill þat hym þose paynes wroghte. (fol. 50v)

Michael Livingston argues that the poem has a specifically anti-war agenda, in that Titus and Vespasian’s war upon the Jewish city in vengeance for the death of Christ is constructed as a just war to end all wars.81 If this is so, the point appears to have escaped Robert Thornton,82 who follows the Siege with another poem that, as Thompson notes, he seems to have considered an ‘adjunct’ to the previous text, giving it a title added in the upper margin to enforce the parallel between them: ‘Here Bygynnys the Sege off Melayne’ (fol. 66v).83 The central action, in which the Christians attempt to retake Milan from the occupying Saracens, is imagined as a war of vengeance, made explicit when Charlemagne’s dream presents the angelic call to arms: ‘þat þou venge alle his [Christ’s] dispyte’ (122), in terms that echo the famous call of the First Crusade to ‘avenge your God, your father, your brother, whom you see rebuked, banished, crucified’. 84 The same theme is evident elsewhere in London Thornton, when the romance of Kyng Richerd þe Conqueroure represents the Lionheart’s mission as ‘to wreke Ihesu’ (fol. 135) and to persevere ‘Till he hadde wonne the holy lande | And slayne the Sowdane with a swerde | And vengede Ihesu Cristis dede’ (fol. 160). The second Charlemagne text, Roland and Otuel, gives every sign of having been selected and perhaps adapted as a companion piece to Melayne: it is another tail-rhyme romance, which features the same Saracen sultan, Garcy; elaborate scribal incipit and explicit titles make 81 ‘The Siege-poet’s answer to the social-political-religious question of whether there is such a thing as a just war is that there was one’; he concludes that the poem is ‘a call to peace and to remembrance’: The Siege of Jerusalem, ed. Michael Livingston (Kalamazoo, MI: Medieval Institute Publications, 2004), TEAMS online edition: Introduction. 82 Similarly, Keiser notes: ‘If the author of Siege intended a nuanced and subtle view of the Jews, Robert Thornton did not recognize that intention’ (‘Robert Thornton: Gentleman, Reader and Scribe’, pp. 95–6). 83 Thompson, Robert Thornton and the London Thornton Manuscript, p. 48. 84 Urban II, quoted by Riley-Smith, The Crusades, p. 23.

190

Charlemagne in Medieval England

clear that it is a story ‘Off Cherlls of Fraunce’, and small details unique to this version of Otinel place the Saracens’ initial acts of aggression against the Christians in Lombardy (instead of Rome), thus mirroring the situation in the opening of The Sege of Melayne. As a further connection, the reference to Ganelon’s future treachery at Roncevaux in the prologue to Otinel is slightly adapted in Roland and Otuel so that it can now refer to the parallel instance of Ganelon’s treacherous behaviour in Melayne.85 However, while this sequence of texts is related by its construction of history as events either leading up to the death of Christ or responding to it, revenge is not the only response the texts call for: the sequence is punctuated by three devotional poems encouraging contemplation on the Passion. The first, ‘A Discourse between Christ and Man’, is placed as a preface to The Northern Passion, inviting the reader to use it as a preparatory exercise before reading the sacred narrative of Christ’s death. It is a complaint from the Cross, based on the verse ‘O vos omnes’ (Lamentations 1:12), in which Christ appeals to sinful Man to behold His sufferings and repent. Man’s reply provides a model response, praying for the grace to think on Christ’s pains at all times. The Northern Passion concludes with a similar prayer, to which is added in Thornton’s copy the following indulgence: ‘and alle þat hase herde this passioun | sall haue a thowsande ȝeris to pardone’ (fol. 50), indicating the conventionally pious context in which the text was designed to operate. Thornton’s sequence of texts thus creates a complex tension between dedicated action and contemplative devotion as the appropriate response to the Passion, and the paired Matter of France romances take their place within this dynamic: two further devotional poems are placed towards the end of the sequence, one after each of the two romances in which Charlemagne battles ‘goddes Enemy’.86 The two poems, like the two romances, are constructed as a pair: both called ‘Cantus’ in a scribal title or explicit; both in the same eightline stanza of which the last line forms a refrain, and to which Thornton 85 Otinel, ed. Guessard and Michelant, 6–15; Roland and Otuel omits the reference to Roncevaux (as does the AN Bodmer version) and increases the number of French dead (from 20,000 or 30,000 to 40,000) to match the 40,000 slain through the treachery of Ganelon in Melayne (169–80, 202–4, 373). 86 Sege of Melayne, 689; Roland and Otuel, 48.

The Middle English Tradition

191

calls attention with unusual double paragraph marks; both focused on the Crucifixion of Christ and composed in elaborate style. The first (which lacks its opening stanza in London Thornton, so whatever title it may have had is missing) is addressed to the Virgin Mary, lamenting the eclipse of her beauty by her grief at the Crucifixion, and concluding with a prayer for her intercession with her crucified son.87 This Marian lyric makes a fitting pendant to The Sege of Melayne, in which Bishop Turpin’s marked devotion to Mary and to Christ crucified plays a central role.88 The second poem, placed after Roland and Otuel, is presented with Latin rubrics that carefully tie it into the sequence of texts, just as is the case with the Siege of Jerusalem, and also link it to the central narrative of the sequence, The Northern Passion. Thornton titles the lyric ‘Passionis Christi Cantus’, begins with the rubric ‘Hic Incipit quedam Tractatus Passionis Domini nostri Ihesu Christi in Anglice’ (fol. 94), and ends ‘Explicit Passio Christi’ (fol. 96).89 It is another poem in the form of a complaint of Christ from the Cross, with the refrain ‘Looke one my woundes, thynke one my passyoun’. The rubrics imply that it functions both as an appropriate closing meditation, answering the prefatory poem placed before The Northern Passion, and as the conclusion of the whole sequence of Passion-centred texts.90 Reading the texts in Thornton’s sequence in relation to each other, it becomes clear that what links them is the typically late-medieval emphasis on the humanity of Christ in his Passion and the feeling response demanded from the reader, whether this response is imaged in Titus’s spontaneous grief on first hearing the story of Christ’s death (Siege of Jerusalem, 181–4), or in Turpin’s literal ‘imitatio Christi’ when he refuses help for his wounds (Sege of Melayne, 1345–50), or in the prayers and exhortations of the three devotional poems. Like the religious texts 87 This poem, with the refrain ‘O florum flos, O flos pulcherime’, is edited by H. N. MacCracken in ‘Lydgatiana 14: ‘O Flos Pulcherrime!’, Archiv 131 (1913), 60–3. 88 For further discussion of this theme, see pp. 386–9. 89 Lydgate’s ‘Complaint þat Crist maketh of his Passioun’, in The Minor Poems, ed. MacCracken, I, 216–21. 90 The remaining texts in the last few folios of the booklet seem to be additional ‘filler’ items: useful and widely available works (Lydgate’s ‘Verses on the Kings of England’ and ‘Dietary’) and brief aphorisms (fols 96–7).

192

Charlemagne in Medieval England

in Lincoln Thornton, they form part of a repertory of resources for Thornton himself, and perhaps a wider local readership among his circle, equally interested in the exercise of lay spirituality.91

Princeton University Library MS Garrett 140 (post-1450): The Sowdone of Babylone This later fifteenth-century, single-text manuscript is ascribed on linguistic grounds to the East Midlands, but offers no evidence of more specific provenance. Carol M. Meale identifies the social status of the intended owner as among ‘the gentry, or professional, or mercantile classes’, and describes the material evidence: The unique manuscript, PUL MS Garrett 140, dating from after 1450, gives some assurance of the modest prestige which attached to its ownership. The volume is of parchment, and no great effort has been made to economise on the use of this more expensive material: the romance is copied in single-column format, in an attractive and competent mixed hand in which secretary forms predominate; the frequent spaces left for rubricated initials have never been filled, but some thought has evidently been given to the lay-out of the text on the page. The work concludes with an explicit written in an elaborate display script, the ascenders of the first line being decorated with strapwork calligraphy. In short, although the manuscript is by no means an exceptional production, it is pleasant to look at, and to handle.92 There is little more to add: owners and readers of the manuscript Sowdone of Babylone have left few traces beyond a conventional ejaculation ‘Jhesu merci, Ladi helpe’ (fol. 33), added in the margin at the point of Ferumbras’s conversion, and later hands have occasionally supplied words perceived as missing. From the evidence described, it is clear that the text was regarded as worthy of some care and expense 91 Thornton has personalized one of the prayers in the Lincoln manuscript, inserting his name for ‘N’ (fol. 177). 92 Carol M. Meale, ‘Patrons, Buyers and Owners: Book Production and Social Status’, in Book Production and Publishing in Britain, 1375–1475, ed. Jeremy Griffiths and Derek Pearsall (Cambridge: Cambridge University Press, 1989), pp. 201–38 (pp. 216, 217).

The Middle English Tradition

193

in the production of the manuscript and in its subsequent preservation, and this perhaps correlates with the signs of innovative literary ambition, learned interests and moral purpose in this Middle English adaptation of the Fierenbras tradition. Despite the widely held critical view of the text as notable chiefly for its crude entertainment value or its ‘wonderful comic register’,93 there is reason to suspect a more serious intention behind the project of producing The Sowdone of Babylone and preserving it in its modestly prestigious manuscript context, and one that is entirely compatible with the reading expectations of the social milieu suggested by Meale. The new prologue, that displaces the chanson de geste’s conventional call for attention and declaration of its matière, addresses itself to God, outlining His plan for creation and Man’s place in it in a solemn, expository manner more typical of the opening lines of a contemporary work of religious instruction, and in the kind of elaborate, polysyllabic style and complex, indeterminate syntax characteristic of Lydgate’s writing.94 As Emil Hausknecht noted, the opening passage ‘may be said to contain the moral of the whole poem’, 95 and it is a moral with an uncompromisingly stern penitential message, as made clear in its key words: ‘synne’, ‘offences’, ‘vengeances’. The history of creation is presented as the repeated thwarting of God’s good intentions, from Man’s first disobedience onwards, and the reader or listener might well expect a catalogue of exemplary tales from biblical or chronicle histories to follow. ‘But for the offences to God idoon | Many vengeaunces haue befalle’ (13–14), the poem observes, before continuing, ‘Whereof I wole you telle of oon, | It were to moch to telle of alle’ (15–16). The ensuing story, adapted from La Destruction de Rome, is thus repackaged as a representative example of divine punishment for human sinfulness. Instead of the French texts’ focus on the loss of the Passion relics at the hands of the wicked Saracens and Charlemagne’s campaign to recover 93 Crofts and Rouse, ‘Middle English Popular Romance and National Identity’, p. 93. 94 See Phillipa Hardman, ‘Lydgate’s Uneasy Syntax’, in John Lydgate: Poetry, Culture, and Lancastrian England, ed. Larry Scanlon and James Simpson (Notre Dame, IN: University of Notre Dame Press, 2006), pp. 12–35. 95 The Sowdone of Babylone, ed. Hausknecht, p. 95. Hausknecht’s own marginal summary of the argument does not quite render the moral of the original (which occupies a single, irregular sixteen-line sentence).

194

Charlemagne in Medieval England

them, the English prologue concentrates entirely on the destruction of Rome, and presents its fall as a consequence of the city’s own sins: While þat Rome was in excellence Of alle Realmes in dignite, And howe it felle for his offence, … And for synne howe it was shente. (17–19, 23)

Rome is not the only great city whose fall is evoked in the poem. Uniquely among the texts of the Fierabras tradition, the Saracen Laban is identified as sultan of Babylon,96 rather than of Spain, and the descriptions of the cargo of the ship sailing from Babylon and bound for the sultan (72–5), and of the many costly materials used in his sacrifices to his gods, call to mind the depiction of the city of Babylon in Revelations 18, where the rich material luxuries of the great city closely match the sultan’s. These added exotic details in The Sowdone of Babylone contribute to the moralizing agenda set up in the revised prologue, and evoke the fall of ‘Babylon the great’ as prophesied and lamented in Revelations 17–18, to form a fitting parallel to the destruction of Rome as portrayed in the romance. The parallel would have been familiar from a traditional interpretation of the biblical reference attributed to St Jerome, equating ‘Babylon the great’ with Ancient Rome;97 and while the implied elision of pagan Rome with the Christian city of Rome might seem problematic, it fits with the 96 Babylon in medieval texts is the name usually given to Cairo of Egypt, as here, but it is identical in form to the name of the biblical Babylon. For further discussion of the title ‘Sultan of Babylon’, see Chapter 5, pp. 335–40. 97 Bibliorum Sacrorum cum Glossa Ordinaria, 6 vols (Venice, 1603), including annotations by St Jerome. Cf. annotations to these verses in the Douay-Rheims translation (1582): ‘Babylon. Either the city of the devil in general; or, if this place is to be understood of any particular city, pagan Rome’; cf. also those on the ten kings prophesied to destroy Babylon: ‘Some understand this of the Goths, Vandals, Huns, and other barbarous nations, that destroyed the empire of Rome’. Medieval exegetes and sixteenth-century annotators were drawing on Early Christian identification of Rome with Babylon, as seen in New Testament usage (1 Peter, 5.13).

The Middle English Tradition

195

sternly moralistic tone of the prologue, and indeed with contemporary arguments that Saracen victories were permitted by God to punish Christian sinfulness. The final lines of the poem link the names of the two cities in a pattern of vengeance: ‘Thus Charles conquered Laban, | The Sowdon of Babyloyne, | That riche Rome stroyed and wan’ (3259– 61). In retrospect, therefore, the romance could in part be understood as a parable of the sinfulness of mankind and the transience of all earthly power: moral themes that feature conspicuously in the typical reading matter of the fifteenth-century ‘gentry, or professional or mercantile classes’ among which Meale places the owners and readers of the manuscript.98 Readers in the second half of the fifteenth century would no doubt have been able to make a further imaginative link between the sack of Rome in the romance and the fall of Constantinople to the Ottoman Turks in 1453, and the widespread fears for the future of Christendom that were felt as a result could only have served to reinforce the appeal of Charlemagne’s fictional conquest of Laban.

Huntington Library MS HM 28,561 (c. 1460): Pseudo-Turpin Chronicle (Turpines Story) The parchment manuscript containing the Middle English PseudoTurpin can be dated on palaeographical evidence to the third quarter of the fifteenth century, and located in the South-West Midlands. Stephen Shepherd describes the manuscript with its ambitious but ‘dialect-style’ decoration in his edition of the text,99 and identifies its commissioner as Thomas Mull of Harescombe, Gloucestershire, whose arms appear in the borders of folios 1 and 88 (p. xix). The precise date of 1460–1 corresponds to the deaths of Thomas in 1460 and of his son William a year later.100 The major contents of the manuscript are a copy of Trevisa’s English translation of Higden’s Polychronicon and a number of 98 See Felicity Riddy’s discussion of gentry and urban reading habits in Sir Thomas Malory (Leiden: Brill, 1987). 99 Turpines Story, ed. Shepherd, pp. xiii–xviii. 100 Shepherd’s dating is based on the incomplete scheme of decoration in the manuscript, assuming that work ceased on the death of the commissioner. However, it should be noted that this kind of decorative tailing off is a common occurrence in manuscripts and might be explained by other hypotheses.

196

Charlemagne in Medieval England

short prefatory items by or attributed to Trevisa (fols 1–319), written by two scribes working in collaborative stints; the remaining two items: a collection of Latin documents on the kings of England, with particular focus on the house of Lancaster (fols 320–25), and the Middle English Pseudo-Turpin (fols 326–37), are both independent productions by different scribes, which might have been later additions to the Trevisa manuscript.101 However, Shepherd argues that the known Lancastrian affiliation of the Mull family would explain the combination of texts. The Latin documents could be read as providing an updated and specifically Lancastrian ‘epitome of English history since the closing date of the Polychronicon in 1387’ (p. xxvi), which takes it up to the middle of Henry VI’s first reign (the latest date mentioned is 1447). The Pseudo-Turpin is a text with Lancastrian interest, chiming with Lydgate’s reference to Henry VI’s descent from Charlemagne in a poem presented to the king at his coronation in France in 1431, and chiming too with the ideological argument behind the Lancastrian/Carolingian compilation of texts and images in BL MS Royal 15 E vi (the Talbot Shrewsbury Book, thought to have been designed for Margaret of Anjou on her marriage to Henry VI in 1445).102 It is also a ‘natural complement to the Polychronicon’, as it is actually referenced in Higden’s text (as translated by Trevisa): ‘who þat woll se more of Charlis lif mot loke of … Turpyn þe archebisshopis bokis’ (p. xxvii).103 Indeed, the text of the Polychronicon makes several references to the PseudoTurpin, most of which relate to Charlemagne himself.104 Beyond the Lancastrian connection, however, one can read a wider argument for the English claim to France, conducted both through the reproduction of genealogical tables to demonstrate the claim (fols 323v, 324v) and 101 As Shepherd notes, no other major Trevisa manuscript contains works not attributed to him (p. xviii). 102 Shepherd points to the connection between the Lancastrian genealogical table (fol. 324v) and the more elaborate version in BL MS Royal 15 E vi (fol. 3), with the presence of Charlemagne text(s) in both manuscripts (pp. xv, lii). For further discussion of the Shrewsbury Book, see Chapter 2. 103 Shepherd notes that two other manuscripts combine a Polychronicon with a Pseudo-Turpin (both in Latin): BL MS Royal 13 D i, and CUL MS Dd. i. 17 (p. xxxiii). 104 Of six references to Pseudo-Turpin in Polychronicon, four are to Charlemagne (see Turpines Story, ed. Shepherd, p. xxxiii).

The Middle English Tradition

197

through the assimilation of English and Carolingian chronicles within the context of world history in a sequence of related texts. All this indicates a readiness to incorporate the Matter of France text within a programme of politically inflected readings of history where the political affiliation of the commissioning or owning family is designed to be on display.105 Here a parallel may be drawn with the earlier combination of British history and Matter of France texts in BL MS Egerton 3028. These Anglo-Norman and Middle English manuscript compilations, in their separate languages, construct a similar combination of British and Carolingian traditions to present an integrated sense of insular history and heritage, effectively appropriating the Matter of France in a cultural analogue to the political claim to the French crown. The programme of cultural appropriation is made clear in these two cases by the manuscript context, but it can be argued that the copying of insular Matter of France texts was in itself an act of cultural appropriation in the fourteenth and fifteenth centuries, claiming the Charlemagne traditions as the inheritance of the English-speaking and insular Francophone subjects of the joint monarchy of England and France as much as of continental French-speaking subjects (though that would have been a nostalgic or obstinately rearguard claim after 1453).

British Library MS Add. 37492 (c. 1475–c. 1500): Firumbras; Otuel and Roland The Fillingham Manuscript, copied on paper by a single scribe106 from an area of the west of England centred on Bristol,107 dates from 105 Not only do the illuminated coats of arms identify the Mull family, but Shepherd speculates that two apparent errors in the translation of the Pseudo-Turpin may be deliberate alterations intended to claim Milo, the father of Roland, as an Englishman and an ancestor of the Mulls, who were descended from Milo, Earl of Hereford (pp. l–li). 106 I agree here with the judgement of Dr Malcolm Parkes, as reported by C. W. Marx in his edition of The Devils’ Parliament and the Harrowing of Hell and Destruction of Jerusalem, Middle English Texts 25 (Heidelberg: 1993), p. 13, n. 6. An alternative perception of three closely collaborating scribes is offered by Mary Isabelle O’Sullivan in her edition of Firumbras and Otuel and Roland, EETS OS 198 (London: 1935), p. xiv. 107 For this localization, see the evidence provided by Richard Firth Green, ‘The Hermit and the Outlaw: An Edition’, in Interstices: Studies in Late

198

Charlemagne in Medieval England

the late fifteenth century.108 It contains a collection of five narrative texts that have, as C. W. Marx observes, ‘some common thematic ground … because of the presence of certain religious themes in the two romances’.109 The manuscript opens with a pair of romances that advertise themselves as Charlemagne texts: Firumbras ends ‘Explicit Kynge Charlys’ while Otuel and Roland begins with a call to hear ‘Off a conquerour | that was y-hote syr Charlemayne’ (3–4). Following Firumbras and Otuel and Roland are, in order: The Eremyte and the Owtelawe, The Devils’ Parliament, and The Myrrour of Mankind. Loss of quires at both ends of the manuscript has left the first and last texts incomplete, and passages from the second and third have also been lost with missing folios;110 however, sufficient remains to allow some investigation of the ‘common thematic ground’ that connects them. The two Fillingham Charlemagne romances have long been recognized as having a particularly obvious religious bent, with their emphasis on Christian conversion and added prayers;111 the tale of the hermit and his brother the outlaw is a lively moral exemplum on the theme of repentance; the Devils’ Parliament is a debate-based narrative presenting the truths of Christ’s incarnation and mankind’s redemption; and the Myrrour of Mankind, more fully described by its title in a contemporary manuscript collection – ‘þe mirrour of vices & of vertues which also

Middle English and Anglo-Latin Texts in Honour of A. G. Rigg, ed. Richard Firth Green and Linne R. Mooney (Toronto: University of Toronto Press, 2003), pp. 137–66 (pp. 144–5). An alternative provenance on the Essex/ Suffolk border is suggested in Devils’ Parliament, ed. Marx, pp. 13–14. 108 For descriptions of the manuscript, see Firth Green, ‘The Hermit and the Outlaw’; Devils’ Parliament, ed. Marx; and Phillipa Hardman, ‘A Note on the Collation of BL MS Add. 37492 (The Fillingham Manuscript)’, Journal of the Early Book Society 11 (2008), 217–21. 109 Devils’ Parliament, ed. Marx, p. 13; Marx argues instead for the texts’ simply reflecting the ‘varied tastes of a compiler’. 110 For the loss of a folio in The Eremyte and the Owtelawe, see Hardman, ‘A Note on the Collation’, pp. 218–20. 111 In a letter dated 17 August 1801 bound in with the manuscript, George Ellis writes to William Fillingham, then owner of the manuscript, of their ‘more or less devotional’ character.

The Middle English Tradition

199

ys clepid þe seuene ages’112 – is a dramatized struggle for man’s soul between the seven vices and virtues. Marx argues that the purpose of the Devils’ Parliament is ‘to reinforce orthodoxy’: The strategy of the text is to show the Devil engaging in recognizable human activities, story-telling and debating, on subjects which are central to Christian doctrine, and therefore to dramatize what is meant by orthodoxy and what is essential to salvation.113 The same purpose may be seen to lie behind the choice of the four texts that accompany it in Fillingham. The Myrrour of Mankind has been identified by Frances McSparran as one of ‘a recurrent set’ of meditative and penitential texts that circulated in the ‘context of lay spirituality’ encouraged by the educational programmes developed in response to the demands of the Fourth Lateran Council (1215) for annual confession and improved religious instruction for the laity.114 The text shows careful attention to these concerns: the debate culminates in a long discussion of repentance, and the narrative is framed by references to the ten commandments, the seven works of mercy and the Creed as the articles necessary for salvation, precisely such essential, orthodox doctrinal items as are stipulated in Archbishop Thoresby’s Lay Folk’s Catechism (1357). Similarly, The Eremyte and the Owtelawe is focused on the necessity for repentance and confession, while the pivotal narrative moment is given particular significance in the context of orthodoxy by its Good Friday setting: As he stode on a Good Fryday He sey moche folke come by the way Barfote they gonne to go. (49–51)

The outlaw questions a woman and learns that they all go barefoot to 112 CUL MS Ff 2.38, fol. 20v. 113 C. W. Marx, The Devil’s Rights and the Redemption in the Literature of Medieval England (Cambridge: D. S. Brewer, 1995), pp. 132, 139. 114 Cambridge University Library MS Ff.2.38, with introduction by Frances McSparran and P. R. Robinson (Cambridge: D. S. Brewer, 1979), p. viii.

200

Charlemagne in Medieval England

church in sorrow for their sins, and resolves to go with her ‘to wetyne what men done thare’ (84). The scene in the church is missing owing to the lost folio, but the context makes clear that the outlaw experiences the traditional custom of ‘creeping to the Cross’ on Good Friday. This liturgical custom was condemned as idolatrous by Lollards as early as the late fourteenth century, as expressed in the text known as The Twelve Conclusions of the Lollards: ‘þe seruise of þe rode, don twyes euery ȝer in oure chirche, is fulfillid of ydolatrie’;115 conversely, veneration of the Cross was specifically required to be promoted in preaching by Archbishop Arundel in his Constitutions and came to be seen as proof of orthodoxy,116 and its central importance in this story of the salvation of a repentant outlaw is thus a clear endorsement of orthodox belief and practice. The same bias towards confirming orthodox religious beliefs and practices can be seen in passages added to the two Charlemagne romances in Fillingham. Firumbras famously ends with a prayer that those who have devoutly heard ‘thys gest’ of the relics of Christ’s Passion, ‘of the spere & the naylys and of the crovn’ (1836–7), should receive God’s blessing and a hundred days’ pardon – an addition that ostentatiously highlights the theme of venerating relics in the inherited Fierabras tradition. Passion relics occupy a similar place to creeping to the Cross in Lollard thought,117 while, on the other hand, relicveneration was endorsed in Arundel’s Constitutions and was invoked to demonstrate orthodox belief.118 Equally offensive to Lollards was the practice of granting pardons (‘neiþer pope neþer bischoppe may graunte 115 Quoted from Roger Dymmok’s Liber contra xii errores et hereses Lollardorum, Cambridge, Trinity Hall MS 17 (1396), in Selections from English Wycliffite Writings, ed. Anne Hudson (Cambridge: Cambridge University Press, 1978), p. 27. The two annual occasions referred to are Good Friday and Easter Sunday (not the feasts of the Holy Cross, as stated by Hudson): see Eamon Duffy, The Stripping of the Altars: Traditional Religion in England, 1400–1580 (New Haven, CT: Yale University Press, 1992), pp. 29–30. 116 Margaret Aston, ‘Lollards and the Cross’, in Lollards and their Influence in Late Medieval England, ed. Fiona Somerset, Jill C. Havens and Derrick G. Pitard (Woodbridge: Boydell Press, 2003), pp. 99–113 (pp. 100, 101). 117 See Selections, ed. Hudson, p. 27. 118 See Aston, ‘Lollards and the Cross’, p. 107.

The Middle English Tradition

201

any pardoun’),119 so this reference to the devout listener’s gaining an apparently unproblematic indulgence, combined with the emphasis on the Passion relics’ centrality in the ‘gest’ of Firumbras,120 retrospectively convert the text into a polemical instrument of orthodoxy, supporting the veneration of relics and the validity of pardons against the threat of Lollard heresy. Otuel and Roland follows Otinel in contrasting Otuel’s miraculous conversion by the Holy Spirit with Clarel’s resistance to conversion, but uniquely adds a speech (quoted pp. 160–1) in which Otuel responds to Clarel’s scornful misrepresentation of Christian beliefs with a summary resembling the articles of the Creed (1331– 42). This outline of the articles of faith (stressing the virgin birth, resurrection and harrowing of hell to counter Clarel’s ‘heretical’ views of the Incarnation and Crucifixion) corresponds to the focus in Thoresby’s Catechism on the Apostles’ Creed as the basic formulation of what the laity had to know. Again, an opportunity afforded by the inherited tradition is exploited to underline the orthodoxy of the representation of the Christian faith so central to this narrative of conversion. In view of this programme of reinforcing orthodoxy in Fillingham, the scribal provenance of the manuscript assumes particular relevance, for as J. A. F. Thomson shows, the Bristol area was a centre of Lollard heresy, with a tradition of heterodoxy that was long established in the area by the later fifteenth century: ‘Even before the end of the fourteenth century, Lollard preachers had been active in the Severn valley, notably in Bristol, and … it seems probable that it was the principal centre from which heresy spread through the neighbouring counties.’ 121 Prosecutions of heretics in Bristol are recorded throughout

119 Quoted from an early fifteenth-century list of alleged Lollard heresies in Cambridge, Trinity College MS B.14.50, fols 30–34, in Selections, ed. Hudson, p. 19. 120 See pp. 297–300 for further discussion of this theme. 121 John A. F. Thomson, The Later Lollards, 1414–1520 (Oxford: Oxford University Press, 1965), p. 20. See also Richard Rex, The Lollards (Basingstoke: Palgrave, 2002), pp. 69–70; Joseph Bettey, The Suppression of the Monasteries in the West Country (Gloucester: Sutton, 1989), p. 2; Joseph Bettey, Morning Stars of the Reformation: Early Religious Reformers in the Bristol Region (Bristol: Avon Local History and Archaeology, 2011).

202

Charlemagne in Medieval England

the fifteenth century122 but, as Thomson notes, they failed to eradicate the heresy: ‘Lollardy had taken too deep a root in the West Country, more particularly in Bristol, for it to be extirpated’ (p. 33). Opposition was also conducted by means of education: in 1435 the bishop of Bath and Wells ordered regular doctrinal instruction of the laity, exactly as laid down by Archbishop Thoresby in the previous century, in an attempt to prevent people’s falling into heresy through ignorance of the Church’s teaching.123 Such a context of prolonged anxiety about the spread of heresy, and awareness of the need to reinforce orthodox religious knowledge among the lay community, seems a close fit with the concerns of the five texts in the Fillingham MS. The selection and, possibly in some cases, adaptation of the texts in line with a project of reinforcing orthodoxy argues for an educational agenda behind the production of the manuscript. Marx assumes a clerical provenance,124 and while this is by no means unlikely, it is equally possible that a lay scribe-owner concerned to maintain and disseminate orthodox beliefs and practices copied the texts for use within a reading circle of domestic and local extent, in a manner similar (though on a smaller scale) to the compiling activities of Robert Thornton or the unknown compiler of CUL MS Ff.2.38.

British Library MS Lansdowne 388 (c. 1475–c. 1500): The Song of Roland The Lansdowne manuscript that contains the unique fragment of the Middle English Song of Roland is a miscellaneous compilation of disparate booklets and papers, formerly in the possession of John Strype (1643–1737), an ecclesiastical historian and biographer. It was part of his large archive of historical materials, now preserved in numerous composite volumes in the British Library. Most of the items he collected relate to the period of the Reformation or afterwards, and material concerning earlier periods is usually in the form of later transcripts; Lansdowne 388 is unusual in that it contains a number of items copied in hands of the late fifteenth century. Its chief contents, however, relate 122 1414, 1420, 1429, 1441, 1449, 1457, 1476, 1499 (Thomson, Later Lollards, pp. 22–46). 123 Ibid., p. 32. 124 The Devils’ Parliament, ed. Marx, p. 14, proposing Clare Priory in Suffolk.

The Middle English Tradition

203

to the martyrologist John Foxe (1516–87), and were probably among the papers lent to Strype by the executor of one of Foxe’s relations, and kept by him after the death in 1701 of the last of Foxe’s descendants. Besides a life of Foxe by his son, and a collection of Foxe’s letters, the compilation includes a transcription in Foxe’s hand of a treatise on the Eucharist, dated 1396, by John Purvey, the Wycliffite scholar and preacher, whose teachings are extensively recorded in Foxe’s Actes and Monuments of 1563.125 It is not possible to know whether the fifteenth-century texts in Lansdowne 388 were also part of the collection of Foxe’s papers or were subsequently bound up with them, but it is at least clear that Foxe had a substantial interest in medieval documents that relate to his major work. The binding of the manuscript, in which all folios have been mounted on guards, makes it difficult to determine the collation of original booklets that have been bound up with collections of separate papers to make the volume. However, while the British Library catalogue entry lists twenty-four distinct items, the first folio provides an early list of contents that names thirteen units, several of which correspond with previous schemes of pagination or item numbering.126 The tenth unit is listed: ‘An old tract of miscellaneous subjects, some in prose, some in verse, beginning wt the Seven Gifts of ye H. Ghost’. It is preceded by ‘Antient Short Annals’ (covering 1066–1240), and followed by ‘The Life of Wm Wickham’, ‘An old Book of ye Consecration of Virgins’, and ‘Vita Caroli 9 Regis Francorum’. All three fifteenth-century units are identified by reference to their antiquity: the ‘Antient’ Latin annals, and the ‘old’ miscellany and ‘old’ book of rituals in Middle English. The list corresponds exactly with the sequence of five discrete manuscript units in the last part of the volume (fols 357–419), if it is accepted (as suggested by Ward and Herbert) that the Song of Roland belongs to ‘the

125 Foxe notes, ‘for so muche as they requyre a longer treatise, we will passe them ouer for this time, mynding to take an other oportunitie by Gods helpe’: John Foxe, The Unabridged Acts and Monuments Online, 1563 edition (Sheffield: HRI Online Publications, 2011), book 2, p. 192; http// www.johnfoxe.org [accessed 25/10/2016]. 126 Independent pagination occurs in: the life of John Foxe (fols 2–51), Foxe’s university letters, etc. (fols 52–147), accusations against the Jesuits (fols 343–56); numbering in the last three items in the volume: ‘No. 7’, ‘No. 8’, ‘No. 9’.

204

Charlemagne in Medieval England

same manuscript (though not in the same hand) as the four preceding articles, all in English’.127 These articles are: (i) a collection of religious enumerations (the seven gifts of the Holy Ghost, the five sins against the Holy Ghost, the eight beatitudes, etc.), fols 368–70v128 (ii) Ypocras (a Middle English version of part of the medical treatise known as Pseudo-Hippocrates, ‘Letter to Caesar’, or Regimen sanitatis, consisting of a text on the ‘four infirmities’, dealing with common maladies of the head, breast, womb and bladder, and another on the four humours), fols 370v–372r129 (iii) Danyellys dremys (a Middle English version of the Somnia Danielis:130 a list of dream topics and their interpretations, described by Steven Fischer as ‘an everyday aid in the divination of real-life experiences’),131 fol. 372v (iv) a religious verse text, ‘Of the holy resurreccione’ (extracted from a Middle English version of La Estorie del Euangelie),132 fols 373–80 Items (iii) and (iv) are incomplete: the Euangelie poem occupies what appears to be a quire of eight folios, and ends with a catchword 127 Ward and Herbert, Catalogue of Romances in the Department of Manuscripts in the British Museum, I, 631. 128 See Margaret Connolly, ‘Preaching by Numbers: The Seven Gifts of the Holy Ghost in Late Middle English Sermons and Works of Religious Instruction’, in Preaching the Word in Manuscript and Print in Late Medieval England, ed. Martha W. Driver and Veronica O’Mara (Turnhout: Brepols, 2013), pp. 83–100. 129 M. Teresa Tavormina, ‘The Middle English Letter of Ypocras’, English Studies 88 (2007), 632–52. The prose introduction in this version of the treatise is apparently unique, and the ordering of the texts is unusual (p. 635). 130 Curt F. Bühler, ‘Two Middle English Texts of the Somnia Danielis’, Anglia 80 (1962), 264–73 (pp. 266–7). 131 Steven R. Fischer, The Complete Medieval Dreambook: A Multilingual Alphabetical Somnia Danielis Collation (Bern: Lang, 1982), p. 7. 132 See ‘The Middle English Evangelie’, ed. Gertrude H. Campbell, PMLA 30 (1915), 529–613.

The Middle English Tradition

205

indicating a missing following quire; Danyellys dremys ends abruptly at the foot of the folio, with none of the scribe’s usual concluding devices, at the end of a sequence of five folios, and is markedly shorter than other comparable texts. Curt Bühler suggests that it may be ‘an extract made by the scribe from a fuller text for his own purposes’ (p. 266), but the fact that it represents a translation from a Latin list of dream topics in alphabetical order ranging from A to C seems rather to indicate that what remains is only the first folio of a longer text.133 All four items are in the same later fifteenth-century mixed cursive hand. The poem known as the Song of Roland is in a similar but distinct hand, and occupies fifteen folios (fols 381–95), with the loss of an unknown number at beginning and end.134 While the codicological evidence is not conclusive (given the state of the binding), the similarities of layout, patterns of rubrication, quality of paper, and level of dirt, damage and damp marks across all five Middle English items suggest that these twenty-eight folios may be the remnants of a modest miscellany manuscript, either produced collaboratively or assembled by an early owner as a ‘made’ compilation. One of the characteristics of the Middle English Song of Roland is its increased focus on religious concerns, which has led to the suggestion of possible clerical authorship.135 It is interesting to note that when Roland alludes to the traditional crusading topos of revenge for Christ’s death, it occurs as an argument supporting the conviction that the embattled Christian knights are assured of heavenly bliss, in language with homiletic overtones: For this day shall we dy, and go no further, but we shall supe ther seintis be many, and crist soulis fedithe, this is no nay. Think he suffrid for vs paynes sore, We shall wrek hem with wepins þer for. (624–8)

133 The thirty-five interpretations include a few anomalous items not corresponding to the Latin, French, German or other English lists cited in Fischer’s collation. 134 The manuscript copy has no title or running head: the title was given by the poem’s first modern editor. 135 Barron, English Medieval Romance, p. 90.

206

Charlemagne in Medieval England

This provides a striking comparison with the accompanying Euangelie poem, a gospel paraphrase that concludes with a sermon based on the Meditations of St Anselm, in which the reader, imagining himself on the Day of Judgement, is exhorted to ‘Thenke also opon þe stoundes | þat ihesu for þe soforde woundes’ as a remedy for sin and assurance of heavenly bliss.136 Further details in the Song of Roland emphasize the knights’ confidence in Christ’s comfort (544, 577, 608, 764, 806–16, 962–3) and imply a reading of the text as exemplifying steadfast Christian trust in salvation. Putting these two longer works with their homiletic colouring alongside items (i) to (iii) suggests that this ‘old tract of miscellaneous subjects’ may once have formed part of a domestic collection of material to serve the secular and spiritual needs of an individual or a household. The careful presentation of the constituent parts of the shorter items, repeatedly using the formula ‘Here suythe’ to direct the reader to each element of the religious lists or the medical treatments, and setting out the topics and interpretations of dreams on separate lines for easy identification, indicates that these were important practical texts, expected to be often consulted;137 the long poems, meanwhile, marked with rubricated divisions for reading, possibly aloud, provide opportunities in narrative for Christian edification. There are signs, too, of common concerns between the practical items and the Song of Roland that can help to construct a more precise sense of the reception of the romance text in the context of this manuscript. It has been noted before that the Middle English Song contains a number of interpolated passages of sententious comment upon the action, either from a character within the narrative or in the voice of the narration,138 and this enhances the text’s potential role as a source of practical wisdom, offering lessons on such topics as friendship, flattery, and good judgement. In addition, the poem deals with the subject of Charlemagne’s ominous dreams in a new way that shares with Danyellys dremys a practical interest in the interpreting of 136 ‘The Middle English Evangelie’, ed. Campbell, lines 1849–50. This section is missing from the incomplete copy in MS Lansdowne 388. 137 Compare the careful layout of similar items in the much larger miscellany volume, CUL MS Ff. 2.38. 138 Stephen H. A. Shepherd, ‘ “I have gone for þi sak wonderfull wais”: The Middle English Fragment of the Song of Roland’, Olifant 11 (1986), 219–36.

The Middle English Tradition

207

dreams: on waking, the king’s first action is to send for ‘the wissest men’, to ask ‘of his dreme hou it be myght’ (107–8), with a long discussion of the possible interpretations and best course of action (109–20). Such concerns, religious, practical, and exemplary, suggest a milieu for the manuscript’s early readership comparable to that of CUL MS Ff.2.38, MS Ashmole 61, and similar vernacular miscellanies: a domestic setting among the urban middle class or provincial gentry families, where it would serve to reinforce traditional, orthodox cultural values and practices.139 Finally, to turn to its possible sixteenth-century reception, if the Song of Roland was in fact found among Foxe’s materials, it might indicate that it was being read in a rather different way. The poem’s representation of Charlemagne’s conflict with the sultan and the defeat of Roland and the peers by the Saracens potentially had relevance for Foxe’s monumental history of the Christian Church. In Actes and Monuments, ecclesiastical history serves Foxe’s agenda, which is to denounce the papacy as ‘principall cause’ of ‘the lamentable ruine of the church’ at the hands of ‘the Turkes and Saracens’.140 As a prime example Foxe uses Louis IX’s crusade of 1248–50, drawing selectively on Matthew Paris’s Chronica Majora to expose the role of Pope Innocent IV, whom Foxe holds responsible not only for the eventual loss of the Holy Land, but for the fatal weakening of the Christian West and the opportunity afforded to the Turks, that ‘Christendome euen to this day may & doth feele & rue’ (p. 321). In his account of the crusade, following Matthew Paris, Foxe notes the ancient animosity between the proud French and the small band of English soldiers; nevertheless, he puts aside national prejudice when he describes the losses inflicted by the Saracens and Turks on the French crusaders, referring to the French as ‘our Christians’ (p. 324). Although he misses no opportunity to blame the pope, Foxe writes with passionate sympathy for the beleaguered and suffering Christians, both in the crusading army and among the Christian populations of the East. In a later section Foxe claims there is plentiful evidence of Turkish persecutions:

139 See, for example, Codex Ashmole 61: A Compilation of Popular Middle English Verse, ed. George Shuffelton (Kalamazoo, MI: Medieval Institute Publications, 2008), Introduction. 140 Foxe, Acts and Monuments, book 4, p. 320.

208

Charlemagne in Medieval England

[I]t is hard to say whether haue bene more cruell against þe Christians, the infidell Emperours of Rome in the primitiue age of the Church, or els these barbarous Turkes in these our latter tymes of the Church now present.141 Such a view might well have been shared by late fifteenth-century readers of MS Lansdowne 388, but, writing in 1567, Foxe could look back over the previous three centuries, covering the loss of the Holy Land to the Mamluk Turks in 1291 and the spread of the Ottoman Empire, culminating in the threat to Vienna in the siege of 1529. Foxe lists the Turks’ many conquests (p. 756), and does not spare the reader gruesome details of the atrocities they are said to have committed against the conquered Christians, stirring up pity in the reader equally for fellow Christians of the Eastern Church in Greece and Roman Catholics in Italy. Foxe’s Protestant convictions occasionally appear in critical marginalia;142 but in general, he accords with Christopher Tyerman’s observation that in the sixteenth century ‘a sense of unity transcended the religious divide in the face of a common enemy, the Turk. … Monks and Calvinists alike sought to extract from crusade histories lessons of faith and devotion.’ 143 This is precisely what Foxe does, mining his sources for accounts of Christian fortitude in response to Turkish atrocities: stories of virtuous virgins and stout soldiers who embraced martyrdom.144 141 Ibid., book 6, p. 756. 142 He notes that an Italian bishop defending himself against Turks with a cross was foolishly trusting in mere wood. 143 Tyerman, God’s War, p. 903. 144 For instance, Foxe gives a ‘memorable example of maydenly chastitie, worthy of all Christians to be noted’, in the story of a young woman captured at the fall of Venetian Negroponte in 1470, ‘who beyng the onely daughter of her father, & noted to be of an exceding singular bewtie, was saued out of þe slaughter, & brought to Mahumete the Turke, to be his concubine: But she denyeng to consent to his Turkish appetite and filthenes, was commaunded therewith to be slayne and murthered, and so dyed she a Martyr, keping both her faith and her body vndefiled vnto Christ Iesus her spouse’ (Acts and Monuments, p. 757). Another memorable example describes the fate of the Christian soldiers of Strigonium in Hungary, who, after suffering humiliating treatment by the besieging Turks in 1543, were offered the chance to save their lives by

The Middle English Tradition

209

If the Middle English Song of Roland was indeed among his collection of documents, it is easy to see how it might have been read by Foxe as an imaginative parallel to these sensational historical stories, figuratively representing the plight of contemporary Christendom at the mercy of the Turks in its portrayal of staunchly faithful Christian knights, designated ‘oure cristyn men’ (828), outnumbered and eventually slaughtered by the Saracens. Such a reading is particularly plausible given that this version of the tradition repeatedly constructs a vision of Roland and his peers as Christ’s chosen, both as victors and, in the end, as martyrs (623–8, 806–16, 822–4, 962–3). And indeed, if it was Strype rather than Foxe who brought the fifteenth-century manuscript together with Foxe’s papers, it is no less interesting in the light it may throw on post-medieval interrelated readings of the history and legends of Christendom.

Texts Extant Only in Printed Editions: The Material Context Caxton’s Prose Romances: The Lyf of Charles the Grete; The Four Sons of Aymon In her study of the persistence of romance, The English Romance in Time, Helen Cooper points out that while ‘the continuities between medieval and Renaissance culture in England are exceptionally strong’, and ‘the transmission of medieval romance to the sixteenth century and beyond’ was very extensive, there were apparent exceptions: Middle English romances of Charlemagne ‘barely make the transition at all’. 145 This is especially surprising in view of the fact that the two Charlemagne texts that took root most firmly in Middle English culture, the chansons de geste of Fierabras and Otinel, each of which was translated from French into English three times in the later medieval period, present stories of individual heroism in battles against the enemy Other not unlike taking service with the sultan, Suleiman I, but who, ‘neither by manasing wordes, nor for any feare of death, could be compelled therunto, of whom certaine which stode stoutly in refusing therof, were presently slayne, whom I may worthely recite in the number and catalogue of holy Martyrs’ (ibid., p. 753). 145 Cooper, The English Romance in Time, pp. 5–6.

210

Charlemagne in Medieval England

those that would prove so lastingly popular in printed romance and chapbook retellings of contemporary heroes such as Guy of Warwick, Bevis of Hampton, or Isumbras. Besides this, William Caxton printed a fashionably updated Middle English prose version of Fierabras, and it was often printed prose romances that provided the bridge for the transmission of medieval romance to the sixteenth century. Yet Caxton’s Lyf of the Noble and Crysten Prynce Charles the Grete (1485), his translation of Jean Bagnyon’s prose romance of Fierebras,146 formed no such bridge: there is no evidence of a second edition, or of any later reprint.147 By contrast, a few years later Caxton published The Right Pleasant and Goodly Historie of the Foure Sonnes of Aymon (c. 1490), his very close translation of the French prose romance Les Quatre Fils Aymoun, which had been adapted from the chanson de geste of Renaud de Montauban in the early fifteenth century and was printed in France about 1482–5. While only one incomplete copy of the first edition of Caxton’s translation survives, it must have had considerable popular success, because further editions were printed by Wynkyn de Worde in 1504 and by Robert Copland in 1554, with reprints in 1582 and 1598: Cooper describes it as one of ‘the most widely known of the early printed prose romances, … both abundantly reprinted and dramatized in the sixteenth century’ (p. 379).148 However, a more specific and elite appeal is implied in Caxton’s stated reason for publishing the translation: that he undertook the work at the request of John de Vere, Earl of Oxford (1442–1513), who wanted it translated as a history ‘of olde tyme passed’ that contains examples of ‘vertues chyvalry’ in the ‘actes and faytes of warre doone and made agaynst the great Emperour and king of Fraunce, Charlemagne, by the .iiii. sonnes of Aymon’ (p. 4). Oxford had played an active role in 146 Hellinga notes that Caxton could have used ‘any one of the five editions’ of Bagnyon’s Fierabras printed in Geneva and Lyon between 1478 and 1485 for his translation (‘From Poggio to Caxton’, p. 92). 147 There is just a tantalizing glimpse in the record of an ‘old’ play text, now lost, called The Twelve Peers of France in the Stationers’ Register for 1586, which Cooper suggests was an adaptation of the same text (The English Romance in Time, p. 414). 148 For details of the dramatic version, see Lost Plays Database, ed. Roslyn L. Knutson and David McInnis (Melbourne: University of Melbourne, 2009), available online at http://www.lostplays.org/.

The Middle English Tradition

211

the Wars of the Roses in opposition to Edward IV (despite Edward’s conciliatory treatment after his father’s execution for treason): imprisoned in the Tower in 1468, he later fled to Margaret of Anjou in France and joined the invasion of England to restore Henry VI in 1470; he fought at Barnet in 1471, and with his two brothers escaped to Scotland and France after the Lancastrian defeat, from where he attempted to stage a return to England. In 1473 he succeeded in capturing St Michael’s Mount off the Cornish coast, where he and his men endured a siege of several months before being forced to surrender in 1474; he was imprisoned at Calais for ten years, but then persuaded the captain of the castle to accompany him in escaping to join forces with Henry Tudor. Oxford fought at the Battle of Bosworth (1485) and was rewarded with high office by Henry VII, which continued under Henry VIII, but his taste for action was undiminished: he saw active service in 1487, 1489, 1492, and 1497.149 With such a career, it is entirely credible that Oxford should have been interested in commissioning a translation of this famous French epic of revolt, which offers a fictional endorsement of his own story in upholding the ‘vertues chyvalry’ and honourable renown of noble rebels in their conflict with the king, and this edifying interpretation would no doubt have appealed to a large readership of Lancastrian sympathizers looking back on the previous decades.150 Caxton’s print of Charles the Grete, which now exists in a single copy, lacks its title page, but the colophon gives the date (1485), while the prologue adds a postscript to its close translation of the original French paratext that indicates the market at which Caxton was aiming with his publication: readers wishing to follow the lead of the alleged ‘persones of noble estate and degree’ (p. 2), who had desired him to produce an English version of the life and history of Charles the Great as the final volume, after the book of King Arthur and the history of Godfrey of Boloyne, to complete the trilogy of Christian Worthies. Caxton 149 S. J. Gunn, ‘Vere, John de, thirteenth earl of Oxford (1442–1513)’, Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004), also online at http://www.oxforddnb.com [accessed 23/08/2016]. 150 As Norman Blake concedes, the argument that Caxton ‘was using them [named patrons] to sell his books … does not necessarily mean that what he wrote about his patrons was fictitious’ (William Caxton and English Literary Culture (London: Hambledon Press, 1991), p. 11).

212

Charlemagne in Medieval England

represents his work of translation as a personal commitment ‘to satysfye the desyre & requeste of my good synguler lordes & specyal maysters and freendes’ (p. 3), creating the sense of a network of literary affinity and implicitly inviting the reader to enter this select circle by investing in the same desire to see the translation. The motivation is constructed as a patriotic and altruistic desire that the trilogy should be available to the English-speaking people to read in ‘our maternal tongue’, for despite the inclusive first-person pronoun, the flattering implication is that ‘we’ already have access to the works – it is others, ‘the moost quantyte of the people … here in this noble royame of englond’, who ‘vnderstonde not latyn ne frensshe’. However, the epilogue puts a somewhat different slant on the genesis of the book. Here Caxton attributes the request for the translation to ‘a good and synguler frend of myn, Maister wylliam daubeny, one of the tresorers of the Iewellys of the noble & moost crysten kyng, our naturel and souerayn lord, late of noble memorye, kyng Edward the fourth’ (p. 251). As the records printed by Herrtage in the introduction to his edition make plain, Sir William Daubeny was indeed clerk of the jewels to Edward IV, but he also held the same office under Richard III in 1483–4, a piece of evidence that Caxton has apparently erased. There is no record of Daubeny holding office under Henry VII, despite the fact that Henry and his wife, Elizabeth of York, were ‘consistently aware of the need to win over Yorkist supporters, including former Ricardians’. 151 In 1485 Daubeny is mentioned, in a document of Henry VII relating to ‘Richard, duc of Gloucester, late, in dede and not of righte, king of England’, as ‘William Daubeney, knight, keeper of the

151 Rosemary Horrox, ‘Elizabeth (1466–1503)’, Oxford Dictionary of National Biography, (Oxford: Oxford University Press, 2004), also online at http://www.oxforddnb.com [accessed 23/08/2016]. Horrox notes: ‘Richard III’s treasurer of the chamber, Edmund Chaderton, was taken into Elizabeth’s service and ended his life as her chancellor.’ As the offices of keeper or clerk of the jewels and treasurer of the chamber were closely related (see Agnes Conway and Edmund Curtis, Henry VII’s Relations with Scotland and Ireland, 1485–1498 (Cambridge: Cambridge University Press, 1932), p. 68), Daubeny and Chaderton presumably held office together under Richard III, but it appears they were not treated equally in the new reign.

The Middle English Tradition

213

juelx with the foresaid pretensed king’. 152 It seems that Daubeny is fatally identified with the past, with the illegitimacy of Richard’s reign; and this perhaps lends credence to the suggestion that he is the same ‘William Dawbeney, late of London, gentleman’ who was later attainted together with Sir William Stanley and others in 1495 for treasonable acts connected with the Perkin Warbeck conspiracy.153 This would explain not only his absence from the conciliatory arrangements in the Tudor households, but also the strange inconsistency between the accounts of the commissioning of the translation in Caxton’s prologue and epilogue: the first mentioning a number of ‘freendes’ with no names, and the second referring only to Daubeny. In light of this friend’s compromised situation in late 1485, Caxton’s inscription of Daubeny’s name in a context of elaborate praise of his former master Edward IV, father of Queen Elizabeth, and the erasure of his service under ‘the pretensed king’ Richard, might have been considered a move towards a reconciliation. It is clear that 1485 was a disruptive year for Caxton, who states that he finished work on his translation on 18 June 1485 in the reign of Richard III, but printed it on 1 December of the same year, in the reign of Henry VII (p. 252). Russell Rutter notes that Caxton is here drawing attention to the ‘disjunction’, and suggests that the printing of Charles the Grete was ‘deferred, probably because of the events surrounding Henry Tudor’s invasion’. 154 It seems likely that the extensive readership implied by the prologue, gesturing towards ‘the moost quantyte of the people’ as potential consumers, as well as ‘persones of noble estate’ and ‘synguler freendes’ like Daubeny, signals Caxton’s anxiety in these uncertain times about securing patronage and recovering his costs;155 152 An order under the Privy Seal to the Treasurer and Chamberlain of the Exchequer, quoted in The Lyf of Charles the Grete, ed. Herrtage (p. xi). 153 Ibid., p. xii. See also S. B. Chrimes, Henry VII (New Haven, CT: Yale University Press, 1999), pp. 84–5. 154 Russell Rutter, ‘Printing, Prophecy and the Foundation of the Tudor Dynasty: Caxton’s Morte Darthur and Henry Tudor’s Road to Bosworth’, in Prophet Margins: The Medieval Vatic Impulse and Social Stability, ed. Edward L. Risden, Karen Rathmell Moranski and Stephen Yandell (New York: Lang, 2004), pp. 123–48 (p. 140). 155 Joerg Fichte argues that ‘the successful sale of Godeffroy of Boloyne [in 1481] encouraged him to publish the other two works [Malory and

214

Charlemagne in Medieval England

it is notable that his prayer at the end of the prologue asks not only for deliverance from ‘dedely synne’ to heavenly bliss, but also that he might ‘come out of dette’ (p. 3). Caxton offers much less discussion of the benefit to the reader in his prologue to Charles the Grete than in the prologues to Morte Darthur and Godeffroy of Boloyne, with their respective stress on exemplary chivalric virtue and zeal for recovery of the Holy Land; but as the translation is explicitly presented as a pendant to those two previous histories, it may be understood to provide the same opportunities. In this respect, it is significant that Caxton’s close rendering of Bagnyon’s original preface makes two additions: the first when he intensifies the account of reading Charlemagne’s great deeds, ‘to the exaltacyon of the crysten fayth and to the confusyon of the hethen sarazyns and myscreaunts, whiche is a werk wel contemplatyf for to lyue wel’ (pp. 1–2, italics added), and the other after Bagnyon’s claim that the book is intended ‘for prouffyte of euery man’, where Caxton adds a note on the table of contents he has provided to enable such profit to be extracted from the work with greatest speed and convenience to each individual user: ‘after the desyre of the redar and herer there shalle be founden in the table all playne the mater of whyche the persone shal haue desyre to here or rede wythoute grete aten[n]dacyon’ (p. 2).

The Taill of Rauf Coilyear how he harbreit King Charlis The single known Scottish Charlemagne romance, The Taill of Rauf Coilyear how he harbreit King Charlis, exists in a unique print copy of thirty-eight pages (NLS H.29.c.9). It has no introductory matter, but is dated 1572, printed by Robert Lekpreuik (or Lapraik) at St Andrews.156 Lekpreuik’s career began in Edinburgh, where from 1561 to 1571 he was the most prolific printer in Scotland. He was the favoured printer for the Kirk, authorized to print parliamentary records, and in 1568, during

Charles the Grete] in 1485, a time when Caxton – in view of the uncertain political situation – had to look for new literary patrons and a wider audience’ (‘Caxton’s Concept of “Historical Romance” ’, pp. 106–7). 156 The spelling of his surname derives from the place-name Leckprevick, near Kilbride, Lanarkshire. See Margaret Scott, ‘Previck and Leckprivick: Onomastic Connections in South-West Scotland’, Nomina 29 (2006), 115–28.

The Middle English Tradition

215

the civil war, was designated ‘our soverane lordis imprentar’. 157 In 1571, Lekpreuik was accused of unlicensed printing, and was forced to flee, first to Stirling and then to St Andrews, where he remained until his return to Edinburgh in 1573; but he was soon prosecuted for printing another unlicensed work, and was imprisoned for several years. His last known work was printed in 1582, and no further records remain.158 Among Lekpreuik’s publications, The Taill of Rauf Coilyear is unusual in its pictorial decoration with two woodcut images.159 On the title page, busts of a prosperously dressed man and woman appear in profile facing each other. The original source of the image, an illustration to Joannes ab Indagine’s Art of Physiognomy, was designed to represent two kinds of nose.160 This woodcut seems to have circulated among the printers working in Edinburgh in the later sixteenth century, who used it indiscriminately as a space-filler.161 The purpose of its reuse by Lekpreuik here, prominently placed on the title-page, may be to represent two of the characters in the narrative: Rauf the charcoalburner and his wife. If so, the focus on the couple who between them 157 T. F. Henderson, ‘Lekpreuik, Robert (fl. 1561–1581)’, rev. Martin Holt Dotterweich, Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004), also online at http://www.oxforddnb.com [accessed 23/08/2016]. See also Emily Wingfield, ‘The Manuscript and Print Contexts of Older Scots Romance’ (unpublished DPhil thesis, University of Oxford, 2010), pp. 243–7. 158 John Morris et al., Scottish Book Trade Index, NLS online resource: http://www.nls.uk/catalogues/scottish-book-trade-index [accessed 23/08/2016]. 159 Decoration in Lekpreuik’s other prints is normally limited to initial capitals and printer’s flowers. 160 Joannes ab Indagine, A Briefe Introduction vnto the Art of Chiromancy and Physiognomy, trans. Fabian Withers (London: J. Day for R. Jugge, 1558); the model was that of Jean de Tournes’ French editions (Lyons, 1549 and 1556). 161 It was used by John Scot in the Works of Sir David Lindsay (1571), printed ‘at the expense of Henrie Charteris’, and in another edition printed by Charteris himself in 1597. Lekpreuik used the cut on the final page of Barbour’s Bruce (1571), another text printed ‘at the expense of Henrie Charteris’. For discussion of the probable relationships between these printers, and between Lekpreuik and John Day, see Wingfield, ‘The Manuscript and Print Contexts’, p. 246.

216

Charlemagne in Medieval England

provide hospitality to the unknown king perhaps implies a reading of the text as a tale of domestic relations, highlighting the theme of the man as master in his own house that is evident in Rauf ’s exchanges with the visitor as well as with his wife. The prosperous appearance of the two individuals accords with the poem’s depiction of the household as a comfortable establishment, where servants stable the horses, and the diners are served on a dais, with excellent dishes of capons and game, and plenty of wine. The other woodcut, placed at the end of the text, shows a man inside a large house with castellated walls and an orchard, pointing a sceptre or offering a baton to another man outside.162 As noted by William Beattie, similar woodcuts appear illustrating King David sending Uriah with a message to Joab in prints produced by Pierre Pigouchet of Poitou,163 and Emily Wingfield argues that the influence of another of Pigouchet’s prints upon the device of the Edinburgh printer Walter Chepman suggests that Chepman may also have appropriated this woodcut image and used it in an earlier edition of Rauf Coilyear, now lost.164 Whether the cut was indeed already adopted as a representation of King Charles and Rauf Coilyear, or whether Lekpreuik was the innovator, in this context it makes an apt visual parallel to the final episode of the narrative,

162 The text is reprinted in Early Popular Poetry of Scotland and the Northern Border, ed. David Laing, rev. W. Carew Hazlitt, 2 vols (London: Reeves and Turner, 1895), I, pp. 212–49. The title page is printed in facsimile, but the final woodcut is omitted, as, according to Laing, ‘it would not have been any ornament, and, besides, has not the slightest allusion to the poem itself ’ (p. 214). 163 William Beattie, ‘Some Early Scottish Books’, in The Scottish Tradition: Essays in Honour of Ronald Gordon Cant, ed. G. W. S. Barrow (Edinburgh: Scottish Academic Press, 1974), pp. 107–20 (p. 115), revising his earlier claim that it ‘has not been found elsewhere’, in The Taill of Rauf Coilyear, printed by Robert Lekpreuik at St Andrews in 1572: A facsimile of the only known copy, intro. William Beattie (Edinburgh: NLS, 1966), p. iv. I am grateful to Emily Wingfield for drawing this revision to my attention. 164 Wingfield, ‘The Manuscript and Print Contexts’, p. 247. Beattie also conjectures that Chepman and Myllar may have produced an early edition of Rauf (The Taill of Rauf Coilyear … Facsimile, p. v).

The Middle English Tradition

217

in which the king bestows upon the newly knighted Rauf the office of Marshal of France (960–5).165 Both cuts may offer some insight on the expected reception of the text: in accord with the stress in Lekpreuik’s title, they focus entirely on the folk-tale element of the composite text and take no note of the interpolated episode concerning the combat between Rauf, Roland and the Saracen. The story of Rauf ’s hospitality and its reward creates comic capital out of the potential for misunderstanding in the gap between courtly etiquette and rustic good manners, but although the integrity of the ‘carll’ awarded him clear moral advantage as master in his own house, the conclusion presents the newly knighted charcoal-burner and his ‘hende’ wife transformed into gentlefolk. Their action in endowing a handsome guesthouse ‘evermair perpetually’ for those in need of shelter (968–72) demonstrates their new status as public benefactors, instead of merely private givers of hospitality as before. The story highlighted by Lekpreuik’s title and woodcuts thus presents a humorous model of the aspirational ambitions that saw many a low-born individual rise to great public office in sixteenth-century courts, and perhaps seek to perpetuate their names with similar acts of endowment of almshouses or schools. When this copy was purchased for the Advocates Library in 1736, it was as part of a composite volume, bound with five texts printed by Wynkyn de Worde.166 It has been assumed that this was a sixteenthcentury Sammelband, rather than a later antiquarian collection,167 though the fact that the pages of Rauf Coilyear show signs of previous sewing for binding either as a single text or in an earlier collection perhaps throws doubt on this assumption. However, the habit of binding small early printed texts together is well attested, and it would 165 In 1514, Robert Stewart, Count of Lennox and Lord of Aubigny by marriage, who had joined the Garde Écossaise of Charles VIII of France in 1493, was indeed created Maréchal de France. 166 Lydgate’s Temple of Glass, Skelton’s Bowge of Court, The Three Kings of Cologne, The Abbey of the Holy Ghost, and Information for Pilgrims unto the Holy Land (The Taill of Rauf Coilyear … Facsimile, introd. Beattie, p. vi). 167 Julia Boffey, ‘ “ The Charter of the Abbey of the Holy Ghost” and its Role in Manuscript Anthologies’, in Medieval and Early Modern Miscellanies and Anthologies, ed. Phillipa Hardman, Yearbook of English Studies 33 (2003), 120–30 (p. 126); Wingfield, ‘The Manuscript and Print Contexts’, p. 251.

218

Charlemagne in Medieval England

be interesting to know which titles, if any, were originally bound with Rauf Coilyear. A useful comparison may be made with an early sixteenth-century manuscript compilation, the Asloan Manuscript (NLS MS 16,500), for although there is no extant manuscript copy of Rauf Coilyear, there is evidence that a cognate if not necessarily identical text once existed in the Asloan MS, and the manuscript context perhaps suggests something of the likely reception of the romance. The Asloan MS is a collection of texts believed to be for the personal use of the scribe, the Edinburgh notary John Asloan, which he copied and accumulated as a series of independent fascicles, and subsequently arranged into two large groupings: religious, historical and moral prose works, followed by verse texts, mostly secular, with a list of contents.168 The list includes several now missing items, one of which is entitled ‘the buke of ralf colȝear’. According to I. C. Cunningham’s analysis of the manuscript, this text is likely to have formed a complete fascicle (p. 129); and so too is the text which was placed next to it in the list: the ‘buke of schir gologruss and schir gawane’, an alliterative romance that is in many ways an Arthurian companion piece to the Charlemagne poem.169 Like Rauf Coilyear, Golagrus and Gawain now exists only in a print copy, which was put out in 1508 by the Edinburgh printers Chepman and Myllar. Cunningham tests and rejects the theory that texts in the Asloan MS could have been copied from Chepman and Myllar prints: there are four extant texts in common, with a possible five others among the lost items (pp. 132–3); but what the parallels do suggest is that Asloan’s procedures in copying texts into separate fascicles resulted in a collection of unbound booklets very similar to the collections of pamphlets that sixteenth-century printer/publishers or purchasers would bind up together into a volume. Thus the context in which Asloan’s ‘buke of ralf colȝear’ was once placed may give us an idea of the kind of texts likely to have accompanied Lekpreuik’s print. The predominant character of Asloan’s collection is its overt 168 I. C. Cunningham, ‘The Asloan Manuscript’, in The Renaissance in Scotland: Studies in Literature, Religion, History and Culture, ed. A. A. MacDonald, Michael Lynch and Ian B. Cowan (Leiden: Brill, 1994), pp. 107–35; Wingfield, ‘The Manuscript and Print Contexts’, pp. 254–62. 169 Wingfield suggests alternatively that Asloan may have copied the two romances in a single fascicle (‘The Manuscript and Print Contexts’, p. 260).

The Middle English Tradition

219

Scottishness, from religious writings by the Scottish divine John Ireland, to tracts on Scottish history, and poems by Scottish writers: William Dunbar, Robert Henryson, Sir Gilbert Hay, Walter Kennedy and Richard Holland are all named in the manuscript.170 All three of the secular literary or historical texts that issued from Lekpreuik’s press besides Rauf Coilyear – Henryson’s Moral Fables (1570), Blind Harry’s Acts and Deeds of the Illustrious and Valiant Champion Sir William Wallace (1570), and Barbour’s Actis and Lyfe of Robert Bruce (1571) – are works by named Scottish writers, and two concern Scottish heroes. All three were commissioned by Henry Charteris, a well-connected Edinburgh bookseller and printer, who was a strong advocate of the need to print Scots literature in Scotland, and was eloquent in defence of the Scots language and Scottish national identity in the prefaces he wrote to his own publications.171 One or more of these works could have been bound in company with Rauf Coilyear, creating a combination that would reflect the Scottish character of the Asloan MS (which also contained a copy of Henryson’s Moral Fables). The thirteen-line alliterative stanza of Rauf Coilyear is shared with two other Asloan texts: Golagrus and Gawain, and Holland’s The Buke of the Howlat.172 Holland’s humorous poem is explicitly concerned with Scotland, whereas the action of the Arthurian romance takes place on a journey to the Holy Land, and Rauf Coilyear is nominally set in France (though, as readers have often observed, the setting of this eccentric romance of Charlemagne is far more like the surroundings of Edinburgh than the environs of Paris). But it is not hard to imagine that the stories of Rauf and Golagrus, robustly defending their sovereign rights in their own home or lands, and told in the same Scots language, might well have appealed to a readership that enjoyed the stories printed 170 The only other writer is Chaucer, wrongly named as the author of Lydgate’s Complaynt of the Black Knight. 171 Joseph Marshall, ‘Charteris, Henry (d. 1599)’, Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004), also online at http://www.oxforddnb.com [accessed 23/08/2016]. 172 For a full list of poems in this stanza and their various rhyme schemes, see Thorlac Turville-Petre, ‘ “Summer Sunday”, “De Tribus Regibus Mortuis”, and “The Awntyrs off Arthure”: Three Poems in the ThirteenLine Stanza’, Review of English Studies 25 (1974), 1–14, Appendix (pp. 12–14).

220

Charlemagne in Medieval England

by Lekpreuik of the heroes of the Anglo-Scottish wars, Sir William Wallace and Robert the Bruce, and that relished the national bias in the presentation of history in the Asloan MS, which includes a text headed: ‘ane tractat of a Part of the yngliss cronikle schawand of þar kingis Part of þar ewill and cursit governance’ (fol. 99).173

Conclusions What the material evidence shows about the Middle English reception of the insular Charlemagne tradition is that the Matter of France was not perceived as different from the rest of the matter of chivalry: these texts can equally embody and transmit the core cultural values of chivalric identity and Christian faith. They appealed to collectors and readers of romance across the social range, from owners of very modest to relatively sumptuous manuscript books. One aspect of their wide appeal was no doubt the adaptability of the narratives, offering opportunities for enhancing different elements, such as exemplary chivalrous conduct, or unequivocally orthodox religious beliefs; or for re-presenting the narrative in a new guise, as a moralizing interpretation of history, or a response to Christ’s Passion; or for appropriating the narrative to serve a new political purpose in times of dynastic and national conflict. However, audiences must also have responded to the timeliness of the subject matter in the fifteenth century, when most of the manuscripts were compiled and the prints published. Narratives dealing with a Christian kingdom threatened by a powerful Saracen empire would allow imaginative engagement with the most pressing anxiety on the international stage: the advance of Ottoman Turkish forces westwards across Europe throughout the century.

173 See M. P. McDiarmid, ‘Rauf Colyear, Golagros and Gawane, Hary’s Wallace: Their Themes of Independence and Religion’, Studies in Scottish Literature 26 (1991), 328–33; Rhiannon Purdie, ‘The Search for Scottishness in Golagros and Gawane’, in The Scots and Medieval Arthurian Legend, ed. R. Purdie and N. Royan (Cambridge: D. S. Brewer, 2005), pp. 95–107.

4 Re-Imagining the Hero: The Insular Roland and the Battle of Roncevaux

T

he best-known narrative of the Charlemagne tradition in the whole of Europe was undoubtedly the material connected with the defeat at Roncevaux, the only defeat the historical Charlemagne suffered during his illustrious career.1 The chanson de geste celebrating the heroic defence of Roland and his companions and the revenge executed by the emperor has dominated critical study of the genre and, since the nineteenth century, has enjoyed particular acclaim as a kind of ‘national’ epic in France itself, appropriated in the early years of modern criticism for a political agenda related to tensions within Europe and specifically between France and Germany. Such political appropriation of this material was not new in the nineteenth century, as we shall demonstrate in this chapter.2 While there was undoubtedly an older oral tradition

1 On the Battle of Roncevaux, see McKitterick, Charlemagne: The Formation of a European Identity, p. 134; Roger Collins, Charlemagne (London: Macmillan, 1998), pp. 66–9. Eastern Frankish leaders, responding to the revolt of Widikund in 782, without Charlemagne’s knowledge, suffered a defeat which Charlemagne avenged (Collins, p. 54; McKitterick, p. 29). 2 For a detailed analysis of political and ideological uses of the Charlemagne material in France, see Robert Morrissey, Charlemagne and France: A Thousand Years of Mythology, trans. Catherine Tihanyi (Notre Dame, IN: University of Notre Dame Press, 2003); for the nineteenthcentury critic Leon Gautier, love of Roland could only serve to increase love of France (‘on ne saurait aimer le Roland sans aimer plus vivement la France’), although he considered the poem to have been written by a Norman (La Chanson de Roland, texte critique accompagné d’une traduction nouvelle et précèdé d’une introduction historique (Tours, 1872), p. vi); see also Di Vanna, ‘Politicizing National Literature’; Joseph Duggan, ‘Franco-German Conflict and the History of French Scholarship on the Song of Roland’, in Hermeneutics and Medieval Culture, ed. Patrick J. Gallacher and Helen Damico (Albany: State University of New York

221

222

Charlemagne in Medieval England

relating these events,3 the oldest extant written version of the chanson de geste is an insular one. The narrative, as noted above in our discussion of the insular reception of the genre, is found in two main strands derived from the chanson de geste and from the Pseudo-Turpin Chronicle, itself an early appropriation of the Roland material, where it is given a particularly religious import and semi-hagiographic tone.

La Chanson de Roland The best known of the all the chansons de geste is often referred to as the Chanson de Roland, the Song of Roland, although it might be more accurate to think in terms of a number of Chansons de Roland. The fundamentals of the narrative in all versions are the same: Charlemagne has been in Spain for seven years and conquered the whole land except Saragossa, which is ruled by the Saracen Marsile, who sends a false message of submission; Charlemagne consequently begins to return to France, leaving his nephew, Roland, in charge of the rearguard. However, Roland’s stepfather, Ganelon, has betrayed them and the rearguard is attacked by a large force, in two waves; only when almost everyone has died does Roland blow his horn to alert Charlemagne, who returns, finds the rearguard destroyed, and avenges his nephew’s death, defeating first Marsile’s army and then that of Marsile’s overlord, Baligant. After the return to France, Ganelon is tried, found guilty, and executed by being pulled apart by four horses. The later rhymed version(s) give a developed role to Roland’s fiancée, Aude, and extend Ganelon’s part in the narrative. The version surviving in Bodleian Library MS Digby 23 is the oldest and the best known today, but does not necessarily contain the narrative which was most widespread in the Middle Ages; there is some evidence to suggest that the Rhymed Roland was better known, and it was this version which provided the intertext for Fierabras.4 Press, 1989), pp. 97–106; R. Howard Bloch, ‘Naturalism, Nationalism, Medievalism’, Romanic Review 76 (1985), 341–60. 3 The oral origins of the genre have been much debated; for a summary of the discussion, see van Emden, ‘ “La Bataille est aduree endementres” ’. 4 Simon Gaunt has pointed out that, while the legend of Roncevaux was widespread, ‘there is little evidence that the Roland as it survives in the Oxford manuscript was widely known in the Middle Ages’ (Gender and Genre in Medieval French Literature (Cambridge: Cambridge University

The Insular Roland

223

The Oxford Chanson de Roland (MS Digby 23) It is the version in the Oxford manuscript which concerns us here as the oldest insular vernacular Charlemagne narrative. While the manuscript dates from later in the twelfth century, the text of the Oxford Chanson de Roland is usually dated around 1100; it is generally considered to be an insular copy of a continental poem,5 though as it is also the oldest surviving version of the Chanson de Roland it is difficult to know how much it had been reworked in the process of being copied. As all the oldest chansons de geste survive only in Anglo-Norman copies we cannot know whether the literary characteristics they share, including brevity of narrative, are because they are insular versions or because they are early texts.6 In the case of the Roland tradition most of the analyses of the relationships between the extant manuscripts place O, the Oxford manuscript, by itself on one side of the stemma; following in particular the detailed work of Césare Segré, it is also now generally accepted that the continental version closest to the Oxford text is that of the FrancoItalian V4 (Venice, Biblioteca Nazionale Marciana MS fr. 4 (225)), the only other surviving witness to the assonanced tradition.7 The Oxford version of the Chanson de Roland is, in many ways, atypical of the genre. While many chansons de geste combine the heroic and the humorous, the Roland is serious throughout. The tidiness and cohesion of the narrative is unparalleled in any other chanson de geste, as Press, 1995), p. 25). This was, however, the version adapted in the Middle English Song of Roland (see below, pp. 235–6); on the Rhymed Roland as intertext for Fierabras, see Ailes, ‘Fierabras and the Chanson de Roland’. 5 While the text abounds in insular graphies, there is no indication of Anglo-Norman affecting the assonance or syllable count – features which would have suggested that it was an Anglo-Norman composition; Short in Duggan, The French Corpus, 1/39 offers two possibilities: that the exemplar of the Oxford Roland pre-dates the fragmentation of the western koinè into Norman and Anglo-Norman, or that it was simply non-insular. 6 See Chapter 1. 7 La Chanson de Roland, ed. Césare Segré, rev. edn with introduction, trans. Madeleine Tyssens, Textes littéraires français 368, 2 vols (Geneva: Droz, 1989), I, 16; cited in The Song of Roland: The French Corpus, 1/33; Duggan, The French Corpus, 1/10–36 offers a summary of the stemmata which have been proposed for the Roland tradition.

224

Charlemagne in Medieval England

is the expert manipulation of formulaic epic discourse. The supporting role given to Aude, Roland’s fiancée, contrasts with the more developed role given to Guibourc in the Chanson de Guillaume and the significant role given to female characters in a number of chansons de geste. By definition the masterpiece of a genre is not a typical example of that genre, and in this case the cohesion and exemplary use of chanson de geste discourse in the Oxford Roland showcase the techniques of the chanson de geste at its best. They are, in part at least, the reason this particular epic has attracted so much critical attention. The presentation of character is also entirely in keeping with the epic mode of narration, though subtler and more expert than in many other surviving texts. While in some ways atypical, the Chanson de Roland exemplifies the dramatic, performative mode of this genre: characters are depicted through their actions, what they say and what other characters say about them; emotions are expressed through external acts (weeping, fainting, laughing) rather than internal analysis; motivations have to be deduced from what is said and done. This does not necessarily mean that characters are shallow or flat, but that the mode of depicting complexity is through external signs and formulaic expressions. In all its French versions it can be read ontologically, with characters representing the good and the bad: the text has a clear villain, Ganelon; a clear enemy, the Saracens; and a hero, Roland. There is also a supporting cast which includes the peers and major Saracens, with Charlemagne as king-pin holding the whole narrative together.8 Yet, in the Oxford version in particular, the hero is not without flaws and the villain not without the potential to have virtues.9 The poet uses formulae to describe both 8 There has been critical dispute over whether Roland or Charlemagne is the central character: see Albert Pauphilet, ‘Sur la Chanson de Roland’, Romania 59 (1933), 161–98; Sarah Kay, ‘Ethics and Heroics in the Song of Roland’, Neophilologus 62 (1978), 480–91. The key point here is the narrative techniques used to depict the main protagonists rather than their relative importance. Roger Pensom, Literary Technique in the ‘Chanson de Roland’ (Geneva: Droz, 1972), favours an ontological reading of the text with Roland representing good and Ganelon embodying evil. Brault describes the emperor as ‘the most complex figure in the Song of Roland’ (The Song of Roland, ed. Gérard J. Brault, 2 vols (University Park, and London: Pennsylvania State University Press, 1978), I, 93). 9 Ailes, The Song of Roland: On Absolutes and Relative Values, pp. 23–77.

The Insular Roland

225

Ganelon and Roland in a way which invites a comparison of the two, and which suggests that Ganelon could have been different. For example, Ganelon is described at his trial as follows:10 Cors ad gaillard, el vis gente color; S’il fut leials, ben resemblast barun. (Roland, ed. Short/Duggan, 3763–4) [His body is robust, his colour handsome; Were he loyal he would certainly seem a good man.]

This both echoes and contrasts with the earlier description of the dead Roland on the battlefield: Cors ad gaillard, perdue ad sa culur. (Roland, ed. Short/Duggan, 2895) [His body is robust, his colour pallid.]

This description of Roland is also found in V4, but not in the Cambridge and Paris manuscripts, which also have this laisse; crucially, no other version contains the echoing description of Ganelon, or indeed such a nuanced presentation of Ganelon, who in the Oxford version presents a noble figure when he appears for trial; all other versions, including V4, have Ganelon twice attempting to flee justice, thus placing more emphasis on negative aspects of his depiction, going so far as to suggest cowardice, far from his presentation in the Oxford text.11 Given the dramatic mode of the genre, with feelings exhibited rather than analysed, the motivations of both Roland and Ganelon have to be inferred from the comments they make, whether in the original argument over whether or not to accept the Saracen offer of submission, in Ganelon’s moment of betrayal, or in Roland’s decision, later reversed, not to sound the horn. That Roland alters his decision, whether because 10 Ibid., p. 66. 11 Gaunt, in Gender and Genre, discusses differences between the Paris manuscript and the Oxford one, noting that the Paris text erases differences between Roland and Oliver and foregrounds Ganelon’s treachery (pp. 37–42). These variations find an echo in those of the Middle English Song of Roland.

226

Charlemagne in Medieval England

of different circumstances or because he has changed his mind,12 in itself suggests that characterization is not simple. Roland is the hero, or at least a hero, but the poet allows others to be critical of him, inviting the audience also to engage with the character in a way which was to change in the Middle English, but which the continental Fierabras exploits. Again, this nuancing of Roland’s character is not found in all versions. Simon Gaunt has compared the Oxford and Paris texts in the depiction of the famous horn scenes: the Paris MS is unique in apparently turning the laisses similaires into a series of laisses in diegetic sequence, but other versions also tone down Roland’s concern with his own reputation as we find it in Oxford version.13 Rhetorical structures and patterns are also used to express Charlemagne’s extreme emotion at Roland’s death. He utters a planctus, over several connected laisses which echo each other (laisses 207–10).14 The rhetorical expression of emotion is part of the discourse of the genre used particularly effectively in the Oxford Roland. Charlemagne’s planctus is retained in all its force in V4 and in some of the manuscripts of the Rhymed Roland.15 Prayers, in particular, in the chanson de geste follow standard patterns. Roland in his mea culpa does not enumerate his sins, yet the poet insists, over three laisses parallèles, on the fact that he cleimet sa culpe (2364, 2368–72, 2382–8).16 The almost liturgical 12 Ailes, The Song of Roland: On Absolutes and Relative Values, pp. 37–46; see also van Emden, La Chanson de Roland, p. 64; Pensom, Literary Technique, p. 137. 13 Gaunt, Gender and Genre, pp. 31–6. Roland’s concern that ‘en dulce France en perdreie mun los’ (O, 1054) is unique to the horn scene in the Oxford text: the line is not found in the Chateauroux-V7 version, or in the Cambridge version (which puts the emphasis rather on the shame Roland’s friends might suffer (397)); even V4, the version nearest to O, expresses concern only that France will lose its reputation (990; 1015 – where Roland says he would rather die than have France be blamed). 14 For a structural analysis of the laments in the Oxford Chanson de Roland, see Paul Zumthor, ‘Étude typologique des planctus contenus dans la Chanson de Roland’, in La Technique littéraire des chansons de geste: Actes du colloque de Liège (Paris: Les Belles Lettres, 1959), pp. 219–35. 15 See the concordance of laisses in Duggan, The French Corpus, 1/85–6. 16 Jean-Charles Payen, Le Motif du repentir au moyen âge (Geneva: Droz, 1967), points out that this is what would be expected at the date of the poem’s composition (pp. 111–16).

The Insular Roland

227

expression of Roland’s penitence, with its allusion to the mea culpa, would be accessible to the audience and enable listeners to relate to Roland’s prayer. Such formulaic expression was also appropriate in the epic mode of discourse. This powerful and dramatic presentation of Roland’s confession and final agony is largely retained in the later manuscripts, so is not unique to the insular version. The use of the rhetorical and the formulaic in the expression of emotion implies the expression of something that is universal and to be shared. The mode of writing is very different from the mode of romance writing, where emotions are expressed in a way which is more individual, with detailed analysis of characters’ feelings. Epic, by contrast, employs the potentially melodramatic externalization of emotion, with tears and fainting. The formulaic discourse of the chanson de geste could be compared to the use of liturgy in worship; the fact that the words are what is expected invites the worshippers to participate in a particular way. The experience is shared. The romance mode puts the emphasis rather on the individual: emotion and prayer are expressed by the individual’s choice of words. Neither mode of expression implies lack of depth or sincerity, but they are related to different narrative modes or genres. As discussed above, some scholars have considered the distinction between romance and chanson de geste meaningless in insular literary tradition, but insular chansons de geste remain part of the wider francophone culture, and retain characteristics of the genre’s discourse.17 Even the abbreviating texts continue to use chanson de geste discourse and modes of characterizing the main protagonists (see Chapter 5 on the abbreviating Fierabras tradition). In the Oxford Roland both Roland and Ganelon, while on one level the hero and the villain, are complex characters, presented to the audience using these dramatic techniques. Charlemagne himself is at the heart of this epic world. The Oxford version of the Chanson de Roland describes Charlemagne at the outset as ‘nostre emperere magnes’, although Charlemagne had never ruled England. The scribe may of course be simply copying his source, and associating himself with the ‘French’, but it might be more than that. There are two references to England in the Oxford Roland, one of which is unique to this version of the narrative. Early in the story the pagan

17 See Chapter 1.

228

Charlemagne in Medieval England

Blancadrin admires Charlemagne for his successes, among which he includes: Vers Engletere passat il la mer salse, ad oés seint Perre en cunquist le chevage. (Roland, ed. Short/Duggan, 372–73) [He crossed the salt sea to England; There he established the head tax for St Peter’s benefit.]

Including England with other conquests, contrary to historic fact, is consistent with the nostre emperere of line 1, and the implication that England was a part of Charlemagne’s empire is found in several chansons de geste.18 Line 373 refers to Peter’s Pence, an annual amount paid to Rome by the English Church. This tribute, or tax, was a cause of friction at times, but also indicated a special relationship between the Vatican and England. It seems that it was already an established, even ancient, custom by the eleventh century.19 Its origins remain obscure, but in the thirteenth century the monks of St Albans (who claimed exemption from the payment) said it had been instituted by Offa, King of Mercia. Our manuscript, about 100 years earlier, is claiming that it was instituted by Charlemagne, thus giving the emperor the credit for England’s particular status with the Vatican. For the scribe/poet/remanieur of the insular Roland this would appear to be an acceptable payment. As Short has said, ‘it is not difficult to read William the Conqueror between these lines, given the reference to the Peter’s Pence tribute which the Conqueror explicitly re-established’.20 Of the other Roland versions 18 It is, for example, explicit in Aspremont, which, as noted above, was also popular in England, as well as in the Pseudo-Turpin Chronicle; see Aspremont, ed. François Suard (Paris: Champion, 2008), pp. 814–17; The Anglo-Norman Pseudo-Turpin, ed. Short, p. 59. As noted in Chapter 1, England is also listed among Charlemagne’s territories in Roland and Vernagu. 19 For a useful summary of the nature of this payment, see Frank Barlow, The English Church, 1000–1066, 2nd edn (London: Longman, 1979), pp. 295–7; see also Frank Barlow, The English Church, 1066–1154 (London: Longman, 1979), pp. 105–6. 20 Short, in Duggan, The French Corpus, 1/286, note to lines 372–3; see also D. C. Douglas, ‘The Song of Roland and the Norman Conquest of

The Insular Roland

229

only the Châteauroux MS and V7 (Venice, Biblioteca Nazionale Marciana MS fr. 7 (251)) have this laisse at all, and they lack the reference to England. The Oxford version is thus aligned with Norman interests in England. The second reference to England is found also in other manuscripts, but with variant readings. When Roland is enumerating his victories, the service he has brought to Charlemagne with his sword, he includes different parts of the British Isles: Jo l’en cunquis e Escoce e Irlande, E Engletere, quë il teneit sa cambre. (Roland, ed. Short/Duggan, 2331–2) [With it I conquered for him Scotland and Ireland, And England, which he considers his bedchamber.]

As noted in the Introduction, a reviser at work on this manuscript, probably in the thirteenth century, also added Wales to the list of conquered territories.21 All the other versions which have this line have it in a slightly different form, including V4, which offers a simpler reading, with England merely one of a list of conquests:22 Si li conquis Ysorie et Irlande,

England’, French Studies 14 (1960), 99–116. Jules Horrent, in La Chanson de Roland dans les littératures française et espagnole au moyen âge (Paris: Les Belles Lettres, 1951), eliminates consideration of this line because ‘il n’était pas connu du Roland primitif ’ (p. 290); see also Jean Dufournet, Cours sur Roland (Paris: Centre de documentation universitaire, 1972), p. 32. 21 See Short’s note to line 2331 (Duggan, French Corpus, 1/312), and the greater detail given in his note to lines 372–3 (1/286). On the competence (or otherwise) and date of the reviser, see Gérard J. Brault, ‘Le Réviseur du manuscrit d’Oxford de la Chanson de Roland’, in Essor et fortune de la chanson de geste dans l’Europe et l’Orient latin: Actes du IXe Congrès International de la Société Rencesvals, Padoue-Venise, 1982 (Modena: Mucchi, 1984), pp. 828–62. Brault considers the revisions could be as early as the late twelfth century, but see Short/Duggan, French Corpus, 1/103–4. 22 Translations of all these lines are ours; translations of the Oxford and Châteauroux/V7 versions are by Duggan and Rejhon.

230

Charlemagne in Medieval England

Et Ingeltere, Sinoples e Garmaise. (Roland, MS V4, ed. Cook/Duggan, 2482–3) [I conquered for him Ysorie and Ireland, And England, Sinople and Garmaise.]23

The Paris MS lacks any reference to Scotland and Ireland, but does include England in its list of conquests, as one among others: Et Engleterre et maint païs estraingne. (Roland, MS P, ed. Rejhon/Duggan, 2650) [And England and many foreign lands.]

This is similar to the readings of the Cambridge text and that of the Lyon MS: Et Angleterre et maint païs lontaigne. (Roland, MS T, ed. van Emden/Duggan, 1959) [And England and many distant lands.] E Engleterre qui siet en lue estrange. (Roland, MS L, ed. Kibler/Duggan, 1523) [And England which is in a foreign place.]

For each of these continental scribes, England is a far-away place, to be classed alongside other strange, foreign lands. The Oxford MS reading of line 2332 is not easy to translate. According to the Anglo-Norman Dictionary,24 ‘cambre’ could mean private domain or treasury: England was the private domain of Charlemagne; again this is clearly a Norman perspective on the status of England. This more obscure reading must be either the lectio difficilior, suggesting an insular or Norman origin for the text, or a scribal alteration by a writer for whom England is not strange but familiar, and for an audience for whom England is home. 23 Ysorie, Sinople and Garmaise are identified in Duggan’s index only as lands ‘conquered by Roland on Charles’s behalf ’. 24 The Anglo-Norman Dictionary is accesible online through http://www. anglo-norman.net [accessed 4/08/2016].

The Insular Roland

231

These references to England as a place which may have been under Charlemagne’s rule, however loosely, are unlikely to have been simply copied unthinkingly by the Anglo-Norman scribe who wrote in his first line ‘nostre emperere’. This phrase, also found in V4 (‘nostre inperer’, line 8), may well have been in his source manuscript, but the scribe retains it. The references to England seem to be deliberately tying England more closely to the empire of Charlemagne. Charlemagne is ‘our’ emperor in that England is not one of the foreign lands but one with a particular (albeit fictive) Carolingian history. The tradition, which seems to have developed early, that the Chanson de Roland was sung at the Battle of Hastings, would reinforce this sense that the Normans in England regarded Charlemagne’s men as their own heroic ancestors.25 If, in the reference to Peter’s Pence, we find a Norman perspective on the Conquest, in the role of Geoffrey of Anjou we seem to have an Angevin one. The house of Anjou has a significant role in all versions of the text: the fictional Thierri d’Anjou is the warrior who takes a stand against Ganelon and defeats his champion. In the Oxford MS Geoffrey is the standard bearer of the king, carrying the oriflamme: Gefreid d’Anjou portet l’orïeflambe: Seint Piere fut si aveit num Romaine, Mais de Munjoie iloec out pris eschange. (Roland, ed. Short/Duggan, 3093–95) [Geoffrey of Anjou carried the oriflamme: It belonged to Saint Peter and was called Romaine But this was changed for the name Monjoie.]

The only other version in which Geoffrey is the standard bearer is V4, where there is no allusion to St Peter.26 Again, this detail may be linked 25 The narratives of William of Malmesbury, Henry de Huntingdon, Gaimar, Wace, Benoît de Sainte-Maure and the Carmen de Hastingae Proelio all recount the singing of a song of Roland at Hastings (Short, ‘Literary Culture at the Court of Henry II’, p. 351); Susan Crane, in ‘Anglo-Norman Cultures in England, 1066–1460’, comments on this as an instance of relatively recently acculturated Norsemen borrowing Charlemagne’s men as their ‘heroic predecessors’ (p. 35). 26 The Lyon MS lacks this line; in the Paris and Cambridge MSS, Charlemagne carries the oriflamme: P, 3559–60 (French Corpus, IV/240);

232

Charlemagne in Medieval England

to William the Conqueror, who, as is well known, invaded England under the vexillum Sancti Petri.27 The reputation of Geoffrey in England was not always positive, but would have been at its height in the years following his marriage to the widowed Empress Matilda in 1128, and, in particular, following the birth of a son, the future Henry II, in 1133.28 A text remanié in England at this time, giving such a role to a ‘Geoffrey of Anjou’, would certainly speak into the political situation in England during the civil war.29 Furthermore, the partisan nature of the text may have been reinforced in a remaniement dating from around 1150: the presence of a Henry, nephew of Richard of Normandy (171), may allude to the creation of Geoffrey and Matilda’s son Henry as duke of Normandy in 1150.30 In the wider French tradition Charlemagne serves as a model for Christian unity, bringing together many peoples under one rule and purpose. As already noted (Chapter 2), Charlemagne’s men in the chanson de geste tradition hail from geographical areas which were not historically under Charlemagne’s direct rule. This presentation of the great emperor as the supreme secular Christian ruler in the struggle against the infidel, the sense that the emperor unites Christendom in its opposition to the Saracen threat and is ‘ours’, emphasizes Charlemagne’s role. Charlemagne is ‘our emperor’ in the AngloNorman Roland manuscript because he is the secular Christian leader. His appropriation as part of insular history is therefore desirable. Peter Haidu has discussed this first line in the context of northern France and T, 2669–70 (French Corpus, V/202); in CV7, 5214, ‘Monjoie’ is the warcry and there is no mention of the oriflamme (French Corpus, III/703). 27 The papal banner can be clearly seen in the Bayeux Tapestry; see Donald Lindsay Galbreath, Papal Heraldry (Cambridge: Heffer, 1930), p. 2. 28 Jim Bradbury, ‘Geoffrey V of Anjou, Count and Knight’, in The Ideals and Practice of Medieval Knighthood, ed. Harper-Bill and Harvey, pp. 21–38. 29 The original ‘Geoffrey of Anjou’ was no doubt the semi-mythical Geoffrey I: see John of Marmoutier, Chronicles of the Counts of Anjou, extracts in English in The Plantagenet Chronicles, ed. Elizabeth Hallam (London: Macmillan, 1988), pp. 19–24; Louis Halphen, Le Comté d’Anjou au XIe siècle (Geneva: Slatkine Reprints, 1974), pp. 6–8. 30 We are grateful to Professor Philip E. Bennett for drawing our attention to this line and the link to Henry. Dufournet, Cours sur Roland, uses this as evidence to date the remaniement before Henry II became king in 1152.

The Insular Roland

233

its relationship with the empire, distinguishing between the ideological and the political. The absence of the pope in the Roland tradition (with the exception of the reference to Peter’s Pence) allows Charlemagne to unite the functions of rex and sacerdos. As Haidu comments, ‘as ideologeme [Charlemagne] represents the fusion of divine devolution and the military component of kingship in the narrative of the poem’. 31 If we transfer the focus from northern France to England the ideological more clearly outweighs the political simply because Charlemagne was not a historical ruler of England. Charlemagne can thus, apparently, combine the function of head of Church and head of State in a conflict, whether literal or ideological, against those who oppose or are opposed by Christianity. Charlemagne in the Oxford Roland, indeed in the wider Roland tradition, has been much studied, and the depiction of the king-emperor in this narrative tradition is almost certainly behind his portrait in both Otinel and Fierabras. For the most part analyses of Charlemagne in the Chanson de Roland do not deal with differences between versions, and much of the depiction of Charlemagne is determined by the narrative. It is quite a complex image, with Charlemagne required to listen to his barons, and sometimes apparently unable to take action; yet he is also a ruler with a special relationship with God: he prays to God, and God communicates with him in dreams and directly enables him to defeat Baligant in single combat.32 All this is largely constant in the different versions. The endings of the texts, however, vary, and the final laisse, unique to the Oxford version, focuses on two aspects of this special relationship with God. In the Oxford MS the whole narrative is open: it begins in medias res and in the last laisse Charlemagne is called by God to undertake another ‘mission’ and go to the rescue of King Vivien, who has been attacked by pagans (3996–97). The battles of Roncevaux, and indeed the whole of Charlemagne’s campaigns in Spain, are thus 31 Peter Haidu, The Subject of Violence: The ‘Song of Roland’ and the Birth of the State (Bloomington and Indianapolis: Indiana University Press, 1993), p. 112. 32 The Song of Roland, ed. Brault, II, pp. 93–6, is concerned largely with the characterization of Charlemagne outside and beyond the Chanson de Roland; on the king and the supernatural, in the context of the sacral function of monarchy, see Dominique Boutet, Charlemagne et Arthur, ou le roi imaginaire (Paris: Champion, 1992), pp. 211–48.

234

Charlemagne in Medieval England

placed in a wider context of an ongoing struggle against the forces of the pagans; as in life so in history, victory against evil is never complete. Charlemagne’s response also presents the cost to the emperor of his commitment to God: ‘ “ Deus”, dist li reis, “si penuse est ma vie”’ (4000). Roland and his companions have paid the ultimate price with their deaths, but in life Charlemagne also continues to pay a price. His response is very human and nuances our reading of the character, reminding us, at the end of the text, that the emperor too suffers, as we have seen in his planctus over the fallen Roland. It is clear that there were some developments of the text after it was imported into the insular context. While it is not possible to prove how much the text was adapted at this early stage, as no detailed evidence of any previous version survives, it is worth noting that some characteristics of the Oxford Roland, notably its brevity and concern for narrative cohesion, are elements found in later Anglo-Norman texts in chanson de geste form.33 This throws into question some of the general assumptions made about the Oxford Roland in its critical history, notably its adoption as a ‘national’ epic, a concept that is surely anachronistic for a time when cultural and linguistic boundaries were not coextensive with polities.

The Song of Roland The Middle English Song of Roland does not, of course, function as an originary text for the tradition of Charlemagne narratives in English in the same way that the Chanson de Roland does for the French cycle du roi. It is simply part of the same fourteenth- and fifteenth-century initiative that produced English reworkings of the later French epics of Fierabras and Otinel and, like them, is of interest as much for what it can reveal about that initiative as for its relation to the original French-language texts behind it. Like the Middle English romance of Tristan, the Song of Roland is a rare narrative witness in English to an obviously familiar story whose main characters were part of insular cultural tradition.34 For example, Chaucer’s Book of the Duchess refers to three exemplary 33 Ailes, ‘What’s in a Name?’ 34 See Phillipa Hardman, ‘The True Romance of Tristrem and Ysoude’, in Cultural Encounters in the Romance of Medieval England, ed. Corinne Saunders (Cambridge: Brewer, 2005), pp. 85–99.

The Insular Roland

235

traitors: Achitophel, Antenor ‘that betraysed Troye’, and (the climax of the triad) ‘the false Genelloun, | He that purchased the tresoun | Of Rowland and of Olyver’ (1118–23). As discussed above, the presence of the Oxford version of the Chanson de Roland (Bodleian Library MS Digby 23), dating from the twelfth century and written in an AngloNorman hand, indicates that the epic poem was known in England long before the composition of this Middle English version in the early fifteenth century; but despite the wide gap of time, as Stephen Shepherd notes, the adaptation is based on a source ‘remarkably close (given the late date of the poem) to that of the Oxford Chanson de Roland’.35 Joseph Duggan, surveying the textual tradition of the Chanson de Roland, gives each version independent status as ‘a literary artifact in its own right’, the product of ‘an interpretive act’, ‘a gesture of homage and renewal’, a refashioning ‘carried out with a purpose, although not always readily apparent to us, in the mind of its maker’.36 This is especially true of the English Song of Roland, which (again, like the English Tristan romance) presents a somewhat eccentric version of the traditional material, and exists in a unique, incomplete copy. However, despite its imperfections, the text still allows us to investigate the ways in which the epic narrative tradition has been refashioned in this late-medieval version.37 The surviving text, which lacks an indeterminate number of folios at both beginning and end, covers material roughly equivalent to laisses 35 Shepherd, ‘The Middle English Pseudo-Turpin Chronicle’, p. 27. 36 The French Corpus, I/38, I/5. 37 Early criticism of the Song of Roland was skewed by unfavourable comparison with the Oxford Chanson de Roland (Smyser, ‘Charlemagne Legends’, p. 96; Barron, English Medieval Romance, pp. 90–1), despite evidence (as Smyser and Barron point out) that it combines details from the earliest tradition with material from later versions, and also borrows from the Pseudo-Turpin Chronicle. It has since been argued that this shows purposeful adaptation of the original materials into a new, different work (Shepherd, ‘ “ I have gone for þi sak wonderfull wais”’ ; Susan E. Farrier, ‘Das Rolandslied and the Song of Roulond as Moralizing Adaptations of the Chanson de Roland’, Olifant 16 (1991), 61–76; Phillipa Hardman, ‘Roland in England: Contextualizing the Middle English Song of Roland’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 91–104).

236

Charlemagne in Medieval England

54 to 128 in the Oxford Roland, from Ganelon’s return to Charlemagne with news of the Saracens’ purported submission, to Roland’s belated readiness to summon aid at Roncevaux: just over a quarter of the complete epic. However, the section of the English poem which remains shows that it involves major rearrangements of the inherited material, and that one consequence of these structural alterations would probably have been a significant abbreviation of the Roland narrative, omitting the episodes after the death of Roland.38 In line with this restricted scope, the text demonstrates a comprehensive programme of rewriting which serves to enlarge and enhance the role of Roland himself as central character and heroic leader.39 Thus the English poem, far more markedly than the French chanson de geste, focuses upon the person of Roland as its hero and the Battle of Roncevaux as its action, producing just the kind of condensed and streamlined narrative described in Mehl’s paradigm of the shorter Middle English romance.40 While the continental development of the Roland tradition in the rhymed versions is characterized by extension and amplification, the insular witnesses, both the early Oxford text and the fifteenth-century English version, can justly be described as ‘laconic’,41 suggesting that in this respect they perhaps share a common insular aesthetic. However, in other respects the English poem demonstrates the pressure of later medieval expectations, both in literary form and style and in wider cultural concerns. As a brief example of its divergence from the epic tradition, the character of prayers uttered in the Song of Roland is more in line with the mode of romance writing as described above (pp. 226–7). Roulond is presented as praying spontaneously in response to particular occurrences: when he first sees the size of the Saracen army, instead of uttering an epic credo and asking for vengeance on Ganelon,42 Roulond prays for divine guidance in the coming battle to 38 See Hardman, ‘Roland in England’, pp. 102–3. 39 Ibid., pp. 94–7. 40 These changes bring the narrative of the Song of Roland closer to the pattern of Middle English romances of Gawain, where one pre-eminent knight, known to all as representative of the court’s reputation and as the embodiment of its values, enacts the central narrative role. 41 The Song of Roland: Translations, trans. Duggan and Rejhon, p. 9. 42 This prayer, not present in the Oxford version, is found in the Châteauroux-V7 version, lines 1544–62.

The Insular Roland

237

ensure the salvation of the Christian knights’ souls (604–14).43 When the Christians slaughter the Saracens without the loss of a single man, he leads them in thanksgiving for their miraculous preservation: ‘we ought to worshippe god myche of his grace.’ | … | euery man tok of his helme & lukyd on hie, lift vp ther hondis and thankid crist, that he sauf and sound defend hem hase. (810, 814–16)

This whole episode has no equivalent in the inherited tradition; the English poem has substantially rearranged and altered what is found in the French texts at this point, to produce a scene of devout humility in victory that has striking echoes of the reaction of Edward III to the French defeat at the Battle of Crécy as reported by Froissart,44 or of the reaction of Henry V to news of his victory at the Battle of Agincourt, as documented in fifteenth-century accounts (and immortalized in Shakespeare’s Henry V).45 If the Song of Roland is here tracing an imaginative connection between the Charlemagne tradition and the successes of the war with France, it can be paralleled in Robert Thornton’s manuscript collection, which brings together romances of Charlemagne with a carol celebrating Henry V’s victory.46 As we have argued, the versification of the Song of Roland, with its unstructured combination of alliterative and rhyming features, may be understood in the context of other poems’ experimental responses to the formal characteristics of Old French epic writing (see pp. 96–7). A further element of epic discourse is also mirrored in the English adaptation: the claim to historical truth, referring to corroborating sources in preserved charters and documents.47 The poem asserts 43 This prayer may be based on Oliver’s despairing speech in the Châteauroux-V7 version, lines 1900–2. 44 John Froissart, Chronicles, trans. John Bourchier, Lord Berners (1523–5), repr. edn, ed. E. V. Utterson, 2 vols (London, 1812), II, 159. 45 Anne Curry, The Battle of Agincourt: Sources and Interpretations (Woodbridge: Boydell Press, 2000). 46 ‘The Rose of Ryse’, in The Early English Carols, ed. R. L. Greene, 2nd edn (Oxford: Clarendon Press, 1977). 47 See, for example, the Oxford Chanson de Roland, lines 1683–5 and 2095–7.

238

Charlemagne in Medieval England

the historicity of its material: ‘Itt is wretyn in storis to remembre euer’ (252). However, the Song of Roland draws not only on multiple versions of the epic Roland to create its own rewriting of the story,48 but also turns to the Pseudo-Turpin Chronicle to supplement the narrative where the epic and chronicle traditions diverge in their account of Ganelon’s return with Saracen treasure. It seems that the translator/adaptor may have appropriated aspects of chronicle style as well as narrative material to form the newly imagined work. For example, each stage of the action is introduced with connecting temporal phrases (‘Then’, ‘When’, ‘And whils’, ‘With that’), establishing the correct order of the unfolding events. The opening words of many chapters in the PseudoTurpin Chronicle are similar (‘Tunc’, ‘Postquam’, ‘Deinde’), and are all preserved in the respective vernacular in the Old French and Middle English translations.49 This is, of course, a typical feature of chronicle style, used to emphasize the chronological accuracy of the narrative of events,50 and Malory also makes extensive use of the technique in the Morte Darthur.51 Edward Kennedy suggests that by echoing the syntax and style of Middle English chronicles such as the prose Brut, Malory ‘wanted his readers to recognize the similarity of his book to

48 Material from the rhymed Roland tradition is found in several episodes, notably Gauter’s expedition (322–67) and the Saracens’ preparations for battle (871–920). 49 Historia Karoli Magni, ed. Meredith-Jones; The Old French Johannes Translation, ed. Walpole; Turpines Story, ed. Shepherd. 50 See, for example, the fifteenth-century ‘continuation’ of the prose Brut, known as Warkworth’s Chronicle, in which passages of connected narrative are differentiated from simple annalistic style by heavy use of temporal phrases such as ‘And when …’, ‘And thenne …’, ‘After that …’ (A Chronicle of the First Thirteen Years of the Reign of King Edward the Fourth, by John Warkworth, D.D., ed. J. O. Halliwell, Camden Society (London, 1839), p. 3). 51 Edward Kennedy claims Morte Darthur is ‘stylistically … similar to what readers of English prose chronicles would have known’: ‘Sir Thomas Malory’s (French) Romance and (English) Chronicle’, in Arthurian Studies in Honour of P. J. C. Field, ed. Bonnie Wheeler (Cambridge: D. S. Brewer, 2004), pp. 223–34 (p. 231).

The Insular Roland

239

the chronicles’ which it was intended to replace.52 Both Malory and the author of the Song of Roland can thus be seen as rewriting their French sources – whether prose romance or chanson de geste – to cater to their fifteenth-century readers’ interest in historical narrative and to reflect something of the chronicle form with which they would have been familiar. In a complementary adaptation, the Song of Roland substantially rearranges the order of events as found in the chanson de geste, to produce a revised story with a linear structure and more transparent operations of cause and effect. This kind of restructuring is not unusual in Middle English romance – indeed, Malory’s treatment of his French sources offers a notable example of the process by which discrete scenes are detached from their contexts in the sources and reconnected as a consecutive narrative.53 Similarly, Helaine Newstead points to ‘the skilful and concise management of the action’ in the stanzaic Le Morte Arthur (adapting the French Mort Artu) produced by the English poet’s omitting digressions and concentrating on ‘the chain of events linked by cause and effect’.54 A good example of this process in the Song of Roland is in the treatment of the Gauter episode. The two oblique references to the excursion of Gautier de l’Hum and the death

52 Ibid., p. 233. See also Thomas H. Crofts, ‘Degrees of Veracity in the Morte Darthur: Elements of Malory’s Chronicle Style’, Journal of the International Arthurian Society 1 (2013), 120–39. 53 This process is described in detail by Eugène Vinaver on pp. lxiv–lxxiii of the introduction to his edition, The Works of Sir Thomas Malory; it is summed up by, for example, John Lawlor: ‘The essential structural difference between the French Arthurian prose cycle and Malory’s work may be expressed as the difference between complex interweaving and a more sequential treatment’ (‘Introduction’ to Sir Thomas Malory, Le Morte D’Arthur, ed. Janet Cowen, 2 vols (London: Penguin, 1969), I, xiii). Cf. Terence McCarthy: ‘[Malory] manipulates several texts at once, selecting from them … and rearranging the chronology to suit his purpose’ (‘Malory and his Sources’, in A Companion to Malory, ed. Elizabeth Archibald and A. S. G. Edwards (Cambridge: D. S. Brewer, 1996), pp. 75–95 (pp. 86–7)). 54 Helaine Newstead, ‘Arthurian Legends’, in A Manual of the Writings in Middle English 1050–1500, I: Romances, ed. Severs, pp. 38–79 (pp. 52–3).

240

Charlemagne in Medieval England

of all his men in the Oxford Roland (800–13, 2040–53)55 are expanded in the Châteauroux–V7 version by the addition of a passage explicitly describing the battle between Gautier and the Saracen king Aumaris (2292–327), creating an extended interlaced narrative. The English poem extricates the material from these three discrete sites and re-forms it into a continuous narrative: Roulond, aware of Saracen forces in the vicinity and suspicious of Gwynylon’s involvement, sends Gauter and 10,000 men ahead on reconnaissance, but they encounter Amaris and his 40,000 Saracens and engage in a fierce battle in which the Christians are all killed except Gauter; heavily wounded and bitterly grieving, he rides back to Roulond and the peers to warn them of the danger and to confirm Gwynylon’s treachery (310–69). However, it is not simply a rearrangement of the narrative sequence: the English poem has created new significance for Gauter’s story by framing it within added passages that contextualize Roulond’s actions and demonstrate his good leadership, as he assesses their situation, isolated in hostile territory and possibly betrayed to the enemy (303–14), and encourages the knights with thoughts of victory and renown, despite his sorrow for the loss of their comrades (370–9). The episode is also relocated in the larger narrative structure, which has consequences for the meaning of the story as a whole. The French epics set Gautier’s mission (to patrol the heights for Charlemagne) at the start of Roland’s undertaking as the peers are volunteering to join him, and Gautier returns at the end of the Battle of Roncevaux when he, Roland and Turpin, all badly wounded, are the only survivors; in the Châteauroux–V7 version, the tragic outcome is heightened by Roland’s disappointed hope of support from Gautier’s (lost) 10,000 men (3442–80). In the English poem, however, Gauter’s mission is delayed until after Charlemagne has left and is on his way to France, and after the account of the Saracens’ preparations for attack, with Amaris’s request to lead the vanguard in the hope of meeting and killing Roulond (234–82). In the French versions, it is Marsile’s nephew Aelroth who makes this request, but, following insular practice, the Middle English text has simplified the narrative by conflating two individuals: Amaris is identified as the Soudan’s nephew both here and when he reports his victory over Gauter (266, 474). As a result of all these alterations, both 55 For the equivalent passages in the Châteauroux-V7 version, see lines 1192–1216, 3420–80.

The Insular Roland

241

Gauter’s mission and Amaris’s defeat of his 10,000 men are read in the specific context of the Saracen threat to Roulond himself; and when Gauter returns with news of the slaughter of all his men, it constitutes not an after-echo (as in the French tradition) but an ominous prefiguring of the catastrophe still to come at Roncevaux. A proleptic narrative device of this kind presupposes for its full effect an existing familiarity with the story matter, and such devices are frequently used in this poem to create a sense of tragic inevitability. For example, the blowing of Roulond’s horn (the most iconic moment in La Chanson de Roland) is here anticipated in a way that assumes the reader or listener is aware of the epic tradition and will recognize the unique network of foreshadowing allusions inserted in this text. First, as he leaves the peers, Charles explicitly forbids any blowing of horns unless to summon his help against the Saracens: he bad no bern be so bold vpon mold, bugle to blow, ne beme to soun, but he the sairsins se all redy boune, ‘And he the hethyn se, and help wold haue, lowd cast vp a cry, and hie vs hym to saue’, and they grauntid so for to do. (236–42)

Next, to calm Charles’s fears over the long delay of the rearguard (a new detail added in this version to explain his anguish), Gwynylon constructs an elaborate fiction about Roulond’s recreational hunting, adapting material relocated from much later in the French texts: your knyghtis behind haue som bores fond, or among the holtis i-herd ryll som hertis; then will Roulond rid among the cleves, he will fell of the fattest, & you flesche bring; ye know þat he louythe well hunting. he will do his will thoughe it to harm turn. (419–25)

This argument effectively pre-empts the following discussion among the knights (433–7) as to whether or not Roulond should sound his horn for aid, by undermining his credit if and when he were to do so. Later, when Oliver asks Roulond to blow for help, recalling the terms

242

Charlemagne in Medieval England

of Charlemagne’s ruling (526–33), the poem replaces the conventional repetition-with-variation of the laisses similaires and the stalemate they produce in the Chanson de Roland with cogent reasoning as Roulond successfully persuades first Oliver and then all the peers that they have no need of help: I se my scheld shyn hole, no pecis out: Thy helme and thy hauberk withouton dout. but our aray be brok, tym it wer aftur help to blow, now I the swer. (537–40) haue ye broken eny bone, or eny harm tid? may ye schew in your sheld eny strokis wid? Is not your compony hole as they come? … blow neuer horn for no help then. (561–3, 567)

As the ‘dredfull day’ (1024) wears on, the English poem inserts a new scene. Roland in the French texts sees so many of his men lying dead that he belatedly proposes sounding the horn to call Charlemagne; in the English poem, Roulond is successfully harrying the enemy when he sees another great Saracen army approaching, ‘vnfought and frech’ (1031), against his own men, who are ‘but few, and hathe fought long, | our horse wery, and we not strong’ (1041–2). This stacks the odds even higher against the Christian knights and gives Roulond good reason to reassess their situation and change his mind about summoning aid. The coherent narrative plotting creates a growing sense of pathos around the anticipated moment when Roulond will blow to summon Charlemagne (although the manuscript copy does not extend to this point), and contributes to a reorientation of the inherited story, away from its focus on Roland’s heroic démesure and towards a greater emphasis on his role as good leader, whose tragic inability to save his men is entirely caused by Gwynylon’s treachery. Gwynylon’s role in the Song of Roland is given equivalent development by the poem’s constant reference to his treachery, naming and blaming him as the evil cause of the peers’ plight even more often than in the French texts, with other added details such as Charlemagne’s assertion that Gwynylon acts out of hatred towards him as well as

The Insular Roland

243

Roulond,56 and the Soudan’s indication, when instructing his troops, that the Saracen strategy has been masterminded by Gwynylon: ‘tak ye no trewes, thoughe ye might, | for gift ne garison as gwynylon hight’ (261–2). Again, innovations in the narrative structure support the English poem’s emphasis on Gwynylon’s treachery. First, material taken from the Pseudo-Turpin Chronicle (1–3, 28–30, 59–76) is interpolated into the account of Gwynylon’s return from Saragossa, to heighten the seductive appeal of his ‘flatring speche’ (6) with the addition of the fair ladies and good wine sent for Charlemagne’s enjoyment by the Soudan. Secondly, where the French texts restrict comment on Ganelon’s treacherous bargain with the Saracens to the omniscient narrative voice,57 in the English poem Gwynylon’s treason is openly suspected by all the barons as soon as Charles mentions his fears for the peers’ safety. At this point, the episode of Ganelon’s trial by combat is relocated from the end of the narrative and, considerably adapted, is placed here, before the Battle of Roncevaux. Unlike the French tradition, where, after the battle, Ganelon is brought from prison to answer the charge of treason, here Gwynylon reacts angrily to the barons’ speculative accusation (392–400) by issuing a formal challenge before the king, with all his kin to support him. He volunteers to fight anyone who calls him traitor or accuses him of taking anything from the Saracens ‘eny harm to done’ (408–14), and swears before God that he made no promise to the Soudan (417–18). The confrontation is simplified, as the feudal nuances that colour the barons’ charges and Ganelon’s defence in the Chanson de Roland are all stripped out, and the roles of Pinabel and Thierry omitted.58 The focus here is exclusively on Gwynylon, who fiercely denies treason and assures Charles there is nothing to fear: ‘For harme of hethen ne dred you neuer’ (427). At this early point in the narrative, while the reader or listener knows of Gwynylon’s betrayal and perjury, Charlemagne does not, and he is represented as reluctantly forced to

56 ‘For thou louys to slee þat I loue best; | And hym thou hatist, and me next’ (155–56). This is contrary to the case in the French tradition. 57 That is, until after Oliver sees the massed Saracen forces awaiting the rearguard (Oxford version, 1024–5). 58 A somewhat similar simplification occurs in Otuel and Roland: see p. 254.

244

Charlemagne in Medieval England

reconcile the two opposed factions (428–32),59 creating a sense of helplessness in the face of the looming tragedy which is even stronger in this Middle English narrative than in the Oxford version.60 All this adds to Gwynylon’s menacing power in a highly charged political situation that is in some ways reminiscent of the final part of Malory’s Morte Darthur. As Helen Cooper notes, Malory ‘makes the threat of disaster more of a constant presence’ in his own work than in any of his sources, a topic that she sees as typical of the English prose romances of the fifteenth century.61 It may be no accident that not only the Middle English prose Pseudo-Turpin Chronicle and Caxton’s Charles the Grete, with its integral translation of the Pseudo-Turpin narrative, but also the two verse adaptations of the disastrous story of Roncevaux, the Song of Roland and the latter part of Otuel and Roland,62 are all in their extant form works of the late fifteenth century, their writers perhaps drawn to the material’s ‘bias towards disaster’ with its apt reflection of the ‘disordered present’.63 In the Song of Roland, changes to the inherited tradition create a narrative of knightly fellowship tragically destroyed by an insider-traitor: a pattern that has an obvious parallel in the destruction of the fellowship of Arthur’s Round Table. Rather than the mutual bond of Roland and Olivier, here it is the fellowship between all the peers that is highlighted. Roulond begins his first speech to the peers with the words: ‘we be 59 The English poem is perhaps adapting the detail of the barons’ fearful capitulation in the French texts: see Oxford version, lines 3793–817; cf. Châteauroux-V7 version, lines 7870–6. 60 Charlemagne’s ability to act in the Oxford text seems limited by legal consideration, see Dominique Boutet, Charlemagne et Arthur, p. 87; on the dual aspect of Charlemagne’s persona: l’exercice personnel de l’autorité and his need to act with the support of his Council, see ibid, pp. 132–3. 61 Helen Cooper, ‘Counter-Romance: Civil Strife and Father Killing in the Prose Romances’, in The Long Fifteenth Century: Essays for Douglas Gray, ed. Helen Cooper and Sally Mapstone (Oxford: Oxford University Press, 1997), pp. 141–62 (p. 146). 62 The section of Otuel and Roland, taken from the Pseudo-Turpin Chronicle, that deals with the material of Roland and the Battle of Roncevaux is given a separate incipit: ‘Here Bygynnyth a Rewful tale | How Rowlond deyde at rouncyuale’ (1976–7). 63 Cooper, ‘Counter-Romance’, pp. 146, 149.

The Insular Roland

245

fellos and frendis’ (304); Gauter sacrifices himself ‘to warn his felos all’ (352); and when Roulond grieves to see the impossible odds they face, it is ‘not for his own sak he soghed often, | but for his fellichip þat he most lovyden’ (600–1). The elegiac note here parallels Arthur’s lament for the inevitable loss of his fellowship, and although the Song of Roland breaks off before the event, the Chanson de Roland provides the opportunity in Charlemagne’s planctus over Roland’s dead body for an elegy to parallel Ector’s lament for Lancelot. Charlemagne’s planctus expresses certainty that Roland is now among the saints in glory, and the French epic depicts angels carrying his soul to paradise at the point of death. While it lacks the ending, the English poem anticipates this heavenly reception in Roulond’s assurance to his knights that they are about to join the fellowship of the saints: For this day shall we dy, and go no further, but we shall supe ther seintis be many, and crist soulis fedithe, this is no nay. (626–8)64

Shortly before this speech, Roulond, alone, prays to God (in the expectation that they will all die in the coming battle) to teach them how ‘our soulis today to send to thy blise riche’ (608). This consciousness of battle as a pathway to salvation is reiterated several times in the course of the fighting, as the knights ‘set þer dyntis | In the worship of hym that fedithe seintis’ (763–4), and Roulond smites a Saracen king, saying ‘criste kep vs cristyn that bene here, | to serue your soper with seintis dere!’ (962–3). This repeated image of heaven as a supper shared with Christ and the saints connects with the vision of peace imagined by Charlemagne as, deceived by Gwynylon, he looks forward with hope to the promised conversion of the Saracen Soudan and a thousand of his greatest allies to the Christian faith, and plans a fortnight and more of feasting to celebrate their friendship (40–5). Charlemagne’s profound desire for peace, summed up in his epigrammatic conclusion: ‘who gothe in woo wintirs full fell, | yet is frendschipe and faithe fairiste at end’ (46–7), offers another topical 64 This speech is based on that of Turpin in the Oxford version, lines 1472– 80, where the context is not, as here, confidence in the strength of the knights’ fellowship, but fear of their flight and loss of reputation.

246

Charlemagne in Medieval England

reflection of the ‘disordered present’ for fifteenth-century readers of the Song of Roland. The topicality of the Morte Darthur for Malory’s disordered England is of course enhanced by the identification of Arthur’s realm with England, and a single but remarkable detail in the Song of Roland seems to suggest that the English adaptor was making a similar claim in relation to Charlemagne and his realm. In the Chanson de Roland tradition, Charles is making his way from Spain to France when he hears Roland’s horn, and, after recovering the bodies, he resumes his journey and eventually arrives at Laon, where Ganelon is tried.65 In the English poem, after the confrontation with Gwynylon, Charles proceeds to ‘Cardoile’ (432), where the narration leaves him (and owing to the fragmentary state of the manuscript, we cannot know whether he would have returned to the peers from Cardoile, or whether, in a final tragic failure, Roulond would summon his help in vain). However, by giving this name to Charlemagne’s destination, the poem makes a startling assertion, for ‘Cardoile’ is the Anglo-Norman form of the placename Carlisle, which is clearly a place in Britain, not in France. Like the writer of the insular Oxford Chanson de Roland, claiming England as Charlemagne’s ‘cambre’ (2332), the English adaptor of the Song of Roland claims Carlisle as the site of his residence, and thus places England at the heart of Charlemagne’s empire.66 The choice of Carlisle, moreover, implies a further claim, for it is also one of the seats of Arthur (as documented in the Morte Darthur and the romances of Chrétien de Troyes, for example), where Arthur sometimes holds court.67 It looks as 65 Laon is the site of Ganelon’s trial and execution in the rhymed versions; in the Oxford version, Charles first mentions Laon (2910) and thereafter Aix as his destination. Duggan argues this ‘may be a reminiscence of earlier versions which, like those that have survived in rhyme, placed the final events at Laon’ (The Song of Roland: Translations, trans. Duggan and Rejhon, p. 463). 66 Although Carlisle was in Scotland at the time of the Domesday Book, by the fifteenth century it had long been part of England, and was the site of the chief fortress of the English Western March. The line naming Laon in the Oxford Roland echoes the earlier mention of England: ‘Cum jo serai a Loün en ma chambre’ (2910). 67 It seems possible that a series of partial echoes may have led the English adaptor from Laon, to Caer Leon (another of Arthur’s seats), to Caer

The Insular Roland

247

if the writer is paralleling Charles with Arthur (who also, according to Malory’s tradition, held power over many realms including both Britain and France), and assimilating the Matter of France to the history of Britain.68 Like the translator of the Middle English prose Pseudo-Turpin Chronicle, claiming Roland for England by identifying his father Milo as ‘a worthi Englysshe lorde’ (342),69 the poet of the Song of Roland thus asserts the centrality of the Matter of France to English-speaking insular culture.

The Insular Traditions of the Pseudo-Turpin Chronicle Although the legend of Roncevaux was popular across Europe, there is little evidence, as we have noted, that the Oxford Roland itself was widely known. Even within England it is not unlikely that the legend was better known through the prose account known as the Pseudo-Turpin Chronicle, whose presence in various libraries attests to its widespread circulation in a number of languages (see Chapter 1). The relationship between the pseudo-chronicle and various vernacular versions of the narrative was a dynamic one: the Latin text apparently appropriated the vernacular tradition regarding Roland, but also fed back into that tradition, including, as we shall discuss, the Middle English Charlemagne romances.

The Anglo-Norman Pseudo-Turpin Chronicle There is no evidence at all that the Anglo-Norman version by William de Briane circulated widely, and it certainly did not feed directly into the Middle English texts. It is, on the other hand, of great interest as a translation and appropriation in its own right. We have discussed above Deol (Cardoile: Carlisle). Malory uses the form ‘Cardolle’ as well as ‘Carleyle’, ‘Carlehyll’ (ed. Vinaver). 68 A comparable assimilation is implied by an extraordinary added phrase in Caxton’s otherwise slavish translation of Bagnyon’s Fierabras (Caxton’s addition in italics): ‘Fyerabras answerd: “I desyre ne praye the of no thynge but that thou sende to me Rolland or olyuer or one of thother knyghtes of the rounde table” (Charles the Grete, ed. Herrtage, p. 56); compare the association of Charlemagne, Arthur and Edward III in BL MS Egerton 3028, discussed in Chapter 2. 69 Turpines Story, ed. Shepherd, pp. xxxix, l–lii, 50, 54.

248

Charlemagne in Medieval England

the fact that of all our texts apart from Caxton’s translations, only this Pseudo-Turpin presents itself as a translation; nonetheless William’s translation is as dynamic as that of the verse accounts, which are presented without explicit reference to their sources and serve rather as replacement texts. The Anglo-Norman translation is related to the Latin C version of the Pseudo-Turpin Chronicle.70 The early French translation made for Renaud de Boulogne seems to come from another manuscript of the same family which circulated particularly in north-eastern France and England.71 It was this Old French text, rather than the Anglo-Norman version, which fed into the Middle English Charlemagne Romances.72 Ian Short has speculated that the survival of a large proportion of C texts in England may indicate a local version of the Chronicle destined in the first place, perhaps, for an Anglo-Norman readership.73 As Short points out, Britain seems to have been an ‘active breeding ground’ for manuscripts of the pseudo-chronicle.74 Short has carried out a detailed comparison of William de Briane’s 70 On the families of the Pseudo-Turpin Chronicle see Historia Karoli Magni, ed. Meredith-Jones, pp. 1–32; Meredith-Jones’s work was subject to critique: for a survey of this, see Ian Short, ‘The Anglo-Norman Translation of the Pseudo-Turpin Chronicle by William de Briane: An Edition and Study’ (unpublished PhD thesis, University College London, 1966), pp. 79–82; and Ian Short, ‘The Pseudo-Turpin Chronicle: Some Unnoticed Versions and their Sources’, Medium Ævum 38 (1969), 1–22; see also Turpines Story, ed. Shepherd, pp. xxxviii–xxxix. MeredithJones’s classification is contested by André de Mandach, Naissance et développement de la chanson de geste en Europe, II: Chronique de Turpin (1963), p. 33; he considered that William was drawing on several different versions, largely on the basis of the biography he constructed for the author. 71 Short, in Duggan, French Corpus, I/6–7; for a more detailed discussion of the connections between the various French translations, see Short, ‘The Anglo-Norman Translation’, pp. 83–5 and 92–106. 72 Ronald N. Walpole, ‘Note to the Meredith-Jones Edition of the Historia Karoli Magni et Rotholandi ou Chronique de Pseudo-Turpin’ , Speculum 22 (1947), 260–2. 73 Short, ‘The Anglo-Norman Translation’, p. 83. 74 Ibid., p. 106.

The Insular Roland

249

text with the different Latin and French manuscript witnesses most closely related to his possible sources.75 The most salient part of this comparison for our purposes is the chapter dedicated to elements which appear to be unique to William’s text. The fluidity of William’s translation, in which he avoids producing a vernacular text with French structures calqued upon the Latin, is remarkable. The text he produces, however, is far from ‘secularizing’; the religious content of the narrative remains a focal point, with William elucidating points he might not have expected his lay patrons to understand, or making explicit what a clerical audience might be expected to have inferred. Thus, for example, he is more precise in the listing of the canonical hours when Charlemagne describes the role of the canons for the pagan king Agoland: the Latin missas et matutinas et oras decantant is expanded to include the later Hours of the day: ‘ne cessunt de chaunter matines, messes e compelins e vesperis pur nous’ (‘they did not cease to chant matins, masses, complines and vespers for us’; 547–9); in a later chapter more detail is also given of the duties of the canons marking the anniversary of Roland’s death.76 As Short sums up, ‘William allows himself to stress and elaborate the didactic aspect of his text, even to the extent of adding his personal moralizing conclusions to those of the Latin … and of improving on the theological discussion of ch. XVII.’77 This ‘improvement’ refers to an elaboration of Roland’s theological discussion with Vernagu developing the doctrine of the Fall. Despite this strong clerical flavour, the dynamic relationship between the PseudoTurpin Chronicle and the vernacular epic tradition which relates the same narratives is also evident. It is not surprising then, to find that William on occasion uses vocabulary which we might expect in a chanson de geste. Short comments on the use of the phrase enausser seynte crestienté (‘to promote holy Christianity’), a more positive way of expressing the motivation of Charlemagne and his men in Spain than the Latin ad expungnanda gentem perfidam (‘to remove the perfidious race/people’), but also one with epic resonances.78 75 Ibid., pp. 86–111. 76 Ibid., p. 122. 77 The Anglo-Norman Pseudo-Turpin Chronicle, ed. Short, p. 9. 78 Short, ‘The Anglo-Norman Translation’, p. 125 – Short refers to La Chanson de Guillaume, Aliscans and Le Charroi de Nimes as epics which

250

Charlemagne in Medieval England

In his prologue to the translation William invites his audiences to take both pleasure and instruction from it. These changes made by the translator elucidate the Latin text, increase its narrative momentum and thus fit it for the different, wider, readership William envisages.

The Pseudo-Turpin Chronicle in Middle English Literature: The Verse Romances Five separate and very different adaptations of the Pseudo-Turpin tradition survive in Middle English literature: the romances of Roland and Vernagu and Otuel and Roland, each of which appropriates extensive selected passages from the Pseudo-Turpin Chronicle to create an episodic narrative focused on the issue of Christian conversion; the Song of Roland, with its story of Ganelon’s wine and women imported from the chronicle into the epic tradition of Roncevaux; the fifteenthcentury translation of the Pseudo-Turpin Chronicle into Middle English prose, and Caxton’s Charles the Grete, closely translated from Bagnyon’s Fierabras, the third part of which reproduces the PseudoTurpin tradition’s account of Roland and Roncevaux from the Speculum Historiale of Vincent of Beauvais. We discuss the three verse romances in detail when considering them in relation either to Otinel (Chapter 6) or to La Chanson de Roland (Chapter 4); here we note their adaptations of the Pseudo-Turpin tradition. The Song of Roland The material imported from the Pseudo-Turpin Chronicle into the Song of Roland seems to have been selected for a specific purpose, to supplement the epic tradition: a process that accords with Stephen Shepherd’s description of the English text as ‘a kind of “researched” compilation’. 79 The Pseudo-Turpin episode in the Middle English poem concerning the Saracens’ duplicitous gifts to Charlemagne of wine use this phrase; the exact phrase appears in Fierabras, line 1582, and at the end of Otinel. Boutet comments on this phrase: ‘Le combat et la conversion sont les deux moyens d’essaucier saint crestienté’ (Charlemagne et Arthur, p. 193); he goes on to point out that, nonetheless, it is normally attacks on Christian lands which actually provoke the Christian aggression. 79 Shepherd, ‘ “I have gone for þi sak wonderfull wais” ’.

The Insular Roland

251

and women is closer to the Old French Johannes Translation than to the standard Latin text,80 but in one respect the episodes in both the English Song of Roland and the Middle English prose Pseudo-Turpin Chronicle differ from all other C-family manuscript versions: they omit a long moralizing explanation that connects the Christians’ sinful fall into drunkenness and fornication with their subsequent deaths. As Shepherd points out, the prose translation preserves ‘enough information to render the moral point at least implicit’, 81 but this is not the case in the Song of Roland, where the narration comments instead on the knights’ contrition and their cursing Gwynylon: they synnyd so sore in þat ylk while that many men wept and cursid þat vile. (75–6)

Indeed, the bias in the retold episode is all towards blaming Gwynylon and excusing the knights, who are represented as if drugged by the wine and unaware of their sinful actions: wyn went betwen þem, non did astert, þat gwynylon to toun brought, euyll hym betid! It swymyd in ther hedis and mad hem to nap; they wist not what þey did, so þer wit failid. when they wer in bed and thought to a restid, they went to the women þat wer so hend, that wer sent fro saragos of sairsins kind. (68–74)

The chronicle material has been selected and carefully interwoven into the epic narrative at three separate points (at lines 1–3, 28–30, 59–76), not to prepare for a moral judgement as in the Pseudo-Turpin tradition, justifying the deaths at Roncevaux as carrying out God’s will, but to amplify Gwynylon’s role as traitor and tempter. The poem elaborates the details of the Saracen ladies’ attractions – they are ‘madins bright in wedis’ (3), and ‘echon of them is a lordis doughtur’ (29). It also 80 Ibid., pp. 223–4. 81 Turpines Story, ed. Shepherd, p. xli. The text reads: ‘many of þe hooste were dronke of þat wynne and toke many of þucke women and so were dede’ (1130–1).

252

Charlemagne in Medieval England

indicates the courtly delights offered both by feminine company – ‘all thes faire ladys with the to pley’ (28) – and by ‘good wyn’ (30), to be enjoyed at a supper where the king is served ‘semly’, ‘with euery thinge þat myght glad his hert’ (66–7). All this contributes to the appeal of Gwynylon’s temptation, which plays on Charlemagne’s weary desire for peace after long years of war. It is an appeal that might have been all too understandable to the poem’s fifteenth-century readers and listeners.

Roland and Vernagu Roland and Vernagu, although sometimes described as no more than a verse rendition of its chronicle source,82 presents a sustained programme of adaptation. The Middle English romance is constructed as a diptych, a two-part narrative in which the action of each half mirrors that of the other (a favourite insular pattern). The first half, based on the opening chapters (I–XIV) of the Johannes Translation of the PseudoTurpin Chronicle, concerns Charlemagne’s liberation of Jerusalem and acquisition of the Passion relics, and his conquest and conversion of Spain. This half is given greater coherence than in the chronicle source by the identification of the Saracen aggressor in Jerusalem with Ebrahim, King of Spain (later mentioned in Johannes, in chapter XIX, as ‘Ebrayns li rois de Sebile’), thus constructing the earlier episode as preliminary to the main action in Spain. The second half, based on chapters XXXIII– XLI, covers the story of the Saracen giant Vernagu. However, a further chapter, taken completely out of sequence, is inserted between the two halves: the ‘Description of Charlemagne’ (chapter XLIX; Roland and Vernagu, 425–60), which functions as a second prologue, introducing the second half. The two halves are both designed to showcase the role of their Christian heroes, Charles and Roland, as champions of God, and each half adds a unique episode to demonstrate this truth. The first half provides a new miracle, a story of grapes ripening out of season: ‘and while charls was in þat stede, | A fair miracle god for him dede’ (305–6). This functions as a narrative postscript to the conquest of Spain, and is designed to crown Charlemagne’s achievements with this miraculous proof of God’s favour. The second half is also extended by a new miraculous incident, in which Roland, having so far failed either 82 Smyser, ‘Charlemagne Legends’, p. 90; Mehl, Middle English Romances, p. 57.

The Insular Roland

253

to convert Vernagu or to defeat him, is informed by an angel it is God’s will that he slay the irredeemable giant. Both Christian heroes, Charles and Roland, are thus shown to be specifically sanctioned by God in their victories over unrepentant Saracen cities and people.

Otuel and Roland We discuss the major structural question in relation to this romance in the Otinel chapter (see pp. 371–7): the combination of a version of the chanson de geste Otinel with material from Johannes. Unlike Roland and Vernagu, Otuel and Roland preserves the Pseudo-Turpin material in much the same sequence as the chapters in the chronicle source (XLII–LXV), except that the episode of the king of Navarre (XXXIII) is inserted; here the ‘Description of Charlemagne’ chapter (XLIX) occurs in the normal place, between the establishment of the archiepiscopal see at Compostela and the lead-up to the Battle of Roncevaux. The English poem transforms the unpromising material about ecclesiastical administration into a proof of Charlemagne’s supreme political power by the simple expedient of repeating his name seven times as instigator of every action, and by alternating the formula ‘Charles comaunded …’ (with the metrical stress falling on ‘Charles’ at the beginning of the line) and the phrase ‘By the hest of Charlemayne’, and finally summing up the episode thus: And so þay dude withoute lete; For charlys hyt hade sette, Durst no man be þeragayne. (1973–5)

This provides a natural preliminary to the following description of Charlemagne, but it also supports the portrayal of his supreme power in the prologue to the whole romance, which announces its subject as ‘a conquerour’ (3) who ‘all hys fomen ouercam’ (34), and finally held his lands ‘in grythe & pes, | withoute warre and fyȝt’ (37–8). The ‘Descriptio’ chapter in turn serves as a preface to the story of Roncevaux, and in this translation the detail that Charles posted an imperial guard every night is explicitly related to the fact that he ‘Euer

254

Charlemagne in Medieval England

douted tresoun’ (2004),83 setting up a context of expectation for the narrative of betrayal to follow. Fear of treason seems to have exerted some pressure on the rewriting of the inherited narrative in Otuel and Roland. Ganelon’s treachery is highlighted in both chronicle and epic traditions, but the conclusion of the story is retold in Otuel and Roland in an entirely new way. Instead of a judicial combat between Thierry (for Charles) and Pinabel (for Ganelon), here Gwynes responds to Terry’s accusation himself, with a personal challenge that acknowledges the ignominious death awaiting him if he is proved a traitor: Gwynes sayd ‘nay, thowe lyxt falsly by thys day, And that schall be well yfownde, thy body anone-ryȝtys to myn! Arme the anon wel afyn! And y wyth a spere y-grownde, But y me defende, y grawnt, so god me amende, Be honged and drawe thys stownde.’ (2693–6)

Having been defeated, Gwynes is publicly declared guilty of treasonably selling the Christians to their enemies (2715–23), and Charlemagne judges him accordingly: ‘Forsothe’, sayd Charlys, ‘hyt ys the lawe, That thow be honged an drawe.’ (2724–5)

As early as the thirteenth century, hanging and drawing was the customary punishment for treason in England, where ‘drawing’ the traitor to execution, usually on a hurdle behind a horse, is what marks it out as such.84 Gwynes then confesses the full extent of his treachery 83 The very similar three-stanza translation of the same Pseudo-Turpin chapter in Roland and Vernagu (425–60), which prefaces the dispute between Roland and Vernagu, reads: ‘Wel wise he was & wiȝt, | & douted of tresoun’ (450–1); the Otuel and Roland version is more pointed. 84 J. G. Bellamy, The Law of Treason in England in the Later Middle Ages (Cambridge: Cambridge University Press, 1970), pp. 18, 26.

The Insular Roland

255

in colluding with the Saracens, who planned to conquer France and kill Charles and all his barons (2727–32). These additions all seem designed to emphasize the exemplary nature of the transgression and the justice of the sentence of execution. Edward III’s Statute of Treasons (1351) had defined the offence of high treason, and Gwynes’s actions as described in Otuel and Roland fall explicitly within the terms of the statutory definition: ‘When a Man doth compass or imagine the Death of our Lord the King, … or be adherent to the King’s Enemies in his Realm, giving to them Aid and Comfort in the Realm, or elsewhere.’85 Similarly, the execution of Gwynes is described with further additional details that connect it with refinements of the punishment for high treason sometimes carried out in England: Tho by the heste of charlys the kyng, The traytour was don to hongyng, And was y-drawe thoruȝ the toun, And after y-honged wel faste. Forsothe tho in haste, Alle quyk he was leten doun, And y-bounde to a stake, And hys bowels out y-take, To brenne hym by-forn. To foure stedys he was y-knyt, By the hondys and by the fet, At the heste of charlyoun. On eche stede sat a knyȝt, And thus he was to-twyt, Gwynes, the falce traytour. (2733–47)

The last part of the sentence, quartering, is represented as in the chronicle and epic traditions by the use of four horses to tear Ganelon’s body asunder, but the previous stages of the execution are described in a manner unrepresented in any of the parallel texts (even among the violent tortures suggested by Charlemagne’s barons in 85 Treason Act 1351: ‘A Declaration which Offences shall be adjudged Treason’ (TNA, 25 Edw. 3, Stat. 5, Ch. 2), http://www.legislation.gov.uk/ aep/Edw3Stat5/25/2/contents [accessed 23/08/2016].

256

Charlemagne in Medieval England

the Châteauroux–V7 version of the Chanson de Roland). The rending of Ganelon’s body might have suggested to the English adaptor the punishment of being ‘hung, drawn and quartered’ that was applied in outstanding cases of high treason from the thirteenth to the fifteenth centuries, very occasionally with the extra penalty of evisceration. Instances of such tortured deaths are not uncommon in Elizabeth I’s reign (as in the case of the Babington Plot, 1586), but they were extremely rare before the sixteenth century. In the late medieval period, it seems that the usual interpretation of the sentence of death for high treason in England (if not a simple beheading) was to be drawn, hanged and possibly quartered, as was the fate of John Ball for his part in the Kentish uprising of 1381,86 and of Thomas Usk, convicted of treason in 1388.87 Evisceration had been part of the sentence in a very small number of egregious cases of treason in the reigns of Edward I and Edward II, notably those of Dafydd ap Gruffydd (1283) and William Wallace (1305), but, as John Bellamy points out,88 the record of the sentence in each case makes an explicit connection between evisceration and sacrilege, as an addition to the other aspects of the treason committed. Interestingly, Gwynes’s treason as represented in Otuel and Roland may be interpreted in similar terms. The rewritten account of Gwynes’s treachery is an almost text-book demonstration of statutory high treason, but it is further freighted with meaning by the identity of the enemy to whom Gwynes betrays his fellow Christians: Saracens portrayed as ‘godys wytherlynges’ (2296, 2645) and ‘deueles/fyndes lemes’ (2184, 2648), in language that makes clear their status as enemies of God and agents of Satan.89 Gwynes is 86 Andrew Prescott, ‘Ball, John (d. 1381)’, Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004), also online at http:// www.oxforddnb.com [accessed 23/08/2016]. 87 Ronald Waldron, ‘Usk, Thomas (c. 1354–1388)’, Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004), also online at http://www.oxforddnb.com [accessed 23/08/2016]. 88 Bellamy, The Law of Treason, pp. 24–40. 89 The rare word ‘wytherling’ (adversary) occurs in King Horn (Bodleian Library MS Laud Misc. 108) in a similar context: ‘þat heþene king, Ihesu cristes wiþerling’; the phrase ‘devil’s limb’ or ‘fiend’s limb’ is commonly used of evil spirits and of humans believed to be in league with Satan (MED, ‘lim’, n.1).

The Insular Roland

257

thus guilty of treason in both the secular and the spiritual realm, and would arguably be liable to the extreme penalty of the law. In light of all these precise references to the legal definition and penalty for high treason, the narratorial concluding comment suggests that this text may be channelling a particular concern with the danger of treason in contemporary England: Forsothe, hit were skele and ryȝt That eueryche traytour were so y-dyȝt, And hadde muche myssaunter. (2748–50)

For readers and listeners in late fifteenth-century England this could have connected both with well-founded fears of ‘disorder’ in the form of political insurrection and also with the concern to maintain religious orthodoxy against the threat of heresy that is so marked in the Fillingham MS texts. Megan Leitch has referred to the emphasis on ‘discourses of treason’ in Middle English prose romances as part of a ‘widespread expression of the English cultural imaginary in the second half of the fifteenth century’, in which the condemnation of ‘treason’ extended beyond the statutory hierarchical categories to include ‘almost any underhanded harmful action, to anyone’. 90 In Otuel and Roland, by contrast, the discourse of treason is restricted. The specificity of Gwynes’s crime of high treason is emphasized not only by the details discussed above, but by the text’s almost exclusive concentration of the lexis of treachery on the account of his actions, without the applications of the terms ‘traitor’ or ‘treason’ in contexts of more general breaches of trust such as Leitch discusses.91 Gwynes is sent on the mission to the Saracens already corrupt in his intentions: ‘But charlys wyst nouȝt | The tresons 90 Leitch, ‘Thinking Twice about Treason in Caxton’s Prose Romances’, pp. 63, 46–7. See also Lydia Fletcher, ‘ “ Traytoures” and “Treson”: The Language of Treason in the Works of Sir Thomas Malory’, Arthurian Literature 28 (2011), 75–88, noting Malory’s ‘addition of English law-derived usages of treason to the foundation of thirteenth-century usages of his source material’ (p. 88). 91 The only use of the term ‘treason’ apart from Gwynes’s treachery occurs in the encounter between Ogier and Clarel in the Otinel section of the text, where it is used to characterize the normal expectation of Saracen

258

Charlemagne in Medieval England

of gwynes thouȝt’ (2032–3). He immediately sells his allegiance to the enemy – ‘Thoruȝe that ylke tresour, | Gwynes bycome traytour’ (2056–7) – and devises the plan to deceive Charlemagne and ‘To sle hym with tresoun’ (2067). The narration identifies Gwynes as the ‘traytour’ (2073, 2093, 2097, 2370) until he is accused of the deaths at Roncevaux by ‘fals tresoun’ (2678, 2685) and brought to justice. The similarly intensified focus on each stage of the bodily punishment inflicted on Gwynes in this version provides a precise contrast with the immediately following account of how Charles has the bodies of Roland, Oliver and the other peers embalmed (2754–74). Danielle Westerhof reads this contrast within a discourse of nobility,92 but an equally significant context in view of this text’s insistently orthodox Christianity may be the discourse of hagiography. Roland, ‘godys knyȝt’ (2418), dies a martyr’s death (2440–1), and Turpin receives miraculous proof of his entry into heaven’s bliss (2481–3). Roland’s body is then embalmed with myrrh (2541–3, 2754–6), echoing the treatment of Christ’s body after the Crucifixion.93 The body of Gwynes, on the other hand, is treated in a way that parallels his fate with that of Judas, the prime exemplar of sin and treason, to whom the narrative explicitly compares him: ‘Forsothe Gwynes tho was | A fals traytour as was Iudas’ (2092–3). Judas, the Bible tells us, ‘being hanged, burst asunder in the midst and all his bowels gushed out’. 94 The specific mention of evisceration in the execution of Gwynes, besides recalling the severity of the law as applied in the worst cases of treason, adds a pointed echo of the death of Judas, the arch-traitor, an echo that would have been readily apparent to readers and listeners, as the scriptural account of Judas’s end

behaviour as treacherous: ‘Off nothyng ne schaltou drede, | Ne off no sarsins tresoun!’ (972–3). 92 ‘Ganelon’s body is destroyed to harmonise it with his inner corruption; by contrast, Roland’s body – the privileged shrine of his noble heart – is lovingly conserved for future burial as a symbol of his excellence’ (Danielle Westerhof, Death and the Noble Body in Medieval England (Woodbridge: Boydell Press, 2008), p. 140). 93 John 19.39–40. 94 Acts 1.18 (Douai-Rheims); Vulgate text: ‘et suspensus crepuit medius: et diffusa sunt omnia viscera ejus’.

The Insular Roland

259

fed into medieval popular culture.95 The grisly detail thus contributes to the contrast with Roland’s martyrdom and the reverent care with which his body is being preserved as if it is to become a ‘corsaint’, a future holy relic. Roland, represented as a martyr in the Pseudo-Turpin tradition, is depicted as a saint at Chartres cathedral, in the Charlemagne window and on the Martyrs’ portal, possibly (as suggested by Rita Lejeune) with a view to establishing a cult of ‘St Roland’.96 Here, however, the contrast between Roland’s revered holy body and Gwynes’s mutilated corpse seems designed rather to accentuate the sacrilegious evil of Gwynes’s treason against God as well as Charlemagne. The lesson embodied in the due processes of law and typified in the fate of Judas is clearly reflected in the narrator’s loyal conclusion: ‘Forsothe, hit were skele and ryȝt | That eueryche traytour were so y-dyȝt’ (2748–9).97 The tone and sentiments of this fifteenth-century response are echoed in the final verse of a late seventeenth-century broadside ballad on the subject of the Gunpowder Plot, applying the lesson of that attempted treason to the popular anti-popery anxieties of its own time: ‘[God] brought Rome’s faction unto Punishment | … | And all that euer Plots, I hope God will.’98 In the example of Otuel and Roland, it could be argued that while verse romance proves itself as capable as prose romance of responding to a narrative ‘bias towards disaster’, here the generic optimism of romance invests the law with the power to restore order.

The Middle English Prose Pseudo-Turpin Chronicle While in general a close translation of its C-family Latin Pseudo-Turpin source,99 the Middle English text nevertheless shows evidence of subtle 95 For instance, it concludes the legendary life of Judas Iscariot inserted into the life of St Matthias, the apostle chosen to fill Judas’s place, in the widely disseminated Legenda Aurea of Jacobus de Voragine. 96 Lejeune and Stiennon, La Légende de Roland, I, 203, 382. 97 This may echo the ending of Fierabras, where the narrator notes approvingly that traitors always come to a bad end (ed. Le Person, 6403). 98 The Pepys Ballads, ed. W. G. Day, 5 vols (Cambridge: D. S. Brewer, 1987), II, 370. 99 Shepherd, ‘The Middle English Pseudo-Turpin Chronicle’; Turpines Story, ed. Shepherd.

260

Charlemagne in Medieval England

adaptation in line with the agenda of other insular appropriations of the Matter of France tradition: in particular, with the parallel case of the Anglo-Norman Pseudo-Turpin Chronicle of William de Briane, described above. A new heroic emphasis is initially signalled in the unique title given to the Middle English text, with its connotations of epic subject matter: ‘the Storye of the Bataille of Rouncivale of Grete Charles the Emperoure’. 100 This emphasis is enhanced by small innovations in the Middle English Pseudo-Turpin that depict the Christians in terms ‘reminiscent of vernacular romance’: Shepherd notes added references to ‘gode’ and ‘worthy’ heroes fighting ‘manfully’ with a ‘tristy’ sword (p. xli). Equally notable is the unique choice of romance vocabulary in the chapter describing Charlemagne (‘De persona et fortitudine Karoli’), to produce a new emphasis on his chivalric credentials: ‘moste douȝttyeste in dedis of armys, moste of knyȝttehode and chyualrye’ (1061–2). There is also increased reference to Roland and Oliver by name, possibly, as Shepherd suggests, ‘out of familiarity with these heroes as they appear in romances and chansons de geste’ (p. 84),101 and again, descriptions of them are phrased in correspondingly chivalric terms: Roland is ‘a grete man and a semely and douȝtty in armys’ (488–9), and Oliver is ‘a manfulle knyght’ (490). Charles is given added heroic prominence when the ‘mysbeleuynge’ Saracens (119, 150) submit to him, wondering at ‘þe manlynes, þe semelynes, þe comlynes of Charlis’ (160), and again when he is said to exhort all the Christian host to ‘fyȝte manfully ayenste þe enemyes of God and þe feyth’ (678–9). All these added details, together with the omission of the moralizing passage in the source that attributes the defeat at Roncevaux to the Christian knights’ sinful indulgence with Saracen wine and women,102 contribute to a reinterpretation of the chronicle tradition in the direction of the insular epics and romances, with their characteristic focus on heroic individual 100 It is notable that four manuscripts of the Chanson de Roland name the text a story of Roncevaux (explicits to MSS V4, C, P, L). See Margaret Jewett Burland, Strange Words: Retelling and Reception in the Medieval Roland Textual Tradition (Notre Dame, IN: University of Notre Dame Press, 2007), pp. 7–8. 101 See Shepherd, ‘The Middle English Pseudo-Turpin Chronicle’, for evidence of influence from heroic and romance expectations in the insular Pseudo-Turpin tradition more widely. 102 Turpines Story, ed. Shepherd, p. xxxix.

The Insular Roland

261

Christian knights championing the faith against the incursions of enemy Saracens. Helen Cooper and Stephen Shepherd read Turpines Story as part of the fifteenth-century textual reflection of contemporary ‘political turmoil’. 103 As discussed in Chapter 3 (pp. 195–7), Shepherd documents the probable connection between this act of translation and the Lancastrian affiliation of the family for whom the manuscript was produced – the Mulls of Harescombe, in Gloucestershire. The implication is that this English family, who claimed descent from the twelfth-century Milo, Earl of Hereford, were in effect appropriating the story of Roncevaux to their own genealogy by the translator’s innovation in identifying Milo, father of Roland, as an Englishman.104 Consequently, for them, Ganelon’s role in betraying Roland would have attracted even greater opprobrium, and it is therefore significant that one of the translation’s few alterations is a detail that changes the way Ganelon’s guilt is presented: an added phrase introducing him as ‘þis false man Ganolion’ (1115). In the standard Latin version, Ganelon is corrupted by the Saracen bribe and agrees to betray the peers in exchange for gold; here, however, Ganelon is presented as already corrupt in his nature as a ‘false man’. The guilt of the treasonous plot is thus transferred from the Saracens (who in the Latin text deceitfully (fraudulenter) offer the bribe), to Ganelon, who accepts their gift as a ‘false man’. His role in the narrative is now clearly signalled to the reader as that of an acknowledged villain from the start.

Caxton’s Charles the Grete As noted above, Caxton’s translation of Bagnyon’s Charlemagne compilation is faithful to the point of slavishness. Like Turpines Story, Bagnyon’s version emphasizes Ganelon’s capacity for treason, despite Charlemagne’s trust in him, before he accepts the Saracens’ bribe, and it interrupts the narrative with an excursus on the theme of avarice in which Ganelon is ‘repreuyd by th’auctour’ for this villainous betrayal of his noble birth, his ‘natural souerayn lord’, and ‘al crysten men’. The lesson is universalized with references to the sin of Adam and the fall 103 Ibid., p. liv. 104 Ibid., p. li; see also our discussion of the Heralds’ Debate in the concluding chapter.

262

Charlemagne in Medieval England

of Troy, but it could also have had contemporary resonance for Caxton. In his own Epilogue he recalls ‘the noble & moost crysten kyng, our naturel and souerayn lord, late of noble memorye, kyng Edward the fourth’, but when he notes that he finished the translation in the reign of Richard III and printed it in the reign of Henry VII, neither of these kings is named with any of the elaborate terms of honour given to Edward IV. However, the same terms are used in respect of Charlemagne, whom Caxton describes in his own Prologue as ‘the noble and crysten prynce Charles the grete’, and who should have commanded Ganelon’s loyalty as his ‘natural souerayn lord’, as the text points out. So, for Caxton, anxieties around the extraordinary dynastic upheavals of the mid-1480s could have cast a nostalgic glow over the reign of Edward IV, imaginatively coupling him with Charlemagne. In the late-medieval cultural imagination, Ganelon occupied a ‘bogey-man’ role similar to that of Guy Fawkes for a later age, as an archetypal traitor in league with a foreign power, ready to betray the nation and subvert its faith. And, just as the identity of the guy burned on the bonfire would change over the years to reflect current hate-figures,105 so it seems likely that the story of Ganelon’s treason would speak to readers and listeners with particular contemporary relevance in the political turmoil of the late fifteenth century.

Conclusions It is evident, despite the survival of only a fragment of a Middle English Song of Roland, that the tradition surrounding the events at Roncevaux was an important strand in the Charlemagne material in England as elsewhere. Of the three major insular strands – the traditions of Roland, Otinel and Fierabras – the Roland has perhaps been translated most dynamically. The bringing together of material from the chanson de geste and the Pseudo-Turpin Chronicle to form new texts in a peculiarly English development also shows a strong awareness of the wider tradition and is, perhaps, a different manifestation of the impulse to connect texts which is seen in the formation of cycles in continental French. As the oldest extant version of this tradition in French is in an insular manuscript, it is 105 Iona Opie and Peter Opie, in The Lore and Language of Schoolchildren (Oxford: Clarendon Press, 1959), cite the examples of the pope, Napoleon, the Kaiser, Hitler, and many others (p. 281).

The Insular Roland

263

difficult to be certain about the degree of alteration that has been carried out to produce the Oxford text, but what is clear is the incorporation of some specifically insular concerns, a development concordant with the level of political engagement that is a characteristic of the chanson de geste. The Pseudo-Turpin material, in both insular vernaculars, shows a continued engagement with religious practice and chivalric concerns, which is very different from the way the tradition developed elsewhere in Europe.106 Despite the (relative) paucity of textual witnesses to the Roland in an insular context, everything points to its foundational importance in the cultural imagination. The other main strands of surviving Charlemagne material in England, Fierabras, to which we will now turn, and Otinel, itself based on Fierabras, have close intertextual links to the Roland tradition.

106 Jean-Claude Vallecalle, ‘La Réception de la Chronique du Pseudo-Turpin’, pp. 465–9; and Le Livre de saint Jacques et la tradition du Pseudo-Turpin, ed. Jean-Claude Vallecalle (Lyon: Presses universitiares de Lyon, 2011).

5 Re-Presenting Otherness: The Insular Fierabras Tradition

T

he narrative of the chanson de geste Fierabras, in continental and Anglo-Norman versions, is focused on the recovery of relics of the Passion, stolen by the Saracen emir Balan from Rome when his army sacked the imperial city. The infidel attack on the Holy City itself, though only referred to briefly in the extant Fierabras, is the cataclysmic event which triggers the whole action. The conflict between the Christian leader Charlemagne and the Saracen emir Balan is prefigured by a single combat between Fierabras and Oliver, which contains echoes of Roland and Vernagu’s conflict in the Pseudo-Turpin Chronicle and of Charlemagne and Baligant’s fight in the Chanson de Roland. Such ‘trials by combat’ were something of a rhetorical set piece in the chanson de geste and, as we have noted, a significant narrative meme in all the insular Charlemagne texts. In Fierabras the emir’s son, Fierabras, challenges the French and Oliver responds; they engage in a single combat which represents the wider battle between Saracen and Christian. Oliver vanquishes Fierabras, who is converted to Christianity.1 However, before Oliver can get back to safety, he is captured by the Saracen army and, in company with others from among the peers, is taken to Balan’s stronghold of Aigremore. Charlemagne sends the remaining peers with a threatening message demanding the release of the prisoners and the return of the relics. Instead, all the peers are taken prisoner by Balan, but then released by Balan’s daughter, Floripas. The peers take over Aigremore, evicting the Saracens before being themselves besieged therein. After various adventures and sorties they are able to send a message to Charlemagne, who comes to their rescue and defeats Balan in a combat which is based on the ideological assumptions that ‘right is might’ and that the representative of God will always win. The narrative ends with the baptism of Floripas and her marriage to Gui de Bourgogne

1 It is probably at this point that the older, lost chanson de geste ended. On the evolution of the legend, see below n. 2; Ailes, ‘A Comparative Study’, pp. 26–60.

264

The Insular Fierabras Tradition

265

and the distribution of the relics to various churches in France; a coda looks forward briefly to the events of La Chanson de Roland. Fierabras texts can be divided into two traditions: the Vulgate tradition, also named by Marc Le Person, the most recent editor of the continental chanson de geste, the ‘version longue’, and the non-Vulgate or abbreviating tradition, the ‘version courte’ according to Le Person’s classification.2 The Vulgate tradition of Fierabras dates from around 1200, and the earliest known Anglo-Norman copy of this text (now lost) belonged to the second half of the thirteenth century.3 The extant fourteenth-century Anglo-Norman copy of the Vulgate Fierabras (Hanover, Niedersächsische Landesbibliothek MS IV 578) and the Anglo-Norman abbreviating remaniement, known as Fierenbras, preserved uniquely in BL MS Egerton 3028, which may have been composed as late as the fourteenth century, testify to the longevity of the genre of chanson de geste in England. The evident popularity of Fierabras among insular readers and listeners in the later Middle Ages is confirmed by the existence of the three independent adaptations of the chanson de geste into Middle English verse.4 The Sowdone of Babylone is 2 For the Vulgate tradition, see the critical editions of Kroeber and Servois (1860) and Le Person (2003). On the classification, see Fierabras, ed. Le Person, pp. 23–56. Le Person’s classification accords with our view of the development of the legend as one that began with a now lost chanson de geste which would have focused on Rome and ended with the combat between Fierabras and Oliver; to this was added the ‘queue postiche’ of the peers’ adventures in Aigremore and the belle sarrasine; the attack on Rome was probably dropped when the remanieur added this new narrative; the extant Destruction de Rome is a ‘prequel’ added to make good that loss; the abbreviating tradition would then be a late development (Ailes, ‘A Comparative Study’, pp. 26–60). André de Mandach proposes a more speculative evolution of the material, which would relate the extant Destruction de Rome more closely to the original, lost chanson de geste and place the abridged version at an earlier stage in the development of the legend by virtue of its having complete the ‘diptyche littéraire’ which we have associated (Chapter 2) with the insular development of the legend (André de Mandach, La Geste de Fierabras: Le Jeu du réel et de l’invraisemblable (Geneva: Droz, 1987), pp. 7–24, 90–1). 3 The Didot MS; for discussion, see p. 149. 4 For evidence of insular manuscript dissemination, see pp. 37–52. References to Fierabras in, for example, Barbour’s Bruce (see Chapter  2,

266

Charlemagne in Medieval England

closely related to the Egerton MS texts of the Destruction de Rome and Fierenbras, representing a distinctly insular branch of the tradition, while the Ashmole and Fillingham texts, though very different from each other, are both ultimately derived from unidentified exemplars that would appear, on the basis of occasional close verbal parallels, to be not far from the Vulgate tradition represented in Le Person’s critical edition of Fierabras. All three Middle English texts share with both AngloNorman texts a distinctly insular version of the central character’s name: instead of ‘Fierabras’, the king of Alexandria is here named Fierenbrace, Fierenbras, Fierembras, Fyrumbras, Firumbras, or Ferumbras. The English versions all treat their source with a degree of creative freedom, but do not undertake the kind of major reordering of events and restructuring of material found in the English Song of Roland. Their main structural techniques are abbreviation and amplification, and what reordering of events does occur tends to be a consequence of the need to restore narrative continuity and tidy up loose details where there has been large-scale omission of material. However, the omissions and additions do in general suggest the same project of reworking the chanson de geste in line with the expectations of Middle English popular romance as demonstrated in the Song of Roland: the excision of digressions and repetitions to create a more clearly organized single narrative, with closer focus on the central protagonists, and the introduction of passages illustrating the orthodox piety and the chivalrous fellowship of the Christian heroes. Further possibilities for innovation are provided by the depiction of Floripas’s passion for Guy of Burgundy in the Fierabras tradition; here also the Middle English poems make alterations that can be read as developments of the narrative’s romantic potential, where all three in various ways enhance the sentimental and reciprocal aspects of the relationship. As the discussion in Chapter 2 has shown, both the Vulgate and Egerton branches of the Fierabras tradition give prominence in their prologues or conclusions to the connection between the loss and recovery of the Passion relics and those churches in France that figure as recipients of the relics. Both Middle English adaptations of the Vulgate version are lacking their opening folios, so how they dealt with the French prologue, with its stress on the custom of the Lendit n.  48) and Skelton’s ‘Ware the Hawke’ attest to the story’s being wellknown in Britain.

The Insular Fierabras Tradition

267

fair at Saint-Denis, is not recoverable; and the Ashmole text lacks its conclusion as well. However, the ending of the Fillingham Firumbras and the parallel case of The Sowdone of Babylone, with its treatment of the relics motif as found in the Egerton version, give two different examples of radical change, showing that the Middle English texts, although retaining the function of the relics in the plot as the object of desire, have completely decoupled the relics from the original focus on the specific cults of Passion relics at different sites in France, for which the chanson de geste had provided authoritative myths of origin. The great question, therefore, is why this story repeatedly captured the imagination of English poets at this time: what opportunities did it offer for engagement with contemporary issues and concerns?5

Insular Texts in the Vulgate Fierabras Tradition Most of the insular witnesses to the narrative belong to the Vulgate tradition.

The Lost Didot Fierabras Our knowledge of an early text of Fierabras contained in the AngloNorman Didot manuscript, which was lost in the fire of Louvain University Library in 1940, is fragmentary.6 It is based on several sources: the 1867 catalogue of sale of the manuscripts of Ambroise Didot,7 including what André de Mandach describes as a ‘transcription maladroite’ of the planctus of Charlemagne when he thinks the peers have been killed;8 extracts published by Léon Gautier in 1897;9 and 5 Alison Wiggins draws attention to the similar history of the manuscript versions of Guy of Warwick, which represent five ‘independent translations into Middle English from one or other of the Anglo-Norman originals’ (‘The Manuscripts and Texts of the Middle English Guy of Warwick’, p. 65). 6 Louvain, Bibiothèque de l’Université catholique MS G.171. 7 Catalogue raisonné des manuscrits de la bibliothèque de M. Ambroise Firmin Didot (Paris, 1867), no. 20. 8 De Mandach, La Geste de Fierabras, p. 136. 9 Léon Gautier, Bibliographie des chansons de geste (Paris, 1897), pp. 97–102; these include a copy of the lines printed in the Catalogue Didot.

268

Charlemagne in Medieval England

the variants given by Alfons Hilka in his edition of the Hanover manuscript.10 De Mandach used the evidence of the Hilka variants to produce an edition of the Gautier fragments in an appendix to his study of the Fierabras tradition.11 The loss of this early manuscript witness is all the more regrettable as it seems to have been quite close to the other thirteenth-century manuscript (Bibliothèque de l’Escorial M. III-21, the base of Le Person’s edition), which lacks nearly 800 lines at the end.12 One of the features of the text which the Didot MS shares with the Escorial text is the order in which we find the laisses recounting the sending of the ambassadors to Balan.13 De Mandach refers to the mise en relief of Richard de Normandie.14 In lines these two manuscripts share, Richard is clearly identified with Richard the Fearless, the first Duke of Normandy: En après cest parole Ricard de Normandie Cest Ricart sanz Pouer od le chier hardie Qui de Faucans fist faire le mestre abeïe. (MS D, fol. 80r; cf. Fierabras, ed. Le Person, 2396–8) [After this speaks Richard of Normandy; This is Richard the Fearless, of the bold face Who had the main abbey at Fécamp built.]15

Transporting the tenth-century duke to serve with the ninth-century Charlemagne would not in itself have raised any questions for the late twelfth-century poet, or for the thirteenth-century copyists. Richard is important in all versions of the narrative and his identification with the almost legendary first duke does not appear to be an invention of the insular scribe. Even so, the importance of the duke of Normandy 10 Chanson de Fierabras, ed. Hilka. 11 De Mandach, La Geste de Fierabras, pp. 131–46. 12 Ailes, ‘A Comparative Study’, pp. 267–9; here we are in agreement with de Mandach, who notes that ‘E[scorial] et D[idot] sont souvent si proches l’un de l’autre que la lecture de chacun peut server à mieux comprendre l’autre’ (La Geste de Fierabras, p. 137). 13 De Mandach, La Geste de Fierabras, pp. 137–40. 14 Ibid., p. 138. 15 Translations of Fierabras and Otinel are our own.

The Insular Fierabras Tradition

269

in the Fierabras tradition may well have contributed to its popularity in England.

The Hanover Fierenbrace The Anglo-Norman text that styles itself ‘de Fierenbrace d’Alisandre et del bone roy Charls’, preserved in Hanover, Niedersächsische Landesbibliothek MS IV 578, belongs to the Vulgate tradition and largely follows the narrative of the continental manuscripts. One of its few changes accords with an insular interest in Richard de Normandie: in the Hanover version alone Richard comes before Basin de Genevois in the list of ambassadors being sent to Balan, thus appearing immediately after the Roland and Naimes.16 There is, as noted above, considerable variation in the order of these laisses and de Mandach may be placing too much emphasis on this when he offers a political interpretation of the different order, as this kind of variation is very common in the chanson de geste; still, it would not be surprising for an Anglo-Norman scribe to give greater precedence to the duke of Normandy, an important character in the narrative.17 The Hanover version shows a tendency to slightly abbreviate: it has 5,828 lines compared to the 6,408 lines of the Le Person edition;18 two folios of the Hanover MS are missing, accounting for some 136 lines of this difference.19 Moreover, it seems that the copyist may have had at least two manuscripts at his disposal: at one point he gives two different redactions of the same passage, corresponding to lines 3980–4028.20 16 De Mandach, La Geste de Fierabras, p. 140. 17 A similar heightening of Richard of Normandy’s role in the version of Fierabras in the Shrewsbury Book, produced in Normandy for an English patron, is noted in Bailey, ‘Fierabras and the Livre de Charlemaine’, pp. 123, 157–8. 18 The Le Person edition is composite, based on MS E (Escorial, M. III-21) but with the ending, lost from MS E, supplied from MS A (Bibliothèque nationale de France MS fr. 12603). The complete text of MS E would probably be longer than Le Person’s 6,408 lines. For full details of all the manuscripts and sigla, see Le Person’s edition, pp. 22–56. 19 Fierabras, ed. Le Person, p. 38. 20 Ibid., pp. 39–41; Chanson de Fierabras, ed. Hilka, p. III; the repeated lines are given in the Hilka edition as ll. 4057a–4102a. Brandin, ‘Le Manuscrit

270

Charlemagne in Medieval England

No satisfactory explanation for this repetition on folio 69r of the material already copied on 68v has been found. Le Person proposes two possibilities: first, that the copyist was working from a text which already duplicated the passage; second, that he sent the gatherings already completed to the illuminator without noting properly where he had finished. Had this been the case we would expect the rest of the manuscript to follow the source of the repeated material rather than the original source manuscript, and this appears not to be the case. Another possibility, not offered by Le Person, is that the copyist was actually working on two copies at once and got mixed up between them. All we can say for certain is that it demonstrates that either the copyist, or the scribe responsible for the manuscript he was copying from, had at least two different versions of the same text before him. It is possible that one of these may have been related to the surviving texts of the abbreviating tradition.21 However, the variants are mostly orthographical, with the exception of one line which is completely different in the two versions,22 and for which there is no parallel in the extant Anglo-Norman abbreviated text in the Egerton MS.23 During the combat between Oliver and Fierabras which forms the dramatic centre of the opening section of the narrative, Charlemagne prays for God’s intervention, at some length, twice in most of the manuscripts.24 In the Hanover MS, and in the two texts of the abbreviating tradition, the first prayer is omitted completely. This may have been an early change in a common source, or the Hanover scribe may have been editing under the influence of his second manuscript source, but it could equally be that two remanieurs chose the same material to omit, being material which did not affect the progress of

de Hanovre’, gives a transcription of fol. 68v with the variants of fol. 69r in footnotes (pp. 491–3). 21 Ida Wirtz, cited in Chanson de Fierabras, ed. Hilka, p. II. 22 Fierabras, ed. Le Person, 4008–9; ed. Hilka, 3422–3. 23 A comparison of variant readings over the duplicate lines suggests the possibility that MS B (BnF f. fr. 1499) may be close to one of the sources, but this is based largely on the presence or absence of particular lines; see the variants listed in Fierabras, ed. Le Person, pp. 519–20. 24 Fierabras, ed. Le Person, 921–6; 1221–87.

The Insular Fierabras Tradition

271

the narrative.25 The second prayer is answered by Fierabras’s defeat and conversion, so is intrinsic to the progression of the narrative; however, although kept, it is considerably shortened, to only thirty-four lines (Hanover fol. 38v, ed. Hilka, pp. 58–62), almost half the length in other Vulgate manuscripts. This does not necessarily indicate any decrease in the importance of the religious element of the text, but it may point to a preference for increased narrative momentum. The combat itself, despite being the focus of the first section of the narrative, is slightly abbreviated in the Hanover MS, mostly by omitting part of the exchange between Oliver and Fierabras at the beginning of the encounter and by abbreviation of descriptive passages.26

The Shrewsbury Book Fierabras Similar techniques are used in the copy of the Fierabras in the fifteenthcentury Shrewsbury Book, BL MS Royal 15 E vi. While not an insular manuscript, as noted above (pp. 39–40), it was made in Normandy under the patronage of an Englishman to be given to the French queen of England, and the continental and insular scribes both exercise the same discipline of abbreviatio, although the Shrewsbury Book is more than a century later than the Hanover MS. This does not imply that a medieval scribe operated like a modern editor trying to produce a ‘best text’ edition, but the common practice does suggest a certain shared approach to the remaniement of the text source. The Hanover MS and, to a greater degree, the Shrewsbury Book, show scribes who bring intelligent consideration to the changes they undertake.27 The scribe of the Shrewsbury Book, while applying similar abbreviating procedures, is much more systematic in what he does, producing a text that is significantly shorter than the standard Vulgate tradition, but preserves the narrative intact.28 Like the Hanover version, the Shrewsbury text 25 For a detailed stemma of the Fierabras tradition, see Ailes, ‘A Comparative Study’, pp. 122–444. 26 Fierabras, ed. Le Person, 613–29; 699–708. The effect of this last omission is to join together two laisses. 27 For a full discussion of the Shrewsbury Book scribe’s ‘consciously modifying’ his source, see Bailey, ‘Fierabras and the Livre de Charlemaine’, pp. 89–160 (p. 108). 28 The text has 4,611 lines, compared to 6,408 in Le Person’s edition.

272

Charlemagne in Medieval England

abbreviates long prayers, repetitive formulaic exchanges, and extensive descriptive passages; at times it cuts whole sequences of laisses, but usually adapts the text around the excisions to maintain the integrity of the narrative. Indeed, its radical programme of abbreviation has more in common with the substantial remodelling of the narrative to be found in the Middle English texts deriving from the Vulgate Fierabras than with the modest interventions made by the scribe of the Hanover Fierenbrace, which, while it does show signs of the characteristic insular tendency towards abbreviation and increased narrative momentum, is otherwise a fairly regular witness to the standard Vulgate textual tradition.

The Ashmole Sir Ferumbras This Middle English text is unusual for the fact that it survives in a single manuscript copy accompanied by two sheets of parchment that contain alternative versions of a number of lines of the text, in the same hand as the manuscript, and is therefore believed to be an autograph work by the translator/poet (see Chapter 3, pp. 183–5 for further discussion of the manuscript). As a translation, it has been described as ‘full and close’, though ‘also more leisurely – perhaps one should say prolix – than its French original’. 29 However, while the Ashmole text for the most part faithfully follows the Vulgate tradition, it also demonstrates the licence to change that we noted as typical of the approach to translation in all our texts. It shares the insular preference for omitting the reprises characteristic of chansons de geste, and also extensively abbreviates two narrative passages: the confrontation of Oliver and Fyrumbras, and the final battle between the French and Saracen armies.30 The switch of metre at line 3411 from septenary couplets to six-line tail-rhyme stanzas (discussed in Chapter 1, pp. 92–3), has led to speculation that a change of source occurred at this point, but the theory is not supported by the evidence of the textual tradition.31 The suggestion that the writer here 29 Smyser, ‘Charlemagne Romances’, p. 84. 30 Fierabras, ed. Le Person, 630–1500; 5825–948. 31 Marcus Konick, in ‘The Authorship of Sir Ferumbras’ (unpublished PhD thesis, University of Pennsylvania, 1953), proposes that Ashmole lines 1–3411 derive from an early text in the ‘abbreviating’ tradition to which the Egerton Fierenbras and The Sowdone of Babylone belong; for refutation of this suggestion, see Ailes, ‘A Comparative Study’.

The Insular Fierabras Tradition

273

shifted to copying from an extant English translation, argued by Konick and quoted by Smyser (pp. 85–6), does not accord with the manuscript evidence: there is no diminution as claimed in the frequency of scribal alterations. It seems that the complex text of Ashmole 33 represents the active engagement of the translator with one or more version(s) of the Vulgate Fierabras. There is considerable evidence, however, of sustained adaptation in the details of the more expansive treatment of the story signalled in Smyser’s description. Unfortunately, the manuscript lacks its first folio, so there is no record of how the poet treated the chanson de geste prologue, but the first few pages of the Ashmole poem as it stands, which show considerable innovation, help to set the reader’s or listener’s expectations. Like most of the English romances, it adds material to increase the narrative continuity and the circumstantial richness of texture of its retelling: the text begins with an expanded account of Charlemagne’s success the previous day, when his knights slew 1,000 fleeing Saracens, and his consequent pleasure, ‘al on murȝhe’ (48), with his barons around him. This heightens the drama of the arrival on the scene of the hostile Saracen challenger, especially as the English poem introduces him with details of his enormous size and strength (51–2), imported from a later point in the French Fierabras,32 repeatedly attributes his aggression not to desire for territorial conquest but to anti-Christian fervour (‘To martyre cristen men & slee ; þat was his desyre’ (55)), and twice states his lack of fear of any Christian man, king or emperor (69, 124). All these changes construct a stereotypical image of Fyrumbras as evil Saracen giant; in a similar simplification, his opponent, Charlemagne, is identified by his most conventional attribute, his white beard (84, 94), to create a strikingly archetypal opposition of the two representative figures. The specific significance of Fyrumbras’s challenge is finally emphasized by the uniquely pointed claim that in sacking ‘Rome þy gode Citee’ and stealing ‘þyn relyqes’ (134–5) he violated Charlemagne’s personal possessions and achievements, ‘þat þou & þyne with strengþe of hond ; in heþnisse sum tyme wonne’ (136). The large-scale confrontation of Saracens and Christians is thus focused

32 The chanson de geste does not mention the Saracen’s physique at this point; Ashmole anticipates details from the description preceding his fight with Oliver (Fierabras, ed. Le Person, 604, 607–10).

274

Charlemagne in Medieval England

and epitomized in the personal opposition of the two larger-than-life individuals, giant Fyrumbras and Emperor Charlemagne. This personalized confrontation in the Ashmole Sir Ferumbras is echoed towards the end of the text (fols 71v–72r), as the narrative sets the scene for the climactic battle between Charlemagne’s army and the forces of Balan. Here the Ashmole MS presents evidence of very extensive revision in the form of additional lines added in the margins. These lines are not included in the edited text, but there is ample scribal evidence to show that they form part of a coherent programme of rewriting.33 The English poem has already extended the French source by adding an account of long conversations between Balan and his brother, the king of Persia, in which they agree their objectives in the coming battle: that Charlemagne be slain and his army defeated, and that Balan’s son Fyrumbras and his daughter Floripas, who have defected to the Christian side, be hanged and burned respectively (5425–30). The scene then switches to Charlemagne’s camp, where he briefly vows revenge for Balan’s mistreatment of his messenger and plans his battle formation. The next scene-change takes us back to Balan’s camp for another boastful speech. However, just before this last scene-change, scribal marks in the margin refer the reader to additional lines in the lower margin, which extend the narrative of Charlemagne’s preparations with a strongly worded speech of encouragement to his men: þat host of Saraȝyns þan he beheld þar by spradde þe large feld, iij myle yn lengþe & brede. Al aloud said he þen: ‘Her ys gret puple of heþene men, god of heuene ous spede. Alþoȝ þay beo mo þan we yn numbre Hure false fayth schal hem encombre & ek hir false dede. þay haue þe wrong & we þe riȝt, We schulleþ sle alle hem yn fyȝt,

33 For details of the manuscript layout, see S. H. Shepherd, ‘Four Middle English Charlemagne Romances’ (unpublished DPhil thesis, University of Oxford, 1988).

The Insular Fierabras Tradition

275

Haue ȝe none drede. Go we þerfor to þam affront & fiȝte we with þo heþene hond & maugre habben þat spare.’ ‘Amen’ saide þay euerechon, Wt þat hert was vp anon & fayn þay wel be þare. þan Char · þat host haþ ynome & esyly rydeþ at o trome, & aȝen þe Sarsyns …34

Despite the loss of the last lines of the passage, what exists here is perfectly clear and makes a significant contribution to the Ashmole version of the Fierabras tradition. It shows Charlemagne in a very positive light, fulfilling the traditional role of the commander encouraging his troops, despite their being heavily outnumbered, and inspiring their enthusiastic response; and it completes the symmetry of the narration, by balancing the earlier insertion of Balan’s speech with his brother. A further set of additional lines is visible, marked for insertion at the end of the page. Little of this material can now be read, but enough remains to indicate that it represents a further challenge added at the end of the Saracen Bruyllant’s boastful speech to Charlemagne.35 At this point it would again complete the symmetry of the narrative, so that the individual Saracen, riding out in front of his army, challenges Charlemagne to ride out in response, to meet him in single combat. While this is to some extent implied in the chanson de geste as originally translated by the Ashmole poet, the added material explicitly highlights the insular preference for named individual figures in single combat representing the two opposed sides in the conflict. Charlemagne’s added speech of encouragement, quoted above, also highlights the opposition of the two faiths in a particularly stark contrast 34 There being no more marginal space, the scribe apparently turned the leaf to finish the new material for the Charlemagne scene at the top of fol. 72v. Unfortunately, these lines are now almost wholly illegible. 35 ‘& if þou wilt nykke þat [with] nay | Com out þe self y þe pray | … | þanne kepe þe neuere … | Bote ic bere body … | …’.

276

Charlemagne in Medieval England

of true and false, right and wrong, that echoes Roland’s famous claim in La Chanson de Roland: ‘Pagans are wrong and Christians are right’. 36 In this it also echoes an earlier conversation between Balan and his counsellor Sortybrant, when they first heard of Charles’s approach with his army, and Sortybrant advised: ‘þay habbeþ þe wrong & we þe riȝt; To-morwe we schulle wyþ hym fiȝt, | & discomfytye hymen echone’ (5241–2). But while Sortybrant’s assurance was rationally based on the material fact of the Saracens’ superior numbers (5239–40), the echo of these lines in Charlemagne’s speech represents the conflict in precisely opposite terms: despite the fact that they are heavily outnumbered, the battle will give the Christians an opportunity to triumph through their faith in ‘god of heuene’. Charlemagne argues that the Saracens will be disadvantaged in battle by their defective faith and their transgressive actions, while the Christians, secure in their right belief, are assured of victory. This argument links to the similar, but more fully developed, assurance that marks Oliver’s earlier prayer as he faces 15,000 Saracens after Fyrumbras’s conversion; the Ashmole text adds to his plea for God’s protection a striking claim of justification: ‘As wys hit was in trewþe & riȝt ; þat he tok þat iornay’ (827). This hints at a concern throughout the text to reinforce a sense of confidence in the justice of the Christian cause in combating Saracens, as shown in the stress on the sacking of Rome and the theft of the sacred relics as infringements of Charlemagne’s personal rights of ownership. A crusading parallel may be seen in the arguments put forward by King Peter I of Cyprus as he travelled around the courts of Europe in the 1360s, including two visits to Edward III, seeking support for his crusade expeditions in the eastern Mediterranean. Peter’s arguments stressed the justice of his cause, since ‘his rightful inheritance and kingdom [he was titular King of Jerusalem] had been stolen’, and therefore ‘on moral, legal and historic grounds, the fight against the “unbelievers” was both necessary and just’. 37 The date of this romance (late fourteenth century) coincides with the period in which Gower’s Confessio Amantis was composed, where the vexed issue of lawful war is directly addressed in Amans’s question: ‘I preie you tell me nay or yee, | To passe over the grete See | To werre

36 ‘Paien unt tort e chrestïens unt dreit’, ed. Short/Duggan, I/1015. 37 Guard, Chivalry, Kingship and Crusade, p. 40.

The Insular Fierabras Tradition

277

and sle the Sarazin, | Is that the lawe?’ (CA, III, 2486–9),38 to which Genius does not give a direct answer, but points out that killing was not part of Christ’s command to preach the gospel, and that Christian gains made in holy wars have subsequently been lost.39 However, Gower elsewhere presents a different take on the issue, arguing in ‘In Praise of Peace’ (1399) that Christian knights are derelict in their duty to defend ‘Cristes feith’, which is under assault every day from the ‘Paiens’, when they ignore papal bulls calling for crusade against ‘the folk paien’ (192– 210),40 and that their right use of arms is to fight against ‘the Sarazins, whiche unto Crist be lothe’ (250). In this case, as Michael Livingston notes,41 the argument probably refers to strictly defensive war against the Ottoman Turks threatening Europe, the ‘mescreantz’ entering Christendom, as Gower sees it, through the ‘open gate’ of a scandalously divided schismatic Church (267–71). Reading the Ashmole text beside Gower’s contemporary engagement with the complex anxieties surrounding the issue of holy war therefore suggests that the Ashmole poet may have been drawn to the Fierabras tradition precisely because it offered an opportunity to stage a conflict between unequivocally offensive Saracens and defensive Christians. As noted above, additions in this version serve to exaggerate the anti-Christian hostility of the Saracen champion, and to emphasize the fact that the city of Rome, which he destroyed, and the holy relics he stole were Charlemagne’s own possessions. Equally pertinent is the change to Fyrumbras’s motive for this expedition: rather than seeking an encounter with the French knights, as in Fierabras, in the Ashmole text his purpose is more 38 The Complete Works of John Gower, ed. G. C. Macaulay, 4 vols (Oxford: Clarendon Press, 1899–1902). 39 The problematical nature of the issue of holy war is signalled by Genius’s referring it to God: ‘Godd do therof amendement, | So as he wot what is the beste’ (CA, III, 2514–15). Amans returns later to the topic of hypocritical churchmen unjustly sending men to slay Saracens (CA, IV, 1659–82). 40 Michael Livingston suggests Gower’s topical reference may be to one or more of Pope Boniface IX’s bulls of 1394, 1398, or 1399, in John Gower: The Minor Latin Works, ed. and trans. R. F. Yeager, with ‘In Praise of Peace’ , ed. Michael Livingston (Kalamazoo, MI: Medieval Institute Publications, 2005), note to l. 208. 41 Ibid., note to l. 250.

278

Charlemagne in Medieval England

general harassment of Christians: ‘[he] soȝte þat contree & al þat cost ; cristenmen to encombre’ (71), a formulation that seems particularly appropriate in a context of fears of invasion by ‘mescreantz’ such as Gower’s poem depicts. The central opposition remains as always that between Fyrumbras and Oliver, but here too additional material highlights the Saracen and Christian identities of the two. Fyrumbras refers to the unnamed Oliver not as ‘vassal’, but as ‘Christene kniȝt’ (402, 440), and not only boasts of his atrocities, but twice gloats that he has thereby destroyed ‘muche of cristendome’ (364, 371); Oliver confirms that he has understood the message: ‘y hure wel by þy sawe | þow hast y-do distruccion ; myche to cristen lawe; | & hast al so y-mad envy ; wiþ christene men to fiȝte’ (378–80). In his own turn, Oliver describes himself as ‘a man heȝ of mod ; Sarasynȝ to yu[e]le array’ (417) who in one day slew three kings ‘þat lyued on ȝour false lay’ (419). These insertions are brought into focus by Oliver’s offer of baptism to the Saracen: where the chanson de geste requires him to change his religious allegiance by renouncing Mahomet and believing in the Lord God, the Ashmole text, in a slightly altered treatment, presents a change of religious culture: ‘þov torndest to crestendome, | & for-soke þy false lay’ (396–7). When at last the wounded Fyrumbras embraces the offer, the Ashmole version uniquely introduces a parallel plan for the conversion of non-Christian lands in general: ‘Cristendom by me schal encressed be ; sykerly if y may scape; | & for payenye, so mot y þe ; ful yuele wil y schape; | þanne schulleþ peynymes cristned be ; & hure lay for-sake’ (760–2). This enlarged proposal speaks to the double agenda that had always theoretically informed crusading intentions: both the recovery or defence of ‘ancestral’ Christian lands and the conversion of Muslims to Christianity.42 Here too we see a parallel with Gower, across both his English works, encouraging combat with offensive Saracens in ‘In Praise 42 See John V. Tolan, Saracens: Islam in the Medieval European Imagination (New York: Columbia University Press, 2002). Tolan traces medieval views on the likelihood of victory and/or conversion as a result of military or missionary efforts or both, from the thirteenth century, when ‘the optimists will at times portray Islam as only slightly different from Christianity and portray Muslims as ready to convert to Christianity’ (p. 172), to the fourteenth century, when Turkish victories and failed missionary projects produced a far more pessimistic view.

The Insular Fierabras Tradition

279

of Peace’ as part of his argument for the healing of rifts between brotherChristians in Europe, while arguing for preaching and conversion as the proper duty of the Church towards the Saracens in their own lands in Confessio Amantis: ‘To slen and feihten thei ous bidde | Hem whom thei scholde, as the bok seith, | Converten unto Cristes feith’ (CA, IV, 1674– 6). Concerns revealed in the programme of alterations and additions in this Middle English translation thus suggest a politically inflected approach to adapting the thirteenth-century chanson de geste in a late fourteenth-century insular context, rewriting the story of Fierabras in the light of complex anxieties about Christian–Saracen hostilities, and offering a projection of renewed, idealistic crusading ambitions. The date of the Ashmole translation invites a parallel reading not only with Gower but also with Chaucer’s engagement with the same concerns in the Canterbury Tales, notably in his account of the Knight and his campaigning history in the General Prologue. As Maurice Keen has argued, the career Chaucer constructs for this fictional crusadeoriented knight (including participation in the several Mediterranean campaigns against Turkish and Mamluk port cities led by Peter I of Cyprus and glorified by Guillaume de Machaut in La Prise d’Alexandrie) represents not an outmoded ideal, but a ‘pattern of virtuous living’ for contemporary English knighthood: a pattern whose topical power derives from the fact that too few real-life members of the knightly class ‘made a serious effort’ to follow its example (just as is the case with the ideal pattern of priestly life afforded by the portrait of Chaucer’s Parson).43 However, many individuals did try to live up to the ideal, for while Edward III avoided official English involvement in plans for crusade to the East throughout his long reign, and nothing in the end came of Philippe de Mézières’ eloquent appeal in 1395 for Richard II to commit to an Anglo-Franco-Burgundian crusade against the Turks,44 historians of English crusading practices have pointed to persistent interest in the crusades throughout the fourteenth century on the level of ‘personal commitment’, particularly among noble families with a long

43 Maurice Keen, ‘Chaucer’s Knight, the English Aristocracy and the Crusade’, in English Court Culture in the Later Middle Ages, ed. V. J. Scattergood and J. W. Sherborne (London: Duckworth, 1983), pp. 45–61 (p. 47). 44 Tyerman, England and the Crusades, pp. 288–301.

280

Charlemagne in Medieval England

tradition of crusading.45 Timothy Guard has identified a good number of individual crusaders, many of whom were following the example of their fathers and grandfathers;46 this is the constituency to which both the portrait of Chaucer’s Knight and the representation of Christian knighthood in the Ashmole Sir Ferumbras are likely to be addressed, not in any sense as specific crusade propaganda, but as providing an idealized reflection of the chivalric identity of the late fourteenthcentury patron or reader.47 The Ashmole translator/poet was apparently not making merely incidental alterations to the source, but seems to have entered into a thorough reimagining of the narrative to suit a new, different set of expectations from those that the French chanson de geste was created to serve. In a smaller way, this reimagining can usefully be seen in relation to the contemporary work of Chaucer as a translator in Troilus and Criseyde, as described by B. A. Windeatt: ‘Chaucer’s own imagination acts in close response to his original story-line, building on it by thickening and enriching with interventions.’48 The effects of these 45 Keen, ‘Chaucer’s Knight’; Tyerman, England and the Crusades; Guard, Chivalry, Kingship and Crusade, p. 119; for discussion of the ongoing projets de croisade see Chapter 2, pp. 120–1. 46 In view of the Devonshire provenance of MS Ashmole 33 (see pp. 183–7), it is interesting that Guard mentions the family of the Courtenays, Earls of Devon, as prominent crusaders. Five of them are recorded as participating in crusades during the fourteenth century (in 1352, 1367–8 and 1390). Members of the Courtenay family played prominent roles at court, on the battlefield, and in the Church in the later fourteenth century: Sir Peter Courtenay (c. 1346–1405) was chief chamberlain of Richard II’s royal household and is buried in Exeter Cathedral; his brother William Courtenay (c. 1342–96) became archbishop of Canterbury (1381–96). While there is no evidence to support the speculation, it is not impossible that this Middle English translation of Fierabras was commissioned by, or intended to be presented to, a member of a prominent local family such as the Courtenays. 47 A fictional version of the crusading family tradition is invoked in The Sege of Melayne, when the dying duke of Normandy commends his son to Roland, to be brought up in the craft of arms to war against Saracens (301–12). 48 B. A. Windeatt, Troilus and Criseyde: A New Edition of ‘The Book of Troilus’ (London: Longman, 1984), p. 11.

The Insular Fierabras Tradition

281

interventions in Chaucer’s Troilus, which once led some readers to label it an early prototype of the novel,49 are equally visible in the Ashmole Sir Ferumbras: greater attention to the social texture of the world of the poem, and a fuller impression of the inner life and developing emotions of the characters. In addition to these, the Ashmole poet consistently amplifies the narrative of his French source with naturalistic details that serve to fill out the sequence of events, giving rational explanations and credible causation for the inherited plot. Even in the opening scene, where, as discussed above, detail is used to set up a sharpened stereotypical opposition of Christian and Saracen, the emotions and motivation of the characters are more fully explained. Fyrumbras not only feels angry (‘engramis’, Fierabras, ed. Le Person, 75, 81), but also resolves to stay until he has ‘foȝt is fille’ (77), swears to make himself the ‘bane’ of Charlemagne (86), and takes grim pleasure, as he sees Charles sitting and eating, in the thought of disrupting his meal (92–3). Some interventions involve change as well as amplification: for instance, from the fragmentary remains of the translation of the first laisse, it seems that (in the absence of the anterior narrative related in La Destruction de Rome) the Ashmole poet rewrote the reference to the battle in which Oliver received his wound so that it matched the different account of it given by Roland in this version (148–52), to explain his refusal to cooperate when Charles seeks a volunteer to fight Fyrumbras.50 In the rewritten Ashmole version, it is Charles and his old knights who had to be saved from Saracen slaughter by the prowess of the young peers, not the other way about, so that when Charlemagne boasts that the old knights fought better, Roland’s angry refusal to fight is more rationally explained in the light of this falsification of the facts: as he accuses 49 The locus classicus for this view is G. L. Kittredge’s essay ‘Troilus’, in his Chaucer and his Poetry (Cambridge, MA: Harvard University Press, 1915), pp. 108–45. 50 The fragmentary lines numbered 38–41 in Herrtage’s edition do not follow on from the French text he prints as lines 1–37 (line 41 repeats line 34, for example), but nor do lines 38–40 translate any material from Fierabras. They do, however, match details in Roland’s account, so it seems that, as in Fierabras, the two Ashmole accounts were identical with each other. There is therefore no reason to assume with Herrtage that the content of the missing opening lines of the Ashmole Ferumbras, or indeed the closing lines, can be supplied from the French Fierabras.

282

Charlemagne in Medieval England

Charles, ‘þov saidest ounriȝt’ (157).51 This produces a characteristic insular bias towards idealizing the hero: the Ashmole Roland retains more of his personal integrity compared with the impression of his counterpart’s peevish recalcitrance in Fierabras.52 A similar bias may lie behind a unique reversal of identities in the single-combat scene between Fyrumbras and Oliver, where in this romance it is the Christian Oliver who chivalrously allows Fyrumbras to recover his sword (678–83), and not the other way about, as in all other versions of the story. The whole encounter between Fyrumbras and Oliver is quite differently produced in the Ashmole version, which, as well as reducing the number of parallel clashes between the two combatants, omits a long sequence of laisses where the chanson de geste describes Fierabras’s arming and the origins of famous swords, uses the epic device of Oliver’s and Charlemagne’s prières du grand péril, and narrates how Oliver drinks the miraculous balm.53 The remaining narrative is characterized by the same amplification of detail noted in Ashmole’s first scene, and indeed, comparison of the revised version with the parallel lines in the draft reveals that the translator/poet returned to this passage to add new material at the second stage.54 The effect of the additions is to increase 51 The treatment of this incident in the Sowdone of Babylone closely follows the pattern of the two Egerton texts, which greatly simplify the issues between Roland and Charlemagne. The passage is lacking in the Fillingham text. 52 See van Emden, ‘The Reception of Roland in Some Old French Epics’, pp. 1–30 (pp. 10–13). 53 The omitted laisses are: XXI–XXII; XXIV–XXV; XXVIII–XXXIII; XXXV– XXXVI (ed. Le Person). Some of these omissions are partially shared with the heavily abbreviated Fierabras text in the Shrewsbury Book (BL MS Royal 15 E vi), but that version preserves the outline of the narrative intact, though abbreviated, and even enhances the account of Fierabras’s three swords (Bailey, ‘Fierabras and the Livre de Charlemaine’, p. 130). 54 Having omitted the incident of Oliver’s drinking the balm before throwing the barrels into the river, Ashmole substitutes a new detail to explain the loss of the balm. On the folio containing their final combat, two additional lines, not present in the draft version, are added vertically in the margin, stating that Oliver’s mighty blow first struck the barrel and shed its balm on the ground, before giving Fyrumbras the decisive wound (742–3). This is remarkable evidence of the translator’s rewriting in order to tidy up loose ends caused by the earlier omission.

The Insular Fierabras Tradition

283

the sense of a developing relationship between the two knights in the course of the episode; for whereas the chanson de geste employs the epic device of repeated tirades from each opponent without much progress, the Ashmole romance creates an ongoing conversation. Oliver’s assumed identity as the ‘semple’ and ‘demeyne’ knight Garyn (354, 382) piques Fyrumbras’s curiosity; he suggests they postpone their combat and talk: ‘“Christene knyȝt”, quaþ Fyrumbras ; “þou art a wonder gome: | þer is non haste in þys cas ; to fiȝt ȝut mowe we come”’ (402– 3). He explicitly enjoys their verbal sparring: ‘“Christene kniȝt”, quaþ Firumbras ; “y haue of þe god game, | Tel me nov riȝt in þis plas ; what ys þy riȝte name” (440–1). When Oliver replies that he really is Garyn, giving a circumstantially detailed fictional life story (a mini-romance plot in itself), Ashmole adds an appreciative comment to Fyrumbras’s response: ‘Fyrumbras gan to smyle: | “Garyn”, quaþ he, “þou art ful bold” ’ (443–53).55 At this point, Ashmole follows the narrative sequence found in the insular version courte (the Egerton text and the Sowdone of Babylone), so that instead of Fierabras repeating his question, Oliver revealing his true identity, and another round of tirades as in the Vulgate tradition, here the armed combat begins between Fyrumbras and ‘Garyn’, and it is only after fierce and prolonged fighting that Fyrumbras, realizing this cannot be an unknown knight, prevails on Oliver to identify himself (570–655). Ashmole capitalizes on this more naturalistic sequence of events to create a sense of the protagonist’s inner life and motivation: Fyrumbras rejoices to hear Oliver’s royal pedigree, which makes him the Saracen king’s ‘per’ (657), for never until now has he met a worthy opponent. Their renewed combat is thus marked with climactic personal significance for the Saracen champion: ‘ac now haþ he so longe y-soght ; ys peer he haþ i-founde’ (670), and when the fight culminates in Oliver’s victory and Fyrumbras’s conversion, Ashmole adds a final detail to the enhanced depiction of their relationship as Fyrumbras begs: ‘“ y pray þe iantail Olyuer ; for-ȝyf me my manace!”’ (769). The Ashmole version as a whole is marked by added references to the ‘gentle’ rank of the protagonists, both Christian and Saracen. For instance, Floripas rebukes the gaoler who refuses to act against her 55 Fierabras, ed. Le Person, 458–65. Ashmole builds on the brief details in the inherited narrative to construct a much fuller version of Garyn’s history; in the chanson de geste Fierabras merely gives a laugh in response.

284

Charlemagne in Medieval England

father’s commandment for answering back to a ‘iantail womman’ (1235),56 and it is to her status that Oliver appeals for help: ‘damesel as þov art gent ; ȝyf ous sum what to dyne’. She demonstrates her gentle heart by her response: ‘Florippe tok wel gret pyte ; of þys iantaile knyȝte’ (1277–8), and his answering promise to fight 200 Saracens on her behalf produces a feeling of admiration for his valour in her private thoughts (1291–2). As Helen Cooper points out, the romance genre is marked by its ‘increasing interest in individuals’ inward lives … especially in its treatment of love’,57 and these insights into Floripas’s emotions and thoughts, which are a significant departure from the adversarial replies of the epic Floripas (Fierabras, ed. Le Person, 2215–18, 2224–6), prepare the way for new notes of lyricism and courtliness in the depiction of male–female relations. Indeed, the whole scene in Ashmole has resonance more of the courtly, witty ‘luf-talkyng’ between hostess and guest at Hautdesert in Sir Gawain and the Green Knight than of the fierce verbal sparring between Floripas and the knights in Fierabras, as is seen in the handsome Berard’s gallant reiteration of Oliver’s pledge to fight for love of her: ‘Comly mayde of kynges kende ; þe corteyst þat i knowe, Fayr of face now beo our frende ; and we schul ben þyn owe; & For þy loue þat art so hende ; we schul boþe ryde & rowe, & þylke þat buþ to þe ounkende ; þay schulleþ be broȝt ful lowe.’ ‘Certis’, saide þat faire flour ; ‘y þanke þe swete wyȝt, þow couþest wel louye paramour ; me semeþ a lady briȝt.’ (1298–303)

The same courtly social milieu is invoked again in a later scene, where the Ashmole version adds new material that creates an impression of aristocratic leisure and entertainment among the knights and maidens in the tower: To þe soper þan wente þay alle þen, þe lordes & eke þe ientail wymen, And made hem murie þat niȝt. amorwe wanne þe sonne hure schon, Togadre þay assemblede hem euerechon,

56 Ashmole omits the gaoler’s inflammatory antifeminist views (Fierabras, ed. Le Person, 2176–83). 57 Cooper, The English Romance in Time, p. 241.

The Insular Fierabras Tradition

285

Lordes and burdes briȝt. (3415–18)

A ‘normalizing’ context is thus provided for Floripas’s suggestion (as in the inherited tradition) that each knight ‘chuse his owe’ among her damsels and enjoy their time as prisoners (3437–40). The love between Floripas and Guy is represented within this courtly context. The account that Floripas gives to Oliver of falling in love with the ‘manlich’ prowess of the ‘iantail kniȝt’ Guy (1407–24) is far more detailed than in the chanson de geste, and, uniquely, her whole account is repeated almost verbatim to Roland as a prologue to the scene of her betrothal to Guy (2072–89), including a poetic trope of their exchanging hearts: Riȝt fro þat day into þis ; myn herte haþ he y-raft. ynow y hadde of ioie & blys ; were his to me-ward laft. (2084–5)

Similar amatory idioms embellish her lament when Guy is captured: Alas! loue, wo dost þou me ; þov sturest al my blod. Alas! Guyon þe loue of þe ; wil do me waxe wod. (2795–6)

A further small but significant innovation marks the betrothal scene in both the Ashmole Sir Ferumbras and the Sowdone of Babylone: unlike all versions of the French epic, the two Middle English romances show Floripas and Guy sealing their betrothal with explicitly mutual kisses: Loueliche þay wente togadre þo ; & cussede i-same an haste, To fermye loue bytwene hem two ; & to makye hem stedeuaste. (Ashmole, 2112–13)58

This detail speaks to the pronounced concern with reciprocal married love that has been noted as characteristic of medieval English romance.59 58 Cf. Sowdone of Babylone, ed. Hausknecht, 1925–38. In the French tradition, Gui remains passive while Floripas embraces him. 59 See Felicity Riddy, ‘Middle English Romance: Family, Marriage, Intimacy’, in The Cambridge Companion to Medieval Romance, ed. Roberta L. Krueger (Cambridge: Cambridge University Press, 2000), pp.  235–52; Gervase Mathew, ‘The Heroine and Marriage’, in his The

286

Charlemagne in Medieval England

The Sowdone of Babylone adds a speech in which Floripas gives herself to Guy in words resembling the marriage service – ‘Myn herte, my body, my goode is thyn’ (1929) – and drinks a pledge to him ‘As to my worthy hosbonde’ (1934). The Ashmole Sir Ferumbras makes a comparable intervention in the later scene of Floripas’s baptism and wedding to Guy, where a speech (now partly illegible) is added in the margin: ‘To wham y pliȝte my trouþe ȝore | To haue & holde for eueremore | On wedlak fre’ (fol. 77v). Just as Floripas is to some extent re-presented in new ‘gentle’ guise, so too is the figure of the emir, her father, reinterpreted to exclude the more grotesque aspects of his role in the epic tradition, and to enhance the sense of conventional emotional reactions. The grimly comic scene where Balan falls over in his anger at hearing of his son’s conversion is omitted in the Ashmole translation,60 and the brief mention of his insane rage at the killing of his envoys is rewritten as a sad lament for the loss of his son, his best knight, his allies and his honour that causes Balan to feel ‘sorwe & care’ (1640–6).61 Even more striking is the new elegiac speech prefaced to the emir’s vow to be revenged on the French knights and Floripas: [He] saide: ‘alas! for Lucafer ; þat was so strong a kniȝt, & for my barons þat wern her ; so noble men & wiȝt! þe flour of Chyualarie now haue y lost ; al for þe loue of one, In wham y trust to alre most ; & heo me haþ by-gone.’ (2339–42)

Court of Richard II (London: Murray, 1962), pp. 129–37; Cooper, The English Romance in Time, pp. 241–51; Hardman, ‘The True Romance of Tristrem and Ysoude’, pp. 97–8. It must be noted that while French romances do not normally share this concern, married love is important in the works of Chrétien de Troyes, and also in the chansons de geste. See Finn E. Sinclair, Milk and Blood: Gender and Genealogy in the chanson de geste (Bern: Lang, 2003). 60 Fierabras, ed. Le Person, 1994–2007. The version of the chanson de geste in BL MS Royal 15 E vi also omits the whole of the laisse; on the interpretation of such scenes as comic, see Philippe Ménard, ‘Le Rire et les mentalités médiévales dans Fierabras: Une réflexion sur la violence’, in Le Rayonnement de Fierabras dans la littérature européenne, ed. Marc Le Person (Lyon: Publication du CEDIC, 2003), pp. 97–108. 61 Fierabras, ed. Le Person, 2542.

The Insular Fierabras Tradition

287

Like the portrayal of Laban in the Sowdone of Babylone (see pp. 343–5), this more nuanced view of Balan is not extended into the final scene, where he refuses baptism with the same fierce curses as in the inherited tradition; however, the role of Fyrumbras as intercessor for his father is enhanced. He appeals to Balan as ‘swete fader’ (5777, 5833, 5839), and asks Charlemagne for a stay of execution with a considered plan for effecting his father’s conversion: & he wolde eft asaye If he miȝte wyþ any colour Brynge him ȝut of his errour, Into þe betere waye. (5760–2)

Assuming that the story of Fierabras was well known in insular culture (as the many references and literary allusions imply), the translator/ poets responsible for the Ashmole and Sowdone versions of the inherited epic narrative could not have expected their audiences to anticipate a different outcome; it seems they were taking advantage of opportunities provided within the plot to reimagine moments of heightened emotion in more reflective and inward mode, in accord with the different expectations of late-medieval English romance. Given the rare status of the Ashmole Sir Ferumbras as a translation/adaptation in progress, the witness of manuscript and poem together gives unusually suggestive evidence of an attempt to produce a new, ‘modernized’ version of a valued inherited text, aimed at a readership among a chivalric, courtly elite.

The Fillingham Firumbras Of the three Middle English Fierabras adaptations, the text that survives in the late fifteenth-century Fillingham MS apparently has least to offer readers and critics. The manuscript has suffered damage, having lost several quires at the beginning, with the result that the text of Firumbras lacks almost half its original length, the equivalent of 3,068 lines in the French Fierabras, covering the duel between Oliver and the Saracen Fierabras, his subsequent conversion, and the sultan’s capture of the Twelve Peers. What remains, dealing with the peers’ escape, the defeat of the sultan and the recovery of the stolen relics, seems to be neither a polished translation of the chanson de geste, nor a particularly

288

Charlemagne in Medieval England

innovative adaptation of the tradition.62 Mehl, in his study of Middle English romances,63 describes it as ‘a rather faithful translation’, and on that account ignores it thereafter in his discussion of Charlemagne romances. A further problem is indicated by the number of unpaired couplet lines: the scribe was apparently working from an imperfect exemplar, colourfully imagined by the editor as having been ‘a jongleur’s manuscript’, ‘torn in places, faded and thumb-marked in others’. 64 According to this account, the Fillingham Firumbras is a fragmentary remnant of a neat transcript of a ‘mutilated’ copy of a workaday English translation of Fierabras. However, despite its problems, an examination of the poem in relation to the French Vulgate text can reveal interesting insights into the making of Middle English romance in the fifteenth century, a period identified by Helen Cooper as a time of topical reappropriation and formal innovation in the history of romance.65 One especially interesting symptom of fidelity in Firumbras to the French chanson de geste is its very unusual choice of metre: alexandrines.66 This suggests a verse form calqued on that of Fierabras, which nevertheless constitutes, in its Middle English romance context, an instance of formal innovation. This pattern of innovation within a frame of faithful translation is a useful way of thinking about the relation of Firumbras to its French source. In so far as Firumbras can be seen as a faithful translation, its closeness permits a detailed comparison with the French Fierabras tradition. Mary O’Sullivan points to the opening lines of the acephalous text as ‘a line for line translation of a French verse text’ (p. xxvi), but placing them beside the corresponding lines from Fierabras shows that while the English text does closely follow the order of the French, it offers a range of translation strategies, from line-for-line equivalence in

62 Barron claims that the redactor ‘treat[s] the narrative individually’, but most of the instances he cites are taken over from the French source (English Medieval Romance, p. 103). 63 Mehl, The Middle English Romances, p. 152. 64 Firumbras, ed. O’Sullivan, pp. xxiv–xxv. 65 Cooper, ‘Romance after 1400’, in The Cambridge History of Medieval English Literature, ed. Wallace, pp. 690–719. 66 See pp. 94–5 for further discussion of the verse form of Firumbras.

The Insular Fierabras Tradition

289

the first couplet, to couplet-for-line rendering, with varying degrees of omissions, doublet variation, and added material. her basnettys and her helmys they lacydyn well And her gode swerdys that were made of stell her noble targys they gonne fonge y dar say that thay dwelleden nouȝt longe. so coraious and wyȝt the gode knyhtes all To and to togedyr wentyn into the halle. ware hem nowe the saresynys & kepe hem well fro the scharpe swordys y-makyd of stel. Les coiffes relachierent, li hyaume sont fermé, Et restrainnent les brans cascuns a son costé. Les escus embraichiez, s’enn iessent aroté Hardi conme lïon, dui et dui ordené Or se gardent paiens; s’el palais sont trouvé, Ja seront o ces branz laidement confessé.67 (Fierabras, ed. Le Person, 3070–5)

The effect of the changes is to emphasize the personal involvement of the narrating voice, to add engaging value terms in respect of the ‘gode’ knights with their ‘noble’ shields, and to connect the description of the knights arming with their ensuing action, going ‘into the halle’ to use their swords ‘made of stell’ against the Saracens. By such small details, the narration is rendered more convincingly circumstantial and involving, and the reader is constructed as a supporter for the Christian knights. Another example shows the English text following the French source equally closely, and again moving from line-for-line to couplet-for-line translation:

67 ‘They relace their coifs, close their helmets, | And fix their swords at their side. | Their shields on their arms they set out | As bold as lions, two by two | Let the pagans guard themselves; if they are found in the palace | They will pay a heavy penance through these blades.’

290

Charlemagne in Medieval England

the day was fayr & clere, the sonne ganne to schyne, the knyghtys token the meddys & the valeys grene, ful noble and vaylant and hardy to fyȝt: the sarsyns schuld abye with grace of god Almyȝt! the Ameravnt balam in hys pleyng sone he behelde in the valeye and seye the knyghtys kome dryuande on her destrers, wel armyd and dyȝt with sperys scharpe & longe, & schyldys bryȝt. (339–46) Li jors fu beauz et clers, ja iert miediz passés. Li conte cevauchierent tout contreval les prés Et Turc et Sarrazins mengiüent a lors trés; N’i a celui d’eux touz ne soit aseürrés. Li amerant Balant s’est aval resgardés Et a coisiz Franchois sus les destriers armés.68 (Fierabras, ed. Le Person, 3356–61)

In this case, however, the word-for-word rendering results in a different reading. Whether mistakenly or by design, the English adaptor has taken the prepositions ‘contreval’ and ‘aval’ (across, down) as phrases referring to the knights riding through ‘the valeys’ and Balam looking towards the ‘valeye’; but the result certainly fits the pattern of expanding the original text to produce a more circumstantial narration. As well as inserting this topographical detail, the poem again adds praise for the ‘noble’ knights, and projects defeat for the Saracens; it then gives a sketch of Balam’s courtly leisure ‘in hys pleyng’ (perhaps adapting the scene of the Saracens dining in their tents), and contrasts that with an heroic, alliterative description of the armed knights, purposefully approaching with their warlike weapons on display. As these two brief examples show, the English text is indeed built on a close reading of the French, though it habitually makes alterations and additions within this framework of faithful adaptation. Some of these

68 ‘The day was fine and clear; midday had passed. | The counts ride across the meadows | The Turks and Saracens were eating in their tents. | There was not one who was not confident. | The Emir Balan looked down | And spied the French, armed, on their horses.’

The Insular Fierabras Tradition

291

changes can be seen in relation to the demands of converting the verse structure of the source into rhyming couplets, with the requirement for a fit between couplet-form and narrative content, but others seem already to show a coherent pattern of small narrative choices; and it is these choices that can reveal the guiding concerns behind the adaptation of the twelfth-century epic into a fifteenth-century Middle English romance. The adaptor of Firumbras was skilled in the art of abbreviatio: without sacrificing anything essential to the main plot, episodes are omitted and references to the excised material are carefully removed from the surrounding narrative; the expansive epic narration is systematically streamlined by cutting its characteristic repetitions and recapitulations. It has been suggested that the alterations must have been present in the French verse source, on the grounds that a Middle English translator was unlikely to have been capable of such an ‘exercise of imagination’,69 though there is no trace of such an abbreviated source among the French manuscript witnesses. On the other hand, we have the evidence of the Middle English Sowdone of Babylone, which is based on an abbreviated Anglo-Norman version of Fierabras but still exhibits its own independent programme of further abbreviation and adaptation. It thus seems safe to attribute the reshaping of the text in the Fillingham Firumbras to its English adaptor. The extant half of Firumbras preserves almost the entirety of the part played in the narrative by Floripas, the beautiful sister of Firumbras. However, in the course of the abbreviation, her original role as the forceful and enterprising Saracen princess is much reduced, including the loss of several of her characteristically fiery speeches. Instead, there is a tendency to heighten the impact of her remaining interventions, as in the incident of the Saracen attempt to fire the tower in which she and the cream of Charlemagne’s best knights are imprisoned, besieged by her father, the emir Balan. At this point Balan’s men are just renewing their attack, using mangonels to smash the battlements and setting fire to the main pillar of the tower (789–92); the knights are in a desperate situation, but Floripas proposes a cunning plan. ‘Lordyngys’, sayd Florype, ‘beȝth of gode chere! Ne hath eche man but o lyffe, selle ȝe hyt dere!

69 Firumbras, ed. O’Sullivan, p. xli.

292

Charlemagne in Medieval England

Y schall turne the fyr and the flames that ben lyȝt Aȝen on the sarisins to brenne well bryght, Thorow crafte that ȝe couthe and queyntyse of gynne.’ And the fyr sche turnyd on the sarisins with-inne. The pauylons and the tentys, they stoden on the grownde, Al to-brent in colys, in frenche as hyt ys fownde. (797–804)

The last phrase deceptively refers to the French source behind the text at precisely the moment when the English work departs from it; for in this passage the account of Floripas turning the fire that threatens the tower against the Saracens and incinerating their tents is unique to the Fillingham text; in all other versions, French and English, she merely quenches the fire. Firumbras, as an abbreviating translation of the Vulgate Fierabras, typically focuses on dramatic speeches, but compresses narrative, sometimes producing versions of events so brief as to be puzzlingly cryptic for a reader who does not know the plot. That is perhaps the case in this passage, with the account of the Saracens setting the tower on fire: ‘þe sarsyns were so wod in yre, | That þe mayster pyler of þe tour flamed alle on fyre’ (791–2). The Fillingham text has omitted the role of Balan’s engineer, who masterminded the Saracen attack on the tower. As a result of these abbreviating tendencies, the fire, no longer an instance of Saracen martial expertise in deploying artificial Greek fire, is here presented simply as a product of the Saracens’ anger, maddened by the failure of their previous attempts to take the tower. But conversely, while the Fillingham Floripas no longer needs to gather the exotic ingredients of camel’s milk and vinegar to make her special quenching mixture, the text retains the idea of her expert knowledge, and presents her ability to divert the fire and use it against the Saracens as a strategic ploy answering their military devices: her ‘crafte’ and ‘queyntyse of gynne’ against their missile-projecting ‘gynnes’ battering the walls of the tower. The incident thus does more than show Floripas frustrating the Saracens’ attempt to burn down the tower: it provides a satisfying counter-attack in which the assailants are made to suffer the destruction intended for the knights. The interesting question about this passage as ‘translation’ hinges on the occurrence of the tell-tale phrase ‘in frenche as hyt ys fownde’ (which appears nowhere else in the text), and whether it simply gestures towards the tradition behind an English version of the story at some

The Insular Fierabras Tradition

293

remove from French texts, or whether it can be read, like Malory’s use of a similar formula, as a self-conscious indicator of deliberate divergence by a translator working from a French source. The key words here are the English ‘turne’ and the French ‘atorner’. In the French, Floripas promises: ‘Je vos cuit ja le feu a noient atorner’ (‘I know how to turn the fire to nothing’; Fierabras, ed. Le Person, 3909). The word ‘turn’ is the only point of contact between these two versions of the incident. In the two other Middle English Fierabras texts that translate this episode (the Ashmole Sir Ferumbras and Caxton’s Lyf of Charles the Grete), no equivalent word is used; instead, the sense of quenching the fire is plainly expressed. The Fillingham adaptor uniquely appropriated the word, whether intentionally or by mistaking its meaning, and gave it a new physical and spatial signification, using it as the springboard for a fresh interpretation of the scene that maximizes Floripas’s agency as female mastermind, displacing the role of the Saracen engineer Mabon in the French source. The part Floripas plays in supporting the French knights against her father on account of her love for Guy of Burgundy is further heightened by a few small, significant alterations. In the first of these, Balan, infuriated by the knights’ attack on his hall, vows revenge especially on his daughter, whom he here uniquely constructs not as a whore (‘la pute’), but as a martyr for love: ‘And Florype for her loue schall be brent’ (90). A similar alteration turns Guy’s resolution to venture out in search of food from a concern for all the starving maidens in the original, to a concentration on his beloved: ‘Me had leuer for her loue, that scho hole betydde, | To suffyr on my body an 100 wovndys wyde’ (223–4). Most striking are the changes to Floripas’s behaviour as she waits for news of Guy’s escape: in the French source she watches with her maidens and shouts defiantly when she sees him, at which all the knights redouble their efforts; but in the English version the scene is reconfigured as an instance of private feeling. Floripas is alone, and her words are overheard by Guy: Mayde floryp loked out at the tour, So red as any rose was here colour. The teres for here lemman, sche let renne stylle. Thanne sey she where he com, y-armed at wylle. Tho sayde floryp, ‘my ioye wexeth blyue, Ȝut schal y haue my lemman to me a-lyue.

294

Charlemagne in Medieval England

Ȝut for al the ameraunt, hym schal y nouȝt mysse, That y schal my dere lemman bothe cleppe & kysse.’ These wordys herde Gy, and sayde to oger denys, ‘The mayde of here wordis ys ful curtays. By swete god of heuene, now y wyl asay, y wyl for here loue a lyte strokes paye!’ (691–702)

The intimacy of the scene is enhanced by the detailed observation: not only Floripas’s conventional rosy complexion, but the fact that she weeps quietly, without wiping her tears. Her joy at seeing her lover is here a personal joy that he is alive, and that despite her father’s opposition she will have him in her arms – a love that Guy immediately reciprocates in his response. This is all quite different from the French source, which is much more focused on the collective exploits of the knights, for whom Floripas’s public words of defiance to her father serve as encouragement to the whole group. Finally, the English adaptor inserts a new ending to the episode of the knights’ excursion. In place of lines expressing the knights’ confidence that Charlemagne will come to their rescue, the English text supplies the conclusion anticipated by Floripas: ‘Floryp the curtays, here tyme sche wyste | And went to here lemman and swythe sche hym kest’ (731–2). Perhaps this new, more sentimental treatment of Floripas’s desire for Guy accords with what Helen Cooper notes in Caxton’s translations from French prose romances: texts that offer ‘an éducation sentimentale, providing an abundance of human interest … alongside action and adventure’, and designed to appeal especially to a market of female readers and patrons.70 Further evidence of this enhanced representation of female

70 Cooper, ‘Romance after 1400’, pp. 703–4. In The English Romance in Time, Cooper argues the case for female readers of romance empathizing with this new ‘exploration of women’s subjectivity’ in ‘the romances’ readiness to enter into the minds of their heroines’, attitudes which she sees as ‘most characteristic of insular writings’ (pp. 218–68, quotations from pp. 220, 226). A parallel ‘sentimentalization’ may be seen in Otuel and Roland, the companion romance in the Fillingham MS, where Belysent expresses her love for Otuel in uniquely romantic terms: ‘y loue the more in hert myn | thanne y do my fadyr and al my kyn | that me to womman bere’ (611–13).

The Insular Fierabras Tradition

295

subjectivity in the Fillingham Firumbras can be seen in the later scene when Floripas is baptized: She kest of her clothys, all folke aforne, and stode ther naked as sche was borne. the good byschope that was of grete pryse Crystenede the mayde & dude the seruise. tho forsoke florype mahoun and hys lay, Toke sche to here ihesu to serue bothe nyȝt & day. Of Ciclatoun and purpre, so was al her wede, And to her thay were brouȝt in dede. And al here othyr maydens that were in the tour, All they weren cristynyd with grete honour. (1735–44)

Charles and the peers then lead her to be married: To haue & to worchype by day & by nyȝt, Syr Gy and the lady were trowthe plyȝt. (1749–50)

In the French Vulgate tradition, Floripas’s naked body is minutely described and its erotic appeal is vividly conveyed by the reaction of the barons, who are all moved by physical desire, even the aged Charlemagne.71 The Fillingham version cuts all the erotic description, and prevents the exclusively male gaze upon Floripas’s body by including all her maidens among the ‘folke’ present at the service. Her nakedness here signals her eagerness for baptism, as she casts off her own clothes, unlike the ceremonial undressing in the French source and, in an addition to the narrative, actively embraces her new Christian faith (1740). The focus on Floripas rather than the knights is enhanced by the inserted detail of the sumptuous new clothes of ‘Ciclatoun and purpre’ that are brought for her to wear, and her wedding to Guy is given added significance by the specificity of their marriage vows (1749–50), echoing 71 Compare the Ashmole Sir Ferumbras: ‘Was non of hem þat ys flechs ne raas’ (5889); on the humour of this scene in the French version see Ailes, ‘Fierabras and the Chanson de Roland’, pp. 15–16; Philippe Ménard, Le Rire et le sourire dans le roman courtois en France au moyen âge (1150–1250) (Geneva: Droz, 1969), p. 80.

296

Charlemagne in Medieval England

Floripas’s baptismal dedication to Christ. All this suggests a coherent programme of change, one maintained to the very end of the narrative. After the conclusion of festivities and Charlemagne’s departure, the writer makes a last addition to his source: all Floripas’s maidens appear again, and Charlemagne, in a final benevolent act, makes honourable provision for their future. A very different sensibility lies behind another set of small alterations in the Fillingham text. Several episodes involve the imprisoned Twelve Peers breaking out of the tower to attack the Saracens, and their heroic feats are uniquely highlighted by momentary instances of black humour. In the first episode, the peers surprise the Saracens at dinner, and use their ‘gode scharp Swerdys’ not to cut them in pieces (‘decoupés’), as in the French source, but to shave their beards (19); and while in Fierabras the text focuses on the violence of the attack, with overturned dinner tables and many Saracens killed, the English writer reinterprets the scene with a dryly witty conceit: ‘where ther neuer sarsyns sythe the world began | bettyr y-seruyd at mete of so fewe men’ (26–7).72 Other instances of this macabre use of irony include Roland’s giving a Saracen ‘a benediccioun’ instead of splitting his skull (400), and his announcing to another Saracen whom he is about to kill: ‘y schall the telle my wylle, | That neuer in al thy lyf schal thou speke ylle’ (661–2). This kind of strategy for distancing the horror of violence with ironic humour is familiar from modern action heroes such as James Bond, and black humour is reportedly also commonly used by combat troops among themselves as a protective device in real-life conflicts. These additions to the Fillingham Firumbras may reveal a comparable purpose in the fifteenth-century uses of romance. The narrative situation in Fierabras pits the peers, besieged within their tower, against the sultan’s army of 100,000 Saracens; in the Fillingham version Roland explicitly draws attention to this disparity: ‘We ne beȝt but xi to go ne to dwelle, | And there ben of the sarsins mo thanne y can telle’ (609–10).73 In light of what Jonathan Phillips calls ‘the remorseless rise of the Ottoman Empire’ in the fourteenth and fifteenth centuries,74 their defeat of the combined Christian forces at Nicopolis in 1396, and 72 This is possibly a reminiscence of the notorious scene in Richard Coeur de Lion. 73 The Ashmole Sir Ferumbras also has this feature, but the French refers to the Saracens’ strength, not numbers. 74 Phillips, Holy Warriors, p. 296.

The Insular Fierabras Tradition

297

their spectacular conquest of Constantinople in 1453, this portrait of a small beleaguered force of Christians holding their own against the might of a Muslim empire resonates strongly with fifteenth-century European fears of Turkish invasion. The use of black humour in these circumstances allows the romance to stage a performance of gallantry, of assumed nonchalance in the face of what is perceived as a terrifying threat to the survival of Christendom. Like some other Middle English adaptations of Matter of France narratives, the Fillingham Firumbras identifies the peers as Christian knights rather than as French barons, and conscripts the reader or listener into supporting them by a network of added or elaborated metanarrative material. Frequent pointed comments show the Christian forces seeking God’s will and receiving God’s grace and protection,75 and first-person exhortations invite readers and listeners to imagine a personal, emotional involvement in the story, and to connect this response with their exercise of traditional pious practices.76 These additions provide an interpretive context for the three episodes centred on the relics that occur in what remains of the Fillingham Firumbras (which is to say, the second half of the story), and all vary markedly from the Vulgate tradition. In the first of these episodes in the Vulgate, as the armed and mounted peers venture out to rescue Guy, Floripas quickly fetches the Crown of Thorns and, one by one, the peers wordlessly put it on over their helmets, before riding off, their morale boosted. Here, the relic seems to function like an amulet: as a quasi-magical object credited with the power to ward off evil. In Firumbras, however, the episode has been expanded and reimagined. As soon as Roland has vowed that the peers will risk their lives to rescue Guy, Floripas, ‘with wel god entent’ (592), brings into the chamber the whole set of relics, which in this text consists of the crown, the spear and the nails. She presents them to the peers ‘in good deuocyoun’ (593), and explicitly recalls Christ’s Passion and the shedding of his blood by Longinus’s spear, at which all the knights fall to their knees, bless themselves and pray in Christ’s name for success that day and protection from shame. Only then do they prepare to ride out. Here, the relics function conventionally as sacred 75 Lines 1, 151–2, 321–4, 367–8, 422–3, 599–600, 999–1000, 1130–2, 1815–18. 76 Lines 45, 52–3, 177–8, 363–4, 414–15, 510–11, 954–5, 1071–2, 1171, 1240, 1834–8.

298

Charlemagne in Medieval England

memorial objects provoking Christian devotion and prayer: the scene has been recast to accord with an orthodox understanding of Christian devotional practice with regard to relics and to remove any potential for reading it as a case of superstitious abuse. The next episode occurs in a heavily abbreviated passage of Firumbras; indeed, the romance omits the majority of the episode. In the French text, Floripas first promises a sight of the relics to the peers if they can successfully defend the tower; later she duly displays them, at which Duke Naimes asks to hold the relics for a while, kisses them piously, and then has the idea to take them to the window and let the sight of them blind the attacking Saracens. In Firumbras, all this is reduced to single incident (1413–22): without reference to Floripas, and in the immediate context of the Saracens’ breaching the walls of the tower, Neymes hurries to fetch the relics (the Crown, nails and spear, now plus the napkin and sudarium), in what appears to be a protective attempt on his part to save the relics from potential destruction. But some Saracens who happen to climb up the tower at that very moment are blinded by the relics. The difference between Naimes’s clever plan in the French narrative and the happy accident in the Middle English version is continued in the duke’s response in each text. In the French, he reacts with joy to this demonstration that the power of the authenticated relics assures their safety from the Saracens. In the Middle English, however, Neymes expresses simple gratitude for the assurance of the relics’ authenticity, and kneels for joy (1419–22). The romance thus avoids any implication that the relics have been appropriated by Floripas or the duke and used as an inducement or as a quasi-magical object to serve their human priorities. Instead, the relics are treated with devotional reverence (shown by Neymes’ ‘humilite’ (1413) as he hastens to fetch them), and the miraculous blinding of the Saracens occurs as a spontaneous manifestation of divine power, which has the effect of confirming the Christians’ faith and their devotion to the nails as true relics of Christ’s Passion; there is nothing here incompatible with orthodox beliefs and teaching. The abbreviating method of the Fillingham Firumbras also produces a striking demonstration of the difference between Christian relics and Saracen idols, for this scene of joyful veneration of the true Passion relics is immediately followed by an episode of iconoclasm. The peers discover a recess containing the three ‘gods’, Termagaunt, Appollyn, and Margot, which they throw into a ditch in contempt of Mahoun and

The Insular Fierabras Tradition

299

in view of the Saracens, provoking the sultan’s outrage and revengeful assault on the tower. This is unlike the corresponding passage in the French text, where the idols are thrown as handy retaliatory missiles against the already attacking Saracens; the change suggests that the English version has been designed as an instructive pair of episodes, juxtaposed in order to emphasize the difference between properly understood Christian relics on the one hand, and idolatrous worship of Saracen gods on the other. The last episode concerning relics in Firumbras is that in which Floripas finally hands them over to Charlemagne. Again, there are significant innovations in the English version. The French text creates an exotic scene, with candles, incense, silken curtains and mysterious fragrances, in which the sage bishop puts the relics to the test, silently holding them aloft until their miraculous levitation demonstrates their authenticity. But in Firumbras any potential likeness to depictions of magical practices is removed; the sensuous effects of perfumes and lights are omitted and the bishop’s testing of the relics is presented as part of the communal devotional response: he prays aloud for the grace of divine authentication, while raising up the relics ‘in the worschyp of god’ (1789). At the climax of the scene a new speech is given to Charlemagne, who humbly echoes the bishop as he prays that God approve his worthiness to keep the stray fragments fallen from the Crown of Thorns as a portable relic,77 and invites the ‘fayre myracle’ of suspending them in air that God performs in answer to his prayer. This is unlike the way the miracle is presented in any other version. Considerable care seems to have been taken in this English romance to establish a context of devout prayer and divine grace in relation to the miraculous power of the relics. The phrase ‘with goode deuocioun’, used earlier of Floripas’s reverent treatment of the relics, is repeated three times in this scene (1780, 1808, 1816) to emphasize the appropriately devout attitude of both the bishop and Charlemagne towards the relics. It appears again in the unique, heavily abbreviated conclusion, where, in a kind of postscript, the whole romance is converted into an opportunity for the listener 77 The immense value attached to such private relics can be seen in the sumptuous reliquaries that were constructed to house them, such as the Holy Thorn reliquary made in Paris in the later fourteenth century for the fabulously wealthy Jean, duc de Berry.

300

Charlemagne in Medieval England

to obtain pardon of 100 days. Readers and listeners are transformed into a congregation as the narrator calls us to prayer (1834), and then in priestly role prays for blessing and pardon on behalf of all who ‘with gode deuocyon’ have heard ‘thys gest … of the spere & the naylys and of the crovn’ (1836–7). There is no link between particular relics and any pilgrimage site; instead, the text is aimed at encouraging an attitude of pious devotion towards all relics of Christ’s Passion, through the perspective of the narration and the exemplary devotional behaviour of the characters within the fiction. Drawing together these three episodes in the Fillingham Firumbras suggests that behind them lies a coherent agenda: the inherited narrative has been remodelled with the specific aim of reinforcing orthodox teaching and correct attitudes towards relics. This agenda reflects the long history of anxiety in the Church about the potential misuse of religious objects for ‘magical’ purposes; but more pressingly, in fifteenth-century England, it works to oppose and disarm heterodox, characteristically Lollard beliefs,78 as expressed in hostility to the veneration of relics. Passion relics occupy a central place in the argument about relics and pilgrimage in the digest of heretical thought known as the Twelve Conclusions of the Lollards, which concludes satirically: ‘for if þe rode tre, naylis, and þe spere and þe coroune of God schulde ben so holiche worchipid, þanne were Iudas lippis, qwoso mythte hem gete, a wondir gret relyk.’79 The words of the prayer added at the conclusion of Firumbras (1835–7) present exactly the same combination of relics of Christ’s Passion for the reader’s or listener’s veneration. The text is putting down a firm, conspicuous marker of traditional religious orthodoxy in the way it uses the Matter of France to highlight the devotional and theological significance of relics. The characteristic features of this reworked text can thus be summarized as effects of sentiment, humour, and orthodox piety; these might all be considered conventional ingredients in the mix of Middle English romance, but what is shown by the programme of rewriting in this version of the Fierabras tradition is that the adaptor who produced the Fillingham Firumbras was employing them in specific ways to 78 See the discussion of the manuscript as a whole, pp. 197–202. 79 Quoted from Roger Dymmok’s Liber contra xii errores et hereses Lollardorum, Cambridge, Trinity Hall MS 17 (1396), in Selections, ed. Hudson, p. 27.

The Insular Fierabras Tradition

301

respond to the cultural, political and religious pressures that were shaping the fifteenth century.

Insular Additions and Responses to the Fierabras Tradition While La Destruction de Rome, which survives as a prequel only in Anglo-Norman versions, may have been an insular development (as discussed in Chapter 1), it is certainly the case that the abbreviating tradition was an innovation that is found only in the insular Fierabras tradition.

The Anglo-Norman La Destruction de Rome As we have argued, it is possible that La Destruction de Rome was an insular development of the tradition, though there is also evidence of a continental, Picard version surviving in three short fragments of sixty-seven lines in total (known as the Geneva fragments) from a manuscript of the late thirteenth or early fourteenth century.80 It is difficult to determine the precise relationship between these different accounts of the pagan attack on Rome: the most likely explanation is that there was an older version of the Destruction de Rome narrative which may have been Picard, from which both the Geneva fragments and the extant Destruction de Rome texts are ultimately descended. The fragments, however, are sufficiently different from the Anglo-Norman texts for the insular tradition to be considered a fresh reimagining of the story. These fragments aside, the Destruction is found only in conjunction with Fierabras, to which it forms a prequel, recounting the events leading up to Fierabras’s challenge on the plains of Morimonde. Most of the action of the Destruction de Rome takes place in and around Rome, thus changing the focus of the Destruction–Fierabras diptych. The main locus of activity is once more the Holy City, and the theft of the Passion relics occupies a significant place in the narration. The creation of this post-written prequel adds to our understanding of the development of chanson de geste narratives. It is established that the focus of the lost older chanson de geste, which concluded with the 80 Ernest Muret, ‘Fragments de manuscrits français trouvés en Suisse’, Romania 44 (1915–17), 215–20; see also de Mandach, La Geste de Fierabras, pp. 98–103; La Destructioun de Rome, ed. Formisano (1981), pp. 22–3.

302

Charlemagne in Medieval England

combat between Fierabras and Oliver and the conversion of the Saracen Fierabras, was much more on the attack on Rome and the theft of the relics.81 The redactor of the new Destruction de Rome may have known this lost text and used it for some of the details of his narrative, though much of the main thread of the account could be found in references left in the extant Fierabras. The Provençal Fierabras also has a prologue that parallels the Anglo-Norman Destruction.82 It is possible that two separate authors developed a new prologue for the text, developing their narratives from what was known of the older version and from references within the Vulgate Fierabras, but it could also be that both accounts share a common source. It is, however, unlikely that this source was the lost chanson, as Balan plays such an important role in the Destruction and may not have figured at all in the lost chanson de geste. The narrative of the Destruction de Rome is not complex. It opens with Fierabras and his sister Floripas in the company of their father, Balan. A sailor brings the message that the Romans have attacked some Saracen merchant ships, massacring the men. The decision is taken to avenge this and to attack Rome. During the preparations Floripas is promised to her father’s warrior Lucafer. The Saracens duly besiege Rome. Things become so desperate that the pope arms himself and joins in the conflict, but on being defeated returns to Rome.83 Meanwhile, Lucafer counterfeits the arms of Savari, one of the Roman leaders, and tricks his way past the outer walls.84 Savari is killed by the giant Estragot. The pope has sent messengers to Charlemagne, who sends Gui de Bourgogne in his vanguard. By the time Gui arrives before Rome the Holy City is in flames, having fallen to the Saracens by treachery. Fierabras himself has killed the pope and made off with the precious relics of the Passion. The French immediately set out to avenge this desecration. In a first encounter between the French and the Saracens, 81 The most important evidence for the content of this older Fierabras is the narrative given in Philippe Mouskès, Chronique rimée des rois de France, ed. Baron F. de Reiffenberg, Ac. roy. de Belgique, 2 vols (Brussels, 1836– 8); on the evolution of the chanson de geste, see n. 2. 82 Der Roman von Fierabras provenzalisch, ed. Immanuel Bekker (Berlin, 1829). 83 For a discussion of the implications of this incident, see Chapter 2. 84 In the Egerton version, Lucafer takes the banner of Savari rather than counterfeiting his arms; see below.

The Insular Fierabras Tradition

303

Oliver receives the wound from which he is suffering at the beginning of Fierabras proper. Both extant manuscripts of the Destruction de Rome ascribe the composition to a certain ‘Gautier de Douai’ and ‘rois Lawis’; the version in the Hanover MS makes direct allusion to a lost chanson de geste which is restored by this poem: Les altres jugelours, kel le vous unt countee, Ne sevent de l’estoire vailant un[e] darree. Le chanchon ert perdu[e] et le rime fausee, Mais Gauter de Douay a la chier[e] membree Et li rois Lawis, dont l’alme est trespassee – Ke li fache pardon la Virgë honoree ! –, Par luy et par Gauter est l’estoire aünee Et le chanchon drescie, esprise et alumee. A Saint Dynis de France primerement trovee, Del rolle de l’eglise escrit’ et translatee. (Hanover Destruction, ed. Formisano, 5–14) [The other jongleurs who have told you it Know nothing of value about this story. The song was lost and the verse falsified, But Gautier of Douai, of distinctive appearance, And King Louis, whose soul has passed on – May the honoured Virgin grant him forgiveness – By him and by Gautier has the story been brought together And the song been put together, and brought to light. At Saint-Denis in France was it first found Written and transfered from a roll in the church.]

There is a clear allusion here to the beginning of Fierabras itself, with its reference to a roll found at Saint-Denis, a cliché which is taken as an authority topos with no basis in fact. André de Mandach reads in this prologue a reference to the composer of the original, and now lost, chanson de geste of the twelfth century and displays considerable ingenuity trying to identify Gautier de Douai and the king, whom he takes to be a king of France.85 However, if we were to take this passage 85 De Mandach, La Geste de Fierabras, pp. 109–28.

304

Charlemagne in Medieval England

literally, it reads far more as a narrative to explain the composition of this ‘prequel’ in an attempt to restore the lost material, than as a claim by the poet of that lost original. If we were to look for a Gautier de Douai (and there are several) it would be in the mid-thirteenth century, not the twelfth; if he existed it would be as the poet of the newly composed ‘prequel’. The prologue is slightly abbreviated in the Egerton MS, which lacks the reference to a lost chanson de geste but retains the identity of the rimeur: Nuls de les autres jugelours qels le vous ont cuntee Ne sevent del estoire vailliant un darré; Mais Gauter de Doway od la chier membré, Et le roy Lawis, dunt l’alme est trepassé – Ke li face pardoun la virge honoré ! – Par luy et par Gauter fu l’estoire auné A Saint Denis de France escrit et translaté. (Egerton Destruction, ed. Brandin, 5–11) [None of the other jongleurs who have told it to you Knows anything of value about this story But Gautier of Douai, of distinctive appearance, And King Louis, whose soul has passed on – May the honoured Virgin grant him forgiveness – By him and by Gautier has the story been brought together At Saint-Denis in France written and translated.]

The reference to Saint-Denis links the Destruction and Fierabras and adds a false imprimatur of authority to the text. The allusion to Gautier could refer to a real trouvère, which would support a Picard, rather than Anglo-Norman, original, or may again be a (fictional) means to add some authority to this new poem.86 Similarly, the reference to the (conveniently dead) King Louis as a possible patron may be a device to suggest the antiquity of the text now presented to us. De Mandach dismisses the possibility that this could be a ‘king of minstrels’ rather 86 For a discussion of this, see also La Chanson d’Antioche, trans. Carol Sweetenham and Susan Edgington (Farnham: Ashgate, 2011), pp. 7–8. There are a number of echoes of the Fierabras tradition in the Chanson d’Antioche.

The Insular Fierabras Tradition

305

than a historical king of France, but this is largely because he is seeking evidence that the Destruction which has come down to us is part of the original lost ‘diptych’ of Destruction–Fierabras; he therefore concludes that there can be no doubt this is a reference to Louis VI of France.87 The evidence of Philippe Mouskès’ Chronique rimée would suggest rather that the lost text, what de Mandach calls the ‘Fierabras primitif’, was not a diptych at all, but a coherent narrative of the destruction of Rome and taking of the Passion relics which would have ended with the combat scene. What we can infer from the prologue is an awareness that there was a lost chanson de geste and that this is an attempt to reconstitute that narrative and thus complete the ‘geste’ of Fierabras. It is likely that the version of the Destruction de Rome in the Hanover MS is older and closer to the original text than that of the Egerton MS, which shows signs of abbreviation consistent with the techniques used in the Egerton Fierenbras. There are ‘fossil’ readings left in the Egerton version which seem to indicate that changes have been made but not carried through consistently. Thus, for example, in the Hanover Destruction when Lucafer asks for Floripas’s hand he boasts he will conquer France as far as Monpellier (248–50). This is echoed in Floripas’s reply: ‘Sire’, dist Floripas, ‘ceo ert au repairier, Quant vous avrés pris France et conquis Monpellier.’ (Hanover Destructioun, ed. Formisano, 285–6) [‘My lord’, said Floripas, ‘It will be on your return When you have taken France and conquered Monpellier’.]

In the Egerton version, Lucafer makes no mention of Monpellier, but Floripas does: ‘Qant il vous ad conquis Rollant et Oliver Et totes les terres desci k’en Monpeler’. (Egerton Destruction, ed. Brandin, 328–9)

87 De Mandach, La Geste de Fierabras, pp. 112–13; he fancifully surmises that the warrior prince would have helped the poet ‘dans ses descriptions de duels, de batailles “en plein champ”, de prises de ponts et de fortresses’ (p. 113).

306

Charlemagne in Medieval England

[‘When you have conquered Roland and Oliver And all the lands from here to Monpellier’.]

Similarly, one of the three messengers sent to Charlemagne for help is named in the Hanover version as Jeffroi (1122–3). No name is given in the Egerton text, but later, when Gui is approaching Rome with his army, the remanieur appears to have forgotten that he had not named the messenger, who now appears pointing out the burning city of Rome to Gui (891). The Egerton remanieur abbreviates both the Destruction de Rome and Fierenbras. This abbreviation is far more radical than the omission of lines of description and combining of a couple of similar laisses which we find in the Hanover MS of Fierabras or the Shrewsbury Book, though such omissions are frequent, with, for example, the Egerton Destruction lacking descriptions of Floripas (Hanover Destruction, 252–62), and of the interior of Balan’s ship and of Floripas’s room (Hanover Destruction, 333–60).88 Additionally, there is some recasting of the narrative, particularly for Fierabras (see below), and some simplification of the narrative in the Egerton Destruction, largely by the omission of the adventures of the messengers on their way to Charlemagne from the pope (Hanover Destruction, 1132–46). It seems most likely that the reason for this omission, as for the omission of descriptive passages, is a focus on the main events of the plot; the result is increased narrative momentum. The main difference between the two accounts of the Destruction de Rome lies in the heraldic trick used by Lucafer to enter Rome. In both the Hanover Destruction and the Middle English Sowdone of Babylone this is a planned ruse, with Lucafer counterfeiting Savari’s arms to trick his way inside the walls, though the emphasis in the Anglo-Norman and Middle English texts is different, with the Hanover text saying little about the preparation and concentrating on its execution, and the Middle English version showing more interest in the planning (Hanover Destruction, 1018–42; Sowdone, 313–27). The Egerton text is unique in presenting Lucafer as an opportunist rather than a clever strategist, as he simply takes Savari’s own banner. There is also some (possibly incidental) change in the depiction of Savari. By shifting focus away from the planning, the Egerton Destruction again achieves increased narrative momentum. 88 For details see ‘La Destruction de Rome et Fierabras’, ed. Brandin, pp. 23–4.

The Insular Fierabras Tradition

307

The fully developed Destruction de Rome survives only in these Anglo-Norman manuscripts, Hanover and Egerton, in significantly different redactions. The redaction in each manuscript corresponds to the treatment in that manuscript of the Fierabras narrative. The longer Hanover Destruction is followed by a text witness to the Vulgate Fierabras, while the shorter Egerton Destruction is paired with the abbreviated Anglo-Norman Fierenbras. The illuminations are also adapted to accommodate the abbreviated narratives, indicating that the illuminator, or the person instructing him, was aware of the textual changes. The manuscripts are clearly closely related in both text and illumination.89 The most likely reason for this would be a common source. It is somewhat problematic, given the clearly close relationship between the Hanover and Egerton MSS, that the two texts of the Hanover MS are of a slightly different date, the Destruction de Rome section being dated to the late thirteenth century and the Fierabras to the early fourteenth; it may suggest only that the illuminator’s model and the copied Destruction section were in the workshop for some time and the Fierabras section then added. As the Egerton MS and the Sowdone also share a source, the common source behind the Egerton and Hanover MSS is at least at one remove from the extant manuscripts. La Destruction de Rome includes some particular themes that may relate to the historical and political context of its composition. These are worth exploring in more detail. Studies of the Fierabras tradition have suggested that there is, in the attack on Rome, a recollection of three historical attacks on the papal city, in 536–7, in 846 and in 1081–4;90 the echoes of the sixth-century attack are found only in the Destruction de

89 See Stones, ‘The Egerton Brut’; see also Chapter 2. 90 Gordon Knott, ‘The Historical Sources of Fierabras’, Modern Language Review 52 (1957), 504–9; A. Graf, Roma nella memoria e nelle immaginazioni del medioevo (Turin, 1882); M. Roques, ‘L’Élément historique dans Fierabras et dans la branche II du Couronnement Louis’, Romania 30 (1901), 161–83; Bédier, ‘La Composition de la chanson Fierabras’; Philippe Lauer, ‘Le Poème de la Destruction de Rome et les origins de la cité léonine’, Mélanges d’archéologie et d’histoire publiés par lécole français de Rome 19 (1899), 307–65; Ailes, ‘A Comparative Study’, p. 16.

308

Charlemagne in Medieval England

Rome and not in any extant version of Fierabras itself.91 Such historic attacks on Rome would certainly have continued to hold some interest, and the memory of the early attacks was kept alive, in the minds of the educated at least, by the inclusion in Gratian’s Decretals, a key element of the teaching of canon law, of letters by the Popes Gregory the Great (590–604) and Leo IV (847–55) regarding the defence of Rome against the Lombards and the Saracens respectively.92 The poet of the Destruction was perhaps prompted to add this prequel by the unsatisfactory beginning, in medias res, of the continental narrative of Fierabras, with its allusions to previous events. However, it may be that the situation of the Papal State at the time of the composition is also being addressed by the poem. From the time of Charlemagne on, relations between the pope and the empire were complex. A succession of donations of land to the pope from the emperor, beginning with Pepin in 754 and followed by that of Charlemagne himself in 774, led to disputes over the extent of papal authority which would affect the relationship between the Church and the empire throughout the Middle Ages.93 Indeed, while the siege of Rome by Aistulf, King of the Lombards, in 756 should perhaps be added to the list of historical attacks on Rome which may be echoed in the Saracen attack (and Frankish response) in the Destruction de Rome, there is a significant 91 Knott pointed out a parallel between an incident in the Destruction and the siege of Palermo by Belisarius, described by Procopius of Caesarea. This would suggest a knowledge of Greek, a level of education much higher than most priests would have had, though not impossible if our text came out of a monastic context; Knott himself admitted that it was not possible to trace lines of transmission or ‘any sound link’ between Procopius and the extant texts (Knott, ‘The Historical Sources’, pp. 506–9). It may be that the author of the Destruction knew the now lost, older version of Fierabras which recounted the attack on Rome and ended with the defeat of Fierabras in combat with Oliver (discussed above), and that his apparent historical knowledge, rather than indicating a high level of erudition, came from that older tradition. 92 Rist, The Papacy and Crusading, pp. 30–1. 93 Daniel Waley, The Papal State in the Thirteenth Century (London: Macmillan, 1961), pp. 1–16; Peter Partner, The Lands of St Peter: The Papal State in the Middle Ages and the Early Renaissance (London: Methuen, 1972), pp. 19–29.

The Insular Fierabras Tradition

309

difference in the contemporary context, for whereas in the eighth century the Franks had relieved the siege of Rome, in the thirteenth century it was the empire itself which threatened the papacy. There was a succession of attacks against the Papal States by different emperors, and popes in turn excommunicated the emperors: in 1210 Viterbo was besieged by Otto IV, who would later be excommunicated by the pope;94 Frederick II’s subordinates attacked Ancona in 1228;95 in 1239 Frederick II himself invaded the Duchy of Spoleto;96 a year or so later Frederick marched against Rome itself.97 During the short reign of Emperor Conrad IV and the disputed succession of Manfred, after the death of Frederick II, Pope Innocent IV tried both negotiation and war; in 1258, during the papacy of Alexander IV, Manfred attacked the March of Ancona.98 The tensions between the empire and the papacy were also linked to the commitment of both to the idea of the crusade. Frederick II had been manoeuvred into undertaking a crusade but postponed the fulfilment of his vow until Gregory IX, in autumn 1227, excommunicated him. This led, somewhat strangely, to his departure on crusade whilst excommunicate.99 Wars between the two powers continued through the 1240s, following a second excommunication of Frederick by the pope. In the thirteenth century, the relationship between pope and empire was perhaps at its nadir. It was clear, moreover, that, from the time of Innocent III onwards, the popes regarded the defence of the Papal States as meritorious. Innocent granted to those who fought against his enemies ‘the same plenary indulgences as for those fighting

94 Partner, The Lands of St Peter, p. 243. 95 Ibid., p. 246. 96 Waley, The Papal State, p. 145. 97 Rist, The Papacy and Crusading, p. 187; Thomas Curtis van Cleve, The Emperor Frederick II of Hohenstaufen (Oxford: Clarendon Press, 1972), pp.  442–54. In a dramatic move Pope Gregory swayed the Romans against opening the gates to the emperor by processing from the Lateran palace to St Peter’s with relics of the Cross and of Saints Peter and Paul (van Cleve, The Emperor Frederick II, p. 444). 98 Waley, The Papal State, p. 153, pp. 162–5. 99 Partner, The Lands of St Peter, p. 246. Frederick would, while still excommunicate, take possession of Jerusalem (Philips, Holy Warriors, pp. 228–39).

310

Charlemagne in Medieval England

Muslims in the East’. 100 Particularly significant here is the kind of language used by the popes when they were calling for support against their political enemies. Innocent III described Markward of Anweiler as ‘morally worse than the infidels, another Saladin and a confederate of Muslims’. 101 The papal historian Rebecca Rist has noted that ‘certainly by February 1241 Gregory regarded his campaign against Frederick as a crusade … [and] … as early as 1239 he had called on his followers to take up the Cross to fight against the emperor in defence of Sts Peter and Paul.’102 A connection between these historical events and the composition of a poem in England or western France some time in the middle of the thirteenth century can only be a matter of speculation, and one of the difficulties in understanding how far the poem can be considered a response to all that was happening in Italy is the imprecise dating of the text.103 Even so, some link between contemporary events and the poem seems fairly likely. First of all, it is evident that what affected the pope affected the whole of Christendom. There were, moreover, specifically English concerns. One key element here seems to have been the pope’s demand for subsidies from England, which gave England a more than general interest in what was going on.104 England was also brought into the dispute over the kingdom of Sicily, with Henry III’s son Edmund a contender.105 Richard of Cornwall’s title as King of the Romans and 100 Rist, The Papacy and Crusading, p. 176. 101 In his letter ‘Quod future sint’; see Rist, The Papacy and Crusading, p. 196; a similar condemnation is found in Turpin’s outburst against Charlemagne in The Sege of Melayne (see Chapter 6). 102 Rist, The Papacy and Crusading, p. 186; see also Norman Housley, The Italian Crusades: The Papal–Angevin Alliance and the Crusades against Lay Christian Powers, 1254–1343 (Oxford: Clarendon Press, 1982). 103 There is no evidence to place the text any earlier than the middle of the thirteenth century, and it may have been later. Neither copy appears to be an autograph, so the composition can be securely assigned to a date earlier than the older of the two manuscripts: before the end of the thirteenth century. 104 Partner, The Lands of St Peter, p. 253; on further levies raised by Urban IV, see p. 262. 105 Both Innocent IV and Alexander IV negotiated with England about this, with Alexander clinching the agreement in 1254; Waley, The Papal State,

The Insular Fierabras Tradition

311

his election as senator for life in 1261 may also have brought events in Rome to the attention of England,106 and indeed, another Richard, Richard de Normendie, plays a role in the relief of Rome in the poem. Perhaps more important than the financial appeals of the pope was the detrimental effect all this had on the crusading movement. As noted above, Frederick II departed on crusade while excommunicate. Louis IX of France, who had a genuine commitment to crusade, was delayed from his departure by his need to support his brother’s candidature to the throne of Sicily, a candidature favoured by the pope more to prevent Frederick II’s illegitimate son Manfred from gaining the throne than for anything else. At the very least, the poem speaks into this situation by presenting an image of a unified Christianity at war against the enemies of Christendom – a war which includes, or perhaps has as its prerequisite, the need to ensure the freedom of Rome. The role played by the pope in the Destruction de Rome provides another possible instance of ideological or political engagement. When Rome is most seriously threatened, the pope arms himself and joins the fight. The issue of clerics participating in warfare is a vexed one, possibly more problematic for modern readers than medieval listeners, though canon law forbade clerics to carry sharp weapons.107 It is clear, for example, that Odo of Bayeux supported his half-brother William the Conqueror at the Battle of Hastings, and he is depicted in the Bayeux Tapestry wielding a club.108 Odo was not a unique figure in p. 153; Partner, The Lands of St Peter, p. 261. 106 Partner, The Lands of St Peter, p. 261. 107 Matthew Strickland, War and Chivalry: The Conduct and Perception of War in England and Normandy, 1066–1217 (Cambridge: Cambridge University Press, 1996), points out that ‘militant clergy who violated canon law by the bearing of arms, were comparatively rare’ (p. 73), yet powerful churchmen might well lead and direct military operations; for some English examples see Michael Prestwich, Armies and Warfare in the Middle Ages: The English Experience (New Haven, CT: Yale University Press, 1996), pp. 168–70. The unease with which a bellicose prelate was viewed can be seen by the injunction placed on Henry Despenser, Bishop of Norwich, after his expedition to Flanders; see Housley, ‘The Bishop of Norwich’s Crusade’, p. 20. 108 He was also one of William’s commanders, according to the testimony of Orderic Vitalis, cited in Strickland, War and Chivalry, p. 73, n. 88.

312

Charlemagne in Medieval England

practice, and in literature the figure of the warrior-archbishop Turpin, fighting alongside Roland and the peers at the Battle of Roncevaux, is one of the well-known characters of the Chanson de Roland.109 While our text has close links to the Chanson de Roland, it is closer in date to Thomas Aquinas, who died in 1274.110 The fact that Aquinas addressed the question in his Summa Theologica indicates that it was at this time a live issue. In reply to the question over whether it is lawful for clerics and bishops to fight, he offers the following uncompromising answer: ‘Warlike pursuits are altogether incompatible with the duties of a bishop and a cleric.’ 111 Aquinas does, however, consider it to be the role of priests and bishops to induce others to fight, giving the example of ‘Charles [who] went to war with the Lombards at the instance and entreaty of Adrian, bishop of Rome’.112 In Aquinas’s dialogical treatise, these are the words of the objection to his argument, and are used to suggest that it was lawful for clerics to fight, but to that objection he further replies that ‘prelates and clerics may … take part in wars, not indeed by taking arms themselves, but by exhorting and absolving them. … It is the duty of clerics to dispose and counsel other men to engage in just wars.’113 This is how the pope in La Destruction de Rome had been behaving until, in desperation, he takes arms to defend Rome himself. We do not mean to argue here that the poet knew the writings of 109 On Turpin as warrior priest, see Edmond Faral, ‘A propos de la Chanson de Roland: Genèse et signification du personnage de Turpin’, in La Technique littéraire des chansons de geste: Actes du colloque de Liège (Paris: Les Belles Lettres, 1959), pp. 271–80; C. A. Robson, ‘The Character of Turpin in the Chanson de Roland’ , Medium Ævum 10 (1941), 97–100. Edmond Faral, La Chanson de Roland: Étude et analyse (Paris: Mellottée, 1934), notes that Turpin defeats the Saracens who seem most inimical to Christianity (p. 221); Ailes, The Song of Roland: On Absolutes and Relative Values, p. 20. 110 Thomas’s own life seems to have been shaped by the tensions between the pope and the emperor, which may have led to his flight to Naples; Josef Pieper, Introduction to Thomas Aquinas (London: Faber, 1962), p. 10. 111 Summa Theologica, II, question XL; The Summa Theologica of St Thomas Aquinas, trans. Fathers of the English Dominican Province, 18 vols (London: Burns Oates and Washbourne, 1911–22), IX, 503–6. 112 Summa Theologica, IX, 504. 113 Summa Theologica, IX, 506.

The Insular Fierabras Tradition

313

Thomas Aquinas (though he may have done), but to demonstrate that this was a concern in the latter part of the thirteenth century. Unlike the poet of the Chanson de Roland, this poet does not simply accept the presence of the warrior bishop, or, in this case, pope. The pope comes face to face with Fierabras himself, is defeated by him, and the Saracen is horrified at discovering that his victim is a priest. Not killing the defeated pope, he adjures him to return to his proper duties: La teste li vousist trencher od s’espé fourbéz; Mais, qant la ventaille li fu du cole oustéz, Fierembras vist la corone qe novelement fu raséz. Qant cil l’a aperceut, si fu espountéz: ‘Hai!’ dit il, ‘veillard, mult m’as vergundéz Jeo quidai aver jouste od roi ou admiréz. Ore me su od un prestre el champe melléz. Meultz te vaudreit en ton cloistre tun sauter solferz Et les seines en berefrois traire et soner Ke en chambe ou en bataille escu ne lance portier.’ Fierembras corust sun destrere seiser: ‘Va tost, si muntéz! Mahon te poet encombrer! Kar, se jeo te ocie, ceo me serreit al reprover.’ Qant l’apostoille l’entent, si prent ajoier. Tantost s’en hast sun bon destrere muntier. Fierembras del ost s’en prent a conveier. Grant curtesie en fist, si le lesse aler. (Egerton Destruction, ed. Brandin, 650–6; see also Hanover Destructioun, ed. Formisano, 810–47) [He wanted to cut off his head with his sharp sword But when his ventail was removed Fierenbras saw his recently shaved tonsure. When he saw this he was shocked: ‘Old man’, he said, ‘you have shamed me. I thought I was fighting with a king or emir But now I have taken the field against a priest; You would be better singing your psalms And sounding out the bells in your belfry Than in chamber or in battle carrying shield or lance.’ Fierenbras ran to take hold of his horse:

314

Charlemagne in Medieval England

‘Quick, mount! May Mohammed bring you down! For if I kill you I will be reproved for it.’ When the pope hears him he rejoices He hastens to mount his warhorse Fierenbras has him taken from the army He showed him great courtesy and lets him go.]

This episode affects the depiction of both the pope and Fierabras. The Destruction poet faced a challenge in the portrayal of Fierabras. This Saracen is destined to convert, and in the extant Vulgate chanson de geste the way is paved for that conversion by his behaviour, outdoing the Christians in courtliness.114 However, he had also led the attack on Rome when the relics were taken and, as the opening laisses of Fierabras recount, had killed the pope. The same chivalrous knight who conducts a single combat with Oliver in Fierabras has previously committed atrocious acts of war. Psychological depth is perhaps not a feature of the chansons de geste, and the composer of the Destruction does not resolve this dichotomy in the presentation of one of his major characters. Fierabras, who considers his honour besmirched by having fought with a clergyman, and who, in the Egerton version, escorts him back to the city of Rome,115 will, on entering that city, kill the same pope when he finds him before his altar, and go on to kill a saintly and elderly monk who has shown him the relics. While Fierabras sparing the pope in battle may balance the depiction of his killing spree when he takes the city, it seems that the function of the first scene is more concerned with what it says about the pope than about Fierabras. In his declaration that no honour would come to him for fighting a priest, Fierabras may be showing some contempt for churchmen, in the spirit of one who knows his own place in the warrior aristocracy, but he also seems to be showing contempt for one who is out of his place. Up to this point the poet has described the pope’s involvement in the combat in positive terms, showing him leading his men like any other chanson de geste hero. Having encouraged the Romans to fight he mounts himself:

114 Ailes, ‘Chivalry and Conversion’, passim. 115 In the Hanover text, Savari and Garin escort him out of the press (ed. Formisano, lines 845–7).

The Insular Fierabras Tradition

315

L’apostoile i monte l’afferant destrer; Un lance tint el maine, ki out bele baner; L’ymage saint Pier i fist putreiter. Vint mil i muntierent, qe burgeis ke chivaler. L’apostoil leve sa maine; de Dieu lor veit seigner. (Egerton Destruction, ed. Brandin, 611–15; see also Hanover Destructioun, ed. Formisano, 815–17) [The pope mounts his warhorse. He holds in his hand a lance, with a fine banner, The image of Saint Peter portrayed on it. Twenty thousand burghers and knights mounted. The Pope raised his hand and made the sign of the Cross, giving God’s blessing.]

This juxtaposition of the function of priest and the function of warrior is precisely the issue for Thomas Aquinas. The criticism by Fierabras is not lessened by coming from the mouth of a pagan; it is rather that even the pagan has a better sense of what is proper than the pope is displaying. On the other hand, before the pope takes the initiative things have been going badly for the Romans, and all medieval warriors would accept the argument he makes that ‘Meulz nous valdreit la defors a honur morir | Qe en ceste cité ad doel lunges languir’ (‘It is better that we should die out there with honour than suffer passively in this city’).116 This is language that medieval chivalric society would understand. The poem is ambivalent about the pope’s actions. A century earlier, the papal annalist had ascribed to Pope Gregory VIII this pacifist ideal that ‘Popes and cardinals may not safely take up arms and wage war: they should only give to the poor and praise the Lord day and night.’117 It seems that Fierabras would have agreed with him. Perhaps the criticism is directed beyond the pope, to those who have not yet come to his rescue. Robert Warm, writing about The Sowdone of Babylone, makes the point that ‘Charlemagne’s intervention is vital. The pope requires assistance, and he receives that assistance from a secular king who is aware of the responsibilities and obligation that he owes 116 Egerton Destruction, ed. Brandin, 607–8; this clear articulation of the warrior ethos is not found in the Hanover text. 117 Quoted in Partner, The Lands of St Peter, p. 219.

316

Charlemagne in Medieval England

to the wider Christian community.’ 118 This observation is just as valid for the Egerton Destruction and its shared source with the Sowdone. The pope’s need for secular support takes us back to the contemporary situation in the Papal States. The fact that the pope himself should be involved in waging war does seem to be open to criticism by the Destruction poet. Whether the poem was originally written in response to the situation in northern Italy in the later thirteenth century or not, it would certainly have resonated with the warrior elite of England at that time, given that the dangers facing the Papal States and the holy city of Rome could be ascribed to lack of secular support.

Insular Fierabras Texts in the Abbreviating Tradition The abbreviated text of Fierenbras, named by Marc Le Person la version courte, is found in only one manuscript in ‘langue d’oil’, namely BL MS Egerton 3028.119 This same abbreviating tradition is the source of one of the Middle English texts, The Sowdone of Babylone. While the Egerton and Hanover MSS present the two narratives of the Destruction de Rome and Fierenbras in a sort of diptych, but nevertheless mark each as a separate text, the Sowdone carries the process further and joins them as one narrative. The abbreviating tradition as extant in Fierenbras and the Sowdone of Babylone radically recasts the narrative in the process of substantial abbreviation. Some of the differences between the Egerton text and the Vulgate Fierabras are shared by the Sowdone of Babylone; others are not. Their common source could be considered to constitute an original Anglo-Norman version of Fierabras as there is no evidence that texts in the abbreviating tradition circulated on the continent. Before we turn to each of the texts individually, it is important to consider the changes which they share and which can be considered Anglo-Norman innovations. Some of the changes to the narrative common to the Anglo-Norman Fierenbras and Middle English Sowdone have the effect of bringing closer together the narratives of the Destruction de Rome and Fierabras. One particularly interesting change, in terms of what it reveals about 118 Warm, ‘Identity, Narrative and Participation’, p. 89. 119 Fierabras, ed. Le Person, pp. 23–56.

The Insular Fierabras Tradition

317

the remaniement of chansons de geste, is the place given to Fierabras’s father, named Laban in these redactions rather than Balan. In the Vulgate Fierabras Balan does not actually appear at all until after the combat scene, an indication that he may not have figured in the older, lost chanson de geste.120 In the Egerton–Sowdone branch Balan/Laban, an important figure in the Destruction de Rome, is introduced into the narrative at a much earlier point of the Fierabras story (Fierenbras, 15; Sowdone, 1683, early in the Fierabras section of the text). This adds greater coherence to the narrative. In the Vulgate, Balan’s appearance in what has been called the ‘queue postiche’, the developed narrative which was added on beyond the point of Fierabras’s conversion, is unexpected, much like Baligant’s sudden appearance in the Chanson de Roland; in the Fierenbras texts Balan/Laban, the leader of the Saracens in the Destruction, is the named leader throughout the text. This argues for a new aesthetic where cohesion has a higher priority. It is not unusual in early chansons de geste for inconsistencies to be left when a text is first remanié. The abrupt introduction of Baligant is not a unique case of what in modern terms would be deemed carelessness.121 In a medieval context it is ‘careless’ only in the sense that it apparently did not matter; in a genre that was largely orally disseminated, a tightly structured narrative was not part of the aesthetic. By the time we come to the Anglo-Norman remaniement it seems that coherence is beginning to be more important. The most drastic change to the narrative in the abbreviating tradition is triggered by having only two peers, Roland and Oliver, taken prisoner by the ambushing Saracens after the combat scene, where in the Vulgate five are taken. The effect of this is twofold: first, more emphasis is placed on the best-known heroes; second, ten peers are sent to the rescue instead of seven. This gives us an extended sequence of laisses parallèles in Fierenbras while each peer in turn is nominated to be sent as a messenger, this despite a tendency in the abbreviating branch to reduce repetition and parallelism. It demonstrates that the 120 On the older version of the narrative and the titles given it see Chapter 2, n. 38. 121 There are a number of examples of inconsistencies in the Chanson de Guillaume, including a move of Guillaume’s main base from Barcelona to Orange; see P. E. Bennett, ‘La Chanson de Guillaume’ and ‘La Prise d’Orange’ (London: Grant and Cutler, 2000), pp. 16–22.

318

Charlemagne in Medieval England

poet-remanieur remains aware of chanson de geste discourse and able to exploit its potential. This is particularly striking in an abbreviating tradition which, as noted above, increases narrative momentum, sometimes at the cost of dramatic presentation. However, the omission of such parallels clearly did offer an easy way to abbreviate the text, so when the messengers arrive at their destination neither the AngloNorman nor the Middle English retains this parallelism when the message is delivered: in each case only Naimes is given the opportunity to speak, as Balan then expresses anger and does not give the others a chance to repeat the message. A similar opportunity for abbreviation comes when Floripas, having rescued the peers from gaol and taken them to her chamber, asks each in turn his name (Fierabras, ed. Le Person, 2882–97). Both the Anglo-Norman and the Middle English texts cut out the repetition. These omissions make the retention, indeed development, of the technique in the sending of the messengers all the more striking.122 In the Vulgate this scene has clear echoes, which may be parodic, of the famous council scene in the Chanson de Roland where peers in turn jump up to offer to be the messenger on the dangerous mission to Marsile, each being rejected by Charlemagne because of the risk involved (ed. Short/Duggan, 244–73).123 Any attempt to explain the development of parallelism in this one case, where it is reduced or lost elsewhere, can only be conjectural, but it may be that the remanieur appreciated the echo of the Roland tradition, without necessarily reading it as humorous. A general paring down of the narrative contributes markedly to increased narrative momentum. Whole incidents are omitted, though some of these are short. Naimes’s declared refusal to be left behind in Aigremore to guard the gate (Fierabras, ed. Le Person, 3324–32) is lacking in this branch of the tradition. The capture of the Saracen Espaulard (Fierabras, ed. Le Person, 3996–4037), which does not actually affect the rest of the narrative, is omitted completely. Richard’s escape from the besieged Aigremore, which is beset by obstacles in the Vulgate, is accomplished with greater ease, and his adventures en route to Charlemagne are omitted (Fierabras, ed. Le Person, 4172–550; compare 122 In the unique manuscript copy of the Middle English Sowdone of Babylone, each peer’s nomination is marked for the rubricator to insert an enlarged initial capital (fol. 42). 123 Ailes, ‘Fierabras and the Chanson de Roland’.

The Insular Fierabras Tradition

319

Fierenbras, ed. Brandin 1284–8). While such changes serve mostly to abbreviate the narrative, another omission (perhaps incidentally) alters the depiction of Floripas, considerably simplifying it, while at the same time omitting an opportunity to raise the theological question of how God’s efficacy is seen at work. In the Vulgate, when the peers are confined in the tower of Aigremore with Floripas and her maidens, Floripas takes the peers to worship her idols, which, she claims, can better supply their needs than the ‘roi’ in whom they believe:124 Mout est petit puissant le roi ou vos cre[e]z; S’eüssiez des essoir les nos dex apelez, A maingier eüssiez et a boivre a plentez. (Fierabras, ed. Le Person, 3265–7) [The king in whom you trust has little power; Had you called upon our gods since yesterday, You would have plenty to eat and to drink.]

When the peers arrive where the idols are kept, to Floripas’s horror, they destroy the images. This presents us with a Floripas who is not yet converted, who continues to believe in her pagan gods and in their efficacy. Yet the Anglo-Norman remanieur omits the whole scene. A further instance of simplification shows the remanieur working across both texts in the joint Destruction–Fierabras tradition, where there are several giants. In the Destruction the giant Estragot is killed (Hanover, 1090–7; Egerton, 581–6). The abbreviated Fierenbras picks this up and streamlines the text by reusing Estragot as the husband (named Effraon in the Vulgate) of the giantess Barak (named Amiote in the Vulgate), replacing Effraon’s combat with the French and his death in the Fierabras by the simple comment concerning Estragot, ‘qe Romeins firent tuer’ (Egerton, 1519; see also Sowdone, 2939–54). Minor changes can simplify the narrative, but show the careful craftsmanship of the insular remanieur. For instance, the peers do not hide their identities on being taken prisoner, as they do in the Vulgate version (Fierabras, ed. Le Person, 2018–42), so there is also no need for them to be identified to Balan later (as in the Vulgate version, Fierabras, ed. Le Person, 3538– 47). Similarly, Lucafer and Floripas are both with Balan when the 124 In the context, this is more likely to be God than Charlemagne.

320

Charlemagne in Medieval England

prisoners are first brought in to him, so there is no need for the separate explanations they are given in the Vulgate tradition (Fierabras, ed. Le Person, 2956–61, 2152). The Egerton and Sowdone texts also share a different account of a game played between Lucafer, Floripas’s Saracen betrothed, and Naimes when Lucafer comes to Floripas’s chamber and find she is sheltering the peers. In the Vulgate, Lucafer challenges the peers to a flame-blowing contest, the aim being apparently to see who can blow the flame further (Fierabras, ed. Le Person, 3016–56); Naimes responds by blowing on the burning brand so that Lucafer’s beard and moustaches are set alight. It is a rather odd episode, in which Lucafer arrives with violence in mind but then engages in a strange game. In the abbreviated version the game is even more bizarre, and involves attaching the burning coal to a needle before blowing; here it is Lucafer who initiates the violence, as he burns Naimes’s beard (Egerton, 899; Sowdone, 2008). All versions end with Lucafer being killed. One can only speculate about why the game is changed; perhaps it is made more exotic in the abbreviated insular version. It certainly creates a significant change in the characterization and motivation of the narrative, as it is Lucafer who now becomes the aggressor, justifying the violent death which is now inflicted on him by Naimes. Some events are slightly reordered in the abbreviated version. In the Vulgate, Floripas kills her governess, tipping her out of the window when the latter recognizes the peers, whom her charge has rescued; in the abbreviated tradition this is brought forward to when Floripas first declares her intention of helping the peers (Fierenbras, ed. Brandin, 1284–8; Sowdone, 2545–64), again creating a tighter structure. Similarly, when Floripas kills the gaoler in order to set free the peers, this has no repercussions in the Vulgate, while in the abbreviating tradition she offers an explanation to her father, accusing the gaoler of feeding the prisoners, and thus gaining his approval for taking charge of the prisoners herself. It seems that some ‘loose ends’ are being tidied up, in the sense of creating a more carefully motivated plot in the reworked version. What is sometimes lost with the increase in narrative momentum is the drama of dialogue or the detail of, admittedly sometimes standard or formulaic, descriptions. This is quite marked in the exchanges between the peers, on their way to treat with Balan, and Balan’s messengers, en route to speak to Charlemagne. In the Vulgate, the actual fighting between the two groups of messengers is framed by two sections of

The Insular Fierabras Tradition

321

dialogue. The first is an exchange between the two groups (Fierabras, ed. Le Person, 2495–516); this is much shorter in the abbreviating tradition (Egerton Fierenbras, 755–66; Sowdone, 1784–96). More is lost by the omission of the debate between the peers after the fight, which includes an argument between Naimes and Roland that shows the latter to be forsené (out of his mind), as Naimes calls him (Fierabras, ed. Le Person, 2547–58). As with the omission of Naimes’s refusal to guard the gate, the sense of tension between Roland, representing the young knights, and Naimes, representing experience, is lost. Extensive use of direct speech, the dramatic mode of presentation favoured by the chanson de geste tradition,125 is here replaced by simple narration. This is, moreover, a considerably abbreviated narration with few details of the actual fight, while in the Vulgate the exchanges and account of the fighting take up a total of over forty lines (Fierabras, ed. Le Person, 2492–534). However, what is gained in the insular abbreviated version, where a mere two lines are given to the whole incident (Egerton Fierenbras, 772–3; Sowdone, 1799–800), is not only economy of narration, but also a more unambiguously positive portrayal of Roland, in line with the treatment of Roulond in the Middle English Song of Roland. The character of Roland is particularly interesting in the Vulgate version, drawing as it does on aspects of Roland’s character which are less heroic and thus, perhaps, offering a particular reading of the Roland tradition.126 He demonstrates impetuosity and hot-headedness, seen most clearly in the quarrel with Charlemagne at the beginning of the Fierabras narrative, when he goes so far as to draw his sword on his king. Having been rebuked by Charlemagne, Roland goes off and sulks and allows his wounded companion Oliver to take up Fierabras’s challenge. Later, when the seven peers are going to the rescue of the imprisoned five peers, Roland endangers their lives by pushing a Saracen into the river Flagot at Mautrible, when Agolafre, the giant guarding the bridge, has been tricked into allowing them to cross (Fierabras, ed. Le Person, 2647–56). Yet Roland is also a man whom others will follow. It is to Roland that Charlemagne turns for help when Fierabras issues his challenge; Roland leads the attempt to prevent Oliver’s capture; 125 Sophie Marnette, Narrateur et points de vue dans la littérature française médiévale: Une approche linguistique (Bern: Lang, 1998), pp. 124–5. 126 Van Emden, ‘The Reception of Roland’, pp. 10–13; Ailes, ‘Fierabras and the Chanson de Roland’, pp. 15–16.

322

Charlemagne in Medieval England

he is the first to be appointed by Charlemagne to rescue Oliver and the other captured peers. He is capable of remorse for his pride when Oliver volunteers: he ‘or feroit la bataille volentiers et de grez’ (‘now he would undertake the combat most willingly’; Fierabras, ed. Le Person, 268), though there is also rivalry between Roland and Oliver for leadership over the peers. Much of this complexity is lost in the abbreviated tradition, but it is not clear how far the loss is incidental to changes in the narrative. The quarrel with Charlemagne is integral to the plot, and remains. However, the change whereby only Roland and Oliver are taken captive has several effects on the way Roland is treated: it brings him into a more prominent position, a change that fits with developments in the Middle English romances, which tend to place more emphasis on the well-known characters. It also leaves Naimes to lead the rescue party without any tension between the two. Naimes is thus given an unambiguous status; as one of the better-known peers it is Naimes, in the abbreviated text, who kills Balan at the end, not Ogier as in the Vulgate. As Roland is not now part of the rescue, the scene in which he endangers all their lives by pushing the Saracen into the river has also gone. All this simplifies Roland’s character and leaves him more heroic, decreasing his tendency to impetuous and foolish, even egotistical, behaviour. Changes in the depiction of Charlemagne may also be the consequence of abbreviating the narrative. In the Vulgate, Charlemagne is far from the ‘symbol of justice’ described by Jessie Crosland.127 He is a man easily persuaded, sometimes to take the wrong action; he is arbitrary, tactless (notably in his dealings with his nephew), and at times inactive. He is a warrior king, but one who manages to get himself trapped inside the outer defences of Mautrible, and who needs the help of his barons to defeat Balan (Fierabras, ed. Le Person, 6026– 9). Two changes shared by both texts in the abbreviating tradition again remove some of this complexity to present a more positive and less nuanced depiction of the king. The shorter texts lack the fighting which breaks out among the barons in the presence of the king in the Vulgate (Fierabras, ed. Le Person, 4636–49); indeed, the whole debate 127 Jessie Crosland, The Old French Epic (Oxford: Blackwell, 1951), pp. 239–40; Crosland is here comparing an idealized Charlemagne with the king as he appears in the Sowdone of Babylone; in fact, however, Charlemagne is frequently less than ideal in the chanson de geste tradition.

The Insular Fierabras Tradition

323

over how to respond to the situation is reduced in a way which removes Charlemagne’s doubts as well as his inability to control his court, presenting a more idealized image of the emperor. Most strikingly, in that it provides a climactic portrayal of Charlemagne as a successful warrior, he defeats Balan unaided in the abbreviating texts. With these changes, perhaps two different desired effects – abbreviation and increased idealization of the emperor – can be achieved with the same alteration to the source text. Both these effects are found consistently as a result of the changes made. Oliver, too, is given a more positive presentation, partly at the expense of Fierabras. In the combat between the two, some of the more chivalrous elements in Fierabras’s behaviour are omitted: his offer to Oliver of the miraculous balm which would heal his wounds and his offer to pretend to be unhorsed are both missing. Although Oliver still helps Fierabras to arm, the dynamic between them is altered as Oliver does not, as he does in the Vulgate, hold Fierabras’s stirrup (Fierabras, ed. Le Person, 689); and as in the Vulgate, Fierabras does dismount when Oliver is unhorsed, but to protect his horse, not out of chivalry (Fierenbras, ed. Brandin, 228–30; Sowdone 1213–21). This loss of Fierabras’s chivalry removes some of the finer nuances of his character. At the same time Oliver is enhanced, in that he now defeats him without any of the advantages such courtesy might have given him. Both the Anglo-Norman and Middle English texts also inform us that Oliver is related to Charlemagne. In Fierenbras Charlemagne addresses him as his nephew (112) and later Oliver tells Fierabras that ‘Charls est mun uncle’ (247); in the Sowdone he tells Fierabras he is ‘Cousyn to kynge Charles the boolde’ (1250). This adds to Oliver’s status, but it is not in keeping with the wider French tradition and suggests an insular willingness to innovate independently. The most striking change in character in the abbreviating tradition is in the depiction of Ganelon. The Vulgate picks up on the implications of the Chanson de Roland that he must have been respected until the point when he argues with Roland, and presents him, three years before Roncevaux, as acting in a noble manner. Thus, at the Battle of Mautrible, when Charlemagne is trapped inside the outer defences, it is Ganelon who, with Fierabras, goes to his rescue, refusing to be tempted to abandon his king and treacherously have a member of his own family crowned instead (Fierabras, ed. Le Person, 5145–7). But where in the Vulgate it is Ganelon’s kinsman Alori who encourages rebellion, in the

324

Charlemagne in Medieval England

insular texts it is Ganelon himself, who is even prepared to be crowned in Charlemagne’s place: Ore oiéz de Genes, le traitre losengier. Il apella ses parenz d’un part conseiller: ‘Seignurs’, dit Genes, ‘un chose vous voil moustrer. Charls est cy enserré, n’i pout eschaper. Paiens le occierunt, car nul nel pout aider. Car lessoms li paien gent od li covenir’. (Fierenbras, ed. Brandin, 1547–52) [Now hear about Ganelon, the lying traitor; He called his family aside for advice: ‘My lords’, said Ganelon, ‘I want to show you one thing; Charles is trapped inside, he cannot get away; The pagans will kill him, for none can help him. Let us leave the pagan peoples to meet with him.’ ] Tho Genelyne saie, the kinge was inne And the yates faste I-stoke, Ther myght no man to him wynne, So was he faste withinne I-loke, To his frendes he gan speke And saide ‘the kinge is dede, And alle xij peres eke. On peyne’ said he ‘to lese myn hede, Let vs hye to Fraunce warde! For I wele be crowned kinge, I shalle you alle wele rewarde, For I wole spare for no thinge.’ (Sowdone, 2967–78)

It is possible that this change, re-ascribing Alori’s treacherous words to Ganelon, could have come about as a result of the fourteenth-century adaptor’s mistaken reading of a Vulgate text,128 but even if this were 128 The adaptor, working on an older text, may not have understood the case system, which had been in decline in Anglo-Norman since the

The Insular Fierabras Tradition

325

so, the alteration fits the direction of change observed in the insular tradition. The Vulgate depicts a reliable and loyal Ganelon, but in these insular texts of the abbreviating tradition Ganelon conforms more to his conventional representation as the arch-traitor.

The Egerton Fierenbras Apart from the changes in characterization noted above, the alterations in the Anglo-Norman Fierenbras in the Egerton MS are chiefly to do with the aesthetics of abbreviatio. Further changes continue the tendency to increased narrative momentum which was already present in its common source with the Sowdone, a trait, moreover, which is characteristic of Anglo-Norman chansons de geste.129 Some of this is achieved by quite simple but effective techniques to cut the amount of repetition, characteristic of chanson de geste discourse; for example, the number of divisions of Charlemagne’s army before the final battle is reduced from ten to four (1624–30).130 The reduction of description, and in particular set-piece description, is striking. Oliver’s arming before the combat is abbreviated to a few lines in Fierenbras, and reduced in the Middle English Sowdone to a single line (1109): A sun esquier ad ses armes demaundéz; Cil l’en ad hastivement gentement arméz. Sun bon destrer li fu avant menéz. Sanz toucher l’estru si s’en est muntéz. Sun fort escu an ad as enarmes enbraciéz. Puis prent sa lance, si s’en est galopéz. (Fierenbras, ed. Brandin, 90–5) [He asked his squire to bring his arms. [The squire] quickly armed him well. His good horse was brought up to him.

twelfth century. In the Vulgate, Ganelon’s relative Alori calls out to him: ‘Guenelon escria’ (‘He called to Ganelon’; 5136). The cas regime is what indicates that it is Alori calling to Ganelon, and not Ganelon to Alori. 129 Ailes, ‘Fierabras and Anglo-Norman Developments’, passim. 130 Compare the Vulgate Fierabras, ed. Le Person, 5765–75. The divisions of Charles’s army are not mentioned in The Sowdone. Sir Ferumbras has five; Firumbras, twenty.

326

Charlemagne in Medieval England

Without touching the stirrup he mounted. He held his strong shield by its straps. Then he took his lance and galloped off.]

Both insular texts lack the detail given in the Vulgate (Fierabras, ed. Le Person, 215–45), with its dramatic confrontation between Oliver and his squire, and Oliver’s address to his horse; this suggests perhaps a more literary, less performative aesthetic. A more complex abbreviating technique is to conflate incidents. The number of sorties made by the peers besieged in Aigremore is reduced from four in the Vulgate tradition to three in the Egerton Fierenbras, where the remanieur has skilfully incorporated two separate narrative incidents into a single episode (Fierenbras, ed. Brandin, 1244–305). We have noted above a streamlining of Richard’s escape, with obstacles removed in both the texts of this tradition. The use of characteristic epic discourse in Fierenbras is tempered by the reduction in parallels and in reprise. The remanieur frequently runs laisses together, omitting the reprise often found at the beginning of laisses. One passage gives an interesting example of how different chanteurs de geste dealt with the reprise. The sequence of laisses III to V in the Vulgate Fierabras makes use of the classic chanson de geste technique of repetition with development. In laisse III, Fierabras first issues his challenge to the watching Christians: Le Sarrazins s’areste soz deuz arbres floris, E regarda aval desouz le pin massis, Et voit le tref Karlon, le roi de Seint Denis, Et l’aigle d’or qui luist con soleil esclarchis: Des loges et des tres voit touz les prés vestis. Kant le paien les voit, mout enn est engramis; Mahonmet en jura et ses seintes marchis Jamais ne finira ses avra estormis. A haute voiz s’escrie: ‘Ahi! roi de Paris! Envoie a moi joster, malvais veillart faillis, De tes barons de Franche trestouz les plus hardis, Rollant et Olivier et s’i viegne Tierris, Et Ogiers li Danois, qui tant par est hardis; Ja n’en refuserai par Mahon jusqu’a sis Et se nes envoiez issi con ge ai requis …

The Insular Fierabras Tradition

327

Je ne m’en tornerai, si seras desconfis …’ A iceste parole est sor l’arbre genchis; Des armes quë il porte s’est mout tost desvestis. (Fierabras, ed. Le Person, 76–89, 91, 95–6) [The Saracen stops by two flowering trees And looked down from under the solid pine And sees the tent of Charles, the king of Saint-Denis, And the golden eagle which shines like the bright sun: He sees the meadows covered with tents and encampments. When the pagan sees them he is very vexed; He swears by Mahommed and his holy grace He will not stop till he has routed them. In a loud voice he cried; ‘Ah, king of Paris! Wicked, failing old man, send to fight against me The boldest of your barons, Roland and Oliver, and let Thierri come, And Ogier the Dane, who is so bold: I won’t refuse up to six of them at once, by Mahommed; And if you don’t send them as I request … I won’t turn back until you are defeated …’ At these words, he leant against the tree; He quickly removed the arms he was wearing.]

The next laisse opens with what at first looks like a reprise, simply picking up at the beginning of one laisse what was already said at the end of the previous one: Le Sarrazins s’arreste desouz l’arbre ramez; (reprise of l. 76) Son cheval arresna a l’arbroiseil rosez; Des armes quë il porte a son cors desarmez ; (reprise of l. 96) (Fierabras, ed. Le Person, 97–9)

We might expect the laisse to then continue with the diegetic sequence, having reached the point where the previous laisse stops. However, it actually develops with a further and more detailed reprise, in which Fierabras reiterates the challenge in slightly different terms, but with strong formulaic echoes across the two laisses:

328

Charlemagne in Medieval England

Et regarda les loges tout contreval les prez. (reprise of ll. 77 and 80) A sa voiz qu’il out haute a durement criez: (reprise of l. 84) ‘Ou es tu Charlemainnes? Mout t’a hui appelez. Envoie a moi joster Olivier tom privez, (reprise of ll. 85 and 86) Ou Rollant ton neveu a corage adurez, (development of l. 86) Ou Ogier le Danoiz, que l’on m’a tant loez. (reprise of l. 87a) Et si li uns n’i ose ne venir ne aler, Si en i viegne[nt] deuz fervestuz et armez; Et se li tiers i vient, bataille avra rangez; Ou li quart, ou li quint, ja n’en iert refussez. Ker par cel Mahonmet ou mes chies est donnez Se .VI. en i venoient de plus haut parentez, (reprise of l. 88) N’en torneroie gen, s’avroie a euz caplez.’ (reprise of l. 91) (Fierabras, ed. Le Person, 100–12)

In terms of traditional analysis of chanson de geste discourse, based on Jean Rychner’s typology of repetitions within the genre, this is at once a reprise and an embryonic laisse parallèle; in terms of what is going on in the continental French it is perhaps more useful to evoke the more flexible understanding of the way repetition works in the genre that we find in the work of Edward Heinemann, who is less rigid in producing a typology of repetitions and concentrates on the idea of internal echoes being created.131 The second part of laisse IV then develops the narrative, with Charlemagne asking who Fierabras is: Kant l’entent l’emperere, s’a le chies enclinnez: Et quant il se redreche, s’a Richart apelez: ‘Sire dus’, che dist Karles, ‘ne me soit pas chelez, Conoissiez vos cel Turc qui si haut a criez? A grant merveille s’est et proissiez et loez: Vers .VI. s’en aatist, ge l’ai bien escoutez.’ ‘Sire’, che dist Richart, ‘ja orez veritez: Ch’est le plus richë hom dont omques fust parlez.’ (Fierabras, ed. Le Person, 116–23)

131 Rychner, La Chanson de geste; Edward A. Heinemann, L’Art métrique de la chanson de geste: Essai sur la musicalité du récit (Geneva: Droz, 1993).

The Insular Fierabras Tradition

329

[And when the emperor heard this he bowed his head And when he raised it, he called on Richard: ‘My lord duke’, said Charles, ‘don’t hide this from me, Do you know the Turk who has called out so loudly? He has a good opinion of himself. If I have understood well he will engage himself against six.’ ‘My lord’, said Richard, ‘now hear the truth: That is the most powerful man ever heard of.’]

Again, this is reprised and developed in the following laisse (ed. Le Person, 130–135a). This whole complex pattern of repetition with alteration, and with no clear diegetic sequence, is completely removed in the Egerton Fierenbras, which reduces the whole to a few lines: Charls oit le noise, Naimes prent apeller: ‘Ki face tiel noise ore me va cunter.’ ‘Par Dieu’, dist duc Neimes, ‘ceo Fierenbras li fier.’ (Fierenbras, ed. Brandin, 49–51) [Charles hears the noise; he calls Naimes ‘Who is making this noise, tell me.’ ‘By God’, said duke Naimes, ‘it is Fierabras the proud.’ ]

Examples of this method of removing repetition, with a loss of chanson de geste discourse, are found throughout the Egerton text: it is a relatively easy way for a remanieur to abbreviate. Indeed, the same technique is found in the abbreviated version of Fierabras in the Shrewsbury Book.132 We noted in our discussion of the importance of the religious conflict in all these texts (Chapter 2) that the Egerton text considerably abbreviates Charlemagne’s long prayer of intercession for Oliver as part of this systematic cutting of the formulaic material which is integral to the discourse of the genre. The text remains a chanson de geste, but the generic markers are less prominent. The process of abbreviation through a reduction in repetition could incidentally bring about some changes in emphasis, and it is not always possible to know if this is simply a sideeffect of the redacting process. It is, for example, difficult to be absolutely certain that changes made to the representation of Charlemagne, such 132 Bailey, ‘Fierabras and the Livre de Charlemaine’, p. 111.

330

Charlemagne in Medieval England

as the omission of his drunkenness which led to a potentially serious quarrel with Roland (Fierabras, ed. Le Person, 159; cf. Fierenbras, ed. Brandin, 66), are not an incidental consequence of abbreviation. They are, however, consistent changes, suggesting deliberate alteration. The addition of a relic of local (i.e. insular) significance, namely the lance, which figures only in the list in the Destruction de Rome and the Egerton Fierenbras, would also give the narrative a different resonance. In these insular texts the locus of Saint-Denis is of less significance, and the inclusion of a relic which, in insular legend, was later given by Charlemagne to a king of England could only accentuate the links between Charlemagne and England which the compiler of the Egerton codex is suggesting through the heraldry. Such changes to the details of the Vulgate narrative add up to a considerable process of insular appropriation.

The Sowdone of Babylone As noted above, both extant Anglo-Norman Fierabras texts are presented in their manuscript context in a diptych arrangement with a version of the post-written prequel, La Destruction de Rome. This pairing of narratives is reproduced in The Sowdone of Babylone, the Middle English text based on the insular abbreviated branch of the tradition, as represented in British Library, MS Egerton 3028. However, the diptych pairing is here converted into a more fully integrated combination, where the material of the two chansons de geste is reworked to produce a single Middle English romance. This new text, entitled The Romance of the Sowdone of Babyloyne and of Ferumbras his Sone who Conquerede Rome, is far from being a close translation of the insular French-language tradition. The insular tendencies towards economy of narration, greater consistency in motivation, and simplified characterization observed in the Anglo-Norman Fierenbras are taken further in the Sowdone, and much new material is added of the kind common to all the Middle English adaptations – circumstantial detail, explicit causative links, representation of emotion – to produce the narrative ‘thickness’ characteristic of Middle English romance. This includes a number of alterations and additions that may suggest a new way of reading the inherited Fierabras tradition. Most striking are several extended or interpolated passages describing the person and the customs of the sultan and his Saracen peoples, but as Smyser notes, ‘The Englishman’s

The Insular Fierabras Tradition

331

hand … may also be evident in a more copious display of “learning” of things Saracen – ritualistic and ethnological – in the body of the English poem as well as in the interpolation’.133 Critics have tended to see these additions in terms of titillating exotica serving to emphasize the otherness of the Saracen enemy;134 but by reading them in the context of the whole Middle English text, they can be seen as evidence of an innovative revision that destabilizes the traditional epic presentation of the Saracen sultan and questions the notion of otherness. As discussed in Chapter 3 (pp. 193–5), the poem starts with four theologically inspired moralizing quatrains, not present in the AngloNorman text, which the English poet has inserted in place of the conventional call for attention at the beginning of the chanson de geste. This introduction to the Destruction de Rome material sets the whole story in a new, wider context. God’s perfectly ordered creation and Man’s intended place in it are carefully explained, only to be abruptly overturned by the evidence of human sinfulness: ‘But for the offences to God I-doon | Many vengeaunces haue be-falle’ (13–14), so that the account of the fall of Rome, destroyed by ‘a Sowdon, that heathen was’, is presented as a representative example of the consequences of sin. As John Tolan demonstrates, the victories of Saracens or Turks over Christians were consistently interpreted, from the seventh century to the sixteenth, as God’s punishment of sinful Christians, with the Muslim victors cast as the scourge of God.135 Indeed, this traditional interpretative strategy was embedded in the language of the liturgy, where the prayers of the votive ‘Missa contra paganos’ in early English missals included a reference to ‘gentem paganam quam pro peccatis nostris super nos cognoscimus preualere’. 136 133 Smyser, ‘Charlemagne Legends’, p. 84. 134 Metlitzki, The Matter of Araby, pp. 188–9; Alan Lupack, ‘The Sultan of Babylon: Introduction’, originally published in Three Middle English Charlemagne Romances, ed. Alan Lupack (Kalamazoo, MI: Medieval Institute Publications, 1990), TEAMS Texts Online (http://www.lib. rochester.edu/camelot/teams/sultint.htm). 135 John Tolan, ‘Looking East before 1453: The Saracen in the Medieval European Imagination’, in Cultural Encounters between East and West, ed. Birchwood and Dimmock, pp. 13–28. 136 ‘The pagan people whom we recognize as prevailing over us on account of our sins.’ See The Leofric Missal, ed. Nicholas Orchard, 2 vols, Henry

332

Charlemagne in Medieval England

‘Pagans’ could be read as Danes in the ninth-century Leofric Missal, but the term was also applicable to other non-Christian adversaries, as shown when the ‘Missa contra paganos’, furnished with additional prayers by Pope Clement V in 1309, was explicitly directed against the Muslim Ottoman Turks by Pope Calixtus III in 1456.137 The same argument is used by the wise sultan of Egypt in The Book of John Mandeville, pointing out the un-Christian manner of life in Christian countries: They break their entire law that Jesus has given them and set out for their salvation. Thus for their sins they have lost all this land that we possess, for because of their sins your God has put them [the lands] in our hands – not through our strength but for their sins.138 In putting this criticism of Christian practices into the sultan’s mouth, the author follows the precedent of a celebrated passage in the PseudoTurpin Chronicle, where the Saracen king Aigolandus withdraws his intention to convert to Christianity on account of the uncharitable treatment of the poor that he perceives in Charlemagne’s society;139 however, the content of the argument voices what became the standard explanation for the failure of crusading ventures in the Holy Land, expressed most powerfully in the papal letter Audita tremendi, issued in response to the fall of Jerusalem in 1187, in which the disaster was attributed to the sins of ‘the whole Christian people’. 140 Jonathan Riley-Smith traces the long-lasting effect of this argument on crusading Bradshaw Society 113–14 (London, 2002); The Missal of Robert of Jumièges, ed. H. A. Wilson, Henry Bradshaw Society 11 (London, 1896). 137 Amnon Linder, Raising Arms: Liturgy in the Struggle to Liberate Jerusalem in the Late Middle Ages (Turnhout: Brepols, 2003), pp. 179–80. In 1961, the votive Mass was renamed ‘In Defence of the Church’ (ibid., p. 120). 138 The Book of John Mandeville, ed. and trans. Iain M. Higgins (Indianapolis: Hackett, 2011), p. 87. We have chosen to cite this modern English translation of the edited early French-language manuscript text, in preference to the later, imperfect, Middle English translations. 139 For discussion of the use of this frequently excerpted passage in latemedieval didactic texts see Turpines Story, ed. Shepherd, pp. xxxiv–xxxvii, 65. 140 Quoted in Riley-Smith, The Crusades, p. 137.

The Insular Fierabras Tradition

333

thought, in the official connection of military success with ‘the spiritual health of all Christianity’, and the consequent stress on the need for repentance as a prerequisite to action in crusade preaching; conversely, as he points out, ‘general councils, summoned to reform Christendom, were often associated with the need to assure crusading success’. 141 The same interpretative strategy is implied in the prologue created for The Sowdone of Babylone. Since the poet omits any reference at this point to the Saracens’ seizure of the relics of the Passion and Charlemagne’s mission to recapture them, as advertised in the opening laisse of the Destruction de Rome, this moralizing narrative of sin and punishment provides an alternative argument. The sin that precipitates the sultan’s actions in the narrative is the opportunistic piracy of the Romans who robbed his dromond of its rich treasure and killed its crew (68–88), and it has terrible consequences: not only is Rome destroyed, but the sultan ‘Conquerede grete parte of Christiante’ (31).142 The historicity of these facts is authenticated in the Middle English text by repeated reference to written sources,143 and the narrator highlights the indubitable cause and effect: ‘That robbery was righte dere boght, | Was never none derrer withouten douȝte’ (111–12).144 Charlemagne’s eventual victory over ‘the Sowdon of Babyloyne, | That riche Rome stroyed and wan’ (3260–1) – for which he explicitly thanks God in the English poem, attributing to God’s power his defeat of the Saracens despite their strength and their ‘Engynes’ (3229–34) – can thus be read as proof of God’s restored favour

141 Ibid., and see pp. 173–5 for discussion of the later papal letter Quia major and the Fourth Lateran Council (1215). 142 This may be adapted from a later line in Destruction de Rome, where Laban orders the destruction of every Christian cleric and religious: ‘Desormais destruerai saint christientéz’ (ed. Brandin, 175). In the Middle English text, however, ‘Christiante’ evidently means Christendom as a geographical entity. 143 Lines 24–8 introduce mention of writings ‘in Romaunce’, ‘bokes of Antiquyte’, and ‘Cronycles’. 144 A similar interpretation is implicitly proposed in the Middle English Song of Roland, where an episode of Christian sin is imported from the PseudoTurpin Chronicle as a prelude to the fateful events at Roncevaux (1–76). A different note of regret occurs in the Destruction de Rome prologues (ed. Brandin, 19–23; ed. Formisano, 33–6).

334

Charlemagne in Medieval England

towards faithful Christians after the purging of Christendom’s sins through the destruction inflicted by the Saracens. However, the text does not adhere to a simple model such as the ‘scourge of God’ in its treatment of the Saracens. The focus on cause and effect creates a narrative momentum of action and counter-action in which escalating violence on both sides is carefully plotted: stress is laid on the sultan’s desire not just to be revenged on Rome (as in the Egerton and Hanover texts) but to destroy Christendom (128, 235), and his additional vow to destroy Charlemagne himself functions as a frame for the action (140, 752), balanced by a complementary stress on Charles’s counter-oath and his determination to expel the sultan and Ferumbras from Christendom (586, 761–6). Wrongs done to the sultan and to Christendom are presented as equivalent when the formula used for the sultan’s sack of Rome (413–18) is repeated for Charlemagne’s destruction of Agremour (779–86), an equivalence which is echoed in their reciprocal insults: ‘hethen houndes’ (935) versus ‘Crystyn dogges’ (956). At the end of the Destruction de Rome part of the text, Ferumbras’s seizure of the relics functions as a plot device, the casus belli for the ensuing Fierenbras part, just as the theft of the sultan’s rich treasure did before; and again, the Middle English poem underlines the causal connection when Floripas passes judgement on the actions of her fellow-Saracens, as she restores the abducted relics of Rome to the victorious Charlemagne: ‘That game was evel bygonnen, | It sithen rewed us alle’ (3139, 3141–2). The balanced structure of the narrative, with its implied collapsing of difference between Christians and Saracens, invites a new reading of the traditional story. Following the omission of any reference to the theme of the loss and recovery of the Passion relics in its rewritten prologue, the Sowdone of Babylone further differs from all other versions in its treatment of the inherited tradition, in that almost all mentions of the relics in the text are erased. What remains is simply the basic framework of theft and restitution necessary for the plot. So the first mention is when Ferumbras seizes the relics, and here the English romance plays down the drama of the event. Egerton creates a scene of atrocity, as Fierenbras hurries to St Peter’s, beheads the pope at the high altar, demands the Passion relics of an aged monk, beheads the monk, seizes the relics and steals sacred vessels from the treasury, before pillaging the city. In the Sowdone of Babylone there are no such acts of deliberate desecration: Ferumbras simply takes the relics as the first part of his

The Insular Fierabras Tradition

335

booty from the sack of Rome (663–70). There are subsequent references to the relics in two formulaic accounts of the Sowdon’s offensive acts: he burned the city and bore away the relics (715–6, 747–8). In each case the English text adds as the crowning insult that he threatens Charles himself. The two recitals of Charles’s retaliatory message to the Sowdon are similarly formulaic: he must return Charles’s nephews and the relics of Rome (1671–2, 1820–1) – in effect, restore Charles’s stolen property. Six further opportunities to mention the relics as found in Fierenbras are omitted (760, 1370, 1462, 1895, 2826, 3226); in the Sowdone of Babylone, the relics play no part in the story of Ferumbras’s conversion or of Floripas’s negotiations with the peers in her tower. The only relics scene to be included is that in which Floripas finally presents them to Charles (3135–50), where the return of the stolen relics marks the satisfactory ending of the story. The English romance emphasizes this structural narrative theme by adding small supporting details: Floripas’s recognition that the original theft caused all the following strife (3139–42); Charles’s thanks to Floripas for keeping the relics (3150); and his charge to Sir Bryer to bring the relics and other treasure safe to Paris (3205–10). The story of the loss and recovery of the relics is concluded when Charles deposits the Cross, crown and nails at Notre Dame, Saint-Denis and Boulogne, but this is narrated as bald fact: the preceding climactic scene at Paris in the Egerton version, with all its contextualizing details about the solemn reception and veneration of the relics by the bishops and clergy, is entirely omitted in the Sowdone of Babylone. Instead, the execution of Ganelon is much elaborated as the final narrative action, and the text concludes with a prayer for the souls of Charlemagne and his barons. This whittling down of references to the relics goes beyond a merely pragmatic intention to delete local references to cults at French pilgrim sites that would have had limited appeal to audiences in England. It is part of the much wider programme of change that the author of the Sowdone initiated, in a highly ambitious and self-consciously literary project, to reimagine the inherited tradition. Further evidence of the text’s innovative approach can be seen by interrogating the title with its naming of the sultan, who is here, uniquely, called the ‘Sowdon of Babyloyne’. The title, which appears at the end of the composite romance, reads like an expansion of the formula found in manuscripts of Fierabras: ‘Ici define la romance de Fierembras de Alisaundre’, simply adding the name of Fierabras’s father as the ‘hero’ of the prequel narrative; but the form of the father’s name

336

Charlemagne in Medieval England

is noteworthy. It is only in this Middle English romance that the sultan is named ‘Sowdon of Babyloyne’, rather than ‘Laban d’Espaigne’, as the sultan is styled throughout both Egerton texts and in the Hanover Destruction, where he is also called ‘ly soldan d’Espaigne’ (1391).145 The English text introduces him as ‘Laban, the kinge of hie degre, | And syr and Sowdon of hie Babilon’ (30), and creates a splendid tableau of all his ships with his arms embroidered on their red silk sails: The Armes displaied of Laban Of Asure and foure lions of goolde. Of Babiloyne the riche Sowdon, Moost myghty man he was of moolde. (133–6)

In the concluding lines of the romance he is named again, ‘Laban, | The Sowdon of Babyloyne’ (3260). He is at one point explicitly recognized as ‘lorde of Spayne’ (1788), but this is clearly just one of his many possessions and does not define him by a universal appellation as it does in the insular French sources.146 Babylon is mentioned near the beginning of both versions of La Destruction de Rome, but in different and subsidiary contexts. In the Egerton text, the cargo ship that proves to be the casus belli is coming from Babylon, a far-off place ‘dejouste la Rouge mer’ (74); in the Hanover version, Babylon is one of the sultan’s Eastern possessions, along with Constantinople and most of the habitable world (74–82). The English text, by contrast, presents Babylon as the centre of the sultan’s world and of his identity. It uniquely states that the sultan was ‘born in Askalon’ (32), the port city of Ashqelon in the Holy Land, a critical stronghold that was in Muslim control for most of the period of the crusades. This location, not far from Egypt, gives a recognizable geographical connection with both his son’s status as ‘Ferumbras of 145 The Hanover Fierabras keeps the Vulgate ‘Balan’; this could be a further indication that the redactor had two different versions in front of him. 146 The Vulgate Fierabras tradition is completely different: Fierabras’s father is known simply as ‘li amirans Balant’ (Fierabras, ed. Le Person, l. 1971), and the one time he is given a territorial epithet, it is ‘Amirant de Nubie’ (2710). This model is followed in both the Ashmole Sir Ferumbras and the Fillingham Firumbras.

The Insular Fierabras Tradition

337

Alexandria’, and his own as Sowdon of Babylon, which was, of course, the usual name for Cairo in medieval texts. The English poem makes clear that ‘Sowdon of Babyloyne’ is not just a title – the sultan’s real ‘home’ is identified as Babylon when Charlemagne threatens him: ‘And but if he will Baptised be | and lefe his fals laye, | Babyloyne shal he never see | For alle his grete aray’ (763–6). Meanwhile, he is residing temporarily in the Spanish city of Agremoure,147 where he expects to receive supplies of exotic luxury goods from his home base in Babylon (68–75). The change in the English text is all the more remarkable because, apart from the sultan’s name, there is no reduction in the importance of Spain in the narrative: quite the reverse. All the action still takes place in Spain, except for the attack on Rome; Spain features several times as an important possession of the sultan (1531–2, 2164), including ‘his’ own cities of Egremour and Mavntreble (719, 2147); and, finally, it is Charlemagne’s conquest and redistribution of all Spain that is named as the crowning achievement of the whole campaign – not, as in other versions, the restoration of the Passion relics. Thus Charles conquered Laban, The Sowdon of Babyloyne That riche Rome stroyed and wan, And alle the brode londe of Spayn. (3259–62)

The concluding lines of the Egerton Fierenbras do not mention Spain: everything is focused on Charles’s return to France with the relics. So given the unusual focus on Spain in everything except Laban’s name in the Middle English Sowdone of Babylone, it is surprising that he is not identified with Spain as in the insular French sources; or, if the absence of this epithet is owing to some cross-referencing between the insular abbreviated version and the Vulgate tradition,148 that is he not (as there) called the emir of Nubia, which would provide the same 147 ‘And in the Cite of Agremare | Vppon the Rivere of Flagote | At þat tyme he soiorned ther’ (33–5). 148 For evidence of such cross-referencing, see, for example, the incident in The Sowdone of Babylone in which Floripas offers her maidens as sweethearts to the peers, and Roland replies (2747–54). This episode

338

Charlemagne in Medieval England

geographical proximity to Fierabras’s territory. There is, of course, epic precedent in La Chanson de Roland, where Charlemagne’s ultimate adversary, the Emir Baligant, resides in Babylon and launches his fleet from Alexandria.149 Moreover, in Middle English romances the sultan of Babylon often appears as a generic Saracen potentate: for example in Octovian Imperator, where the hero’s opportunity to prove his worth is afforded by the sultan of Babylon’s sudden invasion of France. But ‘Sultan of Babylon’ was also a customary title given to the sultans of Egypt in chronicles and in medieval travel writing such as The Book of John Mandeville and Marco Polo’s Description of the World. This is a context worth examining. The sultan of Babylon in Mandeville is very different from the stereotype of the same name in contemporary romance, which exists purely as a function of anti-Christian aggression.150 Mandeville’s sultan, by contrast, is a portrait of a great Eastern overlord who has achieved his position by conquest and force, and can command vast armies from all these lands, but who holds court in a beautiful castle where he is served by thousands of retainers, and can choose for his pleasure the most beautiful and noble women, and where foreign visitors (who must wear Saracen court dress) receive whatever they ask for, as long as it accords with his religious law. This romantic, exotic account is verified by the narrator’s claim that he himself was in the sultan’s service for many years (Mandeville, chapter 6). The narrator’s intimacy with the sultan later permits him a private interview, in which the sultan questions him about the Christian way of life: ‘He asked me how the Christians behaved in our countries, and I said well, thank God; and he told me that truly they did not.’ The sultan then provides a radical critique of both clergy and laity, causing the narrator to exclaim: ‘Alas, what a great scandal it is to our law and to our faith when people who have neither law nor faith rebuke us and reprimand our sins.’ 151 As noted above, there is a long is not present in the Egerton text, but reflects the Vulgate tradition (Fierabras, ed. Le Person, 4050–8). 149 La Chanson de Roland, ed. Short/Duggan, 2614, 2626. 150 For example, see Octovian Imperator, ed. McSparran: ‘Of Babylonye þe hegh sowdan | Werrede vpon Crystene men’ (907–8), threatening: ‘Crystendom schall adoun | Fram euerych man, | And euery kyng ȝelde vp hys croun | To þe soudan’ (945–8). 151 The Book of John Mandeville, ed. and trans. Higgins, pp. 86–7.

The Insular Fierabras Tradition

339

tradition of using external voices to criticize Christian shortcomings, but the interest here is in the combination of that tradition with a description of the court and customs of the sultan of Babylon, for the unique prologue to the Middle English poem that explains the fall of Rome as God’s vengeance for the city’s sinfulness comes immediately before an extended account of the Sowdon of Babylon’s royal lifestyle. The Mandeville narrator makes a comparison between the sultan of Babylon and the even more powerful Great Khan, whose empire he describes at length, and whom he also served as a soldier (chapters 23, 24). The same two Eastern powers appear together in the Prologue of Marco Polo’s Description, where the sultan of Babylon disrupts the Christian merchants’ journey by waging war, while the Great Khan welcomes them and takes the young Marco Polo into his service. Both texts distinguish followers of Muhammad from the people of the Khan who practise idolatry; but interestingly, Debra Strickland observes that in MS Bodley 264, a richly illuminated insular copy of Marco Polo, Mongolians are visually conflated with Muslims, making them into what she terms ‘functional Saracens’.152 Something similar may have happened in the reception of ideas about the rulers and peoples of the East among medieval readers of these widely known texts, producing the kind of amalgamation that we see in The Sowdone of Babylone, where, in addition to conventional epic references to the ‘Trinity’ of Saracen gods (Mahound, Termagant and Appolyn), we find Saracens who are associated with recognizable Islamic belief, and yet at the same time practise exotic idolatrous ceremonies. Ideas of Saracen idolatry are found not only in chansons de geste: they were widely disseminated in the writings of Vincent of Beauvais, and appear in, for example, the courtly, educated context of the works of Alain Chartier, translated into English in the later fifteenth century contemporaneously with the production of the unique manuscript copy of the Sowdone of Babylone.153 By the fifteenth century there was easy access to a more ‘informed’ view of 152 Debra Higgs Strickland, ‘Text, Image and Contradiction in the Devisement dou monde’, in Marco Polo and the Encounter of East and West, ed. Akbari and Iannucci, pp. 23–59 (p. 33). 153 The Treatise of Hope, in Fifteenth-Century English Translations of Alain Chartier’s ‘Le Traite de l’Esperance’ and ‘Le Quadrilogue Invectif’, ed. Margaret S. Blayney, 2 vols, EETS OS 270 (London, 1974), 281 (Oxford, 1980), I, 93.

340

Charlemagne in Medieval England

Islam: both Mandeville and the Golden Legend154 provide accounts of the life of Mahomet and the laws of the Saracens that represent them as monotheists in the Abrahamic tradition, so the persistent portrayal of Saracens as pagan idolaters must speak to a perception of a deeper ‘truth’ than that of factual accuracy. Again, the communal worship of the Church perhaps provides a framework of reference. The liturgy for Good Friday includes a set of general intercessions for mankind,155 both within the Catholic Church and without, and three categories used to be provided for those in need of conversion: heretics and schismatics needed to be rescued from their errors, Jews from their blind unbelief, and pagans from their worship of idols. There was no category for adherents of any other specific faith, and so in praying for the salvation of all mankind on the anniversary of Christ’s sacrificial death for that very purpose, medieval Catholic congregations would have had to accommodate their idea of Muslim beliefs within the framework of the intercessions. And as John Tolan has shown, influential medieval interpretations of Islam did indeed offer alternative models for thinking of Muslims either as pagan idolaters, or as heretical followers of a false preacher and heresiarch.156 Thus it seems possible that the Middle English adaptation casts Laban as the Sowdon of Babylon at least partly in response to the glamour surrounding the sultan and the related descriptions of the Great Khan in available medieval travel writing. As Iain Higgins notes in his edition, The Book of John Mandeville was, from its first appearance in the 1350s, ‘one of the most widely circulated medieval books’ (p. xii), translated into ten languages besides French; and although Marco 154 The Gilte Legende, ed. Richard Hamer and Vida Russell, 2 vols, EETS OS 327 (2006), 328 (2007), II, 940–5; The Golden Legend, trans. Ryan, II, 370–3. Jo Ann Hoeppner Moran Cruz notes: ‘In late medieval England, The Golden Legend was the most popular religious text, the one most often bequeathed in wills; it was commonly chained in parish churches for the clergy and general public to consult’ (‘Popular Attitudes Towards Islam in Medieval Europe’, in Western Views of Islam in Medieval and Early Modern Europe: Perception of Other, ed. Michael Frassetto and David R. Blanks (New York: St Martins Press, 1999), pp. 55–81 (p. 79)). 155 These petitions persisted unchanged in the liturgy until twentiethcentury alterations after the Second Vatican Council. 156 Tolan, Saracens, chapters 5 and 6.

The Insular Fierabras Tradition

341

Polo’s Description survives in fewer copies or vernaculars, Suzanne Yeager points to its reception in England through the medium of Latin translation.157 A learned Latin tradition would not be out of place among the influences on this idiosyncratic Middle English romance. Higgins identifies the ‘fusion of the familiar and the exotic’ as the prime attraction that caused the Book of John Mandeville to be so popular.158 He draws attention to the ‘syncretism’ that characterizes the book’s treatment of difference: ‘not only are other religious practices and beliefs presented as entertaining to hear about; they are sometimes also depicted as analogous to Christian ones’, and ‘this sometimesassimilating openness to otherness extends to customs and manners, which are usually described neutrally and, if explained, are represented as rational’.159 This corresponds to Mary Campbell’s observation of a new perspective in writing the East from the later thirteenth century: ‘Many of the categories under which Home is describable are sensed as applicable to the East as well, and the objects and customs of Home enter the picture explicitly, not only in a fragmentation of similes but in analogues and comparisons.’ 160 As an example, Campbell quotes Marco Polo comparing instances of idol worship with Christian practices regarding saints’ days and feasts, and concludes: ‘Such analogies demonstrate a sense, novel for medieval Europe, that the two worlds are both different and susceptible of relation to one another’ (p. 91). This, it can be argued, is one key to the innovative treatment of the Saracens in The Sowdone of Babylone: a quasi-anthropological interest in the relations between two cultures, seen as both different and not so different. As discussed above, the treatment of the inherited tradition of the loss and recovery of the Passion relics in the Sowdone of Babylone is unusually restrained. But while the Christian relics are reduced in 157 Suzanne Yeager, ‘The World Translated: Marco Polo’s Devisement dou monde, The Book of John Mandeville, and their Medieval Audiences’, in Marco Polo and the Encounter of East and West, ed. Akbari and Iannuci, pp. 156–81. 158 Iain M. Higgins, review of The Egerton Version of Mandeville’s Travels, ed. M. C. Seymour, EETS OS 336 (Oxford and New York: Oxford University Press, 2010), in Speculum 86 (2011), 1123–5 (p. 1123). 159 The Book of John Mandeville, ed. and trans. Higgins, p. xxi. 160 Mary Campbell, The Witness and the Other World: Exotic European Travel Writing, 400–1600 (Ithaca, NY: Cornell University Press, 1991), p. 90.

342

Charlemagne in Medieval England

prominence, increased attention is given to Saracen religious objects and practices. Besides scenes inherited from the epic tradition, featuring the conventional behaviour of a disappointed sultan abusing his idols, new episodes are invented that show elaborate ceremonies to honour or to placate Saracen gods, and, on one occasion, to give religious burial rites to a dead Saracen king. The details surrounding these ceremonies tend to pull in contrary directions: on the one hand, the descriptions of votive objects and ritual actions seem designed to evoke a sense of exotic luxury and pagan sacrifice; on the other hand, interpretations of the rituals’ meaning can imply parallels with Christian practices. For example, there are numerous scenes in which the Sowdon makes offerings to his gods: giving thanks for the sack of Rome with clouds of frankincense and blowing of brass horns (676); petitioning Mars for victory with a precious crown, myrrh, aloes and frankincense (949); or solemnly offering large quantities of gold and silver and slain beasts, with more clouds of frankincense, to Mahounde and all the gods to ensure their aid (1021). Similarly, on three occasions when the Sowdon has denied his gods and threatened to burn them, he is persuaded to sue for pardon with exotic offerings of oil, honey, milk, frankincense, and 1,000 bezants (2447, 2521, 2787). The detail is partly reflecting literary influence: Laban’s invocation of ‘rede Mars Armypotente’, with his sacrifice, vow, and prayer for victory, shows clear echoes of the pagan setting in Chaucer’s Knight’s Tale with Arcite’s address to Mars. It is also echoing descriptions of the East in travel writing: the offering of milk, for example, recalls Marco Polo’s description of the sacrifices made by the Great Khan to his gods (chapter 52). But some details, such as offerings of gold at the altar and the ritual use of incense, are also normal elements of Christian religious practice, in literary tradition as well as in medieval life. This parallel is made more explicit when the narrative explains, after the Sowdon’s penitential offerings, that ‘the prestis assoyled him of that synne’ (2453), and that the Sowdon knelt to ask forgiveness and obtained pardon ‘throgh prayere and specialle grace’ (2526), using phrases that seem borrowed from the practice of Christian shrift through repentance, confession and absolution. There is an even more interesting conflation of religious references in the account of the burial of the king of Barbary according to the ‘right of Sarsenye’ (2269). The exotic ritual involves the use of ‘brennynge fire and riche oynemente’ (2270), but also includes a familiar-sounding sung service: they ‘songe the Dirige of Alkaron, | That bibill is of her laye’

The Insular Fierabras Tradition

343

(2271–2). This attempt to describe Saracen practice by parallel references to Christian and Islamic texts: the biblical Psalm from the Office for the Dead, and the Koran (whose status in Islam is explicitly compared to the Bible in Christianity, without any obvious prejudice), produces an effect of the kind noted by Higgins in describing Mandeville’s ‘syncretism’ (p. xxi) and by Campbell in her account of the increasing awareness of possible connections between Home and the East, evident in medieval writing through the use of ‘analogues and comparisons’ (p. 90). Even when there is no suggestion of parallel customs, the Middle English poem’s added details convey a sense of lively interest and a desire to explain and interpret the alien culture. For instance, the Saracens’ exotic victory celebrations are described at some length, culminating with their chant ‘Antrarian, Antrarian’, which the narrator interprets: ‘That signified Ioye generalle’ (689–70). And when the Sowdon summons an enormous army of peoples of different nationalities and races, the narrator interprets what might sound like just a barbarous detail as an intelligible practice in a military context: ‘He made hem drinke Wilde beestes bloode, | Of Tigre, Antilope and of Camalyon, | As is here use to egre here mode, | Whan þai in werre to battayle goon’ (1007–10). The narrative stance here is precisely that of the experienced traveller explaining the customs of distant lands. As noted above, though, such details of Saracen practices do not occur in isolation: they are part of a larger-scale evocation of the Sowdon’s court and customs in added passages that can be said to hinge on the concept of ‘worthiness’: a concept that is applied to the values both of Charles’s Christian world and of the Saracen world, but which is used with extraordinary frequency in relation to the Sowdon himself. The first of these passages occurs immediately after the prologue, and replaces lines which in the Egerton text describe Laban’s holding a great feast. The Egerton description is given in terms that would immediately have conveyed a familiar image of luxury, but which make no value judgements. In the English poem by contrast, the worthiness of the Sowdon and his entourage is repeatedly asserted, while the elaborate springtime setting, with its strong Chaucerian allusions, establishes an idea of courtly sophistication through the conventional links it sets up between the Spring season, love, and the Sowdon’s elegant hunting party. In line with this different presentation, the prologue makes no mention of Laban’s wickedness or of the atrocities he committed, as in

344

Charlemagne in Medieval England

the Egerton text, but stresses instead his achievements in winning Rome and conquering a large part of the Christian world. The other passage occurs at the point of transition between the two paired Anglo-Norman texts. Here, the Sowdone poet elaborates four lines in the Egerton text, on Charlemagne’s praising his old knights at the expense of the young peers, to produce a more general encomium lauding hard-won worthiness and knighthood, reminiscent of the conventional openings of Middle English romances praising the values of olden time (919–38). This is followed by the sultan’s dramatic invocation to Mars (939–62), and then by another conventionally ‘poetical’ Springtime opening (963–78). Here worthiness and knighthood are again invoked in a context of love and honour, and are specifically associated with ‘this worthy Sowdan’ and ‘his worthy sone Ferumbras’, and careful recognition is given to the Saracens’ honourable martial achievements (979–90). Thus the Mars passage, with its literary, classical, and heroic pretensions, comes between two parallel assertions of the worthiness and knighthood of both Christian and Saracen heroes, and the whole sequence serves to reinstate the earlier, positive image of the Sowdon for the second narrative in a diptych of two parallel narratives. In each of them Charlemagne is juxtaposed with a sultan of Babylon who is constructed as a worthy opponent, and their separate outcomes are paired at the end: Charles’s conquest of Laban the destroyer of Rome, and his conquest of Spain. But this is also a diptych of two worlds, whose differences do not exclude real points of connection. This might not seem surprising in relation to Fierabras, a prominent example of the epic stereotype of the noble Saracen who ultimately converts to Christianity, and who is portrayed in this text as a model of chivalrous behaviour even as he challenges Charlemagne,161 but it requires a radical reinterpretation of Laban’s role, for Laban is a figure who is usually seen in unequivocally negative terms. However, the Middle English poem does not make a complete break with tradition: the more positive, nuanced aspects of the presentation of the Sowdon and his culture that have been considered here clash violently with the retained account of his vicious refusal to convert, cursing his children by ‘almyghty Sathanas’, his soul’s finally being fetched to Hell (3165–90), and with the conventional images of Saracens 161 In The Sowdone Ferumbras’s extended challenge is couched in elaborate knightly terms (1059–74).

The Insular Fierabras Tradition

345

as fiendish creatures, ‘horible and stronge as devel of helle’ (1004–6, 3096–8).162 Moreover, there is no question of the text’s orthodox presentation of Christianity as the one true faith: the explicitly religious aspect of the conflict between the Saracen and Christian contingents is emphasized in the Middle English text when the Roman Duke Savaris anticipates victory as a manifestation of Christ’s power overcoming false gods (196–9), and again when the death of 10,000 maidens at Laban’s command is constructed as glad martyrdom (230–1). In both instances the reader’s investment in Christian triumph is assumed. Yet The Sowdone of Babylone shows clear signs of more thoughtful engagement with the Saracen Other than is found in most Middle English popular romances, and we can surely trace in this the influence of what seems to have been a sophisticated programme of reading in contemporary literature such as Chaucer’s Knight’s Tale, with its detailed portrayal of pagan beliefs and customs, and travel writing such as Mandeville, with its openness to difference in religion and culture. Both of these texts would appeal to a milieu not only learned but courtly.163 The Sowdone of Babylone perhaps represents the English translator/poet’s recognition that the insular Fierabras tradition offered an opportunity to explore the problematic issue of relations between Christendom and the Islamic world in a freshly imagined rewriting of the Anglo-Norman texts. Such a project may be rather more ambitious than the appropriations of the tradition seen in the other insular Fierabras texts, but it stems from the same innovative attitude – the readiness to exploit a licence to change things – that to some degree characterizes them all.

162 In The Sowdone the giant Saracen king Astrogot is not only boar-headed, but is ‘a develes sone, | Of Belsabubbis lyne’ (356–9). 163 Higgins notes that Mandeville explicitly presents itself as the equal of Latin learning, and addresses itself to a courtly audience (The Book of John Mandeville, ed. and trans. Higgins, pp. xvi–xvii).

6 Re-Purposing the Narrative: The Insular Otinel Tradition

T

he French Otinel tradition is represented far more sparsely than Fierabras in the corpus of manuscripts: only one continental and one Anglo-Norman witness survive as more or less complete manuscripts, and there are significant lacunae in the continental version.1 There are also two fragments, the longer of which (known as the ‘Mende’ fragment) is Anglo-Norman;2 it is difficult to ascertain the dialect of the shorter fragment, which runs to a mere four lines. Still, evidence of the reception of the narrative through adaptations, imitations and allusions in insular Charlemagne-related texts points to a wider circulation than this sparse corpus might suggest. The little critical attention the chanson de geste has received has not been generous to the text. The rather dismissive evaluation of its editors was that ‘Il n’y faut chercher aucune de ces beautés natives qui éclatent dans les plus anciennes chansons de geste. C’est une oeuvre de la décadence.’ 3 The only major study of the poem, by Paul Aebischer in 1960, is more concerned with its origins than its literary value.4 A few more recent studies have seen some

1 Vatican City, Biblioteca Apostolica Vaticana MS Reg. Lat. 1616 (cont.); Cologny Geneva, Bodmer Library MS 168 (AN). 2 The four lines of the short fragment are given in full in the edition: Otinel, ed. Guessard and Michelant, p. viii; the edition can be read on Gallica, the digital library of the Bibliothèque nationale de France. The short fragment uniquely represents Fernagu as Otinel’s father, not his uncle. The longer, ‘Mende’, fragment is published in Langlois, ‘Deux Fragments épiques: Otinel, Aspremont’, and corresponds to lines 636–929 of the edited text. 3 Otinel, ed. Guessard and Michelant, p. viii; Pierre Le Gentil was no less scathing in ‘Reflections sur la chanson d’Otinel’, Cultura Neolatina 321 (1961), 66–70. 4 Aebischer, Études sur Otinel; the question of the legend’s origins has not been resolved.

346

The Insular Otinel Tradition

347

merit in its use of intertextual echoes and its treatment of some specific themes.5 Otinel opens with the Saracen messenger, Otinel, arriving at Charlemagne’s court. There he announces that his king, Garsie, has attacked Rome and aggressively calls upon Charlemagne to convert. Otinel challenges Roland to individual combat, and during the subsequent fight between them, as representatives of their respective religions, God sends His Holy Spirit to convert Otinel, who is baptized and subsequently betrothed to Charlemagne’s daughter Belissant. Otinel joins Charlemagne’s army as they set off to make war against Garsie. Roland, Oliver and Ogier capture the noble Saracen Clarel in a skirmish but on their way back they are attacked by 1,500 Saracens, who rescue Clarel and take Ogier prisoner, putting him in the charge of Affamie, Garsie’s daughter. Meanwhile, Otinel has set out in search of the missing peers and engages in battle. He fights with Clarel, whom he defeats and kills. A general battle follows and Ogier, who has escaped from his captors, joins in. Otinel takes on Garsie, who is defeated and taken prisoner. Finally, Otinel marries Belissant and becomes king. Even this brief summary of the narrative reveals that Otinel shares major concerns of other chansons de geste of the cycle and on first reading seems to use the same narrative conventions and types: interfaith combat and conversion, the noble Saracen, the belle Sarrasine. Its links to Fierabras have long been accepted, and were a major factor in the dismissal of the text as a relatively unimportant work, treated as a paler version of the original.6 An examination of how the major elements of this text are treated does, however, reveal that while Otinel may appear conventional, it provides a critical response to traditions of the genre as expressed in other texts of the cycle du roi, and specifically to the tradition of Fierabras. Most notable is the different way in which 5 Jean-Claude Vallecalle, Messages et ambassades dans l’épopée française médiévale: L’Illusion du dialogue (Paris: Champion, 2006), pp. 161–4; 241–372; Bernard Ribémont, ‘Ius gentium, droit féodal et art du jongleur: Otinel et l’esthétique judiciaire’, in Chanter du geste: L’Art épique et son rayonnement: Hommage à Jean-Claude Vallecalle, ed. Marulène PossamaïPerez and Jean-René Valette (Paris: Champion, 2013), pp. 401–14; Ailes, ‘Otinel: An Epic in Dialogue with the Tradition’. 6 Gautier, Les Épopées françaises, III, p. 398; on the intertextual relationships between all three of the main narrative strands, see Chapter 1.

348

Charlemagne in Medieval England

conversion is treated, with Otinel converting not as a result of being defeated in combat, but rather when he is apparently on the point of vanquishing Roland, the conversion being the result of the action of the Holy Spirit in response to Charlemagne’s prayer.7 The Mende fragment follows the text in the Bodmer MS version of Otinel closely; this AngloNorman text, like the Hanover version of Fierabras, substantially follows the continental model. A line-by-line comparison of the two more or less complete manuscripts (including the Mende fragment as simply another witness to the Anglo-Norman version) reveals differences between the two manuscript versions, as we shall discuss below, but no consistent pattern of change. The underlying language of the chanson de geste is not Anglo-Norman and there is no way of knowing with certainty whether the continental or Anglo-Norman scribe has instigated the changes.8 By contrast, the Middle English response to Otinel again survives in three separate versions which, despite the complex theories surrounding the Auchinleck romances and the Fillingham Otuel and Roland,9 are all strikingly different and constitute three independent reworkings of the chanson de geste. All three English versions name the Saracen champion Otuel (variant spellings: Otuell, Ottuell, Otwel, Otuwel), never using either the form Otes or its diminutive Otinel as found in the French texts.10 Thus it may be significant that the Anglo-Norman version, despite using these forms throughout most of the text, begins by using the form Otuel in its first seventy lines,11 perhaps indicating an insular preference followed in all three English versions, just as all three English Fierabras texts follow the insular preference for the form Fierenbras. The two French-language versions are closer to each other than any of the English Otinel romances is to any other text. Even so, the French versions do show some marked differences, and comparison on these 7 See Ailes, ‘Otinel: An Epic in Dialogue’. 8 On the language of Otinel, see Chapter 1, n. 24. 9 For discussion, see Chapter 1, pp. 76–9. 10 ‘Otes’ appears in Thornton Otuel as the name of a French knight (67, 1127). 11 We are grateful for Diane Speed for allowing us to use her transcription of the Anglo-Norman text; line numbers are based on this transcription; however, quotations are taken from the manuscript, now digitized and available at http://www.e-codices.unifr.ch/en/cb/0168/211r.

The Insular Otinel Tradition

349

points with the Middle English versions demonstrates the fluidity of the textual relations between all the extant poems. (i) The opening action is set at Easter in the continental version and on Holy Innocents’ Day in the Anglo-Norman; both Auchinleck Otuel and Fillingham Otuel agree with the latter12 (ii) the king of Spain is called Garsile in the continental version, but in the Anglo-Norman his name is Marsilie/Garsie;13 while Fillingham Otuel follows the continental version, both Auchinleck Otuel and Thornton Otuel echo the Anglo-Norman14 (iii) a whole laisse describing Otinel’s encounter with Corsuble in the continental version is absent from the Anglo-Norman; Thornton Otuel omits the content of the laisse, but Auchinleck Otuel includes it15 (iv) the continental version concludes with Otuel’s wedding and investiture, which are described at length, while the AngloNorman ends with a brief account of Garsie’s capture; Fillingham Otuel and Auchinleck Otuel echo this brief

12 Thornton Otuel omits the date. 13 The Anglo-Norman text appears to distinguish between Marsilie, ruler of Spain in the Chanson de Roland tradition, and Garsie, ruler of Spain and lord of Otinel; the continental text treats them as one. This makes little sense; the continental reading could be considered a scribal error or as part of a pattern of contradiction in the poem, possibly perceived as an error and corrected by the Anglo-Norman scribe. The fact that the Fillingham manuscript has the same confusion suggests that a manuscript close to the extant Vatican manuscript was circulating in England; see Ailes, ‘Otinel: An Epic in Dialogue’. The form ‘Garsilion’ in the continental text is the cas regime form of ‘Garsile’. 14 Auchinleck Otuel interprets Marsile as one of Garsie’s lands (19, 63); Thornton Otuel names ‘Kynge Merthill’ (46) as Charlemagne’s enemy before Garcy’s messenger arrives (58), approximating to the AngloNorman version. 15 Otinel, ed. Guessard and Michelant, laisse 55 (lines 1864–915). Seven folios are missing from the Fillingham MS at this point.

350

Charlemagne in Medieval England ending,16 but Thornton Otuel accords with the continental version.

There are some common omissions shared by the three Middle English romances, all of which lack text equivalent to laisses 37 and 46,17 but further omitted laisses (present in both versions of the chanson de geste) affect only one or two of the romances (Auchinleck Otuel omits the equivalent of laisses 8, 11–12, part of 29, 34, 47, 60; Fillingham Otuel omits laisses 8, part of 34, 47; Thornton Otuel omits laisses 45, part of 47, 51). On the other hand, Purdie’s evidence for textual connection between Auchinleck Otuel and Fillingham Otuel (see above, p. 77) is supported by the presence of narrative material not present in either French-language text, or in Thornton Otuel, when Roland and Oliver (Roland and all the peers in Fillingham) offer to take the battle with Clarel in Charlemagne’s place before Otuel claims the fight.18 All this suggests that variant versions of Otinel were in circulation in England, and that insular translator/poets adapted freely from what was available.19 Nevertheless, there are some hints of specific insular preferences. For instance, in Fillingham Otuel, as soon as Charles has been identified by his ‘hore berde’ (70), Otuel responds by threatening to singe his ‘hore lokkys’ (82), and the same incident appears in Thornton Otuel (79–84, 94–6). Both romances are here in accord with the Anglo-Norman Otinel rather than the continental version, possibly suggesting an insular taste for such broad effects.20 In another episode 16 Otuel ends imperfectly in the Auchinleck MS, at the end of a quire after which an unknown number of folios is missing; however, the narrative up to this point closely parallels Fillingham Otuel in its focus on Garcy, suggesting that it too followed the Anglo-Norman version, and few lines are missing (equivalent to Fillingham Otuel, 1677–91). 17 Lines 1394–406 and 1586–98 in the continental Otinel, ed. Guessard and Michelant; lines 1354–69 and 1512–25 in the Anglo-Norman Bodmer MS version. 18 Auchinleck Otuel, 1239–48; Fillingham Otuel, 1295–306. 19 For wider discussion of this topic, see Field, ‘Patterns of Availability and Demand’. 20 There may be an echo here of the episode in La Destruction de Rome involving the Saracen Lucafer and the peers in a game of fire-blowing – a contribution to the constantly evolving process of intertextual allusion in reworking the texts of the Charlemagne tradition.

The Insular Otinel Tradition

351

shared by these three versions, Otuel directs his scornful insult for the ‘veillard’ who is too old to fight not to Duke Naimes, as in the continental tradition, but to Charlemagne himself.21 The insular texts here reinterpret the conventional epic motifs of Charles’s great age and white hair, making them function as deceptive signs of apparent impotence, to which the Saracens respond with hubristic and boastful scorn, thus providing assurance to the informed reader or listener of their eventual defeat, and adding further to Charlemagne’s reputation as the unyielding focus of Saracen aggression. As was the case with Fierabras, Otinel provides no obvious ‘hook’ for readers in England – nothing analogous to the theme of the Lendit relics for readers in France. The question, then, must be why did this story repeatedly capture the imagination of English poets in the fourteenth and fifteenth centuries? To some extent, an answer is provided by the unusual brevity of Otinel as a chanson de geste (its nineteenth-century editors unkindly remark that ‘le plus grand mérite du poème … est d’être très court’).22 As Mehl has shown, Middle English romance in general demonstrates a preference for brevity so marked as to construct the category ‘shorter romance’ as ‘a typical English variety of romance’. 23 The simple structure of the narrative, with its central focus on Otinel’s conversion and the deferred resolution of his marriage to Belissant until after the conquest of Garsile, presents a ready-made example of the basic pattern identified by Mehl – ‘a short romance with plenty of incident and a central hero’ (p. 58) – without the need for the kind of drastic abbreviation required to create a romance of this type from a much longer or more episodic French source. Shorter romances also lend themselves to a project that seems to have been a prominent aim in the production of vernacular books for secular use at this time: the education of the young.24 Certainly, the story of Otinel offers valuable

21 There is a possible intertextual echo in the continental version of the scene in the Fierabras tradition when Roland suggests leaving Naimes to guard the gate because of his age; however, on this occasion, the three insular Otinel texts evince a preference for greater focus on Charlemagne’s role. 22 Otinel, Guessard and Michelant, p. viii. 23 Mehl, Middle English Romances, p. 39. 24 See Hardman, ‘Popular Romances and Young Readers’.

352

Charlemagne in Medieval England

opportunities for illustrating key elements in the contemporary educational agenda. A further answer relates to the longstanding characterization of Otinel as a derivative work, whether created to capitalize on the success of the Matter of France by a poet who ‘vouloit rattacher sa chanson aux poèmes déjà célèbres qui racontoient la conquête de l’Espagne’,25 or, as argued above (pp. 347–8), composed as a response to Fierabras with deliberate echoes of the earlier chanson de geste.26 The popularity of Fierabras in England may have assured a positive reception of Otinel. It is, moreover, a remarkable fact that each of the three Middle English Otinel romances occurs paired with a companion romance of Charlemagne, as the second element in a diptych arrangement of texts: Roland and Vernagu and Otuel (Auchinleck MS), Sege of Melayne and Roland and Otuel (Thornton MS), and Firumbras and Otuel and Roland (Fillingham MS).27 Two of these companion texts, Roland and Vernagu and Sege of Melayne, have also been considered as possibly custom-made accretions to the Charlemagne tradition,28 while the climactic single-combat scene in the eccentric late romance of Rauf Coilyear has been shown to be ‘particularly influenced by some version of the Otinel episode’.29 This prompts the argument that all these independent but ‘attachable’ romances were created to feed an insular appetite for repeated engagement with the same archetypal fantasy, in which the Carolingian hero overcomes the Saracen aggressor in single combat.30 In the fourteenth and fifteenth centuries especially, this was a reassuringly optimistic fantasy, and one that Otinel embodies in perhaps its purest form.

25 Otinel, ed. Guessard and Michelant, p. vii. 26 See Ailes, ‘Otinel: An Epic in Dialogue’. 27 See Ailes and Hardman, ‘Texts in Conversation’. 28 Mehl, Middle English Romances, p. 57; Barron, English Medieval Romance, p. 96. 29 Diane Speed, Medieval English Romances, 2 vols (Durham: Durham Medieval Texts, 1993), I, 198. 30 Sege of Melayne has at its centre the single combat between Charlemagne and Darnadowse (997–1092). All three of these Otinel-related romances will be discussed at the end of this chapter.

The Insular Otinel Tradition

353

The Anglo-Norman Otinel None of the three French-language manuscript witnesses to Otinel offers a flawless text; each of them lacks lines found in another manuscript which are needed for the sense of the text. In his edition of the Mende fragment, Langlois noted its closeness to the text of the Bodmer MS, in a comparison based solely on the variants given in the edition.31 A detailed comparison of the fragment and the Bodmer MS confirms this closeness, but also shows that there are probable errors in both manuscripts. For example, in both complete manuscripts, when Roland is in combat, his Saracen opponent puts up his shield to defend himself, with the result that his shield is cut by Roland’s sword: [Et] Rollant sake Durendal le vaillant. Un colp li veit doner de mantenant; Meis li paen jette l’escu devant, Trestut li trenche quanke l’espée enprent, Fort se conbat, mès ne li valt nient. (Otinel, ed. Guessard and Michelant, 873–7; Bodmer MS, fol. 215vb) [And Roland draws his valiant [sword] Durendal. He wishes to strike him a blow; But the pagan thrusts his shield in front, He cuts it all, wherever the sword goes. He fights well, but it is of no use to him.]

The Mende fragment lacks line 875, and thus it reads as though it is Roland who is struggling in the fight; that is clearly a nonsense, as the pagan goes on to submit to him. On the other hand, a little later the Mende fragment appears to have the best reading. This time it is Oliver who is in combat, against the Saracen Balsami. In both complete manuscripts it seems that Oliver strikes the first blow against the Saracen: Oliver juste al rei de Ninivent, A Balsami qui at grant hardement, Sur sun escu ù out un liun peint, Mès Oliver le fiert si dreitement

31 Langlois, ‘Deux fragments épiques’, p. 436.

354

Charlemagne in Medieval England

Sur la ruele … (Otinel, ed. Guessard and Michelant, 839–43; Bodmer fol. 215va) [Oliver combats the king of Nineveh Balsami, who is so bold On his shield where there is painted a lion But Oliver strikes him so accurately On his round shield …]

The Mende fragment has an additional line after l. 840: ‘Li saracin le fiert ireemant’ (‘The Saracen strikes him in anger’), which makes it clear that it is Balsami who strikes Oliver on his shield, which bears the noble heraldic device of a lion; Oliver then responds with a parallel blow on the Saracen’s round shield. While the reading of the other manuscripts is not nonsense, the fragment makes a great deal more sense. The uncertain filiation and shifting allegiances across the various versions of the narrative in French and English, and the small number of extant textual witnesses to the tradition in French, make is difficult, if not impossible, always to know where changes took place. For example, the Anglo-Norman text is shorter than the continental one: the Guessard and Michelant edition (1859), based on the continental manuscript, is 2,133 lines long, compared to the 1,908 lines of the AngloNorman manuscript.32 It is worth noting that the general tendency is for Anglo-Norman texts and manuscripts to be shorter than their continental counterparts, so the probability is that the Anglo-Norman scribe abbreviated, but without a more stable manuscript stemma and other manuscripts it is difficult to be sure that this is the case, rather than the continental manuscript representing an expanded tradition.33 However, the fact that the lack of some lines in the Anglo-Norman manuscript renders the text nonsensical, or difficult to understand, suggests that the omissions could be the result of careless copying and not a programme of abbreviation at all. Thus when Otinel is about 32 The continental manuscript itself is incomplete and the lack has been provided in the edition from the Anglo-Norman Bodmer manuscript, so the difference may be greater than these statistics suggest. 33 On the length of Anglo-Norman chansons de geste in general, see Ailes, ‘What’s in a Name?’; Ailes, ‘Fierabras and Anglo-Norman Developments’.

The Insular Otinel Tradition

355

to join in a conflict between the French and the Saracens, Belissant, Charlemagne’s daughter, hands Otinel, her betrothed, the standard of King Gaifer (1104–5); Otinel spurs his horse, and then meets Roland, whom he addresses: Et Otes broche le bon corant destrier. Rollant encontre à l’issir d’un [viver]; De sa parole la va contralier: ‘Sire Rollans, venez vos [de] peschier?’ (Otinel, ed. Guessard and Michelant, 1105–8) [And Otinel spurs his good, swift horse He meets Roland coming away from a fish-pond He begins to harangue him: ‘Lord Roland, have you been fishing?’]

In the Anglo-Norman text, the line in which Otinel spurs his horse, with its change of grammatical subject, is missing, so that here it is Roland who is the subject of the line and who, nonsensically, asks the ironic question ‘Sire, fait il, venez vos de peschier?’ (Bodmer MS, fol. 217ra).34 A little later, when the pagans are retreating to their stronghold as night approaches, the continental manuscript tells us that that the pagans have entered the city: L’ost se desoivre chascuns à sa partie. Paien s’en entrent en la cité garnie. (Otinel, ed. Guessard and Michelant, 1249–50) [The army disperses, each to his own place. The pagans enter the defended city.]

These lines are lacking in the Anglo-Norman text (fol. 218ra), creating some narrative discontinuity. But perhaps the most interesting case of 34 In the continental manuscript ‘Rollant’ is the oblique case form, making it clear that Roland is being met and addressed. In the Anglo-Norman text the form is retained (‘Rollant’ rather than ‘Rollans’) but it appears to be Roland who goes on to speak. The difference between the texts suggests the Anglo-Norman scribe has misunderstood the syntax, possibly because the case system disintegrated earlier in Anglo-Norman than in continental dialects.

356

Charlemagne in Medieval England

missing lines making apparent nonsense of the text is when there is an exchange between Charlemagne and the pagan Clarel: Karles li rois s’est de l’eve aprimez; Clarel le voit, si s’est haut escriez: ‘Qui estes vos, qui de là cheminez?’ Dit l’emperere: ‘Biaus amis, que volez?’ ‘Je sui roi Karles, por quoi le demandez?’ Respont Clarel: ‘Jel te dirai assez: Je maudi l’oure que tu fuz onques nez.’ (Otinel, ed. Guessard and Michelant, 1340–6) [Charles the king goes near to the water; Clarel sees him and calls out aloud: ‘Who are you, who goes there?’ The emperor said, ‘Good friend, what do you want? ‘I am king Charles, why do you ask?’ Clarel replies, ‘I will tell you: I curse the hour when you were born.’]

In the Bodmer MS, the line which introduces Clarel’s response to Charlemagne is missing (line 1345), so that it seems as if Charlemagne is speaking the words which should be ascribed to Clarel (the speaker goes on to threaten to grant his empire to the pagan Florien, so this is clearly Clarel speaking). However, it may be rather that this is simply omission of the verbum dicendi to create a more dramatic exchange, requiring active reading to note who is speaking (a modern editor would indicate the change of speaker by punctuation).35 These cases suggest abbreviation arising from errors of copying or understanding. Elsewhere, abbreviation, if such it is, is not consistent, nor is it carried out systematically through the poem. Often it is simply a matter of a few lines of direct speech lacking, or two lines in one manuscript represented by a single line in the other; such minor differences between manuscripts are common in the chansons de geste and could be the result of elaboration in the continental text as much 35 There are other places where apparently careless copying has made the Anglo-Norman text unclear; for example, lines 1730–2 of the edited text are lacking in the Anglo-Norman manuscript, so there is no introduction to the newly armed knight Hardouin, who is here defeating a Saracen.

The Insular Otinel Tradition

357

as abbreviation in the Anglo-Norman.36 A few differences between the redactions affect linguistic embellishments. For example, the anaphora of the Anglo-Norman text in these lines: L’escu li brise suz la bucle d’argent L’escu li false li clavel en estent (Bodmer MS, fol. 217vb) [He breaks the shield on its silver boss; He shatters the shield, the links open]

corresponds to just one line in the continental version: L’escu li perce souz la boucle d’argient. (Otinel, ed. Guessard and Michelant, 1193) [He pierces the shield below its silver boss.]

But where lines begin with the same words, this loss of a line could easily be a case of eye-skip. Variations of this kind are not peculiar to the insular nature of the manuscript, but relate to the instability of the medieval text. In a few passages the Anglo-Norman text is noticeably longer, and given the general Anglo-Norman tendency to abbreviate, these instances demand attention. One such occurs when Roland is armed in preparation for combat with Otinel,37 and here the additions suggest deliberate enhancement. In the continental version, Roland is given his armour, helmet and shield, and here the most interesting difference is the added information that the helmet he dons was that of Goliath, won by Charlemagne: ‘Co fu le healme Golias le jehant | Charles le prist quant il occist brachant’ (Bodmer MS, fol. 212va). Finally, he is given his sword: ‘On li aporte Durendal la trenchen’ (Otinel, ed. Guessard and Michelant, 307). The Anglo-Norman version splits this line, so that 36 For example, lines 1376–8 of the edited text are lacking in the AngloNorman manuscript; they consist of the second line of Charlemagne’s call for his arms (‘Si irai combater au paien desfaez’) and the comment that Otinel jumps up to address him; however, the following line makes the change of speaker clear. 37 Otinel, ed. Guessard and Michelant, 294–314; Anglo-Norman Bodmer MS, fol. 212rb.

358

Charlemagne in Medieval England

Roland receives Durendal immediately after putting on his helmet (‘puis li aportent durendal le trenchant’; ‘then they bring to him Durendal, the sharp’). Then there are four additional lines about the sword’s excellence, while the later line recounting the same detail is retained, but without naming the sword: ‘puis li aportent un fort espe tranchant’ (‘then they bring to him a strong sharp sword’). The result is that Roland is now equipped with two swords. In three places the Anglo-Norman text diverges further from the continental version, corresponding to the following lines in the edited text: 1128–65, 1323–35 and parts of the final battle scene corresponding to lines 1695–820.38 The first of these, also part of a battle scene, gives a much longer description of the combat from which Engillers must be rescued by Ysoré. In the continental text, Engillers appears abruptly in the narration: ‘Malement est Engillers atornez’ (Otinel, ed. Guessard and Michelant, 1141). The Anglo-Norman narrative is more satisfactory in that the need for him to be rescued by Ysoré is made apparent by the description of the fighting he has been engaged in. The second passage is even more interesting in that it suggests a greater interest in the religious aspect of the conflict and in the (constructed) religious practices of the pagans, entirely in keeping with the way Otinel is converted earlier in the text. We have noted above (Chapter 2) that during the exchange between Clarel and Otinel before they meet in combat Clarel is more insulting to Christianity and shows more knowledge of Christian doctrine (Bodmer MS, fol. 218rb). This is followed by more elaboration with a developed arming sequence as Clarel rises in the morning, similar to the earlier sequence given to Roland when he is armed in preparation for combat with Otinel. As noted above, Roland bears Goliath’s helmet, with its history of Carolingian conquest, and as if to create chivalric equivalence, Clarel bears the helmet of Priam, emphasizing his status. While this suggests a possible parity with Roland, Clarel’s alterity is maintained by the detail that this helmet is made not of steel or silver but ‘de la teste d’un serpent’ (Bodmer MS, fol. 218va). Even more significant, the snake’s-head helmet is inscribed with the names of his gods:

38 The final battle is longer and more detailed in the Anglo-Norman text than the continental one: the 125 lines of the edited text become 181 lines in the insular version (Bodmer MS, fols 220va–221vr).

The Insular Otinel Tradition

359

Escrit i sunt Jovin e tervagant E mahumet en la guise d’un enfant Cil sunt li deu kil recleime sovent Par ces quide il aler seurement. (Bodmer MS, fol. 218va) [ Jove and Tervagant are inscribed there And Mohammed, like a child. These are the gods he frequently invokes. He believes he can be safe because of them.]

As he leaves he is commended to his gods by his beloved, who promises an offering of 1,000 marks for his success (the last line appears to be a prayer): Dist alphamie, ‘A Mahun te comant Apolin sue victorie li consent De mil mars d’or te frai acressement.’ (Bodmer MS, fol. 218va) [Afamie said, ‘I commend you to Mohammed: May Apollo give his consent to his victory I will enrich you by 1,000 marks of gold.’]

Following Clarel are his men – Saracens and Persians, Arabs and Turks – and an idol of Mohammed: Mahumet levent en un char vertant Ultre li passent la fort eue bruiant Sur un halt pui le leissant en estant Forment latachent a chaenes d’argent K’il ne chete nariere navant Trestuit laurent e prient humlement Que vertu i face chascun i fet present Tut li plus povre i offri un besant. (Bodmer MS, fol. 218va) [They raise Mohammed on a travelling cart They cross the strong noisy water with it They leave it standing on a high hill They attach it strongly with silver chains

360

Charlemagne in Medieval England

That it would not follow backwards or forwards They all praise him and humbly pray Each one makes an offering that he might do well there Even the poorest offers a besant.]

Such added evocations of the pagan gods may be compared to similar innovations in The Sowdone of Babylone (see p. 342). A little later, when Clarel challenges Otinel, the latter responds with a défi couched in religious terms. In the continental text, Otinel refers back to Clarel’s invocation of his god: Dist Otinel: ‘Mult t’es ore vantez, Mult as tes diex et prisiez et loez; Mès, par celui qui en croiz fu penez, Ne mengerai devant qu’eres tuez, Car Jhesu Crist a mult grant poestez, N’autre de lui ne doit estre aorez. Je vos defi, de moi bien vos gardez.’ (Otinel, ed. Guessard and Michelant, 1444–50) [Otinel said: ‘you have boasted much, You have praised and honoured your gods But, in the name of him who hung on a cross I will not eat until you are killed For Jesus Christ has great power He alone should be worshipped I defy you, defend yourself ’.]

The Anglo-Norman version is quite different, with an additional line that explicitly links victory and faith, attributing the outcome of their combat to the power of Christ: ‘Par deu fait otes, culvert, vos i mentez Si jo combat tu esteras matez Kar de jhesu averai la poestez Nautre de lui nen iert deu apelez Dehez en ait mahun e fiertez E vos meinie que par lui vos clamez De mun espe vos ferrai neelez Par cel seingnur qui en croiz fu pendez

The Insular Otinel Tradition

361

Tumber vos frai sa cest colp matendez (Bodmer MS, fol. 219rb) [‘By God’, said Otinel, ‘you lie If I fight you will be defeated For I will have the power of Jesus. None other than he shall be called God May Mohammed be cursed and [his] reliquaries And followers, whom you call upon in his name. I will chisel out patterns on you with my sword. By the Lord who was hanged on the cross I will bring you down if you await this blow.’]

Following the arming of Roland and the congé granted by the king, a few lines unique to the Anglo-Norman version, while apparently of little importance in themselves, seem more significant when we read them in the light of this later passage, as here Roland is also commended to God: Apres lui vunt puceles e enfant Ki tuit li dient a jhesu te comant Seinte marie te seit de mort garant. (Bodmer MS, fol. 212vb) [After him came maidens and children Who all say ‘I commend you to Jesus, May holy Mary protect you from death.’]

In the context of an apparently greater concern in the Anglo-Norman than the continental manuscript with religious matters, minor changes such as the use of the term ‘li convers’ (fol. 218ra; fol. 218rb) where the continental manuscript simply has ‘cil’ (‘he’, 1258) or gives Otinel’s name (1295) may have some significance, especially in light of the fact (as discussed elsewhere in this volume) that the theme of conversion is heightened in the Middle English Fillingham MS texts of both Firumbras and Otuel and Roland.39 One other peculiar reading of the Anglo-Norman manuscript is 39 The word convers seems to have been used more frequently to refer to lay brothers in a monastery than to ‘converts’ in the modern understanding. The Tobler-Lommatzsch Altfranzösisches wörterbuch, II, 816, cites only

362

Charlemagne in Medieval England

worth noting, if only because it attracts a rare positive response from the editors, who print the lines in their notes. The exchange of words between Garsile and Otinel, which occupies nineteen lines in the edited text, is considerably shorter in the Anglo-Norman, which lacks Otinel’s pleadings with Garsile to convert (thus losing one of the parallels with Fierabras, where Fierabras similarly pleads with his father Balan). In the edited text, Otinel’s appeal to his former lord to convert is preceded by a fairly commonplace taunt: ‘Sire Garsile, lessiez vos la mellée? La vostre gient lessiez mult effréé[e] Et sanz seignor, dolente et esgarée! Hé! mauvès rois, ta vie est afinée: Mult mal véistes onques ceste jornée, Se Dex me sauve et ma trenchant espée.’ (Otinel, ed. Guessard and Michelant, 2010–15) [‘My lord Garsy, are you leaving the fight? Your men are in great fear Without a lord, saddened and lost. Ah! Evil king, your life is at an end. That you saw this day is a curse on you If God saves me and my sharp sword.’]

Instead of this taunt, in the Anglo-Norman text Otinel tells Garsie that he will have to eat humble pie, in a few lines described by the editors as ‘assaisonnée d’un gros sel’:40 ‘Pur deu, dit il, dite mei, Sire reis, Devez anuit conreer ces franceis Alez vos querre le cras lard as peis Nel mangerent pur mil mars d’orkeneis Altre mes feites, co est manger a burgeis.’ (Bodmer MS, fol. 222ra) [‘By god’, he said, ‘tell me, lord king,

two instances of convers meaning proselyte; see also Godefroy, 286 c and I 190 a. 40 Otinel, ed. Guessard and Michelant, p. 91.

The Insular Otinel Tradition

363

Should you tonight look after these French, Will you look for fat bacon and peas? They would not eat it for a thousand Orkney marks: Make a different dish; that is to eat like a bourgeois.’ ]

Narrative differences are largely limited to changes in the combats. The change of one word, when the French are apparently in retreat, produces a change in the meaning, for where the continental text reads ‘Ariere sont noz François reculez’ (‘Our French moved back’; 1720), in the Anglo-Norman text it is the pagans who are being forced back: ‘Par grant efforz ont paiens reculez’ (‘with great efforts the pagans moved back’; Bodmer MS, fol. 221rb). Perhaps the insular scribe could not accept that the Saracens might have the ascendancy, even for a short time; this is a reaction that seems to have been shared in some instances by the Middle English adaptors of the Charlemagne tradition (see p. 282). We have also noted above the more abrupt ending of the AngloNorman text. The Anglo-Norman version diverges from the continental one from around line 2052 of the edited text. Here the continental text describes a final brief combat where Ogier, distracted by the sight of a pagan fleeing, leaves the group of peers who are escorting Garsile to the emperor; the text then returns to the Saracen king, telling us that he is imprisoned by Charlemagne and dies in his prison ‘à duel et à vilté’ (‘in suffering and humiliation’; 2076). The narrative continues with the marriage of Otinel and Belissant and Charlemagne’s departure for Paris, leaving Otinel in charge of Garsile’s former lands. All of this is lacking in the Anglo-Norman text. Roland and Otinel take Garsile directly to Charlemagne and we are told simply that ‘Il len a Paris sa cite mene’ (‘He took him to Paris, his city’; Bodmer MS, fol. 222rb). The final few lines of the text tell us that the Saracen city was taken by nightfall, and enjoin us to pray for any such messenger. The text ends part way down the second column of the last folio and there is no evidence of any material damage. Invocations to prayer are, moreover, not unusual at the end of a chanson de geste. However, it is possible that a previous copy of the narrative had some damage to its final folio which has been tidied up, either by the Bodmer scribe or an earlier copyist. While it is impossible to know whether this difference represents an insular abbreviation or an expansion in the continental text to tie in the loose ends of the narrative, the Anglo-Norman version shows interesting independence. The call to prayer at the end is certainly far from the usual formulaic exhortation:

364

Charlemagne in Medieval England

Quant l’um orra de itel messagier Bien devent tuit pur salme prier Ke si aida paiens a traverser. (Bodmer MS, fol. 222rb) [When one hears of such a messenger Everyone should pray for his soul That he should be helped to thwart pagans.]

This turns our attention back to Otinel himself and to the beginning of the poem, when he arrives at Charlemagne’s court as a messenger.41 It suggests a remanieur who is aware of the convention of finishing such texts with a pious invocation, and whose sense of narrative structure is not bound by the conventional, but who is also perhaps not so concerned with the intertextual parallels with Fierabras (which ends, like the continental Otinel, with a marriage and the handing over of the conquered lands to the converted Saracen). It is not just difference that can be striking. One continuity across the French-language Otinel texts is the ongoing depiction of Charlemagne as ‘nostre emperere’, a phrase which is found on a number of occasions in both manuscripts. Charlemagne’s actual empire has long since disappeared from sight; in French-language texts he is emperor of the West or of Christendom. It is indeed clear in Garsie’s challenge to Charlemagne that he assumes Charlemagne to have control over Normandy and England, which he is prepared to cede back to him if he submits: Il te dorra aver e manantie Ensurketut te larra normendie E dengleterre les porz e la navie (Bodmer MS, fol. 211vb) [He will give you goods and dwelling places Moreover he will leave you Normandy And the ports and navy of England.]

Though the edited text has corrected these lines (148–50) using the 41 A parallel concern with the role of messengers is shown in the Auchinleck Otuel (see below, p. 368).

The Insular Otinel Tradition

365

Anglo-Norman manuscript, the allusions to England and Normandy are found in the continental manuscript. They recall the lines in the Chanson de Roland which imply that Charlemagne’s empire included the whole of Christendom (see pp. 228–31). While such an implication would seem to place England under the authority of a foreign power, the allusions must be read in the context of a myth which ignores the actual limits of the empire.42 There is certainly no implied link here to the Holy Roman Empire as it existed in the fourteenth century, when the Bodmer MS was copied. The Anglo-Norman scribe need have no reservations about including England as part of Charlemagne’s empire; it is simply that England is part of Christendom. Indeed, the implication that there was one Christian ruler over England and France (which Garsie says he will keep for himself) might rather appeal to England in or on the verge of a war with France, a war undertaken with the ultimate aim of establishing just such a dual monarchy. It is not dissimilar to the way England is included as part of Charlemagne’s empire in the Oxford Roland, and may be read as part of the evidence for a seemingly unproblematic naturalizing of the Matter of France in insular textual culture. One other peculiarity of the Anglo-Norman text is the presence of two laisses written in alexandrines (ll. 17–46). There is no obvious reason for the short-lived change of metre from the decasyllabic line of the rest of the text. The first laisse (17–26) is quite different in expression (but not in content) from the continental text; it is also in this section that Garsie is conflated with Marsile. It raises the possibility that there was a second source manuscript, now lost, which had a version of the narrative in alexandrines. Overall it seems that there may be greater interest in the AngloNorman text in arming sequences, in some of the combat scenes and in describing constructed religious practices of the Saracens. As with the 42 Gabriele, An Empire of Memory, writes of Charlemagne’s ‘expanding empire’ that the list of Charlemagne’s conquests in the Roland ‘most likely represent a contemporary eleventh-century understanding of Charlemagne’s legendary empire, encompassing virtually the whole of Christendom’ (p. 33); see also the section on ‘L’Empire et la chrétienté’, in M. Combarieu du Gres, L’Idéal human et l’expérience morale chez les héros des chansons de geste, 2 vols (Aix-en-Provence: Publications Université de Provence, 1979), I, 226–37, where she concludes that ‘pour tous l’empire se confond avec la chrétienté (p. 237).

366

Charlemagne in Medieval England

abbreviations discussed above, it is not possible to say with certainty that these are elaborations, but there is clear evidence here of ongoing, and possibly increased, interest in the inter-faith combat aspect of the narrative, the ‘holy war’ meme. Further, as noted above, an interest in the exotic practices of the religious Other connects with innovative adaptations observed in the Middle English Sowdone of Babylone. Both these features may owe something to a heightened awareness of the Islamic Other in light of, first, the ‘emergence of the aggressive sultanate of Egypt’ in the thirteenth century,43 and then the growing threat from the Ottoman Turks throughout the fourteenth century. In all these cases, the insular adaptors of Charlemagne texts in both languages seem particularly sensitive to the material’s potential for appropriation in new political contexts. Meanwhile, the instances of possible intertextual echoes between the different narratives suggest a tradition of constant reworking and cross-influences.

The Auchinleck Otuel Of the three Middle English versions of Otinel, this is the most thoroughly adapted to the demands of insular consumers outlined above. The epic prologue in which the French text makes its claim as a new geste of Charlemagne and the twelve peers to set beside the famous story of the traitor Ganelon and Roncevaux is entirely rewritten; instead, the story is advertised as telling of battles ‘þat was sumtime bitwene | Cristine men & sarrazins kene’ (5–6). Parallel verse paragraphs follow, describing a king of France called Charles who ‘meintenede cristendom ariȝt’ (7–14), and a king of Lombardy called Garsie who plotted night and day to ‘bringe cristendom to nout’ (15–32). Garsie’s unprovoked imperial aggression is elaborated in two further paragraphs setting out his ambition to conquer all Christendom and turn it into ‘heþennesse’ with the help of fifteen heathen subject kings, all sworn to attack Charles as ‘chef of cristene gynges’ (33–53). Particular care has been taken to incriminate Garsie. First, his belief in ‘maumettrie’ is represented as culpable unregeneracy: he ‘forsok god and seinte marie’ (25–6). This perhaps draws on the widespread notion of Islam as a heretical offshoot of Christianity, and thus enables Garsie to be represented as if he were a Christian renegade, guilty of apostasy, rather than simply as adherent 43 Tyerman, God’s War, p. 715; see also Phillips, Holy Warriors, pp. 258–75.

The Insular Otinel Tradition

367

to an alien faith.44 Secondly, the injustice of Garsie’s aim to take Christendom by force is underlined by the narration, as ‘He was lord of londes ynowe’ (38). The scene is thus set for a reading of the Otinel story in terms not of the wider Matter of France, but of a paradigmatic struggle between the forces of good and evil for the future of Christendom. However, this is not to cast Otuel as a cartoon-like simplification. The serious attention to religious faith indicated by the account of Garsie’s explicit rejection of Christ’s redemption (23–4) is evident also in Otuel’s unique treatment of the Saracen champion’s conversion. Here the poem gives a detailed account of the Christians’ prayers, Christ’s response, and their thanksgiving, with emphasis on the role of God’s grace and Otuel’s will in bringing about his change of heart (566–98, 633–4). Otuel’s reconciliation with Roland (the ‘seiȝtnesse bitwene þo kniȝtes’ (570) for which Charles prayed) is demonstrated in a display of forgiveness and brotherly love, enhanced by explanatory details: ‘Eyther forȝaf oþer his loþ | Nas non of hem wiþ oþer wroþ’ (603–6).45 It is hard to avoid the impression of didactic intent behind these additions. Indeed, several innovations in the romance appear to have been designed to maximize the educational impact of the story. The inherited narrative falls naturally into two halves: Otinel’s challenge and conversion, and Charlemagne’s campaign against Garsie. In the Auchinleck Otuel, the diptych structure is clearly signalled with a large capital and a new call to attention at the beginning of the second half: Lordinges, boþe ȝinge & olde, Herkneþ as we formest tolde, How þe werre was fol hyȝe, Bitwene king charles & king garsie. (669–72)

44 As John V. Tolan argues, citing Peter of Cluny and other medieval theologians, ‘Islam was not, for these authors, a separate religion, … it was merely one variety of heretical error’ (Saracens, p. 165). 45 This detail is contextualized within a discussion of rituals of feudal love and friendship by Marcel Elias, in ‘Emotional Rhetoric in the Middle English Charlemagne Romances: The Failures of Crusading Practices in the Otuel Texts’, a paper delivered at the 49th International Congress on Medieval Studies, Kalamazoo, 2014. There is a parallel with Fyrumbras asking Oliver’s forgiveness in the Ashmole Sir Ferumbras (see p. 283).

368

Charlemagne in Medieval England

Each half is arranged to present a lesson in correct chivalric behaviour. In all versions, Otinel as Garsie’s messenger is afforded protection by Charlemagne in accordance with the rules of chivalry, but in the Auchinleck Otuel the motif is elaborated so as to provide a demonstration of the inviolability of the rule, even under extreme provocation. Six new instances are added in which Charles reiterates the principle that ‘noman be so wood, | For to don hym oþer þan good, | A kinges messager for he is’ (141–3; also 193–6, 205–8, 220, 329–32, 355–8, 363–70), and even though Otuel’s outrageous words and behaviour cause Charles ‘to meuen his blod’ (356), he still commands that the messenger be ‘onoured als a kniȝt’ (370). The second half begins with the episode of the three knights (Roland, Oliver and Oger) who capture the Saracen king Clarel; uniquely in the Auchinleck Otuel, Clarel reminds the knights as he surrenders to them, ‘To ȝou it were lutel honour | To sle me þat nabbe no socour’ (821–2), and this chivalric rule is put into practice when the three knights decide to let him go: ‘Hit where sschame to ous, iwis, | To sslen a man þat ȝolden him is’ (861–2). These self-conscious references to accepted ideals of knightly conduct can be seen to accord with the view of modern historians that chivalry is, in essence, the ‘restraint of violence amongst fellow warriors’ of noble rank: an ideal identifiable in theory and in practice from the eleventh century at least.46 Clarel later encounters Otuel, and again the Auchinleck version takes the opportunity to illustrate honourable chivalric behaviour in a series of linked additions to the source: Clarel approaches the French camp ‘in fourme of pees’ (1137), and agrees to return and fight Otuel the next day. He asks for an assurance of safe conduct: ‘Woltou sikere me on hond, | þat no man of king charles lond | Schal do me no vileynie’ (1185–7) and Otuel promises him in the same form of words (1194–5). This promise is reiterated in the narration (1215–18) when Clarel returns with safety: ‘þo was clarel fol trist, | For to segge what him lust’ (1221–2). The romance 46 David Crouch, The Birth of Nobility: Constructing Aristocracy in England and France, 900–1300 (London: Longman, 2005), p. 27, citing John Gillingham, ‘Kingship, Chivalry and Love: Political and Cultural Values in the Earliest History Written in French, Geoffrey Gaimar’s Estoire des Engleis’, in Anglo-Norman Political Culture and the Twelfth-Century Renaissance, ed. C. W. Hollister (Woodbridge: Boydell Press, 1997), pp. 33–58, and Strickland, War and Chivalry, pp. 19–30.

The Insular Otinel Tradition

369

concludes with a final lesson in chivalry which Otuel delivers to Roland, Oliver and Oger, putting into practice the text’s earlier teaching on how to treat a yielded enemy, as he counsels them to deal honourably with the defeated Garsie: Nou he haueþ þis ȝift iȝiue, I rede þat we laten him liue. Bifore þe king he schall be brouȝt, For gode, we nulle slen him nouȝt. (1719–22)

The emphasis on correct procedure continues in the last speech of the incomplete text, where Otuel uniquely spells out the feudal details of Garsie’s submission to Charlemagne (1729–38). The whole scene is quite unlike the imprisonment of Garsie in the French tradition, and although each of the English romances rewrites this episode to some extent, neither of the other two has anything resembling Auchinleck’s marked concern with chivalrous practice. Otuel, like Fierabras, is modelled on a conflation of the ‘hostile challenger’ and the chanson de geste stereotype of the ‘noble Saracen’; but in this Middle English adaptation the stereotype has been systematically enhanced. Otuel is portrayed in terms appropriate to a chivalrous Saracen from the very start, as ‘a sarasin douȝti and good’ (69), and ‘a douȝti kniȝt’ (88), and he plights his troth to fight Roland the following day with such perfect formality that Charles is astonished at his ‘gode wille’ (309–14). The supreme proof of Otuel’s commitment to the rules of chivalry occurs in the course of the highly stylized account of his one-to-one combat with Roland (443–554), when first Roland and then Otuel accidentally kills his opponent’s horse:47 uniquely in this version, the episode is reconfigured to provide a model of courtesy in battle. Roland chivalrously stays his hand until Otuel has recovered from his fall (465–8), and Otuel, impressed by this noble behaviour, consciously follows Roland’s example (489–94). Later, after Otuel’s miraculous conversion, he fights with Charlemagne against the Saracens and, once again, a single-combat episode serves to focus the wider conflict. Otuel, now a Christian, confronts the Saracen Clarel, and in this version alone 47 This scene appears to be modelled on the horse-killing and dismounting episode in Fierabras: see Ailes, ‘Otinel: An Epic in Dialogue’.

370

Charlemagne in Medieval England

their preliminary conversation is carefully constructed to show both knights acting in accordance with the code of chivalry, and addressing each other with exemplary politeness, instead of exchanging curses and insulting each other’s religion as in all other versions (1143–50, 1179– 1204). Despite his chivalry, Clarel (unlike Otuel) remains obdurate; indeed, uniquely in this version, he proposes they undertake their battle explicitly as a trial by combat between Otuel as champion of Christ and Clarel as champion of the Saracen gods, as Otuel reports: [He] seide, algate he wolde preue, Þat ich am in mis beleue, Þerefore he profreþ him to fiȝt To wite wheþer is more of myȝt, Ihu þat is louerd min, Or mahoun & apolyn. (1265–8)

However, it remains the case that in this version of the Otinel story every opportunity has been taken to demonstrate the finer points of chivalrous behaviour, whether in extreme situations between opponents on the battlefield, or in the court by repeatedly honouring the protected status of an enemy messenger, no matter how outrageously provocative he is. These instances slot into a wider added emphasis in Otuel on chivalric values, embodied in the knights’ behaviour not only in courteous actions and polite speech, but also in expressions of knightly shame which are attributed to Charles and the peers on many occasions in confrontation with the enemy Saracens (170, 277–80, 543, 861–2, 1389). All this indicates a specific reprogramming of the narrative. It may be read as contributing to an educational project: the construction of a pattern of correct knightly behaviour, not simply as an ideal, but as a programme of practical action, which is effectively conveyed through the exemplary conduct of the individual hero confronting his enemy in one-to-one combat.48 Like Malory’s Morte Darthur over a century later, Otuel seems in this respect to offer a narrative enactment of the codified behaviour laid out in the handbooks of chivalry that sometimes accompany romances or chansons de geste in manuscript compilations.49 48 See Hardman, ‘Popular Romances and Young Readers’, pp. 154–9. 49 An outstanding example is the Shrewsbury Book, BL MS Royal 15 E vi.

The Insular Otinel Tradition

371

Middle English romance, with its typical focus on a central hero, clearly lends itself to such a project, while the predominance of incident also noted by Mehl as characteristic of the genre is enhanced in the Auchinleck Otuel where the inherited narrative offers opportunities. For example, there is a newly imagined account of Oger’s escape from prison with the help of a clever young squire (1621–62). Here too the educational bias appears, as the incident is introduced as a lesson on God’s providence: ‘Herkneþ, what hede good to him nam | & how he out of prisoun kam’ (1623–4). Otuel thus constitutes a systematic reappropriation of the Otinel tradition for a specifically didactic purpose, aiming to reinforce patterns of chivalric ideals and etiquette in an exemplary narrative of good triumphing over evil that can also function as a manual of good practice.

The Fillingham Otuel and Roland This romance is unique in its treatment of Otinel in that it realizes the potential for cyclical development inherent in the prologue of the French tradition, where reference is made to Charlemagne’s loss of the twelve peers at Roncevaux through Ganelon’s betrayal of them to Marsilie. In Otuel and Roland the story of Roncevaux, together with further material from the Pseudo-Turpin Chronicle, is incorporated into the narrative to produce a continuous history of Charlemagne’s exploits from the conversion of Otuel and the defeat and conversion of Garcy, to the conquest of Spain and Navarre, to the disaster of Roncevaux, ending with Charlemagne’s defeat of Beligans and the punishment of the traitor Ganelon. The scholarly theories surrounding the origins of this composite text and its relation to Roland and Vernagu are discussed in Chapter 1; discussion here is centred on the extant text as an autonomous Middle English romance and how it relates to the French source texts. Despite its dual origins, and the possibility that the two halves may be by different writers, the romance has the appearance of homogeneity on account of, first, the similar verse forms used in the two parts,50 and secondly, a unique set of descriptive rubrics imposed on the whole work to subdivide it into nine passus or chapters. Following 50 Both halves (lines 1–1697 and 1698–2786) use twelve-line tail-rhyme stanzas with varying four-stress and three-stress lines. The rhyme schemes in the halves are, respectively, aabaabccbddb and aabccbddbeeb.

372

Charlemagne in Medieval England

the prologue and its introduction to the first part of the Otinel story, a six-line rubric announces the next ‘pas’: Here bygynneth a batayle snelle Off Rowland and off Otuel that wonderlyche was in fyȝt, and howe Otuel ycristened was. herkenyth nowe a mery pas and of a stronge fyȝt. (225–30)

Similar rubrics introduce: a battle of Roland, Oliver and Oger with four Saracen kings (662–73); a battle of ‘these thre kniȝtes’ with 6,140 Saracens ‘for cristendom’ (863–8); a battle of King Clarel and Otuel ‘for the lawe’ (1157–62); a battle of Charles and King Garcy and how Garcy was christened (1559–64); a battle of Charles and Ebrahim ‘for … cristendom’ (1692–7); ‘a Rewful tale | How Rowlond deyde at rouncyuale’ (1976–7); and a battle of Charles and Turpin against Saracens (2583–8). The whole text is summarized in the explicit: Here endeth otuel, Roulond & Olyuer, And off the twelf dussypers That deyden in the batayle off Runcyvale. (2781–3)

This final rubric presents the text as a diptych composed of two major narrative sections: the Otinel story and the tale of the Battle of Roncevaux. Some effort has been made to link them together and to incorporate the other episodes taken from the Pseudo-Turpin Chronicle that fill the space between them.51 At the end of the Otinel story, the narrator refers to Charles and his peers living ‘in warre many ȝerys’ and promises to tell of their other battles (1686–91), thus leading to the next episode, which is further connected by a unique added reference to the preceding material: ‘Afftyr Garcy nouȝt long hyt nas’ (1698). Similarly, in the last passus, the account of Charles’s victory at Saragossa is linked to the Otinel story by a unique added passage in which Otuel is inserted into the Pseudo-Turpin tradition, fighting alongside Charles and Turpin 51 Equivalent to Pseudo-Turpin Chronicle, chapters 18–19 (see Historia Karoli Magni, ed. Meredith-Jones).

The Insular Otinel Tradition

373

to avenge the deaths of Roland and Oliver (2622–30).52 However, this ‘diptych’ structure is overlain by the series of nine passus divisions that construct the text as a single sequence of coherent episodes. The passus rubrics, despite their ‘minstrel’ idiom in calling for attention,53 divide the text in a similar way to the chapter headings in the chronicle tradition, and the Middle English Pseudo-Turpin provides a contemporary model for comparison in the section of text equivalent to Otuel and Roland, 1692–2786.54 The ten headings given in the chronicles are reduced in the romance to just three, which correspond to those referring to battles in the Pseudo-Turpin, and the same focus on battles is evident in the five rubrics that cover the Otinel material. The whole text is thus presented as a catalogue of Charlemagne’s battles against Saracens, in line with the repeated emphasis in the opening lines of Otuel on telling of ‘batailles bolde’ (2, 4). However, while in Otuel the battles are exploited as opportunities for teaching chivalry, in Otuel and Roland battles are interpreted as demonstrations of God’s grace in action. As the rubrics make clear, what is at stake in all the battles is the Christian faith, and the Middle English author loses no opportunity to enhance the text’s reference to the Christian identity of Charlemagne’s warriors.55 Both the Christian and Saracen leaders appeal to their gods 52 There is possibly a distant connection with the rhymed Roland tradition, where in the Châteauroux–V7 version of La Chanson de Roland, the French knight Oton features in the episode of the capture of Ganelon. 53 O’Sullivan notes the ‘minstrel influence’ evident in a similar interpolation in Firumbras, ed. O’Sullivan, 54–5. 54 These are: ‘The bataille of the visoures’; ‘How þat Charlis wente to Seinte Iames and made Compostilla a see’; ‘The discripcion of King Charles persone’; ‘The treson of Ganalon and þe batail of Rounsivale’; ‘The paine of Rowlande and þe deþe of Marcerie and fliȝte of Bigalande’. The rest of this text is lacking, so the remaining chapter headings are taken from the opening Tituli: ‘The sowne of Rowlondis horne, his confession and dethe’; ‘The maneris and the liberaliteis of gode Rowlonde’; ‘The vicion of Turpyne and lamentacion of Charlis for the deþ of Roulond’; ‘How the sonne stoode thre dayes and the slawȝtre of four thowsonde Sarzynes’; ‘How the dede bodies were saltted and spiside; … how Roulonde was beried with oþer many’. 55 See additional references to the knights as ‘crysten’ at (for example) lines 755, 918, 998, 1762, 2152, 2721.

374

Charlemagne in Medieval England

for help in battle, but the narrative leaves no doubt as to which is the true God: in added episodes, Garcy is shown progressively losing all faith in the power of his gods to help (1148–53, 1247–58, 1529–70),56 while Charles prays for continuous daylight until he has won the battle and is told by an angel that his prayer will be answered (2559–76), as it soon is, with repeated reference to the ‘grace that god hym sent’ (2661– 72).57 Indeed, Charles attributes all his previous victories over heathen kings to the grace of God (1289). The battle between the Christian Otuel and the Saracen Clarel has the potential in all versions to be read as a symbolic trial of their two faiths; here Otuel rides out ‘With the grace of god almyȝt’ (1372), and having wounded Clarel uniquely points out that his gods have not helped him (1485–7). This point is repeated by the narration when Clarel is killed – ‘tho men myȝt se that ys god Mahoun | was but of lytyl myȝt’ (1521–2) – thus turning the outcome of the battle into a general proof. Otuel in this version acts with the fervour of the newly converted: he calls his fellows to arms ‘In the name of the holy gost’ (1044) and proposes a new battle to be revenged ‘of godys sone’ (1068). Thus it comes as no surprise when Otuel reappears in the last battle of the composite romance to assist Charles and Turpin in their fight against the Saracens. This unequivocal presentation of Christian certainties throughout the text clearly lends itself to the agenda of reinforcing orthodox Christian beliefs and practices that can be seen to inform the compilation of the whole of the Fillingham MS. The last battle, in which Charlemagne slays every Saracen in revenge for the deaths at Roncevaux, combines with the execution of Ganelon and the burial of Roland and the peers to give the romance a comprehensive finale. However, the earlier battles lead up to three interim conclusions, each of which features the conversion of Charlemagne’s Saracen enemies: Otuel, Garcy, and the inhabitants of Cordoba. In every case the Fillingham version enhances the account in the source. Otuel not only renounces his gods but adopts a new life as a knight of Christ (576–7), and needs no further instruction: he is ‘that knyȝt | that couthe goddys lawe’ (594–5). The defeated Garcy 56 This is ultimately based on the conventional chanson de geste motif of the Saracen king abusing his idols. It is notable that a similar sequence of the Saracen’s progressive loss of faith occurs in the Sowdone of Babylone. 57 The miracle of the sun standing still in Pseudo-Turpin occurs without Charlemagne’s prayer.

The Insular Otinel Tradition

375

is neither imprisoned nor dies, but asks to become a Christian, and Turpin gives him baptism and instruction, to save his soul from hell (1671, 1680–5). At Cordoba, not only the Almansor but all the inhabitants convert, ‘and all were cristened in a thro, | with grete Ioye & solempnite’ (1854–5). Like the added christening of all Floripas’s maidens in the Fillingham Firumbras, these innovations indicate an interest in the idea of conversion that goes beyond the conventional motifs of the ‘noble Saracen’ and the ‘belle sarrasine’, and perhaps reflects something of the optimistic fantasy that informs the contemporary account of Christian and Saracen relations in the Beauchamp Pageant (c. 1483–4),58 where Richard Beauchamp, Earl of Warwick, is shown on a pilgrimage to the Holy Land. There, like ‘John Mandeville’, he enjoys the hospitality of a Saracen lord who has close knowledge of Christianity, and in this case has secretly converted: ‘in his hert, thowe he durst nat utter his concept, yet he feithfully beleved as we do, rehersyng by ordre the articles of our feith’ (pp. 88–9). However, the proximity of these scenes in the Pageant to the following sequence, showing Earl Richard’s involvement in putting down the Oldcastle revolt of 1414, produces a juxtaposition of Saracen conversion and English heresy that indicates the predominant anxiety in the late fifteenth-century context: ‘traiterous heretikes’ who purposed ‘to have destroied the Church of Englond’ (pp. 96–7). It may be the case that a reading of Otuel and Roland in the Fillingham MS could allow the idea of conversion to play into the agenda of the compilation: conversion from suspect heterodox or heretical views to acceptable orthodoxy. One innovative passage in the romance supports such a reading. When Otuel claims the right to fight Clarel, he reports verbatim their earlier dispute: Clarel’s subversive attack on Christian beliefs, and his own orthodox defence (1319–42).59 While this exchange is unlike any other version of Otinel, it has some resemblance to the extensive disputation between Roland and Ferracutus in the PseudoTurpin tradition, for instead of merely disparaging Christianity, Clarel denies the doctrines of the incarnation and redemption by declaring that Christ was born of a woman (so was merely mortal) and was crucified on account of his false sayings. Otuel responds by outlining the key articles of faith: the virgin birth, redemption, resurrection, 58 The Beauchamp Pageant, ed. Sinclair. 59 The passage is based on the version in the Auchinleck Otuel (1259–68), but alters and extends it.

376

Charlemagne in Medieval England

harrowing of hell, sending of the Holy Spirit and ascension. Clarel (like Ferracutus) is not converted; however, it is this proof of Otuel’s orthodoxy that persuades Charlemagne to let him take the fight as champion of the Christian faith: ‘Otuel, as tou sayst, it ys. | Go fyȝt in godys name!’ (1347–8). Diane Vincent points to unique additions in the Middle English Pseudo-Turpin that represent Roland undertaking the fight with Ferakutte ‘in þe name of God’ and ‘for the feythe of God and for þe defence þerof ’, and argues that, besides the widespread view of Islam as a species of heresy, several details in the text work to identify Ferakutte with English heretics, so that the disputation between Roland and the Saracen ‘could have reminded readers of contemporary exchanges between ecclesiastical authorities and those suspected of heresy’. 60 Though far less elaborate, the dispute in Otuel and Roland similarly has potential to evoke comparison between Clarel’s offensively rationalist opinions and the critical questioning of orthodox matters of faith associated with Lollard heretics. Thus Otuel becomes a symbolic representative of both the dream of Islam’s conversion to Christianity,61 and the fear of home-grown Christian heresy that lies behind the Fillingham compilation’s agenda of religious orthodoxy. Finally, readers may also have found Otuel and Roland resonated with the wider context of international politics. Its presence in the Fillingham MS considerably post-dates the defeat of the Turks at Belgrade in 1456 (which, as Jonathan Phillips argues, ‘did much to stem the Ottoman advance for the next fifty years at least),62 and in a period of preparations for campaigns in Spain, in which English forces played a part,63 leading to the conquest of Granada in 1492. In the romance, the stories of Saracens driven out of Lombardy and conquered in Spain are heavily pointed by narrative comments drawing attention to divine protection 60 Diane Vincent, ‘Reading a Christian–Saracen Debate in FifteenthCentury Middle English Charlemagne Romance: The Case of Turpines Story’, in The Exploitations of Medieval Romance, ed. Ashe et al., pp. 90–107 (p. 106). 61 The ‘dream of conversion’ is explored in John Victor Tolan, Sons of Ishmael: Muslims through European Eyes in the Middle Ages (Gainesville: University Press of Florida, 2008), pp. 66–7. 62 Phillips, Holy Warriors, p. 306. 63 Phillips notes the involvement of Henry VII, who sent men and ordered prayers throughout the land (ibid., p. 308).

The Insular Otinel Tradition

377

of the Christian forces, and this would have given contemporary readers the opportunity to connect the fictional accounts with their own hopes and prayers for the security of Christendom and for successful reconquest in Granada.

The Thornton Roland and Otuel While the Sege of Melayne has been repeatedly discussed in modern criticism, very little attention has been given to its accompanying text in the Thornton MS, ‘The Romance of Duke Rowlande and Sir Ottuell of Spayne’ (here abbreviated as Roland and Otuel).64 It has long been recognized that they belong together: their first editor, Sidney J. Herrtage, argued that the Sege of Melayne, possibly based on a lost original Anglo-Norman composition, was intended to form an introduction to the story of Otinel ‘in the same way as the Destruction of Rome is introductory to Fierabras’. 65 Analysis of the Thornton texts suggests that the copyist/compiler of the collection was fully aware of the connection, for particular care has been taken at the beginning of Roland and Otuel to emphasize the common geographical setting in the two romances. Lombardy is mentioned in Roland and Otuel earlier than in the French source, and in this version alone, Lombardy (instead of Rome) is made the site where the Saracens slaughtered Charles’s own Christian subjects. This creates a clear parallel with the Sege of Melayne, in that both texts now represent Lombardy as Christian territory under brutal attack from Saracen aggression, rather than, as in the French Otinel, a Saracen possession where the Sultan Garcy has his stronghold. In the first half of the fifteenth century this was a crucial difference: Western Christians had seen their Orthodox neighbours in Bulgaria and Serbia conquered in the rapid expansion of the Ottoman Empire, so that by the end of the fourteenth century Catholic Hungary was directly in the line of Muslim advance. Further details in Roland and Otuel feed a

64 Diane Speed discusses Roland and Otuel as a nostalgic fiction of ideal Christian chivalry in ‘Chivalric Perspectives in the Middle English Otuel Romances’, in Medieval Cultural Studies: Essays in Honour of Stephen Knight, ed. Ruth Evans et al. (Cardiff: University of Wales Press, 2006), pp. 213–24. 65 The Sege off Melayne [etc.], ed. Herrtage, p. x.

378

Charlemagne in Medieval England

reading of the romance in the light of Western Christian anxiety about potential invasion. A small but unique addition to the source in Roland and Otuel prepares the ground early in the narrative for this perspective. In the description of Charlemagne and his peers preparing for war, two lines explain their defence policy both at home and abroad: ‘for to kepe þe heythyn here, | & stryue there goddes Enymys’ (47–8). Only in this version is there any mention of resisting or repelling the heathen ‘here’ – at home, on our own territory.66 Not only does this pose a new double agenda with regard to the Saracens (fighting to destroy the false faith and also defending the Christian homeland); it also actually gives priority to the task of opposing the non-Christian invaders. A little later Otuel boasts to Charles of the vast extent of the Saracen sultan’s empire and, again uniquely in this text, stresses his continuing advance, laying waste everything in his path: ‘This noble kynge of grete pouste | He distruyes bothe londe & see, | reghte in to fermonye’ (214–16). Nearly all the place names in this passage are distorted, and it is not clear where ‘Fermonye’ might be, but the phrase ‘reghte in to fermonye’ clearly indicates somewhere far away from the sultan’s base and, importantly, a place he was probably not expected to reach. These details add immediacy and urgency to Charles’s words in the next episode, when he consults the peers in response to the sultan’s depredations: ‘Sir Garcy with his stronge powere | Distruyes my landes both ferre & nere, | Mi cites brekes & bristes’ (664–6). Charlemagne’s two-pronged defence policy is maintained with added details throughout: the campaign in Lombardy is proposed ‘For to distruye there godes Enemy’ (655, 684), but at the same time the peers’ advice to their king is couched in terms of defensive action: ‘to fende of, lorde, vs thynke it gude, | to batayle are we priste’ (671–2), where the sense of ‘fending off ’ implies repelling the enemy attack upon Charles’s own lands and people. Again, in the final episode, in a passage largely rewritten in Roland and Otuel, Charles responds to Garcy’s stereotypical vow not to eat or drink until ‘he struyed hade Cristen blode’ (1349) by considering what he can do with God’s help ‘For to kepe the heythyn laye’ (1354–6) – a turn of phrase that suggests an image of Christendom fighting to preserve itself from the invasive Saracen religion. The references to ‘Franceis’ and ‘douce France’ in Otinel have been 66 See OED, ‘Keep’, v., †6, †7; ‘Kep’, v., 1.

The Insular Otinel Tradition

379

almost entirely erased from the text in Roland and Otuel, producing a streamlined, single-issue narrative of Christian versus Saracen. Much has been made of the fact that the narration of the Sege of Melayne frequently refers to Charlemagne and his host as ‘oure’ king, our knights, our men; while not so numerous, the same usage occurs, uniquely among the Otuel texts, in Roland and Otuel. It does undoubtedly serve to involve and include the reader and listener in the concerns and allegiances of the narrative heroes: here, specific references to the identity of ‘our’ side classify the knights as Christian (rather than French), and the audience is thus directly enrolled in the struggle against the enemies of the faith. However, this is not the whole story. In Roland and Otuel the possessive ‘oure’ always occurs in the context of a retained or added reference to Charlemagne’s army fighting against Turkish forces, creating a sense of immediacy and involvement in the confrontation between the Christians and this particular subset of the Saracen armies. In the fifteenth century there would have been an obvious resonance with fears in Europe about the Ottoman Turks and the expansion of their empire; as Tyerman puts it, it was ‘a question of cultural and political survival against what appeared … an inexorable force’.67 Memories of the devastating defeats inflicted by the Turks on Christian coalition armies at Kosovo in 1389 and at Nicopolis in 1396 would have fuelled present fears when the Ottomans resumed their advance further into Europe in the fifteenth century, and the contemporary context gives a special prominence to one of the uniquely ‘Turkish’ episodes in Roland and Otuel (1450–64). This is a point in the fighting when it seems as if ‘oure’ army will be defeated: the Saracen king and his 30,000 Turkish foot-soldiers force them back so decisively and maintain the offensive so unremittingly that they ‘almoste wonn þe felde’ (1461); ‘oure’ army is saved only when reinforcements arrive in the nick of time. The threat of defeat is sustained so long that it seems designed to evoke a frisson of anxiety in the reader or listener, particularly if the repeated mention of the Turkish infantry were to spark a memory of the tactics by which the Christian forces were disastrously defeated at the Battle of Nicopolis. Kelly DeVries describes how the Turkish infantry played a crucial role, resisting the Franco-Burgundian cavalry charge again and again, until the rest of 67 Tyerman, God’s War, p. 837; on crusading post-Nicopolis, see also Housley, Contesting the Crusades, pp. 132–40.

380

Charlemagne in Medieval England

the Ottoman troops were able to defeat the exhausted knights with a concerted counter-attack.68 Thus in Roland and Otuel the eventual victory of Charlemagne’s army over the Turks and other Saracen forces, when Otuel, Roland, and Oger the Dane ride to the rescue, offers to the fifteenth-century audience an opportunity to reimagine the confrontation between the Ottoman Empire and European Christendom.69 By identifying with ‘oure folkes’ (1528), as they are presented in this final battle, fighting fiercely against 20,000 Turks, the reader or listener could map onto the victory of Charlemagne and his peers a satisfying reversal of the previous century’s humiliating defeats. Apart from these reinterpretations of the Otinel tradition to enhance its relationship with the Sege of Melayne and to reflect the topicality of its portrayal of Christian versus Saracen conflict, Roland and Otuel tends to adhere more closely than the other two Middle English Otinel romances to the inherited narrative. However, it does make a few alterations which serve to emphasize the role of Otuel as eponymous hero of the romance. It is Otuel (not Roland or the other peers, as in the French texts) who is presented as taking the initiative in the final defeat of Garcy: Otuel personally pursues and captures Garcy and sends him to Charles (1567– 75).70 This climactic scene is prepared for by a small innovation earlier in the text, for when Otuel embraces Christianity and accepts Belisent as his bride, he defers their wedding, uniquely, until ‘I hafe tane Myn Eme garcy’ (658).71 Roland and Otuel apparently follows the continental version of the final laisses, as the romance includes the wedding; however, in a change to the traditional ending, the wedding takes place in Paris, and the conclusion focuses on Otuel’s future life at Charlemagne’s court 68 Kelly DeVries, ‘The Battle of Nicopolis’, Medieval History Magazine  2 (October 2003), accessed online through http://deremilitari.org/ articles/ [accessed 23/08/2016]. See also Tyerman, God’s War, pp. 852–8. 69 This would have been a familiar strategy, for, as Matthew Gabriele has shown, the equation of Charlemagne’s empire with the whole of Christendom was an early feature in the development of his myth (An Empire of Memory). 70 In Fillingham Otuel it is Oliver who performs these actions; in Auchinleck Otuel the scene is rewritten. 71 The French texts and Fillingham Otuel refer to the death of Garsie at this point; Auchinleck Otuel refers to his being slain or taken; but only in Roland and Otuel does Otuel take the responsibility on himself.

The Insular Otinel Tradition

381

as the fellow of Roland and Oliver (rather than on his remaining to rule Garcy’s conquered lands while Charles and the peers return to France, as in Otinel).72 Again, this change is anticipated by parallel details in the betrothal scene, where Charlemagne uniquely promises fellowship with Roland and Oliver and invites Otuel to ‘Duelle & be a pere’ (636), so that the final stanza of the romance declares the exact achievement of all the hero’s rewards: ‘And þus he duellys & es a pere, | Rowlande felawe and Olyuere’ (1591–2). An already focused narrative structure in the short chanson de geste Otinel is here even more specifically designed as the romance of Otuel, where the ending of the first half of the story sets the agenda that Otuel must achieve in the second half. The Thornton text is exceptionally well provided with scribal incipit and explicit formulae, each naming it ‘Þe Romance of Duke Rowlande & of sir Ottuell of Spayne’ (fols 82, 92);73 revealingly, however, a final concluding line adds ‘Explycit sir Otuell’, drawing attention to Otuel’s role as the protagonist whose history shapes the narrative. This is the only one of the English versions to follow the Anglo-Norman text in its description of Otinel’s alienating appearance and hostile behaviour when he challenges Charlemagne’s barons, rolling his eyes, ruffling his moustache, and looking like a fierce lion;74 and this gives particular significance to his paradigmatic transformation when he is finally described not as a great king,75 but simply as ‘A gud Cristyn man’ (1593).

Insular Additions and Responses to the Otinel Tradition The existence of three entirely different Middle English Charlemagne romances, Roland and Vernagu, the Sege of Melayne and Rauf Coilyear, all apparently created as independent responses to the Matter of France

72 The chanson de geste Otinel is again modelling itself on Fierabras in this pattern of narrative closure. 73 Each rubric also adds a secondary reference to ‘Cherlls of Fraunce’ (fol. 82), ‘Charlles’ (fol. 92). 74 Roland and Otuel, lines 172–4. This passage from the Anglo-Norman version is printed by Guessard and Michelant (lines 119–21) to supply a lacuna in the continental Otinel. 75 See Otinel, ed. Guessard and Michelant, lines 2126–31.

382

Charlemagne in Medieval England

tradition and particularly to Otinel,76 testifies to the profound cultural impact of this narrative corpus in Britain. As the following discussions will show, all three, but especially the two ‘new’ fifteenth-century romances the Sege of Melayne and Rauf Coilyear, have thoroughly assimilated the characteristic memes of the insular Charlemagne tradition: not only the subject of ‘holy war’ and the repeated narrative developments on the theme of the noble or converted Saracen, but also the topos of the frustrated leader threatening his gods, for example, and the elaboration of narrative opportunities for instruction, both religious and chivalric. Further, these romances also replicate the insular tendency to reinterpret the Matter of France tradition in response to contemporary political concerns.

The Sege of Melayne The Sege of Melayne has been much discussed in romance criticism: variously characterized as ‘a true crusading poem’;77 a political romance focused on Christian unity;78 a character study of the battling Bishop Turpin;79 a parody of Matter of France epics;80 or an embodiment of aristocratic chivalric identity.81 Critics have been drawn to its vigorous narrative style and dramatic action; they have also been intrigued by the element of mystery in its unique status: is it a version of a lost French 76 Janet Cowen notes that ‘certain episodes’ in the Sege of Melayne ‘have a general resemblance to parts of the texts in the Ferumbras group, which also feature the temporary capture of a group of French knights and a siege’ (‘The English Charlemagne Romances’, p. 159). This general resemblance can probably be attributed to the text’s assimilation of the characteristic memes of the Matter of France, as discussed here. 77 Mehl, Middle English Romances, p. 154; Stephen H. A. Shepherd, ‘ “ This grete journee”: The Sege of Melayne’, in Romance in Medieval England, ed. Maldwyn Mills, Jennifer Fellowes and Carol M. Meale (Cambridge: D. S. Brewer, 1993), pp. 113–31. 78 Warm, ‘Identity, Narrative and Participation’. 79 Three Middle English Charlemagne Romances, ed. Lupack, p. 107. 80 Elizabeth Berlings, ‘The Sege of Melayne – A Comic Romance; or, How the French Screwed Up and “Oure Bretonns” Rescued Them’, in Boundaries in Medieval Romance, ed. Cartlidge, pp. 57–70. 81 Crofts and Rouse, ‘Middle English Popular Romance and National Identity’, pp. 94–5.

The Insular Otinel Tradition

383

Charlemagne text, or an original English one? As a unique text with no known source, it requires a different approach from the comparative analysis appropriate to the Otinel and Fierabras romances. While looking at the poem in its wider manuscript context in the Thornton MS indicates one way a fifteenth-century reader might engage with the Sege of Melayne,82 other ways are suggested by examining in detail its relation to the Otinel tradition and in particular to its companion romance, Roland and Otuel. As in Roland and Otuel, Lombardy is represented in the Sege of Melayne as a Christian land which the Saracens attack, though here it is ruled by the Christian lord of Milan, Sir Alantyne. The Sege of Melayne opens with the people of Lombardy complaining about the abuses committed by the heathens in what is presented as the last stage in an irresistible invasion (16–24). The narrative plots the advance of the sultan northwards through Italy: he has overthrown the authority of the pope and of secular rulers in Rome, colonized Tuscany with planted heathen communities, and now, with no one able to stop him, he has arrived at Lombardy. This was obviously a nightmare scenario. Jean Froissart used a similar imaginative projection in his Chronicles, to heighten his account of the response to the Ottoman threat that ended in the Battle of Nicopolis. Froissart states that the king of Hungary wrote to ask the French king’s aid against the Great Turk, Bajazet, claiming that Bajazet had threatened to come and fight the king of Hungary ‘ou mylieu de son pays et chevaucheroit si avant que il venroit à Romme et feroit son cheval mengier avoine sur l’autel Saint Pierre à Romme, et tenroit là son siége impérial’ (‘in the myddes of his [own] realme, and to go fro thens to the cytie of Rome, and wolde make his horse to eate otes vpon the high auter of Saynt Peter, and there to holde his see imperyall’).83 This last threat parallels the sultan’s subversion of the pope’s power in the Sege of Melayne, while the vivid image of desecration is matched in the romance by the sultan’s tearing down Rood crosses and statues of the Virgin in churches and abbeys, where 82 See Chapter 3, pp. 187–92; Phillipa Hardman, ‘The Sege of Melayne: A Fifteenth-Century Reading’, in Tradition and Transformation in Medieval Romance, ed. Field, pp. 71–86. 83 Oeuvres de Froissart, ed. M le baron Kervyn de Lettenhove, 25 vols (Brussels: Devaux, 1867–77), vol. 15: Chroniques, 1392–1396 (1872), pp. 216–17; English translation: Chronicles, trans. Berners, II, 636.

384

Charlemagne in Medieval England

he set up ‘Mawmettes’ in their place (the idols that Muslims were widely believed to adore as gods). Both texts reflect and feed the same fear of cultural and political obliteration. Froissart frames the king of Hungary’s plea for aid in just these terms: ‘affin que sainte crestienneté ne fuist foulée, ne violée’ (p. 217), and represents Bajazet gathering ‘tous les seigneurs de sa loy’ in order to ‘destruire la sainte crestienneté’ (p. 310). Froissart also saw the parallel between the traditions of the Matter of France and the contemporary situation. Writing about the Ottoman defeat of the Christian coalition forces at Nicopolis, caused as he sees it by the disastrous pride of the French contingent in refusing to wait for the rest of the army before attacking the Turks, he makes an epic claim for its effect and significance: Or regardés la grant folie et le grant oultrage; car, se ils euissent attendu le roy de Honguerie et les Hongres où bien avoit soixante mil hommes, ils euissent fait ung grant fait; et par euls et par leur fole oultre-cuidance et orgueil fut toute la perte, et le dommage que ils recheuprent, si grant que depuis la bataille de Ronchevauls où les douze pers de France furent mors et desconfis, ne receuprent si grant dommage. (ed. Kervyn de Lettenhove, XV, 315–16) Lo, beholde the great foly and outrage, for if they had taryed for the kynge of Hungery, who were threscore thousande men, they had been lykely to haue doone a great acte: and by them and by their pride all was lost, and they receyued suche dommage, that sythe the batayle of Rounseualx, where as the xii. peres of Fraunce were slayne, crystendome receyued nat so great a dommage. (trans. Berners, II, 669) Clearly, at the turn of the fourteenth century the Matter of France could be used to articulate and explore the most pressing anxiety about the future of the Christian world. The exploration of anxiety is unusually frank in the Sege of Melayne. Faced with the sultan’s ultimatum – to convert to Saracen beliefs and keep all his lands, or be killed with his entire family – Alantyne takes twenty-four lines to consider and pray about these alternative choices. He recognizes that by forsaking his faith he would imperil his soul, but

The Insular Otinel Tradition

385

still asks God’s advice, ‘Whethire þat me es better to doo, | The hethyn lawe to torne too, | Or my lyfe in lande to tyne’ (82–4). It is not hard to see fears for the survival of Christendom behind this apparently personal dilemma. The same anxiety is more extensively and dramatically played out in Charlemagne’s dilemma as to which of the two objectives named in Roland and Otuel he should pursue: to stay at home and protect his own kingdom, or to attack the enemies of Christendom abroad? The alternatives are caricatured in the opposed views of the traitor Ganelon, on the one hand, who counsels staying at home and, if necessary, even submitting to the sultan in order to retain his kingdom (181–4, 595–600, 649–57), and on the other, the enthusiast Turpin, who denounces Charles as a heretic, God’s enemy, worse than a Saracen, a lost soul, an apostate, and the potential destroyer of Christendom, for listening to Ganelon’s treacherous advice instead of responding immediately to the Lombards’ plea for help (687–708). Once Charles has confessed himself in the wrong and joined Turpin in Lombardy, an unusual scene shows him fighting the Saracen Darnadowse, having refused Roland’s typical offer to fight on his behalf with this strikingly personal explanation: ‘Sen ilke a man feghtis for his saule | I sall for myn do mede’ (1016–17). In the course of the fight, Darnadowse tries to convert him, but Charles is resolute: ‘Neuer sall I forsake my lay’ (1058). This incident, which in an important sense is the turning point of the narrative, seems designed to ‘correct’ the dangerous alternatives explored in Alantyne’s prayer and Ganelon’s counsel with a model of renewed commitment to Christian identity as the means to salvation, on both a personal and a political level. Charles’s references to his ‘saule’ and his ‘lay’ explicitly answer Turpin’s earlier denunciation of him as a ‘saule for lorne’ (705) who ‘forsuke thi laye’ (707). Not only does Charles win spiritual merit by resisting the temptation offered by Darnadowse, but he also wins the defeated Saracen’s sword, which is presented as a token of future temporal victories beyond the limits of his own kingdom: ‘And thare he wane þe saraȝene swerde, | And certis þat with one the erthe | He conquered many a lande’ (1087–9). True to the insular pattern, this crucial resolution episode takes the form of a one-to-one confrontation between Charlemagne and a Saracen challenger. Darnadowse is the epitome of the ‘noble Saracen’, chivalrous, eager to win renown, generous in his admiration of Charles, and courteous in battle: he discards his own sword when he sees his opponent’s is broken. It thus comes as a shock that once Charles has struck him dead, he is referred

386

Charlemagne in Medieval England

to simply as ‘this fend’, and his dead body is butchered by Charles: ‘He broches hym so boldely | That his hert blode sekerly | rane to oure kynges hande’ (1084–6). Even more shocking, Charles does this ‘with his awenn brande’ (1083) – the very sword that Darnadowse had courteously cast away in the fight. The suddenness of this switch in register is echoed in Roland and Otuel, where the noble Saracen king Clarel, once slain by Otuel, is summarily consigned to hell: ‘His saule went vnto Mahoun | than by those gates gayne’ (1340–1).84 In both cases the tone underlines the didactic significance of the single combat between Christian champion and chivalrous Saracen in an extreme form. It shows that the only issue that differentiates ‘us’ in the text from ‘the Other’ is faith, but that this is a total difference: salvation or damnation. According to the beliefs informing the text, if Darnadowse dies without baptism, no matter how noble, courteous or praiseworthy his behaviour, he is damned, and no different from a fiend.85 Moreover, as Turpin vehemently points out to Charlemagne, the same fate awaits him and all Christendom if he makes the wrong choice. Turpin’s unusual outspokenness and violent behaviour in this romance have been the focus of much critical discussion, particularly his abusive outburst directed at the Virgin Mary,86 which has been read as ‘blasphemous’, presenting Turpin as indistinguishable from a ‘pagan’.87 84 In the French tradition Clarel curses Mahomet as he dies: a motif reinterpreted in the Fillingham version as proof of Mahoun’s inefficacy (1521–2). 85 Similarly, the account of Garcy’s coronation as emperor (829–76) elaborates the luxury of the occasion and the rich gifts in familiar courtly terms, but undermines all with a final instance of the sultan’s ‘haythen’ custom, claiming that Saracen chivalry is so outweighed by lechery ‘Þat þay myghte noghte wele spede’ (876). 86 Turpin’s dedication to the Virgin Mary (733, 1044) is echoed in a unique detail in Roland and Otuel, in which the newly converted Otuel makes a vow to ‘mylde Marie | that I hafe now chosen to my lady, | þat es so mylde of mode’ to ‘distruye þe heythyn blode’ in Lombardy (643–8). 87 Crofts and Rouse, ‘Middle English Popular Romance and National Identity’, pp. 91–2. This article provides a useful summary of the views of Mills, Shepherd, Warm, and Akbari on this issue, but despite invoking the ‘Middle English popular-romance context’ as the appropriate interpretive framework, fails to look beyond the obvious parallel with the

The Insular Otinel Tradition

387

It is important to note that as instances of ‘saint-abuse’ were a normal occurrence within medieval culture,88 aggressive behaviour of this kind would not have branded Turpin a pagan for contemporary readers and listeners. However, equally important is the evidence that such scenes provide of the insular habit of reinterpreting characteristic Matter of France narrative memes. We have already seen how the conventional epic motif of the defeated sultan abusing his gods, which is a prominent feature in the Fierabras tradition, was reworked within that tradition to represent the depth of Charlemagne’s anguish in moments of crisis – recall his threat to empty all the abbeys in France and destroy altars and crucifixes if Mary does not protect Oliver in his fight against Fierabras (Fierabras, ed. Le Person, 921–6; see Chapter 2), and his threat to demolish God’s altars in France and abandon the holy relics if God permits Oliver’s death (Fierabras, ed. le Person, 1213–19). Naimes, like the sultan’s counsellor, persuades Charlemagne to cease his threats and pray to God (Fierabras, ed. Le Person, 927–30). In the Fillingham Firumbras a new scene on the same model is created when Charles later laments the fate of his knights in Balan’s tower, and swears that if they are lost, he will eradicate the practice of Christianity in France, and will never worship God: y ne schal suffre in fraunce no bellys ryng, In Chyrch ne in chapel no prest mas to synge, ne in none oþer plas halywater to spryng, No non bokes, godys name worchipyng. The vygours and þe autar þat in holy chyrch beȝth found, I schal hem adoun falle and bete to þe grounde. Be my dussepers ouercom, to god a gyfte y ȝeue, Ne schal I neuer worschyp god whyles þat y leue. (1115–22)89

enraged Balan in the Sowdone of Babylone and reiterates Mills’s view that Turpin is being ‘presented as a pagan’ (p. 92, n. 50). 88 Suzanne Conklin Akbari, ‘Incorporation in the Siege of Melayne’, in Pulp Fictions of Medieval England, ed. McDonald, pp. 22–44 (p. 27); Hardman, ‘The Sege of Melayne’, pp. 81–2. 89 In this instance, Firumbras takes the role of the counsellor reasoning with the distraught king (1123–8).

388

Charlemagne in Medieval England

The specificity of Charles’s oath accords with the preference for circumstantial detail seen in all the Middle English rewritings of chansons de geste, but here it works to create a stronger frisson of anxiety for contemporary readers and listeners. They would hear a dangerous echo of the fears of cultural obliteration staged in Froissart’s imagined scenario of desecration in the heart of Christendom, and in the Sege of Melayne when Arabas destroys the Rood figures of Christ and Mary that belong in every church and abbey as signifiers of their Christian identity (25–30). The Fillingham Firumbras is not presenting Charlemagne as a pagan; it is building on the chanson de geste motif, where the narrative meme of the Saracen ruler abusing his gods has been adapted to put a similar spotlight on the extreme emotion of the Christian leader, thus intensifying and sustaining the structural parallel between them. By adding a new scene of this kind, Firumbras heightens the narrative tension at a crucial point, just as Charlemagne is about to come to the aid of the peers; but it also uses the language of repudiation to register what is felt to be at risk of happening if this conflict between Christian and Saracen forces should come to the worst: a systematic demolition of Christian culture. In this context, the scene in which Turpin passionately rebukes the Virgin Mary in the Sege of Melayne may be read as part of the same tradition of reinterpretation. Like Balan in Fierabras, blaming Mahoun for the loss of his men at Mautryble,90 Turpin blames Mary for the death of the Christian knights who were fighting under her protection: A! Mary mylde, whare was thi myght, Þat þou lete thi men thus to dede be dighte, Þat wighte & worthy were? Art þou noghte halden of myghtis moste, Full Conceyuede of þe holy goste? Me ferlys of thy fare. Had þou noghte, Marye, ȝitt bene borne Ne had noghte oure gud men thus bene lorne Þe wyte is all in the. Thay faughte holly in thy ryghte, Þat þus with dole to dede es dyghte,

90 The Ashmole Sir Ferumbras gives a close translation of the Vulgate Fierabras at this point (4923–50).

The Insular Otinel Tradition

389

A! Marie, how may this bee? (547–58)

Unlike Balan, however, Turpin does not deny God or renounce his faith, or even turn his back on Mary, but he continues his emotional complaint: wringing his hands, he ‘flote with Marye euer amange | For þe losse of oure menȝee’ (563–4).91 Turpin is not exhibiting a loss of faith, or even a wavering faith;92 it is his belief in Mary’s power (and especially in the power of her Immaculate Conception)93 that fuels his complaint at her failure to exercise it, and as soon as his crisis of emotion has passed, Turpin again calls on Mary (606, 733, 900, 903, 1043). The point of this reinterpretation of the Saracenabuse-of-gods trope seems to be to provide (as for Charlemagne in Fierabras) externalized evidence of extreme grief, and Turpin’s abuse of Charlemagne for having become ‘ane Eretyke’ (672) by refusing to wield his God-given sword against the Saracens (745–50) can be read as a further reworking of the motif. Turpin excommunicates Charles with an accusation in the same terms as his rebuke to Mary: ‘If Cristyndome loste bee | Þe wyte bese casten one the, | Allas þat þou was borne!’ (697–9), and Naymes persuades Charles to be reconciled to Turpin (772–81) in an exchange patterned on the same characters’ roles in Fierabras. All this indicates how thoroughly the narrative meme has been assimilated in the insular tradition. Both Thornton romances use the Matter of France with an eye to the dangers, political and spiritual, posed by the presence of the Ottoman Turks uncomfortably close to home, and as has been shown, both involve readers and listeners with the Christian cause by the systematic 91 The OED cites this instance of the verb flite, meaning ‘to pray in the language of complaint, or remonstrance’. 92 Caxton’s brief comment on Charlemagne as ‘Inpacyent of a perfayte fayth’ when he makes his threatening prayers is not applicable to Turpin (Charles the Grete, ed. Herrtage, p. 65, translating Jean Bagnyon’s Fierabras). 93 The phrase used in line 541 to refer to Mary’s conception without original sin is patterned on the expression used in the Apostles’ Creed for Christ’s incarnation: ‘conceptus est de Spiritu Sancto’. For further contextual documentation of this issue in the Sege of Melayne, see Hardman, ‘The Sege of Melayne’, p. 79.

390

Charlemagne in Medieval England

use of the first-person plural pronoun to refer to ‘oure kynge’ and so on. It is certainly a remarkable usage in the Sege of Melayne: there are forty-one instances of this first-person possessive, and another four references to ‘us’ and ‘we’,94 including this conscription of the reader’s or listener’s response to the death of ‘many a saraȝene’ (1095): We may thanke gode þat is in heuen Þat lent vs myghte & mayne. (1100–1)

However, unlike Roland and Otuel, the emphatic identification with ‘oure folke’ (1201) does not exclusively indicate ‘the Christen folke’ (1088): it can also mean ‘franche folke’ (1063), ‘folke of Fraunce’ (1296); and Charles, King of France, and his lords are all identified by their bearing the arms of France, the fleur-de-lys (94, 281), as Charlemagne is shown in MS Egerton 3028. The making of the Thornton MS is contemporary with the final phase of the Hundred Years’ War, occupying time between the victories of Henry V (commemorated in a lyric in the manuscript)95 and the eventual loss of English possessions in France by the end of the conflict. It is therefore worth remembering both that Edward III incorporated the arms of France into his own royal arms of England, to signify his claim to the French throne, at the outbreak of hostilities;96 and that when the young King Henry VI of England was crowned king of France in 1430, his coinage bore the arms of France as well as of England.97 Thus we can perhaps read the unproblematic inclusion of these French references in the Sege of Melayne as a parallel annexation of French heroic tradition: the king of France is now ‘oure kynge’, and French peers bearing the fleur-de-lys are ‘oure folke’. In this, as in its confident reuse of the epic meme of the Saracen abusing his gods, the Sege of Melayne shows how thoroughly the Matter of France was appropriated into Middle English popular romance. 94 Pace Crofts and Rouse, who count twenty-four instances of these pronouns (‘Middle English Popular Romance and National Identity’, p. 90). 95 ‘The Rose of Rys’, fol. 110v. 96 See Chapter 2, pp. 141–4. 97 Thomas Woodcock and John M. Robinson, The Oxford Guide to Heraldry (Oxford: Oxford University Press, 1990), p. 188.

The Insular Otinel Tradition

391

Roland and Vernagu Like the Sege of Melayne, the romance of Roland and Vernagu has no extant immediate source in the French-language Matter of France tradition, but can equally be considered in relation to the Otinel tradition, as a composite narrative produced to function as a prequel to the story of Otinel’s challenge and conversion. The chief sourcetext for Roland and Vernagu is the same Johannes Translation of the Pseudo-Turpin Chronicle that forms the basis of the ‘batayle grym | Off charlys and of Ebrayn’ and the ‘Rewful tale | How Rowlond deyde at rouncyuale’ in Otuel and Roland (i.e. lines 1698–end), and which the author of the Song of Roland used to augment his narrative.98 The Middle English Roland and Vernagu follows its chronicle source in outline, but there are important changes, both at the structural level and in details, that can give us clues to its particular use of the material. Structurally, the romance consists of two separate sequences taken from the source and spliced together. This produces first a brief narration of Charlemagne’s coming to the aid of the exiled patriarch of Jerusalem and the emperor of Constantinople, with a fuller account of his conquest of the whole of Spain, followed by a lengthy one-to-one combat between the Saracen challenger, Vernagu, and Roland, Charlemagne’s champion and spokesman for the Christian faith.99 Thus the ‘historical’ victories are effectively restaged in quasi-symbolic mode in the one-to-one episode: the favourite insular representation of Christian–Saracen encounter. Unique details underline the fact that the action of Roland and Vernagu is constructed in terms of an archetypal Christian-versusSaracen conflict. In the opening lines, the register of Charles’s territorial possessions and titles is extended to include the specific role ‘lord of al christendome’ (15), while in other small but telling additions the peers are described as ‘þe cristen’ (554), and the defeated Saracen giant 98 For discussion of the scholarly theories surrounding the relation of Roland and Vernagu, Otuel and Roland and the Johannes Pseudo-Turpin, see Chapter 1, pp. 76–9. 99 The chronicle chapters that are omitted from the romance include the Agolond narrative, with its dispute between Charlemagne and the Saracen king for the right to Spain, but also its representation of flawed Christian values, which was perhaps not in line with the agenda of Roland and Vernagu. Some details of Charles’s dispute with Agolond are transferred to Roland’s dispute with Vernagu.

392

Charlemagne in Medieval England

Vernagu prays to his gods for revenge on Roland, ‘þis cristen hounde’ (856). Similarly, Vernagu himself is identified as ‘þe painim’, while his geographical origin in Syria and his kinship with Goliath, as given in the chronicle source, are erased, leaving him open to interpretation as representing the Muslim enemies of Christendom anywhere. Given the much more detailed account of Charles’s conquest of Spain, it is significant that in this romance the Saracen king who exiled the patriarch of Jerusalem is uniquely identified as the king of Spain, and that Vernagu, sent by the sultan of Babylon, taunts Charles for the weakness of his knights by referring with scorn to his success in Spain: ‘Sir, þo þou won Spain, | Hadestow non better þo?’ (513–14). Spain is thus established as the geographical reference point of the whole composite romance. The setting in Spain is particularly appropriate here, for the prolonged disputation in which Roland explains Christian doctrine is extended, uniquely, by his explicit attempt to convert the Saracen giant and, as Geraldine Heng notes, ‘Historically it is Spain, not the Holy Land, that is the locus of conversion efforts’. 100 In passages added to the Pseudo-Turpin material, Roland and Vernagu begin their dispute with unusual friendliness: Vernagu is grateful for Roland’s courteous action in placing a stone as pillow for his head, and offers friendship (641–9); Roland begins the questioning between them in response to his love (653); Vernagu questions Roland with a polite prayer, ‘Al so þi god þe spede’ (666);101 and as he offers his last explanation, Roland exhorts Vernagu: ‘O Vernagu, vnderstand, | Herken now to me’ (774–5). Roland’s words reflect the optimistic view in the twelfth and thirteenth centuries of the chance of converting Muslims to Christianity: twelfth-century popes encouraged conversion of the Muslim population of Spain, and the Spanish Order of Santiago was established with the explicit aim ‘to lead the Saracens to the practice of Christianity’, as stated in its rule.102 Yet, despite such encouragements, as Christopher Tyerman observes, ‘the few attempts at conversion 100 Heng, Empire of Magic, p. 343. 101 See Hardman, ‘Popular Romances and Young Readers’, pp. 157–8. These additions contribute to the chivalric agenda; see, too, the unique reference to knightly shame (506–8). 102 Heng, Empire of Magic, p. 343, citing Benjamin Z. Kedar, Crusade and Mission: Eastern Approaches toward the Muslims (Princeton, NJ: Princeton University Press, 1984), p. 60.

The Insular Otinel Tradition

393

amounted to little’, and the successes of the Reconquista, which by the mid-thirteenth century had reduced Islamic Spain to the Kingdom of Granada, were the result not ‘of conversion, but of conquest and, in places, expulsion.’ 103 Indeed, the very idea that rational debate might convert a Muslim to Christianity was beginning to lose credit. John V. Tolan points to the view of the Florentine friar Riccoldo da Montecroce, whose widely read, early fourteenth-century polemic Contra legem Sarracenorum asserts that Saracens are ‘impervious to reason and can only be conquered by force’.104 This powerful stereotype seems to inform the striking added scene in Roland and Vernagu where an angel assures Roland that it is his duty to fight and kill the Saracen giant, for all the preachers in the world would never succeed in turning him into a Christian (806–14). The whole exchange between Roland and Vernagu in a sense mirrors the tension between the idealistic dream of conversion and the actual experience of the Reconquista. Charlemagne’s legendary conquest of all the cities of Spain and Roland’s exemplary defeat of the irredeemable Saracen could thus be read in this fourteenth-century romance as fictional counterparts of the Christian successes in Muslim Spain in the previous century; and this mapping would promote a consciousness of divinely sanctioned Christian supremacy, filtered through a world imagined in terms of moral polar opposites. For despite the fact that, as in the chansons de geste, the opposed identities ‘heþene’ and ‘cristen’ are not simply equivalent to ‘bad’ and ‘good’ in all senses (Charlemagne, the king of Spain, and Vernagu are all equally characterized as ‘stern’, ‘stout’, and ‘douȝti’ knights), the Saracens in Roland and Vernagu are unequivocally cast as guilty aggressors, and therefore in the wrong (23–34), with the angel’s message providing final divine judgement (814), whereas Charles and Roland are explicitly acting in defence of God’s ‘riȝt’ (802). This moral polarization resembles the representation of Christians and Saracens in Otuel, the romance that follows Roland and Vernagu in 103 Tyerman, God’s War, pp. 670, 659; see also Phillips, Holy Warriors, pp. 307–9. 104 Tolan, Saracens, p. 254. For an analysis of the friar’s more ambivalent account of Islam elsewhere in his writings, see Rita George-Tvrtkovic, A Christian Pilgrim in Medieval Iraq: Riccoldo da Montecroce’s Encounter with Islam (Turnhout: Brepols, 2013).

394

Charlemagne in Medieval England

the Auchinleck MS. Both narratives seem to have been adapted not only to suit the typical expectations of Middle English romance, but also to provide parables of ethical action, and the order in which they appear is significant. While Otuel completes the story of Roland and Vernagu on the literal level by staging Otuel’s revenge for the death of his uncle Vernagu, at the same time on a parabolic level it ‘corrects’ the story, replacing the model of Christian overcoming Saracen in mortal combat with another of Christian–Saracen reconciliation through conversion. This ‘correction’ works together with the clear educational agenda that can be seen in both Auchinleck romances: promoting chivalric values of honourable conduct between knights, and especially towards enemies.

The Taill of Rauf Coilyear The Taill of Rauf Coilyear is included in this discussion of insular responses to the Otinel tradition because, as Diane Speed has argued, while the Scottish text is obviously indebted to the meme of the converted Saracen as found in both the Fierabras and Otinel traditions, it shows a more sustained debt to the Otinel narrative.105 Like The Sege of Melayne, this eccentric Charlemagne romance has proved more attractive than most to modern editors and critics.106 Part of its appeal seems to lie in its difference from the other Matter of France texts and its apparent amenability to interpretation as critiquing or subverting the tradition. W. R. J. Barron, for example, sees in the romance ‘the conventions of chivalric combat burlesqued’ (p. 228), while Diane Vincent describes its ‘satirical treatment’ of the Saracen conversion motif, and states of the Roland/Magog exchange: ‘The whole dialogue is clearly parodic.’107 Others find critiques of social conventions: Dieter 105 Otinel ‘affords the greater number of parallels’ (Speed, Medieval English Romances, I, 198). 106 Herrtage’s edition (EETS) was followed by: F. J. Amours (Edinburgh: Scottish Text Society, 1897); William Hand Browne (Baltimore: Johns Hopkins Press, 1903); Priscilla Bawcutt and Felicity Riddy (Edinburgh: Scottish Academic Press, 1987); Elizabeth Walsh (New York: Lang, 1989); Alan Lupack (Kalamazoo, MI: Medieval Institute Publications, 1990); Diane Speed (Durham: Durham Medieval Texts, 1993). 107 Vincent, ‘Reading a Christian–Saracen Debate’, p. 103. Glenn Wright discusses readings of the scene as parody and notes the lack of ridicule in its treatment of the material (‘Convention and Conversion: The Saracen

The Insular Otinel Tradition

395

Mehl interprets it as a criticism of the artificiality of courtly ceremony ‘because it is not always justified by a higher moral standard’ (p. 102); Nancy Mason Bradbury reads Rauf as a voice of social protest, asserting ‘truths and perspectives often lacking in chivalric romance’;108 and Alan Lupack finds in the conversion scene ‘a satiric comment on Roland’s failure in religious awareness’. 109 However, as Douglas Gray judiciously notes, the romance ‘does not end with a moralitas’, and, pointing out the author’s ‘even-handed’ treatment of ‘gentils’ and ‘commounis’, he concludes: ‘It is difficult to see “Sir Rauf ” carrying a banner for democracy or even for the upwardly mobile newly rich. The poem has fun with the mores of the chivalric romance and its heroes, but it does not seem to be an attack on them.’110 Indeed, Rauf Coilyear’s sophisticated manipulation of literary and cultural conventions resists any attempt at a simple moralizing interpretation, but its combination of two separate and entirely different stories, the ‘king incognito’ folk tale and the ‘additional story line’ of Christian–Saracen combat, poses a question that has long puzzled readers. As Ad Putter puts it: ‘It is far from obvious why the poet intertwined [the two].’ 111 The story of the King and the Collier has strong folk-tale connections, with analogues from cultures across the world, though the closest is a text several times mentioned by medieval writers together with

Ending of the Taill of Rauf Coilyear’, Al-Masaq: Islam and the Medieval Mediterranean 14.2 (2002), 101–12). 108 Nancy Mason Bradbury, ‘Representations of Peasant Speech: Some Literary and Social Contexts for The Taill of Rauf Coilyear’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 19–33 (p. 32). 109 Lupack, Three Middle English Charlemagne Romances, TEAMS online edition, introduction. 110 Douglas Gray, Later Medieval English Literature (Oxford: Oxford University Press, 2008), p. 520. Similarly, Diane Speed concludes: ‘It is difficult to take the serious moments too seriously, and perhaps we are not meant to. The human condition at large is presented as a wonderful mixture of the noble and the ridiculous’ (Medieval English Romances, I, 202). 111 Ad Putter, ‘Ralph the Collier’, in Heroes and Anti-Heroes, ed. Cartlidge, pp. 145–58 (p. 147).

396

Charlemagne in Medieval England

Rauf Coilyear: John the Reeve, preserved in the Percy Folio MS.112 The most notable difference between the tale type in Rauf Coilyear and its insular analogues is the identity of the king: all other instances feature an English king (Edward I in John the Reeve) rather than Charlemagne, and set the action in England, not France. The other major difference is that the analogues conclude with the knighting of the protagonist: only in Rauf Coilyear does the newly knighted churl proceed to act out his new status in chivalric combat. Putter argues that the action of the whole two-part romance is designed to allow Rauf, like a hero of romance, to demonstrate specifically knightly virtues: largesse through his hospitality and loyalty through keeping his promises.113 However, this does not entirely answer Putter’s original question: we might still ask why a Matter of France-type narrative was chosen for the ‘additional story line’ rather than any other suitable story of challenge? The answer may be found by taking a different approach to the puzzle: instead of looking at the Christian–Saracen encounter as a chivalric addition to the folk-tale story that, as Diane Speed puts it, ‘makes it necessary to identify the king as Charlemagne and set the entire action in France’, 114 we might rather see the whole text as a romance of Charlemagne that incorporates an adapted folk-tale motif into its narrative. This approach seems to fit better with the way the text unfolds. Not only is the Carolingian setting carefully established and maintained from the beginning,115 but the opening stanzas resemble those of two chivalric romances featuring the Arthurian court: The Awntyrs of Arthur at the Terne Wathelyne and Golagrus and Gawain. All three combine 112 For detailed discussion, see Speed, Medieval English Romances, I, 198–201; Rachel Snell, ‘The Undercover King: Rauf Coilyear and its English Analogues’, in Medieval Insular Romance: Translation and Innovation, ed. Judith Weiss, Jennifer Fellows and Morgan Dickson (Cambridge: D. S. Brewer, 2000), pp. 133–54. 113 Putter, ‘Ralph the Collier’, p. 158. On the other hand, Rauf at court is not a heroic figure: quickly fearful when he sees Wymound/Charles (709– 10), abjectly regretting his actions (719–25), quaking at the king’s recital (732–3). 114 Speed, Medieval English Romances, I, 201. 115 References are made to: Charles (1), Paris (5), the ‘douze peres’ (10), Roland and Oliver (326), Bishop Turpin (343), Saint-Denis church (352), the fleur-de-lys (668).

The Insular Otinel Tradition

397

two separate stories within the structure of the romance, and use the same alliterative verse form of thirteen-line rhymed stanzas. In all three, the great king is on a journey, attended by a splendid royal and noble company (in Golagrus and Gawain the destination is the Holy Land, in Rauf Coilyear it is Paris, and in the Awntyrs of Arthur they are travelling to Carlisle for the hunting). In each case, the king’s progress is interrupted by terrible weather or inhospitable terrain, an unexpected encounter ensues that provokes discussion of chivalric values, and a further encounter with a hostile opponent is resolved by single combat and final reconciliation.116 The poems imply a common model of chivalric narrative, equally available for use with the Matter of Britain and the Matter of France. The ‘holy war meme’ that Crofts and Rouse identify as marking Matter of France romances (p. 88) is evoked when Charlemagne explains why Rauf should not be hung: I hald the counsall full evill that Cristin man slais, For I had myster to have ma, And not to distroy tha That war worthie to ga To fecht on Goddis fais! (747–51)

Fighting God’s foes is the proper expression of knightly values in Matter of France texts, so it is quite predictable that the climactic encounter in Rauf Coilyear should be between Christian and Saracen opponents; and given the insular concentration on Fierabras and Otinel, it is not surprising that (as has often been observed) the encounter has strong echoes of the combat scenes in those two favourite chansons de geste.117 In the second part of the romance, the action springs from Rauf ’s determination to honour his pledge to meet with Roland on the moor.

116 Stephen Shepherd notes the formal similarity between these romances and compares the ‘defiant aboriginality’ of their challenger figures: Rauf, Golagros, and Galeron of Galloway, in ‘ “Heathenic Catechesis” and the Source of Awntyrs B’, Medium Ævum 81 (2012), 1–17 (p. 7). 117 Speed, Medieval English Romances, I, 197–8; Metlitzki, The Matter of Araby, p. 120.

398

Charlemagne in Medieval England

As Rauf waits for Roland to keep his appointed time , he sees a knight approaching: He sa cummand in thra The maist man of all tha That ever he had sene. (801–3)

Rauf takes this knight for Roland, but the reader can easily identify him as a Saracen not only by his riding a camel, but also by his resemblance to Fierabras, famous for his giant stature. Once he knows his identity, Rauf responds with typically partisan heroic fervour: Thow sayis thow art ane Sarasine? Now thankit be Drichtine That ane of us sall never hine, Undeid in this place. (852–5)

When the Saracen and Rauf (whose horses, like Otinel’s and Roland’s, were killed in their first course) have fought to a standstill, Roland arrives and intervenes (865–8), and in the encounter between the Saracen and Roland the parallels with Otinel are particularly striking.118 The Saracen reveals that, like Otinel, he comes as a hostile messenger from a Saracen king, both to avenge his slain kin (‘Thow slane hes oft thyself of my counsingis, | Soudanis and sib men that the with schame socht’ (897–8)), and to challenge Charles: We sall spuilye yow dispittously at the nixt springis, Mak yow biggingis full bair – bodword have I brocht – Chace Charlis your king fer out of France; Fra the Chane of Tartarie, At him this message wald I be, To tell him as I have tauld the, Withoutin plesance. (900–7)

118 Shepherd argues for a similarly marked set of parallels between Otinel and the second part of The Awntyrs of Arthur (‘Heathenic Catechesis’, pp. 13–14).

The Insular Otinel Tradition

399

He boasts of being a habitual slayer of Christians, again echoing Otinel and Fierabras: ‘Sa thrive I,’ said the Sarasine, ‘to threip is my thocht, Quha waitis the Cristin with cair, my cusingis ar thay; My name is Magog, in will, and I mocht, To ding thame doun dourly that ever war in my way.’ (912–15)

Nevertheless, Roland seeks to convert Magog (874–7, 888–94), with the offer of a rich wife (921–33) as in Otinel. Like Otinel, the Saracen at first refuses, then unexpectedly converts, to Roland’s great joy (934–48). The newly converted Saracen, exactly like Otinel, joins with Roland in a new fellowship of three knights, where Rauf takes the place of Oliver: Thay swoir on thair swordis swyftlie all thre And conservit thame freindis to thair lyfis end, Ever in all travell to leif and to die. (949–51)

Finally, Magog is christened ‘Gawtere’ (echoing the renaming of Fierabras as ‘Florent’) and, like Otinel, marries the lady.119 Like Otinel, the poem displays a concern with correct chivalric behaviour, both in court (humorously represented by Rauf ’s belligerent sensitivity to the etiquette of hospitality (79–87, 118–30, 144–67)), and in battle (conveyed by Charles’s instructions, on knighting Rauf, as to his new role of vassal (754–70), and by Rauf ’s sense of his obligation to live up to it (784–90)). Putter sees parallels between the scenes of Christmas hospitality and the ‘ethos of chivalry’ in both Rauf Coilyear and Sir Gawain and the Green Knight.120 A further parallel may be seen in both romances’ engagement with the insular Matter of France traditions, especially with Otinel and Fierabras. For example, the Green Knight as a potentially hostile challenger has marked resemblances to Fierabras in his giant size yet elegant appearance, and to Otinel in the fearsome way he rolls his eyes; the amorous lady hostess/‘gaoler’ with her offer of a life-saving 119 The name ‘Gautier’ occurs in Otinel as one of the twelve peers (e.g. l. 1982). 120 Putter, ‘Ralph the Collier’, pp. 148–52, 154–5.

400

Charlemagne in Medieval England

magic girdle has clear echoes of Floripas.121 Both romances provide evidence that the Matter of France was an integral part of the horizon of expectations shared between Middle English narrative poets and their audiences, who would have been able to recognize the allusions to these very well-known traditions and to appreciate the poets’ playful appropriations of them. Rauf Coilyear is thought to date from the late fifteenth century, not necessarily many years before the Edinburgh notary John Asloan copied a version of the story into his manuscript collection (c. 1515). It may well be that the strongly national Scottish consciousness evinced by the contents of the whole collection had a bearing also on the composition of Rauf Coilyear. Speed argues that the heraldic decoration of Charles’s hall (663–87) may ‘suggest the arms of Charlemagne as well as those of Scotland’, and ‘may also invoke the traditional close association of France and Scotland’.122 Certainly, the association of Charlemagne with fleurs de lis (‘Flowris with Flourdelycis’, 670) would by then have been accepted.123 The choice of the Matter of France for this sophisticated variant of an Arthurian romance model may thus be seen as a political statement of independence from England and alliance with France. It must have seemed especially timely to Asloan in the early years of the sixteenth century. The temporary stability between the adjacent kingdoms following the peace treaty of 1502 and the marriage in 1503 of James IV of Scotland to Margaret of England, daughter of Henry VII, 121 For further discussion, see Phillipa Hardman, ‘Dear Enemies: The Motif of the Converted Saracen and Sir Gawain and the Green Knight’, Reading Medieval Studies 25 (1999), 59–74. 122 Speed, Medieval English Romances, I, 314–15. 123 The absence of any eagle (as in MS Egerton 3028, discussed in Chapter 2) is strange; the double tressure flory counter-flory had been associated with the royal house of Scotland since the middle of the thirteenth century and, as the heraldist John Maldon states, ‘the most often quoted explanation for the use of the double tressure is that it represents the protection offered to Scotland by France’ (‘The Double Tressure’, in Emblems of Scotland (Dunfermline: The Heraldry Society of Scotland, 1997), pp. 13–19); Malden himself questions this explanation, proposing that ‘it may perhaps be a sign of Kingship’ (p. 13). On the appearance of the double tressure in the Scottish royal arms, see Bruce A. McAndrew, Scotland’s Historic Heraldry (Woodbridge: Boydell Press, 2006), pp. 24–5.

The Insular Otinel Tradition

401

had come to an end in 1512, when Henry VIII joined the anti-French Holy League, and James invaded England to uphold the ‘auld alliance’ between France and Scotland: an invasion that ended with James’s death at the Battle of Flodden (1513). Rauf Coilyear’s combination of Scots language with Matter of France narrative tradition, and of French place names with Scottish-seeming locations, could be thought to construct a literary equivalent of the ‘auld alliance’. Such a project indicates not only the longevity of the insular Matter of France but also its continuing openness to adaptation and reappropriation. Like The Sege of Melayne, the other insular text that appears to be an original English-language addition to the Charlemagne tradition, Rauf Coilyear constitutes a reading of the existing insular tradition that marks a full appropriation of the material, the final step in its ‘translation’ from one cultural zone to the other. As has been pointed out by Roger Ellis, translation ‘demonstrates the adequacy of the target language for the presentation of material hitherto confined to the source language’. 124 It was not, of course, that English was not seen as adequate for romance composition, but that the Matter of France could be seen as being properly expressed in the language of French. Translation of the Frenchlanguage texts into English, however, was an appropriation of the material which paved the way for the composition of English-language Charlemagne romances.

124 Roger Ellis, ‘Introduction’, in The Medieval Translator IV, ed. Roger Ellis and Ruth Evans (Exeter: University of Exeter Press, 1994), pp. 1–19 (p. 3). See also Copeland, Rhetoric, Hermeneutics and Translation, pp. 221–9.

Conclusion The Insular Afterlife of the Matter of France

H

elen Cooper’s observation that the Charlemagne romances in England ‘barely made the transition’ from the Middle Ages into post-medieval culture is, as far as can be judged from the extant evidence, an accurate diagnosis for the Middle English verse texts.1 Unlike Guy of Warwick or Sir Isumbras, both also notable warriors on behalf of Christendom, Roland and Oliver did not have their heroic victories over Saracen opponents perpetuated in print, either as romance narratives or in chapbook retellings.2 Even the protagonists of Caxton’s prose romance The Four Sons of Aymon, which was well known in the sixteenth and seventeenth centuries and frequently reprinted, did not thereafter join King Arthur’s knights and Robin Hood’s men in the repertoire of popular heroes in English culture, although the lost play The Four Sons of Aymon (c. 1581) was apparently still being performed in 1624.3 Stories of Charlemagne and Roland were re-presented to insular audiences in the form of Ludovico Ariosto’s Orlando Furioso, translated by Sir John Harington (1591). The first stanza sets the familiar scene, ‘when the Moores transported all their might | On Affrick seas the force of France to breake’, seeking revenge against ‘the Romane Emperour Charlemaine’ (I, 3–4, 8). Harington’s annotations on ‘The Historie’ of each book explain and comment on the connections to the ‘historical’ Charlemagne he discerns in Ariosto’s highly complex romance,4 but 1 Cooper, The English Romance in Time, p. 6. 2 As noted earlier, however, it is possible that the appearance of The Four Sons of Aymon in sixteenth-century lists of popular tales may hint at lost chapbook editions. 3 Cooper, The English Romance in Time, p. 417. No performance records exist for the other lost Charlemagne play, The Twelve Peers of France (before 1586): see Lost Plays Database, ed. Knutson and McInnis. 4 Ludovico Ariosto’s Orlando Furioso Translated into English Heroical Verse by Sir John Harington (1591), ed. Robert McNulty (Oxford: Clarendon Press, 1972). Further editions were published in 1607 and 1634.

402

Conclusion

403

the favourite insular stories of Roland and Otuel and of Oliver and Ferumbras do not feature (though there is a recollection of Roland’s fight with Vernagu in the story of Orlando’s protracted single combat against the almost-invincible Spanish knight Ferraw). Orlando Furioso, dramatized by Robert Greene (c. 1590), was possibly also one of the influences on Charlemagne, or The Distracted Emperor,5 an anonymous drama (c. 1604) preserved only in a seventeenth-century manuscript collection of plays (BL MS Egerton 1994),6 but which was apparently performed more than once.7 The play’s prominent theme of corruption and treachery is staged through the actions of Ganelon, paralleling his heightened role in the Middle English retellings of the story of Roncevaux, while his ambition to supplant Charlemagne as king of France reflects the similar desire of Ganelon in the insular Fierabras tradition,8 though it is impossible to know how far, if at all, these insular developments influenced the play’s adaptation of the Charlemagne legend. There is no evidence of the insular Charlemagne legends being adapted for drama in the period when one might most have expected to find them, when Europe was again threatened by Ottoman advances 5 Cooper describes the play as ‘independent of both traditions’, Caxton’s translation and Ariosto (The English Romance in Time, p. 414); however, Suzanne Leedham points out details that suggest some influence from Orlando Furioso, in ‘The Impact of Charlemagne and Roncevaux in English Literature and Culture from the Fifteenth to the Twenty-First Centuries’ (unpublished PhD thesis, University of Reading, 2012), pp. 87–8. 6 Leedham discusses the play and its manuscript context in terms of its contemporary political significance (‘The Impact of Charlemagne’, pp.  82–90). See also Curtis Perry, Literature and Favoritism in Early Modern England (Cambridge: Cambridge University Press, 2006), pp. 148–54. 7 Evidence is provided by annotations in the manuscript: see Charlemagne, or The Distracted Emperor, ed. John Henry Walter (Oxford: Oxford University Press, 1938), p. vii. 8 Fierenbras (Egerton MS), lines 1547–56; Sowdone of Babylone, lines 2967–78. In the continental versions, Aloris makes the suggestion, which Ganelon rejects. See Marianne Ailes, ‘Ganelon in the Middle English Fierabras Romances’, in The Matter of Identity in Medieval Romance, ed. Phillipa Hardman (Cambridge: D. S. Brewer, 2002), pp. 73–85 (pp. 82–4).

404

Charlemagne in Medieval England

in the later seventeenth century.9 Daniel Vitkus notes that early modern Turk plays express ‘anxious interest in Islamic power’ at this time, ‘when the Ottomans posed a sustained threat to Christian rule in Europe’, 10 but the role of Matter of France legends as ‘conductors’ for anxieties about the inroads of the Ottoman Turks had given way to more up-todate literary engagements with the confrontation of East and West, Christian and Turk, in a Protestant England which had ‘established diplomatic, cultural and commercial relations with the Ottoman Empire’.11 Yet some of the characteristic ‘memes’ endured: William Davenant’s operatic treatment of The Siege of Rhodes (1656) represented the siege as ‘the triumph of the Christian few over the overwhelming hordes of Turks’ while subverting ‘notions of stereotypical alterity’ by 9 The victory of the Holy League in the Battle of Lepanto (1571) temporarily halted Turkish expansion in Europe (though Cyprus had fallen to the Ottoman Empire); further inroads in the second half of the seventeenth century into Hungarian territory and campaigns against Austria culminated in the Battle of Vienna (1683), where a new Holy League defeated the Turks. According to Virginia Aksan, ‘The 1699 Karlowitz treaty meant the end of Ottoman control of Hungary’ and ‘inaugurated a new Ottoman approach to diplomacy’; Aksan, ‘War and Peace’, in The Cambridge History of Turkey, vol. 3: The Later Ottoman Empire, 1603–1839, ed. Surayia N. Faroghi (Cambridge: Cambridge University Press, 2006), pp. 81–117 (p. 97). 10 Three Turk Plays from Early Modern England, ed. Daniel Vitkus (New York: Columbia University Press, 2000), p. 3. Bridget Orr notes the topicality of the Turk plays in ‘the paucity of plays with Turkish themes’ after 1683 (Empire on the English Stage, 1660–1714 (Cambridge: Cambridge University Press, 2001), p. 86). 11 Mark Hutchings, ‘The Stage Historicizes the Turk: Convention and Contradiction in the Turkish History Play’, in English Historical Drama, 1500–1660: Forms outside the Canon, ed. T. Grant and B. Ravelhofer, Early Modern Literature in History (Basingstoke: Palgrave, 2007), pp. 158–78 (p. 171). Hutchings examines the 1453 loss of Constantinople, repeatedly symbolized in variations of the Irene myth, ‘registering both its cultural force and, simultaneously, its negotiability’ as a ‘signifier’ post-1453 for relations with the Turks (p. 159). We are grateful to Dr Hutchings for discussion of this topic. See also Matthew Dimmock, New Turks: Dramatizing Islam and the Ottomans in Early Modern England (Aldershot: Ashgate, 2005).

Conclusion

405

treating Solyman without ‘antipathy toward the infidel’.12 This can be read as a contemporary parallel to the earlier insular use of the ‘noble Saracen’ as a means of engagement with the Islamic Other. Ferumbras and Otuel seem to have been unknown to later insular culture until the researches of antiquaries in the early nineteenth century. Most important was George Ellis, whose three-volume Specimens of Early English Metrical Romances (1805), and especially the single-volume edition published in Bohn’s Antiquarian Library (1848), popularized Middle English verse romances and introduced modern readers for the first time to Ferumbras (in the Sowdone of Babylone), Otuel (in the Auchinleck Otuel and the Fillingham Otuel and Roland), and Roland and Vernagu, drawing their attention to the texts’ debt to the Pseudo-Turpin Chronicle. J. O. Halliwell’s preface to the Bohn edition explains Ellis’s success: the romances, ‘which had daunted all but the few initiated’, through Ellis’s retellings with extensive quotations ‘became the friends and companions of thousands’ (p. iii). However, this was a poisoned popularity, as Ellis probably laid the ground for later dismissive academic judgements with his ridicule of the romances’ absurdities. National prejudice, though absent in medieval texts (as we have seen), perhaps also has some bearing on the later demise of the Matter of France. Ellis’s attempt to reclaim the legends for Britain by arguing that they originated as Celtic stories, and were probably first written by French poets in England (pp. 17–23), does not seem to have convinced: John Ashton, for example, while citing Ellis’s Specimens as ‘the best book of all on the subject’, avoided the Charlemagne texts altogether in his own retelling of medieval romances.13 Those rewriting heroic literature for young British readers in the late nineteenth and twentieth centuries often retold the story of Roland at Roncevaux

12 Janet Clare, Drama of the English Republic, 1649–1660: Plays and Entertainments (Manchester: Manchester University Press, 2000), pp. 181–2. 13 John Ashton, Romances of Chivalry (London, 1887), p. vi. Ashton selected romances that ‘give us a wonderful insight into the manners and customs of our own country’, and ‘carefully avoided those relating to Charlemagne, believing that the Carlovingian Romances ought to be made into a series of their own’ (p. vii).

406

Charlemagne in Medieval England

(sometimes noting its ‘foreign’ status),14 but even when they retold a wider range of Charlemagne texts, the story of Ferumbras was only rarely included, and that of Otuel never.15 As we have seen, academic interest in the English Charlemagne legends lagged far behind most other Middle English romances, both for editors/anthologizers and for critics. The same was true for the French chansons de geste most popular with insular adaptors, Fierabras and Otinel: Fierabras only recently received a published modern critical edition (2003), and Otinel is still lacking one.16 The Sege of Melayne, Sowdone of Babylone, and Rauf Coilyear had some attention as 14 For example, Andrew Lang’s The Book of Romance (London: Longmans, 1902) includes ‘The Battle of Roncevalles’, a close retelling of the Chanson de Roland up to the death of Roland, alongside numerous tales from Malory’s Morte Darthur and ‘The Story of Robin Hood’, but in the preface Lang makes a distinction between the native and foreign heroes: ‘A child can see how English Robin is, how human, and possible and goodhumoured are his character and feats, … while the deeds of the French Roland … are exaggerated beyond the possible’ (p. ix). H. E. Marshall’s Stories of Roland Told to the Children (London: T. C. and E. C. Jack, 1907) retells the Chanson de Roland in ten stories, claiming it for English readers as the song sung at the Battle of Hastings. For discussion of Adair Forester, The Children’s Story of Roland (London: Harrap, 1942), as a ‘reappropriation at a time of international war’, see Leedham, ‘The Impact of Charlemagne’, pp. 176–80. 15 For example, Renaud of Montauban, trans. Robert Steele (London: George Allen, 1897) and Huon of Bordeaux, trans. Robert Steele (London: George Allen, 1895), building on the recent EETS reprints of Caxton’s and Lord Berners’ translations; Alfred J. Church, Stories of Charlemagne and the Twelve Peers of France (London: Seeley, 1902), also drawing on Caxton and Berners for tales of the Four Sons of Aymon, Fierabras, Huon of Bordeaux, with the Chanson de Roland and, unusually, Rauf Coilyear, ‘a genuinely English production’ (preface); Agnes Herbertson, adapting the Chanson de Roland and Fierabras in Heroic Legends Retold for Children (London: Blackie, 1908); Jennifer Westwood, Stories of Charlemagne (London: Bodley Head, 1972), with tales of Berthe of the great foot, Aspremont, the Four Sons of Aymon, Roland, Oliver and Roncesvalles, Huon of Bordeaux. 16 A critical edition of Otinel is currently in preparation by Jean-Baptiste Camps.

Conclusion

407

the liveliest and most individual of the Middle English Charlemagne romances, notably in Alan Lupack’s edition of all three (1990),17 but the Matter of France as a whole in Middle English texts was mostly sidelined until an upsurge of critical interest beginning in the 1990s and increasing into the twenty-first century. The topics of these texts chimed with twentieth-century critical concerns such as post-colonialism, racial and cultural Otherness, and, in the Fierabras texts, gender. Consequently, studies of both French and English Charlemagne material have tended to concentrate on the theme of crusading, the representation of Saracens, and the role of Floripas.18 Our work has of course benefited from the insights of these studies, but we have taken a different approach to the corpus of texts, recognizing that despite their longevity, the narrative traditions changed over time, and the changes reveal the cultural imprint of the environment in which each text was rewritten. Our close focus on each text and, where appropriate, its manuscript context produces a history of continuity and change, in which the Charlemagne traditions in their various translations – translations in the widest sense of appropriations and adaptations – can be seen to reflect some of the most pressing concerns of late-medieval insular society. In a sort of mirror image, it is hard not to see current critical interest in these narratives of cultural confrontation between Western and Eastern powers, or between adherents of Christianity and Islam, as reflecting something of present-day concerns over global tensions and conflicts. Most recently the concerns of our texts with the presence of the Saracen on the borders of Christendom may resonate with the concerns of the Western world faced in 2016 with an ‘Islamic State’ which includes in its fiveyear plan the re-Islamization of Spain. But while current anxieties may 17 Three Middle English Charlemagne Romances, ed. Lupack. 18 For example, Metlitzki, The Matter of Araby; Warm, ‘Identity, Narrative and Participation’; Rouse, ‘Crusaders’; Norman Daniel, Heroes and Saracens: Interpretation of the ‘chansons de geste’ (Edinburgh: Edinburgh University Press, 1984); Calkin, ‘Saracens’; Heng, Empire of Magic; Jacqueline de Weever, Sheba’s Daughters: Whitening and Demonizing the Saracen Woman in Medieval French Epic (New York and London: Garland, 2008); Bennett, ‘The Storming of the Other World’; Hans Erich Keller, ‘La Belle Sarrasine dans Fierabras et ses dérivés’, in Charlemagne in the North, ed. Bennett et al., pp. 299–307.

408

Charlemagne in Medieval England

possibly sensitize modern readers’ responses to textual representations of cultural conflict figured as holy war, this does not imply reading easy parallels between medieval texts and twenty-first-century situations. Reading medieval texts may challenge our modern world view in general ways, but, as crusades historian Jonathan Phillips warns, ‘the medieval and modern contexts are wholly different and for that reason such shadows [of the crusades] need to be treated with real care … to avoid disaster’. 19 Each age makes its own use of history and of inherited legends. In the fifteenth-century Débat des hérauts d’armes de France et d’Angleterre, written to assert the superiority of the French, the French Herald cites Charlemagne and his conquest of England as evidence.20 The debate did not go unanswered: in the sixteenth century John Coke replied with The Debate between the Heralds of England and France, which rebuts the claims of the French Herald by suggesting that, as Duke of Brabant, ‘Charlemayne was a Dowcheman’, and by counter-claiming that Roland was actually English: Roulande, the noble knight was of the Nacion of England … for he was borne in Armorica, which beying conquered by Maxymian kyng of England, gave the same to Conan Mereodoke his cosyn and to his heyres, to hold of the kynges of Englande, for ever naming it Lytel Brytayne … the sayd Rowlande was lineally extracted of the nacion of England.21 According to Coke, the Bretons were ‘copled’ with ‘noble mens doughters of England’, explaining Roland’s insular heritage. This insular appropriation of the material, using it as a weapon in contemporary Franco–English rivalry, is quite contrary to the ethos of the medieval texts and manuscripts, where the Charlemagne myth is used to unite 19 Phillips, Holy Warriors, p. 355. 20 Le débat des hérauts d’armes de France et d’Angleterre suivi de The Debate between the Heralds of England and France by John Coke, ed. Léopold Pannier and Paul Meyer, SATF (Paris, 1877), pp. iv, 13, 19; John Coke, The Debate between the Heraldes of Englande and Fraunce (1550). An extract from the French text is included in Vernacular Literary Theory, ed. WoganBrowne et al. 21 Coke, The Debate between the Heraldes, pp. 69–70. The insular tradition of Roland’s English ancestry is discussed in Chapter 4, p. 247.

Conclusion

409

Europe, and particularly England and France, against the non-Christian threat. Additionally, as we have shown, in the later Middle Ages the insular narrative tradition of Charlemagne, at different times and in various contexts, channelled contemporary concerns with: relics and their power towards the great goal of salvation; political stability in terms of monarchical legitimacy and good governance; the need to reinforce the values of chivalry; relations with the cultural and religious Other; religious orthodoxy and the fear of heresy; the threat of invasion and fears for the survival of Western Christendom. At the same time, the rewritten texts were responding to the pressure of changing expectations in relation to narrative form. Epic discourse in insular chansons de geste was adapted in ways that parallel major characteristics of Middle English popular romance, while some features of popular romance noted by Ad Putter,22 such as the use of repeated structures to create meaning, have their equivalents in elements of epic discourse. This points to ongoing and complementary developments in the writing of texts in both insular vernacular languages and both narrative genres. It remains to be seen what use the modern age will make of the tradition: creative re-engagement remains rare among insular writers and few have so far attempted a large-scale adaptation. In his ambitious appropriation of the Charlemagne material, Charlemagne and Roland (2007), Allan Massie follows the example of Caxton, placing his treatment of the Matter of France as the third in a trilogy of related narrative texts.23 Where Caxton held up the three Christian Worthies (Godfrey, Arthur, Charlemagne) as models for emulation, Massie presents (through the narrating voice of the tutor to Frederick II) lessons for the young emperor drawn from the empires of Rome, Arthur, and Charlemagne. Unlike Caxton, though, Massie treats his sources with postmodern freedom, drawing in this third novel on Einhard, the 22 Ad Putter, ‘Story Line and Story Shape in Sir Percyvell of Gales and Chrétien de Troyes’s Conte du Graal’ in Pulp Fictions of Medieval England, ed. McDonald, pp. 171–96. 23 Allan Massie, The Evening of the World: A Romance of the Dark Ages (London: Wiedenfeld and Nicholson, 2001); Arthur the King (London: Wiedenfeld and Nicholson, 2003); Charlemagne and Roland: A Romance (London: Wiedenfeld and Nicholson, 2007); Massie terms all three ‘The Matter of Eternal Rome which is the Matter of Europe’ (Evening of the World, title page).

410

Charlemagne in Medieval England

Pseudo-Turpin tradition and Ariosto’s Orlando Furioso, to produce a blend of researched historical detail and extravagant fantasy.24 Yet despite their pertinence to the concerns of today, there have been no modern adaptations of either Fierabras or Otinel.25 New cultural adaptations, however, are taking other branches of the Charlemagne tradition in new directions: Sir Christopher Lee has translated the Charlemagne, Roland and Roncevaux material into the medium of rock music in two Wagner-inspired concept albums, Charlemagne: By the Sword and the Cross (2010) and Charlemagne: The Omens of Death (2013),26 while 2012 saw the unexpected development of a new tradition in Walsall in the West Midlands, built on the possible link between once-known stories of Renaud de Montauban and the local heritage artefacts known as Bayard’s Colts.27 Elements of the original story, particularly Bayard and Maugis, are combined with numerous other popular traditions in the spirit of nineteenth- and 24 Massie’s novel is analysed in Leedham, ‘The Impact of Charlemagne’, pp. 121–44. Leedham also discusses other reworkings of Charlemagne material in American fantasy fiction (pp. 102–20). 25 A partial exception may be made for the free translation of Fierabras into modern English verse (aimed at general readers as much as students of the Middle Ages) by Michael A. H. Newth, Fierabras and Floripas: A French Epic Allegory (New York: Italica Press, 2010). As the sub-title indicates, Newth speculatively builds on hints in Le Person’s introduction (p. 188) and in Brewer’s Dictionary of Phrase and Fable (1898; under ‘Fierabras’) to re-present the chansons de geste of La Destruction de Rome and Fierabras as a continuous allegory in four stages (Vanity, Submission, Desires, Deserts), to be read on three levels: national, religious, personal (pp. xviii–xix). 26 In an interview published on the internet, Christopher Lee claims: ‘The album is based entirely on the historical events that happened during the life of Charlemagne. We tried to keep any creative license to a bare minimum’; http://skullbanger.net/2013/06/22/my-interview-with-sirchristopher-lee-discussing-his-new-charlemagne-album-omens-of-death/ [accessed 23/08/2016]. 27 Bayard’s Colts is the name of a collection of wooden staves dating from the eighteenth century or earlier, each of which bears a carved finial in the shape of a human, animal or grotesque head, or an abstract shape. See F. N. Bowler, ‘Bayard’s Colts’, Folklore 81 (1970), 266–7. Until recently they were on display at Walsall Museum.

Conclusion

411

twentieth-century village mumming plays to produce a series of street plays for annual performance, uniquely linked to the local community.28 Such initiatives as the Bayard’s Colts project have consciously social aims concerned with community identity, and in this they resemble the engagement of the Middle English and Anglo-Norman Charlemagne texts with the politics and social concerns of their own times. Questions of identity – linguistic and religious, cultural and political – were raised by the poets and translators of the medieval narratives, though there are no easy answers to these concerns for their age or for our own. The ways in which the Matter of France was translated do not imply a rejection of France or the French language, but an appropriation of the Matter, situating it within the wider identity and concerns of Christendom, while increasingly doing so in the ascendant vernacular of English; and as translations, the Middle English texts take different, but almost always dynamic, approaches. However, the pattern of insular appropriation is far more complex than a narrative of cultural transfer from French to English. In researching these texts we discovered evidence for a more widespread dissemination of the French epic tradition in England than had hitherto been suspected. We found a persistent history of cultural reappropriation of the tradition and adaptation of the stories to new circumstances: particularly striking was the discovery that these insular texts, French-language (Anglo-Norman) as well as Middle English, were put to specifically political and religious use in their new contexts. Far from marking them as inferior derivatives of poorly assimilated originals, the many variations in the later texts (in particular the Middle English ones) are evidence of a vigorous and continuing creative engagement with the inherited tradition.

28 For details and videos of the plays, see https://vimeo.com/ channels/320507.

Appendix The Corpus: Texts and Manuscripts For convenient reference we list here the core texts in the insular Charlemagne corpus with details of the extant manuscripts, and the editions used in this volume.

La Chanson de Roland – The Anglo-Norman ‘Oxford’ version Oxford, Bodleian Library MS Digby 23 Digital images of the complete manuscript text are available at: http://image.ox.ac.uk/ show?collection=bodleian&manuscript=msdigby23b Edited by Ian Short in La Chanson de Roland: The French Corpus, 3 vols, ed. Joseph J. Duggan (Turnhout: Brepols, 2005), I Translated into modern English in The Song of Roland, trans. Joseph J. Duggan and Annalee C. Rejhon (Turnhout: Brepols, 2012), pp. 27–168 Duggan’s edition includes all French-language manuscripts of La Chanson de Roland. In our discussion we refer to the following continental manuscripts by their accepted names and sigla: V4: Venice, Biblioteca Nazionale Marciana MS fr. Z 4 (225) V7: Venice, Biblioteca Nazionale Marciana MS fr. Z 7 (251) P: Paris, Bibliothèque nationale de France MS fr. 860 T: Cambridge, Trinity College MS R.3.32 L: Lyon, Bibliothèque municipale MS 743 C: Châteauroux, Bibliothèque municipale MS 1

The Song of Roland – Middle English version of the Chanson de Roland London, British Library MS Lansdowne 388 Edited in The English Charlemagne Romances, Part II: ‘The Sege off Melayne’ and ‘The Romance of Duke Rowland and Sir Otuell 412

Texts and Manuscripts

413

of Spayne’ … with a fragment of ‘The Song of Roland’, ed. Sidney J. H. Herrtage, EETS ES 35 (London, 1880)

The Pseudo-Turpin Chronicle – The Anglo-Norman translation by William de Briane London, British Library MS Arundel 220 Edited in The Anglo-Norman Pseudo-Turpin Chronicle of William de Briane, ed. Ian Short, ANTS 25 (Oxford, 1973) Turpines Story – The Middle English translation of the PseudoTurpin Chronicle San Marino, CA, Huntington Library MS HM 28,561 Digital images of the manuscript, including the opening folio of this text, are available at: http://dpg.lib.berkeley.edu/ webdb/dsheh/heh_brf?CallNumber=HM+28561 Edited in Turpines Story: A Middle English Translation of the Pseudo-Turpin Chronicle, ed. Stephen H. A. Shepherd, EETS 322 (Oxford, 2004) Roland and Vernagu – Middle English romance based on the Pseudo-Turpin Chronicle Edinburgh, National Library of Scotland MS Adv. 19.2.1 (the Auchinleck MS) Digital images of the complete manuscript text are available at: http://auchinleck.nls.uk/ Edited in The English Charlemagne Romances, Part VI: ‘The Taill of Rauf Coilyear’ with … ‘Roland and Vernagu’ and ‘Otuel’, ed. Sidney J. H. Herrtage, EETS ES 39 (London, 1882) Fierabras – The Anglo-Norman ‘Vulgate’ text Hanover, Niedersächsischen Landesbibliothek MS IV 578 Images of folios 57r, 89r and 94r are included in Helmut Härtel and Felix Ekowski, Handschriften der Niedersächsischen Landesbibliothek Hannover, 2 vols (Wiesbaden: Harrassowitz, 1982), plates VI, XI, XII

414

Appendix

Edited in Chanson de Fierabras, ed. Alfons Hilka and André de Mandach (Neuchâtel: Pré-publication de Neuchâtel, 1981) Variant readings from the Hanover MS are given in the critical apparatus of Fierabras: Chanson de geste du XIIe siècle, ed. Marc Le Person, CFMA 42 (Paris: Champion, 2003); this edition is based on the continental Escorial MS M. III-21.

Sir Ferumbras – Middle English romance based on the ‘Vulgate’ text Oxford, Bodleian Library MS Ashmole 33 Digitized images of a specimen folio and the parchment wrapper are available at: http://medievalromance.bodleian.ox.ac. uk/A_rough_draft_and_fair_copy_of_Sir_Firumbras Edited in The English Charlemagne Romances, Part I: Sir Ferumbras, ed. Sidney J. H. Herrtage, EETS ES 34 (London, 1879) Firumbras – Middle English romance based on the ‘Vulgate’ text London, British Library MS Add. 37492 (the Fillingham MS) Edited in ‘Firumbras’ and ‘Otuel and Roland’, ed. Mary Isabelle O’Sullivan, EETS 198 (London, 1935) Fierenbras – The Anglo-Norman ‘version courte’ text London, British Library MS Egerton 3028 (the Egerton MS) Digital images of the manuscript, including all folios with illuminations, are available at: http://www. bl.uk/catalogues/illuminatedmanuscripts/record. asp?MSID=6654&CollID=28&NStart=3028 Edited in ‘La Destruction de Rome et Fierabras, MS Egerton 3028 de Musée Britannique’, ed. L. Brandin, Romania 64 (1938), 18–100 La Destruction de Rome – The Anglo-Norman ‘Vulgate’ text Hanover, Niedersächsische Landesbibliothek MS IV 578 An image of folio 22v is included in Helmut Härtel and Felix Ekowski, Handschriften der Niedersächsischen

Texts and Manuscripts

415

Landesbibliothek Hannover, 2 vols (Wiesbaden: Harrassowitz, 1982), plate V Edited in La Destructioun de Rome, ed. L. Formisano (Florence: Sansoni, 1981)

La Destruction de Rome – The Anglo-Norman ‘version courte’ text London, British Library MS Egerton 3028 (the Egerton MS) Digital images of the manuscript, including all folios with illuminations, are available at: http://www. bl.uk/catalogues/illuminatedmanuscripts/record. asp?MSID=6654&CollID=28&NStart=3028 Edited in ‘La Destruction de Rome et Fierabras, MS Egerton 3028 de Musée Britannique’, ed. L. Brandin, Romania 64 (1938), 18–100 The Sowdone of Babylone – Middle English romance based on the Anglo-Norman ‘version courte’ texts of La Destruction de Rome and Fierenbras Princeton, NJ, Princeton University Library MS Garrett 140 Edited in The English Charlemagne Romances, Part V: The Sowdone of Babylone, ed. Emil Hausknecht, EETS ES 38 (London, 1881) Otinel – The Anglo-Norman text Cologny Geneva, Bodmer Library MS 168 Digital images of the complete manuscript text are available at: http://www.e-codices.unifr.ch/en/fmb/cb-0168 An edition of this text is in preparation, ed. Jean-Baptiste Camps. Extracts and variant readings from this text are given to supplement the edition of the continental text (Vatican City, Biblioteca Apostolica Vaticana MS Reg. lat. 1616) in Otinel: Chanson de geste, ed. F. Guessard and H. Michelant, APF (Paris: Vieweg, 1859).

416

Appendix

Otuel – Middle English romance based on Otinel Edinburgh, National Library of Scotland MS Adv. 19.2.1 (the Auchinleck MS) Digital images of the complete manuscript text are available at: http://auchinleck.nls.uk/ Edited in The English Charlemagne Romances, Part VI: ‘The Taill of Rauf Coilyear’ with … ‘Roland and Vernagu’ and ‘Otuel’, ed. Sidney J. H. Herrtage, EETS ES 39 (London, 1882) Roland and Otuel – Middle English romance based on Otinel London, British Library MS Add. 31042 (the London Thornton MS) Edited in The English Charlemagne Romances, Part II: ‘The Sege off Melayne’ and ‘The Romance of Duke Rowland and Sir Otuell of Spayne’ … with … ‘The Song of Roland’, ed. Sidney J. H. Herrtage, EETS ES 35 (London, 1880) Otuel and Roland – Middle English romance based on Otinel and the Pseudo-Turpin Chronicle London, British Library MS Add. 37492 (the Fillingham MS) Edited in ‘Firumbras’ and ‘Otuel and Roland’, ed. Mary Isabelle O’Sullivan, EETS 198 (London, 1935) The Sege of Melayne – Middle English romance with no known French source London, British Library MS Add. 31042 (the London Thornton MS) Edited in The English Charlemagne Romances, Part II: ‘The Sege off Melayne’ and ‘The Romance of Duke Rowland and Sir Otuell of Spayne’ … with … ‘The Song of Roland’, ed. Sidney J. H. Herrtage, EETS ES 35 (London, 1880)

Texts and Manuscripts

417

The Taill of Rauf Coilyear – Middle English romance with no known French source No extant manuscript; a unique print copy: Edinburgh, National Library of Scotland, H.29.c.9 The complete text is reproduced in The Taill of Rauf Coilyear, printed by Robert Lekpreuik at St Andrews in 1572: A facsimile of the only known copy, intro. William Beattie (Edinburgh: National Library of Scotland, 1966) Edited in The English Charlemagne Romances, Part VI: ‘The Taill of Rauf Coilyear’ with … ‘Roland and Vernagu’ and ‘Otuel’, ed. Sidney J. H. Herrtage, EETS ES 39 (London, 1882)

Bibliography Primary Sources Adenet le roi, Les Enfances Ogier, ed. Albert Henry, Les Oeuvres d’Adenet le roi, III (Bruges: Rijksuniversiteit te Gent, 1956) The Anglo-Norman Pseudo-Turpin Chronicle of William de Briane, ed. Ian Short, ANTS 25 (Oxford, 1973) Ariosto, Ludovico, Ludovico Ariosto’s Orlando Furioso Translated into English Heroical Verse by Sir John Harington (1591), ed. Robert McNulty (Oxford: Clarendon Press, 1972) Aspremont, ed. François Suard (Paris: Champion, 2008) The Auchinleck Manuscript, ed. David Burnley and Alison Wiggins (NLS, 5 July 2003), version 1.1, http://auchinleck.nls.uk/ editorial/history.html The Auchinleck Manuscript: National Library of Scotland Advocates’ MS 19.2.1, with introduction by Derek Pearsall and I. C. Cunningham (London: Scolar Press, 1977) Bagnyon, Jehan, Histoire de Charlemagne, ed. Hans-Erich Keller (Geneva: Droz, 1992) Barbour’s Bruce, ed. Matthew P. McDiarmid and James A. C. Stevenson, 2 vols, STS, 4th series, 15 (Edinburgh, 1985) The Beauchamp Pageant, ed. Alexandra Sinclair (Donington: Watkins, 2003) Bibliorum Sacrorum cum Glossa Ordinaria, 6 vols (Venice, 1603) Cambridge University Library MS Ff.2.38, with introduction by Frances McSparran and P. R. Robinson (Cambridge: Scolar Press, 1979) Caxton, William, The Lyf of Charles the Grete, ed. Sidney Herrtage, EETS ES, 36, 37 (London, 1880–1) —— The Four Sons of Aymon, ed. Octavia Richardson, EETS ES 44, 45 (London, 1884–5) La Chanson d’Antioche, trans. Carol Sweetenham and Susan Edgington (Farnham: Ashgate, 2011) Chanson de Fierabras, ed. Alfons Hilka and André de Mandach (Neuchâtel: Pré-publication de Neuchâtel, 1981) La Chanson de la Première Croisade en ancien français d’après Baudri 419

420

Bibliography

de Bourgueil: Édition et analyse lexicale, ed. Jennifer Gabel de Aguirre (Heidelberg: Universitäetsverlag Winter, 2015) La Chanson de Roland: The French Corpus, ed. J. J. Duggan et al., 3 vols (Turnhout: Brepols, 2005) La Chanson de Roland, ed. Césare Segré, rev. edn with introduction, trans. Madeleine Tyssens, Textes littéraires français 368, 2 vols (Geneva: Droz, 1989) Charlemagne, or The Distracted Emperor, ed. John Henry Walter (Oxford: Oxford University Press, 1938) Chartier, Alain, The Treatise of Hope, in Fifteenth-Century English Translations of Alain Chartier’s ‘Le Traite de l’Esperance’ and ‘Le Quadrilogue Invectif’, ed. Margaret S. Blayney, 2 vols, EETS OS 270 (London, 1974), 281 (Oxford, 1980) Codex Ashmole 61: A Compilation of Popular Middle English Verse, ed. George Shuffelton (Kalamazoo, MI: Medieval Institute Publications, 2008) Coke, John, The Debate between the Heraldes of Englande and Fraunce (1550) The Complaynt of Scotland, ed. A. M. Stewart, STS, 4th series, 11 (Edinburgh, 1979) Le Couronnement de Louis, ed. E. Langlois, 2nd edn (Paris: Champion, 1984) The Crusade and Death of Richard I, ed. R. C. Johnston, ANTS 17 (Oxford, 1961) Le débat des hérauts d’armes de France et d’Angleterre suivi de The Debate between the Heralds of England and France by John Coke, ed. Léopold Pannier and Paul Meyer, SATF (Paris, 1877) ‘La Destruction de Rome et Fierabras, MS Egerton 3028 de Musée Britannique’, ed. L. Brandin, Romania 64 (1938), 18–100 La Destructioun de Rome, ed. L. Formisano (Florence: Sansoni, 1981) La Destructioun de Rome, ed. L. Formisano, ANTS Plain Texts 8 (London, 1990) ‘La Destruction de Rome: Première branche de la chanson de geste de Fierabras’, ed. Gustav Gröber, Romania 2 (1873), 1–48 The Devils’ Parliament and the Harrowing of Hell and Destruction of Jerusalem, ed. C. W. Marx, Middle English Texts 25 (Heidelberg: Winter, 1993) Documents on the Later Crusades, 1274–1580, ed. and trans. Norman Housley (Basingstoke: Macmillan, 1996)

Bibliography

421

The Early English Carols, ed. R. L. Greene, 2nd edn (Oxford: Clarendon Press, 1977) Early Popular Poetry of Scotland and the Northern Border, ed. David Laing, rev. W. Carew Hazlitt, 2 vols (London: Reeves and Turner, 1895) Der festländische Bueve de Hantone, ed. Albert Stimming (Dresden: Gedruckt für die Gesellschaft für romanische Literatur, 1911–20) Fierabras: Chanson de geste du XIIe siècle, ed. Marc Le Person, CFMA 42 (Paris: Champion, 2003) Fierabras, ed. A. Kroeber and G. Servois, APF (Paris: Vieweg, 1860) Firumbras and Otuel and Roland, ed. Mary Isabelle O’Sullivan, EETS OS 198 (London, 1935) Foxe, John, The Unabridged Acts and Monuments Online, 1563 edition (Sheffield: HRI Online Publications, 2011), http//www. johnfoxe.org Froissart, Sir John, Chronicles, trans. John Bourchier, Lord Berners (1523–5), repr. edn, ed. E. V. Utterson, 2 vols (London: Rivington, 1812) —— Oeuvres de Froissart, ed. M le baron Kervyn de Lettenhove, 25 vols (Brussels: Devaux, 1867–77) Gaimar, Estoire des Engleis, ed. Alexander Bell, ANTS 14–16 (Oxford, 1960) The Game and Playe of the Chesse, ed. William Axon (London: Elliot Stock, 1883) Gaufrey, ed. F. Guessard and P. Chabaille, APF (Paris, 1859) Geoffrey of Monmouth, Historia Regum Britanniae, The History of the Kings of Britain, ed. Michael D. Reeve, trans. Neil Wright (Woodbridge: Boydell Press, 2009) The Gilte Legende, ed. Richard Hamer and Vida Russell, 2 vols, EETS OS 327, 328 (Oxford, 2006, 2007) Girart de Vienne, ed. Wolfgang G. van Emden, SATF (Paris, 1978) The Golden Legend: Readings on the Saints, trans. William Granger Ryan, 2 vols (Princeton, NJ: Princeton University Press, 1993) Gormont et Isembart: Fragment de chanson de geste du XIIe siècle, ed. Alphonse Bayot, 3rd edn, CFMA 14 (Paris: Champion, 1931) Gower, John, The Complete Works of John Gower, ed. G. C. Macaulay, 4 vols (Oxford: Clarendon Press, 1899–1902) —— John Gower: The Minor Latin Works, ed. and trans. R. F. Yeager,

422

Bibliography

with In Praise of Peace, ed. Michael Livingston (Kalamazoo, MI: Medieval Institute Publications, 2005) The Historia Ierosolimitana of Baldric of Bourgueil, ed. Steven Biddlecombe (Woodbridge: Boydell Press, 2014) Historia Karoli Magni et Rotholandi, ou Chronique du Pseudo-Turpin, ed. C. Meredith-Jones (Paris: Droz, 1936; repr. Geneva: Slatkine, 1972) The History of the Holy War: Ambroise’s ‘Estoire de la Guerre Sainte’ , ed. Marianne Ailes and Malcolm Barber, 2 vols (Woodbridge: Boydell Press, 2003) Huon of Bordeaux, trans. Robert Steele (London: George Allen, 1895) Indagine, Joannes ab, A Briefe Introduction vnto the Art of Chiromancy and Physiognomy, trans. Fabian Withers (London: J. Day for R. Jugge, 1558) Laneham, Robert, A Letter (1575), facsimile reprint (Menston: Scolar Press, 1969) Langtoft, Pierre de, La Règne d’Édouard Ier, ed. Jean-Claude Thiolier (Paris: Université de Paris XII, 1989) The Leofric Missal, ed. Nicholas Orchard, 2 vols, Henry Bradshaw Society 113–14 (London, 2002) Le Livre de Saint-Jacques et la tradition du Pseudo-Turpin, ed. JeanClaude Vallecalle (Lyon: Presses Universitaires de Lyon, 2011) Lydgate, John, The Minor Poems of John Lydgate, ed. H. N. MacCracken, 2 vols, EETS OS 107, 192 (London, 1911, 1934) —— ‘Lydgatiana 14: ‘O Flos Pulcherrime!’, ed. H. N. MacCracken, Archiv 131 (1913), 60–3 Malory, Sir Thomas, The Works of Sir Thomas Malory, ed. Eugene Vinaver, 2nd edn, 3 vols (Oxford: Clarendon Press, 1967) —— Le Morte D’Arthur, ed. Janet Cowen, 2 vols (London: Penguin, 1969) —— The Works of Sir Thomas Malory, ed. Eugene Vinaver, 2nd edn (Oxford: Oxford University Press, 1977) Mandeville, John, The Book of John Mandeville, ed. and trans. Iain M. Higgins (Indianapolis: Hackett, 2011) ‘The Middle English Evangelie’, ed. Gertrude H. Campbell, PMLA 30 (1915), 529–613 The Missal of Robert of Jumièges, ed. H. A. Wilson, Henry Bradshaw Society 11 (London, 1896)

Bibliography

423

Mouskès, Philippe, Chronique rimée des rois de France, ed. Baron F. de Reiffenberg, Ac. roy. de Belgique, 2 vols (Brussels, 1836–8) Octavian, ed. Harriet Hudson, in Four Middle English Romances, 2nd edn (Kalamazoo, MI: Medieval Institute Publications, 2006) Octovian Imperator, ed. Frances McSparran, Middle English Texts 11 (Heidelberg: Winter, 1979) The Old French Johannes Translation of the Pseudo-Turpin Chronicle: A Critical Edition, ed. Ronald N. Walpole, 2 vols (Berkeley: University of California Press, 1976) Otinel: Chanson de geste, ed. F. Guessard and H. Michelant, APF (Paris: Vieweg, 1859) Paris, Matthew, La Vie de Saint Auban, ed. Arthur Robert Harden, ANTS 19 (Oxford, 1968) Le Pèlerinage de Charlemagne, ed. and trans. Glyn S. Burgess (Edinburgh: British Rencesvals Publications, 1998) [Le Pèlerinage de Charlemagne], The Journey of Charlemagne to Jerusalem and Constantinople, ed. J. L. G. Picherit (Birmingham, AL: Summa, 1984) The Pepys Ballads, ed. W. G. Day, 5 vols (Cambridge: D. S. Brewer, 1987) Piramus, Denis, La Vie Seint Edmund, ed. Florence L. Ravenel, Bryn Mawr College Monographs 5 (Bryn Mawr, PA: Bryn Mawr College, 1906) The Plantagenet Chronicles, ed. Elizabeth Hallam (London: Macmillan, 1988) Les Quatre Fils Aymon, ou Renaud de Montauban, ed. and trans. Micheline de Combarieu du Grès and Jean Subrenat (Paris: Gallimard, 1983) Renaud of Montauban, trans. Robert Steele (London: George Allen, 1897) Rigord, Oeuvres de Rigord et de Guillaume le Breton, Historiens de Philippe Auguste, ed. H. François Delaborde, Société de l’Histoire de France (Paris, 1882) Der Roman von Fierabras provenzalisch, ed. Immanuel Bekker (Berlin, 1829) ‘The Sege off Melayne’ and ‘The Romance of Duke Rowland and Sir Otuell of Spayne’ … with a fragment of ‘The Song of Roland’ , ed. Sidney J. H. Herrtage, EETS ES 35 (London, 1880)

424

Bibliography

Selections from English Wycliffite Writings, ed. Anne Hudson (Cambridge: Cambridge University Press, 1978) The Siege of Jerusalem, ed. Michael Livingston (Kalamazoo, MI: Medieval Institute Publications, 2004) Sir Ferumbras, ed. Sidney J. H. Herrtage, EETS ES 34 (London, 1879) Skelton, John, The Complete English Poems, ed. John Scattergood (New Haven, CT: Yale University Press, 1983) The Song of Roland, ed. Gérard J. Brault, 2 vols (University Park and London: Pennsylvania State University Press, 1978) The Song of Roland: Translations of the Versions in Assonance and Rhyme of the ‘Chanson de Roland’, trans. Joseph J. Duggan and Annalee C. Rejhon (Turnhout: Brepols, 2012) The Sowdone of Babylone, ed. Emil Hausknecht, EETS ES 38 (London, 1881) Supplementary Lives in some Manuscripts of the Gilte Legende, ed. R. Hamer and V. Russell, EETS OS 315 (Oxford, 2000) The Taill of Rauf Coilyear, printed by Robert Lekpreuik at St Andrews in 1572: A facsimile of the only known copy, intro. William Beattie (Edinburgh: NLS, 1966) ‘The Taill of Rauf Coilyear’ , with the Fragments of ‘Roland and Vernagu’ and ‘Otuel’, ed. S. J. H. Herrtage, EETS ES 39 (London, 1882) Three Middle English Charlemagne Romances, ed. Alan Lupack (Kalamazoo, MI: Medieval Institute Publications, 1990) Three Turk Plays from Early Modern England, ed. Daniel Vitkus (New York: Columbia University Press, 2000) La Traduction du Pseudo-Turpin du manuscript Regina 624, ed. Claude Buridant (Geneva: Droz, 1976) Turpines Story: A Middle English Translation of the Pseudo-Turpin Chronicle, ed. Stephen H. A. Shepherd, EETS OS 322 (Oxford, 2004) La Vie de Seint Clement, vol. I: Text: 1–7006, ed. Daron Burrows, ANTS 64 (London, 2007) Wace, Roman de Brut, ed. and trans. Judith Weiss (Exeter: University of Exeter Press, 1999) Warkworth, John, A Chronicle of the First Thirteen Years of the Reign of King Edward the Fourth, by John Warkworth, D.D., ed. J. O. Halliwell, Camden Society (London, 1839)

Bibliography

425

Windeatt, B. A., Troilus and Criseyde: A New Edition of ‘The Book of Troilus’ (London: Longman, 1984)

Secondary Sources Aebischer, Paul, Études sur Otinel (Berne: Editions Francke, 1960) Ailes, Adrian, Royal Arms of England (Reading: University of Reading, 1982) —— ‘Heraldry in Medieval England: Symbols of Politics and Propaganda’, in Heraldry, Pageantry and Social Display, ed. Coss and Keen, pp. 83–104 Ailes, M. J., ‘Doon de Nanteuil and the Epic of Revolt’, Medium Ævum 52 (1983), 247–57 —— ‘Some Observations on the Grouping of Epics of Revolt in Manuscripts and Compilations’, Reading Medieval Studies 10 (1984), 3–19 —— ‘Faith in Fierabras’, in Charlemagne in the North, ed. Bennett et al., pp. 126–33 —— ‘From Epic to Chronicle and Back: The Pseudo-Turpin Chronicle and the chanson de geste Fierabras’, in Thirty Years of Medieval Studies at the University of Reading, 1965–95: A Celebration, ed. Anne Curry (Reading: University of Reading, 1995), pp. 17–24 —— ‘Deux fragments inconnus de Maugis d’Aigremont’, Romania 116 (1998), 415–30 —— ‘Fierabras and Anglo-Norman Developments of the chanson de geste’, in Epic Studies: Acts of the Seventeenth International Congress of the Société Rencesvals for the Study of Romance Epic, Storrs, CT ( July 2006), ed. Leslie Zarker Morgan and Anne Berthelot = Olifant, special issue, n.s. 25.1–2 (2006 [2009]), 97–110 —— ‘Gui de Warewic in its Manuscript Context’ in Guy of Warwick, ed. Wiggins and Field, pp. 12–26 —— ‘The Anglo-Norman Boeve de Haumtone as a chanson de geste’, in Sir Bevis of Hampton in Literary Tradition, ed. Jennifer Fellows and Ivana Djordjević (Cambridge: D. S. Brewer, 2008), pp. 9–24 —— What’s in a Name? Anglo-Norman Romances or chansons de

426

Bibliography

geste?’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 61–76 —— ‘Témoins fragmentaires de la geste de Maience et la reception du cycle en Angleterre’, in La Geste de Doon de Mayence, ed. Boutet, pp. 115–36 —— ‘Deux manuscrits de la chanson de geste de l’automne du moyen âge: Un renouvellement du genre?’, in Mélanges Roussel, ed. Françoise Laurent (forthcoming) Ailes, M. J., and P. E. Bennett, ‘Deux nouveaux fragments de poèmes sur Guillaume d’Orange’, in Proceedings of the Oxford Congress, ed. Ailes et al., pp. 87–105 Ailes, Marianne, ‘A Comparative Study of the Medieval French and Middle English Verse Texts of the Fierabras Legend’ (unpublished PhD thesis, University of Reading, 1989) —— ‘Chivalry and Conversion: The Chivalrous Saracen in the Old French Epics Fierabras and Otinel’, Al-Masaq: Studia AraboIslamica Mediterranea 9 (1996–7), 1–21 —— ‘The Date of Fierabras’, Olifant 19 (1994–5 [1999]), 25–71 —— ‘Fierabras and the Chanson de Roland: An Intertextual Diptych’, Reading Medieval Studies 28 (2002), 3–22 —— ‘Ganelon in the Middle English Fierabras Romances’, in The Matter of Identity in Medieval Romance, ed. Phillipa Hardman (Cambridge: D. S. Brewer, 2002), pp. 73–85 —— The Song of Roland: On Absolutes and Relative Values, Studies in Mediaeval Literature (Lampeter: Mellen, 2002) —— ‘Pseudo-Turpin Chronicle’, in Encyclopedia of the Medieval Chronicle, ed. Graeme Dunphy (Leiden: Brill, 2010), pp. 1454–5 —— ‘Otinel: An Epic in Dialogue with the Tradition’, Olifant 27 (2015), 9–39, http://scholarworks.iu.edu/journals/index. php/olifant/issue/archive Ailes, Marianne, and Phillipa Hardman, ‘How English are the English Charlemagne Romances?’, in Boundaries in Medieval Romance, ed. Cartlidge, pp. 43–55 —— ‘Texts in Conversation: Charlemagne Epics and Romances in Insular Plural-Text Codices’, in Insular Books: Vernacular Miscellanies in Late Medieval Britain, ed. Raluca Radulescu and Margaret Connolly (London: British Academy Publications, 2015), pp. 31–47

Bibliography

427

Ailes, Marianne, and Suzanne Leedham, ‘Le Pseudo-Turpin en Angleterre’, Cahiers de recherche médiévales et humanistes 25 (2013), 495–514 Ailes, Marianne, Anne E. Cobby and Philip E. Bennett (eds), The Proceedings of the Oxford Congress of the International Société Rencesvals, 2012 (Edinburgh: British Rencesvals Publications, 2015) Akbari, Suzanne Conklin, ‘Incorporation in the Siege of Melayne’, in Pulp Fictions of Medieval England, ed. McDonald, pp. 22–44 —— Idols in the East: European Representations of Islam and the Orient, 1100–1450 (Ithaca, NY: Cornell University Press, 2009) Akbari, Suzanne Conklin, and Amilcare A. Iannucci (eds), Marco Polo and the Encounter of East and West (Toronto: University of Toronto Press, 2008) Aksan, Virginia, ‘War and Peace’, in The Cambridge History of Turkey, vol. 3: The Later Ottoman Empire, 1603–1839, ed. Surayia N. Faroghi (Cambridge: Cambridge University Press, 2006), pp. 81–117 Alexander, Jonathan, and Paul Binski (eds), Age of Chivalry: Art in Plantagenet England, 1200–1400 (London: Royal Academy of Arts, 1987) Allaire, Gloria, ‘Literary Evidence for Multilingualism: The Roman de Tristan in its Italian Incarnations’, in Medieval Multilingualism, ed. Kleinhenz and Busby, pp. 145–53 Ashe, Laura, Ivana Djordjević and Judith Weiss (eds), The Exploitations of Medieval Romance (Cambridge: D. S. Brewer, 2010) Ashton, John, Romances of Chivalry (London, 1887) Aston, Margaret, ‘Lollards and the Cross’, in Lollards and their Influence in Late Medieval England, ed. Fiona Somerset, Jill C. Havens and Derrick G. Pitard (Woodbridge: Boydell Press, 2003), pp. 99–113 Bailey, Jade, ‘The Portrayal of Charlemagne in Text and Image in British Library MS Royal 15 E vi’, paper given at Leeds International Medieval Congress, July 2012 —— ‘Fierabras and the Livre de Charlemaine in British Library MS Royal 15 E vi: An Edition and Study’ (unpublished PhD thesis, University of Bristol, 2014) Bailey, Matthew, and Ryan D. Giles (eds), Charlemagne and his

428

Bibliography

Legend in Early Spanish Literature and Historiography, Bristol Studies in Medieval Cultures (Cambridge: D. S. Brewer, 2016) Baker, Denise N. (ed.), Inscribing the Hundred Years’ War in French and English Cultures (New York: State University of New York Press, 2000) Barbero, Alessandro, Charlemagne: Father of a Continent, trans. Allan Cameron (Berkeley: University of California Press, 2004) Barker-Benfield, B. C. (ed.), St Augustine’s, Canterbury, 3 vols, Corpus of British Medieval Library Catalogues 13 (London: British Library in association with the British Academy, 2008) Barlow, Frank, The English Church, 1000–1066, 2nd edn (London: Longman, 1979) —— The English Church, 1066–1154 (London: Longman, 1979) Barron, W. R. J., English Medieval Romance (London: Longman, 1987) Bawcutt, Priscilla, ‘ “ My bright buke”: Women and their Books in Medieval and Renaissance Scotland’, in Medieval Women: Texts and Contexts in Late Medieval Britain, ed. Jocelyn WoganBrowne et al. (Turnhout: Brepols, 2000), pp. 17–34 Beadle, Richard, The Lyell Lectures, given at Merton College, Oxford, 2013 Beattie, William, ‘Some Early Scottish Books’, in The Scottish Tradition: Essays in Honour of Ronald Gordon Cant, ed. G. W. S. Barrow (Edinburgh: Scottish Academic Press, 1974), pp. 107–20 Bédier, Joseph, ‘La Composition de la chanson Fierabras’, Romania 17 (1888), 22–51 —— Les Légendes épiques: Recherches sur la formation des chanson de geste, 4 vols (Paris, 1908–13) Beer, Jeanette, ‘Patronage and the Translator: Raoul de Presles’s La Cité de Dieu and Calvin’s Institution de la religion Chrestienne and Institutio religionis Christianae’, in Translation and the Transmission of Culture between 1300 and 1600, ed. Jeanette Beer and Kenneth Lloyd-Jones (Kalamazoo, MI: Medieval Institute Publications, 1995), pp. 91–142 Beer, Jeanette (ed.), Medieval Translators and their Craft, Studies in Medieval Culture 25 (Kalamazoo: Western Michigan Publications, 1989) Bell, David N., The Libraries of the Cistercians, Gilbertines and

Bibliography

429

Premonstratensians, Corpus of British Medieval Library Catalogues 3 (London: British Library in association with the British Academy, 1992) Bellamy, J. G., The Law of Treason in England in the Later Middle Ages (Cambridge: Cambridge University Press, 1970) Benjamin, Walter, ‘The Task of the Translator’, in The Translation Studies Reader, ed. Lawrence Venuti (London: Routledge, 2000), pp. 75–83 Bennett, Michael, ‘France in England: Anglo-French Culture in the Reign of Edward III’, in Language and Culture in Medieval Britain, ed. Wogan-Browne et al., pp. 320–33 Bennett, Philip E., ‘The Storming of the Otherworld, 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 in Celebration of his Seventieth Birthday, ed. Philip E. Bennett and Wolfgang G. van Emden (Reading: Société Rencesvals British Branch, 1984), pp. 1–14 —— ‘La Chanson de Guillaume, poème Anglo-Normand?’, in Au carrefour des routes d’Europe: La Chanson de geste, Xe Congrès International de la Société Rencesvals (Aix-en-Provence: Publications du CUER MA, 1987), pp. 259–81 —— ‘L’Épique dans l’historiographie anglo-normande: Gaimar, Wace, Jordan Fantosme’, in Aspects de l’épopée romane, ed. van Dijk and Noomen, pp. 321–30 —— ‘La Chanson de Guillaume’ and ‘La Prise d’Orange’ (London: Grant and Cutler, 2000) Bennett, Philip E., Anne Elizabeth Cobby and Graham A. Runnalls (eds), Charlemagne in the North: Proceedings of the Twelfth International Congress of the Société Rencesvals, Edinburgh, 4th to 11th August 1991 (London: Grant and Cutler, 1993) Berlings, Elizabeth, ‘The Sege of Melayne – A Comic Romance; or, How the French Screwed Up and “Oure Bretonns” Rescued Them’, in Boundaries in Medieval Romance, ed. Cartlidge, pp. 57–70 Berman, Antoine, L’Épreuve de l’étranger (Paris: Gallimard, 1995) —— ‘Translation and the Trials of the Foreign’, in The Translation Studies Reader, ed. Lawrence Venuti, 2nd edn (London: Routledge, 2004), pp. 276–90

430

Bibliography

Besamusca, Bart (ed.), Cyclification: The Development of Narrative Cycles in the chansons de geste and Arthurian Romances (Amsterdam: Royal Netherlands Academy of Arts and Sciences, 1994) Bettey, Joseph, The Suppression of the Monasteries in the West Country (Gloucester: Sutton, 1989) —— Morning Stars of the Reformation: Early Religious Reformers in the Bristol Region (Bristol: Avon Local History and Archaeology, 2011) Birchwood, Matthew, and Matthew Dimmock (eds), Cultural Encounters between East and West: 1453–1699 (Newcastle: Cambridge Scholars Press, 2005) Bisaha, Nancy, Creating East and West: Renaissance Humanists and the Ottoman Turks (Philadelphia: University of Pennsylvania Press, 2004) Blaess, Madeleine, ‘L’Abbaye de Bordesley et les livres de Guy de Beauchamp’, Romania 78 (1957), 511–18 Blake, Norman F., Caxton and his World (London: André Deutsch, 1969) —— William Caxton and English Literary Culture (London: Hambledon Press, 1991) Bloch, R. Howard, ‘Naturalism, Nationalism, Medievalism’, Romanic Review 76 (1985), 341–60 Boffey, Julia, ‘ “ The Charter of the Abbey of the Holy Ghost” and its Role in Manuscript Anthologies’, in Medieval and Early Modern Miscellanies and Anthologies, ed. Phillipa Hardman, Yearbook of English Studies 33 (2003), 120–30 Boutet, Dominique, Charlemagne et Arthur, ou le roi imaginaire (Paris: Champion, 1992) —— La Chanson de geste: Forme et signification d’une écriture épique au moyen âge (Paris: Presses Universitaires de France, 1993) Boutet, Dominique (ed.), La Geste de Doon de Mayence dans ses manuscrits et dans ses versions (Paris: Champion, 2014) Bowers, John M., ‘Chaucer after Retters: The Wartime Origins of English Literature’, in Inscribing the Hundred Years’ War, ed. Baker, pp. 91–125 Bowler, F. N., ‘Bayard’s Colts’, Folklore 81 (1970), 266–7 Bradbury, Jim, ‘Geoffrey V of Anjou: Count and Knight’, in The

Bibliography

431

Ideals and Practice of Medieval Knighthood, ed. Harper-Bill and Harvey, pp. 21–38 Bradbury, Nancy Mason, ‘Representations of Peasant Speech: Some Literary and Social Contexts for The Taill of Rauf Coilyear’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 19–33 Brandin, L., ‘Le Manuscrit de Hanovre de la Destruction de Rome et de Fierabras’, Romania 28 (1899), 489–507 Brault, Gérard J., ‘Le Réviseur du manuscrit d’Oxford de la Chanson de Roland’, in Essor et fortune de la chanson de geste dans l’Europe et l’Orient latin: Actes du IXe Congrès International de la Société Rencesvals, Padoue-Venise, 1982 (Modena: Mucchi, 1984), pp. 828–62 —— ‘The FitzEdmund Arms (Heralds’ Roll 586–89) and the French Epic Renaut de Montauban’, The Coat of Arms 12 (1997), 2–6 —— ‘Adenet le Roi and the Coats of Arms of Charlemagne and his Twelve Peers in a Medieval Roll’, in ‘Contez me tout’: Mélanges de langue et littérature médiévales offerts à Herman Braet, ed. Catherine Bel, Pascale Dumont and Frank Willaert, La république des lettres 28 (Louvain: Peeters, 2006), pp. 115–28 —— ‘Adenet le roi et l’héraldique médiévale’, Olifant 25 (2006), 141–9 Brieger, Peter, English Art, 1216–1307, Oxford History of English Art (Oxford: Clarendon Press, 1957) Bühler, Curt F., ‘Two Middle English Texts of the Somnia Danielis’, Anglia 80 (1962), 264–73 Burland, Margaret Jewett, Strange Words: Retelling and Reception in the Medieval Roland Textual Tradition (Notre Dame, IN: University of Notre Dame Press, 2007) Burnley, J. D., ‘Late Medieval English Translation: Types and Reflections’, in The Medieval Translator, ed. Ellis (1989), pp. 37–53 Burns, Robert I., Moors and Crusades: Collected Studies (London: Variorum, 1978) Burrell, Margaret, ‘The Voyage of Charlemagne: Cultural Transmission or Cultural Transgression?’, Parergon, n.s. 7 (1989), 47–54 Busby, Keith, Codex and Context: Reading Old French Verse Narrative in Manuscript, 2 vols (Amsterdam: Rodopi, 2002)

432

Bibliography

Butterfield, Ardis, The Familiar Enemy: Chaucer, Language and Nation in the Hundred Years’ War (Oxford: Oxford University Press, 2009) Calin, William, The French Tradition and the Literature of Medieval England (Toronto: University of Toronto Press, 1994) Calkin, Siobhain Bly, Saracens and the Making of English Identity: The Auchinleck Manuscript (New York: Routledge, 2005) —— ‘Romance Baptisms and Theological Contexts in The King of Tars and Sir Ferumbras’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 105–19 —— ‘Saracens’, in Heroes and Anti-Heroes in Medieval Romance, ed. Cartlidge, pp. 185–200 Campbell, Emma, and Robert Mills, ‘Introduction: Rethinking Medieval Translation’, in Rethinking Medieval Translation, ed. Campbell and Mills, pp. 1–20 Campbell, Emma, and Robert Mills (eds), Rethinking Medieval Translation: Ethics, Politics, Theory (Cambridge: D. S. Brewer, 2012) Campbell, Mary, The Witness and the Other World: Exotic European Travel Writing, 400–1600 (Ithaca, NY: Cornell University Press, 1991) Camps, Jean-Baptiste, ‘Otinel et l’Europe: Éléments pour une histoire de la diffusion de la geste’, in Proceedings of the Oxford Congress, ed. Ailes et al., pp. 137–56 Carolus-Barré, Louis, and Paul Adam-Even, ‘Les Armes de Charlemagne dans l’héraldique et l’iconographie médiévales’, in Mémorial d’un voyage en Rhenanie de la Société nationale des antiquaires de France (Paris, 1953) Cartlidge, Neil (ed.), Boundaries in Medieval Romance (Cambridge: D. S. Brewer, 2008) —— Heroes and Anti-Heroes in Medieval Romance (Cambridge: D. S. Brewer, 2012) Catalogue of Manuscripts in the British Museum, n.s., vol. I, part 1: The Arundel Manuscripts (London, 1834) Catalogue raisonné des manuscrits de la bibliothèque de M. Ambroise Firmin Didot (Paris, 1867) Chrimes, S. B., Henry VII (New Haven, CT: Yale University Press, 1999)

Bibliography

433

Church, Alfred J., Stories of Charlemagne and the Twelve Peers of France (London: Seeley, 1902) Clare, Janet, Drama of the English Republic, 1649–1660: Plays and Entertainments (Manchester: Manchester University Press, 2000) Clifton, Nicole, ‘The Seven Sages of Rome, Children’s Literature, and the Auchinleck Manuscript’, in Childhood in the Middle Ages and the Renaissance: The Results of a Paradigm Shift in the History of Mentality, ed. Albrecht Classen (Berlin: De Gruyter, 2005), pp. 185–201 Coates, Alan, English Medieval Books: The Reading Abbey Collections from Foundation to Dispersal (Oxford: Clarendon Press, 1999) Cobby, Anne E., ‘Religious Elements in Le Voyage de Charlemagne à Jérusalem et à Constantinople’, in Au carrefour de routes d’Europe: La Chanson de geste, Xe Congrès international de la société Rencesvals pour l’étude des épopées romanes (Strasbourg, 1985), 2 vols (Aix-en-Provence: Université de Provence, 1987), I, 367–82 —— Ambivalent Conventions: Formula and Parody in Old French (Amsterdam: Rodopi, 1995) Coldiron, A. E. B., ‘Visibility Now: Historicizing Foreign Presences in Translation’, Translation Studies 5 (2002), 189–200 —— ‘William Caxton’, in The Oxford History of Literary Translation in English, I, ed. Ellis, pp. 160–9 Coldiron, Anne, English Printing, Verse Translations, and the Battle of the Sexes, 1476–1557 (Farnham: Ashgate, 2009) Cole, Andrew, Literature and Heresy in the Age of Chaucer (Cambridge: Cambridge University Press, 2008) Collins, Roger, Charlemagne (London: Macmillan, 1998) Connolly, Margaret, ‘Preaching by Numbers: The Seven Gifts of the Holy Ghost in Late Middle English Sermons and Works of Religious Instruction’, in Preaching the Word in Manuscript and Print in Late Medieval England, ed. Martha W. Driver and Veronica O’Mara (Turnhout: Brepols, 2013), pp. 83–100 Conway, Agnes, and Edmund Curtis, Henry VII’s Relations with Scotland and Ireland, 1485–1498 (Cambridge: Cambridge University Press, 1932) Cooke, W. G. and D’A . J. D. Boulton, ‘Sir Gawain and the Green

434

Bibliography

Knight: A Poem for Henry of Grosmont?’, Medium Ævum 68 (1999), 42–54 Cooper, Helen, ‘Counter-Romance: Civil Strife and Father Killing in the Prose Romances’, in The Long Fifteenth Century: Essays for Douglas Gray, ed. Helen Cooper and Sally Mapstone (Oxford: Oxford University Press, 1997), pp. 141–62 —— ‘Romance after 1400’, in The Cambridge History of Medieval English Literature, ed. Wallace, pp. 690–719 —— The English Romance in Time: Transforming Motifs from Geoffrey of Monmouth to the Death of Shakespeare (Oxford: Oxford University Press, 2004) Copeland, Rita, ‘The Fortunes of “Nor verbum pro verbo”: or why Jerome is not a Ciceronian’, in The Medieval Translator, ed. Ellis (1989), pp. 15–35 —— Rhetoric, Hermeneutics and Translation in the Middle Ages: Academic Traditions and Vernacular Texts (Cambridge: Cambridge University Press, 1991) Coss, Peter, and Maurice Keen (eds), Heraldry, Pageantry and Social Display in Medieval England (Woodbridge: Boydell Press, 2002) Cowen, Janet, ‘The English Charlemagne Romances’, in Roland and Charlemagne in Europe, ed. Pratt, pp. 149–68 Crane, Susan, ‘Anglo-Norman Cultures in England, 1066–1460’, in The Cambridge History of Medieval English Literature, ed. Wallace, pp. 35–60 Crofts, Thomas H., ‘Degrees of Veracity in the Morte Darthur: Elements of Malory’s Chronicle Style’, Journal of the International Arthurian Society 1 (2013), 120–39 Crofts, Thomas H., and Robert Allen Rouse, ‘Middle English Popular Romance and National Identity’, in A Companion to Medieval Popular Romance, ed. Radulescu and Rushton, pp. 79–95 Crosland, Jessie, The Old French Epic (Oxford: Blackwell, 1951) Crouch, David, The Birth of Nobility: Constructing Aristocracy in England and France, 900–1300 (London: Longman, 2005) Cruz, Jo Ann Hoeppner Moran, ‘Popular Attitudes towards Islam in Medieval Europe’, in Western Views of Islam in Medieval and Early Modern Europe: Perception of Other, ed. Michael Frassetto and David R. Blanks (New York: St Martins Press, 1999), pp. 55–81

Bibliography

435

Cunningham, I. C., ‘The Asloan Manuscript’, in The Renaissance in Scotland: Studies in Literature, Religion, History and Culture, ed. A. A. MacDonald, Michael Lynch and Ian B. Cowan (Leiden: Brill, 1994), pp. 107–35 Curry, Anne, The Battle of Agincourt: Sources and Interpretations (Woodbridge: Boydell Press, 2000) —— ‘War or Peace? Philippe de Mézières, Richard II and AngloFrench Diplomacy’, in Philippe de Mézières and his Age: Piety and Politics in the Fourteenth Century, ed. Renate BlumenfeldKosinski and Kiril Petkov (Leiden: Brill, 2012), pp. 295–320 Curry, Anne, et al., ‘Languages in the Military Profession in Later Medieval England’, in The Anglo-Norman Language and its Contexts, ed. Richard Ingham (York: York Medieval Press, 2010), pp. 74–93 Daniel, Norman, Heroes and Saracens: Interpretation of the ‘chansons de geste’ (Edinburgh: Edinburgh University Press, 1984) Dannenbaum, Susan, ‘Anglo-Norman Romances of English Heroes: “ancestral romances”?’, Romance Philology 35 (1981/2), 601–8 Dean, Ruth, and Maureen Boulton, Anglo-Norman Literature: A Guide to Texts and Manuscripts (London: ANTS, 2000) De Mandach, André, Naissance et développement de la chanson de geste en Europe, 6 vols (Geneva: Droz, 1961–93) —— ‘A Royal Wedding-Present in the Making: Talbot’s Chivalric Anthology (Royal 15 E vi) for Queen Margaret of Anjou, and the Laval-Middleton Anthology of Nottingham’, Nottingham Mediaeval Studies 18 (1974), pp. 56–76 —— La Geste de Fierabras: Le Jeu du réel et de l’invraisemblable (Geneva: Droz, 1987) Dennison, Lynda, ‘ “Liber Horn”, “Liber Custumarum” and other Manuscripts of the Queen Mary Psalter Workshops’, in Medieval Art, Architecture and Archaeology in London, ed. Lindy Grant, British Archaeological Association Conference Transactions 10 (London: Maney, 1990), pp. 118–34 DeVries, Kelly, ‘The Battle of Nicopolis’, Medieval History Magazine 2 (October 2003), accessed online through http://deremilitari. org/articles/ Dijk, Hans van, and Willem Noomen (eds) Aspects de l’épopée romane: Mentalités, idéologies, intertextualités (Groningen: Forsten, 1995)

436

Bibliography

Dimmock, Matthew, New Turks: Dramatizing Islam and the Ottomans in Early Modern England (Aldershot: Ashgate, 2005) Di Vanna, Isabel N., ‘Politicizing National Literature: The Scholarly Debate around La Chanson de Roland in the Nineteenth Century’, Historical Research 84 (2011), 109–34 Douglas, D. C., ‘The Song of Roland and the Norman Conquest of England’, French Studies 14 (1960), 99–116 Duffy, Eamon, The Stripping of the Altars: Traditional Religion in England, 1400–1580 (New Haven, CT: Yale University Press, 1992) Dufournet, Jean, Cours sur Roland (Paris: Centre de documentation universitaire, 1972) Duggan, Joseph, ‘Franco-German Conflict and the History of French Scholarship on the Song of Roland’, in Hermeneutics and Medieval Culture, ed. Patrick J. Gallacher and Helen Damico (Albany: State University of New York Press, 1989), pp. 97–106 Elias, Marcel, ‘Emotional Rhetoric in the Middle English Charlemagne Romances: The Failures of Crusading Practices in the Otuel Texts’, paper delivered at the 49th International Congress on Medieval Studies, Kalamazoo, 2014 Ellis, George, Specimens of Early English Metrical Romances, rev. edn (London: Bohn, 1848) Ellis, Roger, ‘Introduction’, in The Medieval Translator IV, ed. Roger Ellis and Ruth Evans (Exeter: University of Exeter Press, 1994), pp. 1–19 —— ‘Patronage and Sponsorship of Translation’, in The Oxford History of Literary Translation, I, ed. Ellis, pp. 98–115 Ellis, Roger (ed.), The Medieval Translator: The Theory and Practice of Translation in the Middle Ages (Cambridge: D. S. Brewer, 1989) —— The Oxford History of Literary Translation in English, vol. I: To 1550 (Oxford: Oxford University Press, 2008) Eriksen, Stefka Georgieva, Writing and Reading in Medieval Manuscript Culture (Turnhout: Brepols, 2014) Eversheim, Walter, ‘A Citizens’ Prize for Distinguished Service on Behalf of European Unification’, http://www.aachen. de/de/stadt_buerger/pdfs_stadtbuerger/pdf_karlspreis/ karlspreis_03/eversheim03_en.pdf

Bibliography

437

Faral, Edmond, La Chanson de Roland: Étude et analyse (Paris: Mellottée, 1934) —— ‘A propos de la Chanson de Roland: Genèse et signification du personnage de Turpin’, in La Technique littéraire des chansons de geste: Actes du colloque de Liège (Paris: Les Belles Lettres, 1959) Farrier, Susan E., ‘Das Rolandslied and the Song of Roulond as Moralizing Adaptations of the Chanson de Roland’, Olifant 16 (1991), 61–76 Fein, Susanna, and Michael Johnston (eds), Robert Thornton and his Books: Essays on the Lincoln and London Thornton Manuscripts (York: York Medieval Press, 2014) Fewster, Carol, Traditionality and Genre in Middle English Romance (Cambridge: D. S. Brewer, 1987) Fichte, Joerg, ‘Caxton’s Concept of “Historical Romance” within the Context of the Crusades: Conviction, Rhetoric and Sales Strategy’, in Tradition and Transformation in Medieval Romance, ed. Field, pp. 101–13 Field, Rosalind, ‘The Anglo-Norman Background to Alliterative Romance’, in Middle English Alliterative Poetry and its Literary Background, ed. Lawton, pp. 54–69 —— ‘Romance in England, 1066–1400’, in The Cambridge History of Medieval English Literature, ed. Wallace, pp. 152–76 —— ‘Waldef and the Matter of/with England’, in Medieval Insular Romance: Translation and Innovation, ed. Judith Elizabeth Weiss, Jennifer Fellows and Morgan Dickson (Cambridge: D. S. Brewer, 2000), pp. 25–39 —— ‘Romance’, in The Oxford History of Literary Translation in English, I, ed. Ellis, pp. 296–331 —— ‘Patterns of Availability and Demand in Middle English Translations de romanz’, in The Exploitations of Medieval Romance, ed. Ashe et al., pp. 73–89 Field, Rosalind (ed.), Tradition and Transformation in Medieval Romance (Cambridge: D. S. Brewer, 1999) Finlayson, John, ‘Definitions of Middle English Romance’, Chaucer Review 15 (1980/81), 44–63, 168–81 —— ‘The Context of the Crusading Romances in the London Thornton Romances’, Anglia 130 (2012), 240–63

438

Bibliography

Fischer, Steven R., The Complete Medieval Dreambook: A Multilingual Alphabetical Somnia Danielis Collation (Bern: Lang, 1982) Fischer, Walther, ‘Zur me. “Romanze Sir Ferumbras”: das Verhältnis des ersten Entwurfs. (V. 331–759) zur Reinschrift’, Archiv für das Studium der neueren Sprachen und Literaturen 142 (1921), 25–54 Fletcher, Lydia, ‘ “ Traytoures” and “Treson”: The Language of Treason in the Works of Sir Thomas Malory’, Arthurian Literature 28 (2011), 75–88 Fletcher, R. A., ‘Reconquest and Crusade in Spain, c. 1050–1150’, in The Crusades: Essential Readings, ed. Thomas F. Madden (Oxford: Blackwell, 2002), 51–67 Fleury, Rohault de, Mémoire sur les instruments de la Passion (Paris : Lesort, 1870) Forester, Adair, The Children’s Story of Roland (London: Harrap, 1942) Formisano, Luciano, ‘Alle origini del lachmannismo romanzo: Gustav Gröber e la redazione occitanica del Fierabras’, Annali della Scuola Normale Superiore di Pisa, Classe de Lettere e Filosofia, 3rd series, 9 (Pisa, 1979), pp. 247–302 Freeman, Charles, Holy Bones, Holy Dust: How Relics Shaped the History of Medieval Europe (New Haven, CT: Yale University Press, 2011) Fresco, Karen L., and Charles D. Wright (eds), Translating the Middle Ages (London: Ashgate, 2012) Friis-Jensen, Karsten, and James M. W. Willoughby (eds), Peterborough Abbey, Corpus of British Medieval Library Catalogues 8 (London: British Library in association with the British Academy, 2001) Frolow, Anatole, La Relique de la vraie croix (Paris: Institut Français d’Etudes Byzantine, 1961) Furrow, Melissa, Expectations of Romance: The Reception of a Genre in Medieval England (Cambridge: D. S. Brewer, 2009) —— ‘Chanson de geste as Romance in England’, in The Exploitations of Medieval Romance, ed. Ashe et al., pp. 57–72 Gabriele, Matthew, An Empire of Memory: The Legend of Charlemagne, the Franks, and Jerusalem before the First Crusade (Oxford: Oxford University Press, 2011) Gabriele, Matthew, and Jace Stuckey (eds), The Legend of

Bibliography

439

Charlemagne in the Middle Ages: Power, Faith, and Crusade, The New Middle Ages (New York: Palgrave Macmillan, 2008) Galbreath, Donald Lindsay, Papal Heraldry (Cambridge: Heffer, 1930) Gaunt, Simon, Gender and Genre in Medieval French Literature (Cambridge: Cambridge University Press, 1995) Gautier, Léon, La Chanson de Roland, texte critique accompagné d’une traduction nouvelle et précèdé d’une introduction historique (Tours, 1872) —— Les Épopées françaises: Étude sur les origines et l’histoire de la littérature nationale, 4 vols (Paris: Société Générale de Librairie Catholique, 1878–92) —— Bibliographie des chansons de geste (Paris, 1897) George-Tvrtkovic, Rita, A Christian Pilgrim in Medieval Iraq: Riccoldo da Montecroce’s Encounter with Islam (Turnhout: Brepols, 2013) Gillespie, Alexandra, Print Culture and the Medieval Author: Chaucer, Lydgate, and their Books, 1473–1557 (Oxford: Oxford University Press, 2006) Gillingham, John, ‘Kingship, Chivalry and Love: Political and Cultural Values in the Earliest History Written in French, Geoffrey Gaimar’s Estoire des Engleis’, in Anglo-Norman Political Culture and the Twelfth-Century Renaissance, ed. C. W. Hollister (Woodbridge: Boydell Press, 1997), pp. 33–58 Graf, A., Roma nella memoria e nelle immaginazioni del medioevo (Turin, 1882) Gray, Douglas, Later Medieval English Literature (Oxford: Oxford University Press, 2008) Green, Richard Firth, ‘The Hermit and the Outlaw: An Edition’, in Interstices: Studies in Late Middle English and Anglo-Latin Texts in Honour of A. G. Rigg, ed. Richard Firth Green and Linne R. Mooney (Toronto: University of Toronto Press, 2003), pp. 137–66 Greenblatt, Stephen, ‘Culture’, in Critical Terms for Literary Study, ed. Frank Lentricchia and Thomas McLaughlin, 2nd edn (London: University of Chicago Press, 1995), pp. 225–32 Gres, M. Combarieu du, L’Idéal human et l’expérience morale chez les héros des chansons de geste, 2 vols (Aix-en-Provence: Publications université de Provence, 1979)

440

Bibliography

Grigsby, John L., ‘Le Voyage de Charlemagne, pèlerinage ou parodie?’, in Au carrefour des routes d’Europe: La Chanson de geste, Xe Congrès international de la société Rencesvals pour l’étude des épopées romanes (Strasbourg, 1985), 2 vols (Aix-en-Provence: Université de Provence, 1987), I, 567–81 Grillo, Peter R., ‘Vers une édition du texte français de l’Historia Jerosolimitana de Baudri de Dol’, in Autour de la Première Croisade: Actes du Colloque de la Society for the Study of the Crusades and the Latin East (Clermont-Ferrand, 22–5 June 1995), ed. M. Balard (Paris: Publications de la Sorbonne, 1996), pp. 6–16 Guard, Timothy, Chivalry, Kingship and Crusade: The English Experience in the Fourteenth Century (Woodbridge: Boydell Press, 2013) Haidu, Peter, The Subject of Violence: The ‘Song of Roland’ and the Birth of the State (Bloomington and Indianapolis: Indiana University Press, 1993) Halphen, Louis, Le Comté d’Anjou au XIe siècle (Geneva: Slatkine Reprints, 1974) Hanna, Ralph III, ‘The Growth of Robert Thornton’s Books’, Studies in Bibliography 40 (1987), 51–61 —— ‘Reconsidering the Auchinleck Manuscript’, in New Directions in Later Medieval Manuscript Studies, ed. Derek Pearsall (York: York Medieval Press, 2000), pp. 91–102 —— London Literature, 1300–1380 (Cambridge: Cambridge University Press, 2005) Hardman, Phillipa, ‘Reading the Spaces: Pictorial Intentions in the Thornton MSS: Lincoln Cathedral MS 91 and BL MS Add. 31042’, Medium Ævum 63 (1994), 250–74 —— ‘Dear Enemies: The Motif of the Converted Saracen and Sir Gawain and the Green Knight’, Reading Medieval Studies 25 (1999), 59–74 —— ‘The Sege of Melayne: A Fifteenth-Century Reading’, in Tradition and Transformation in Medieval Romance, ed. Field, pp. 71–86 —— ‘The True Romance of Tristrem and Ysoude’, in Cultural Encounters in the Romance of Medieval England, ed. Corinne Saunders (Cambridge: D. S. Brewer, 2005), pp. 85–99 —— ‘Lydgate’s Uneasy Syntax’, in John Lydgate: Poetry, Culture, and

Bibliography

441

Lancastrian England, ed. Larry Scanlon and James Simpson (Notre Dame, IN: University of Notre Dame Press, 2006), pp. 12–35 —— ‘A Note on the Collation of BL MS Add. 37492 (The Fillingham Manuscript)’, Journal of the Early Book Society 11 (2008), 217–21 —— ‘Popular Romances and Young Readers’, in A Companion to Medieval Popular Romance, ed. Radulescu and Rushton, pp. 150–64 —— ‘Domestic Learning and Teaching: Investigating Evidence for the Role of “Household Miscellanies” in Late-Medieval England’, in Women and Writing, c. 1340–c. 1650: The Domestication of Print Culture, ed. Anne Lawrence-Mathers and Phillipa Hardman (York: York Medieval Press, 2010), pp. 15–33 —— ‘Learning Lessons in Middle English Romance’, Reading Medieval Studies 37 (2011), 1–13 —— ‘Roland in England: Contextualizing the Middle English Song of Roland’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 91–104 —— ‘Knight, King, Emperor, Saint: Portraying Charlemagne in Middle English Romance’, Reading Medieval Studies 38 (2012), 43–58 —— ‘Bodleian Library, MS Ashmole 33: Thoughts on Reading a Work in Progress’, in Middle English Texts in Transition, ed. Simon Horobin and Linne Mooney (York: York Medieval Press, 2014), pp. 88–103 Harper-Bill, Christopher, and Ruth Harvey (eds), The Ideals and Practice of Medieval Knighthood III: Papers from the Fourth Strawberry Hill Conference, 1988 (Woodbridge: Boydell Press, 1990) Härtel, Helmut, and Felix Ekowski, Handschriften der Niedersächsischen Landesbibliothek Hannover, 2 vols (Wiesbaden: Harrassowitz, 1982) Heinemann, Edward A., L’Art métrique de la chanson de geste: Essai sur la musicalité du récit (Geneva: Droz, 1993) Heinermann, Theodor, ‘Zeit und Sinn der Karlsreise’, Zeitschrift für romanische Philologie 56 (1936), 497–562 Hellinga, Lotte, ‘From Poggio to Caxton: Early Translations of Some

442

Bibliography

of Poggio’s Latin Facetiæ’, in Makers and Users of Medieval Books: Essays in Honour of A. S. G. Edwards, ed. Carol M. Meale and Derek Pearsall (Cambridge: D. S. Brewer, 2014), pp. 89–104 Heng, Geraldine, ‘The Romance of England: Richard Coeur de Lyon, Saracens, Jews and the Politics of Race and Nation’, in The Post-Colonial Middle Ages, ed. Jeffrey J. Cohen (New York: St Martin’s Press, 2000), pp. 135–72 —— Empire of Magic: Medieval Romance and the Politics of Cultural Fantasy (New York: Columbia University Press, 2003) Herbertson, Agnes, Heroic Legends Retold for Children (London, Blackie, 1908) Higgins, Iain M., review of The Egerton Version of Mandeville’s Travels, ed. M. C. Seymour, EETS OS 336 (Oxford and New York: Oxford University Press, 2010), in Speculum 86 (2011), 1123–5 Hinkle, W. M., The Fleurs de Lis of the Kings of France, 1205–1488 (Carbondale and Edwardsville: Southern Illinois University Press, 1991) Hopkins, Andrea, The Sinful Knights: A Study of Middle English Penitential Romance (Oxford: Clarendon Press, 1990) Horrent, Jules, La Chanson de Roland dans les littératures française et espagnole au moyen âge (Paris: Les Belles Lettres, 1951) —— Le Pèlerinage de Charlemagne: Essai d’explication littéraire (Paris: Les Belles-Lettres, 1961) Housley, Norman, The Italian Crusades: The Papal–Angevin Alliance and the Crusades against Lay Christian Powers, 1254–1343 (Oxford: Clarendon Press, 1982) —— ‘The Bishop of Norwich’s Crusade, May 1383’, History Today 33/5 (1983), 15–20 —— The Later Crusades, 1274–1580: From Lyons to Alcazar (Oxford: Oxford University Press, 1992) —— Contesting the Crusades (Oxford: Blackwell, 2006) Hutchings, Mark, ‘The Stage Historicizes the Turk: Convention and Contradiction in the Turkish History Play’, in English Historical Drama, 1500–1660: Forms outside the Canon, ed. T. Grant and B. Ravelhofer, Early Modern Literature in History (Basingstoke: Palgrave, 2007), pp. 158–78 Ingham, Richard, ‘Anglo-Norman: New Themes, New Contexts’,

Bibliography

443

in The Anglo-Norman Language and its Contexts, ed. Richard Ingham (York: York Medieval Press, 2010), pp. 1–7 Ingledew, Francis, ‘Sir Gawain and the Green Knight’ and the Order of the Garter (Notre Dame, IN: University of Notre Dame Press, 2006) James, M. R., The Ancient Libraries of Canterbury and Dover (Cambridge: Cambridge University Press, 1903) Jefferson, Judith A., and Ad Putter (eds), Multilingualism in Medieval Britain (Turnhout: Brepols, 2013) Johnston, Michael, ‘Constantinian Christianity in the London Manuscript: The Codicological and Linguistic Evidence of Thornton’s Intentions’, in Robert Thornton and his Books, ed. Fein and Johnston, pp. 177–204 Jones, Catherine M., An Introduction to the chansons de geste (Gainesville: University Press of Florida, 2014) Jubb, Margaret A., ‘Enemies in the Holy War, but Brothers in Chivalry: The Crusaders’ View of their Saracen Opponents’, in Aspects de l’épopée romane, ed. van Dijk and Noomen, pp. 251–9 Kay, Sarah, ‘Ethics and Heroics in the Song of Roland’, Neophilologus 62 (1978), 480–91 —— ‘The Nature of Rhetoric in the chanson de geste’, Zeitschrift für romanische Philologie 94 (1978), 305–20 Kedar, Benjamin Z., Crusade and Mission: Eastern Approaches toward the Muslims (Princeton, NJ: Princeton University Press, 1984) Keen, Maurice, ‘Chaucer’s Knight, the English Aristocracy and the Crusade’, in English Court Culture in the Later Middle Ages, ed. V. J. Scattergood and J. W. Sherborne (London: Duckworth, 1983) Keiser, George R., ‘Robert Thornton: Gentleman, Reader and Scribe’, in Robert Thornton and his Books, ed. Fein and Johnston, pp. 67–108 Keller, Hans Erich, ‘La Belle Sarrasine dans Fierabras et ses dérivés’, in Charlemagne in the North, ed. Bennett et al., pp. 299–307 Kelly, Luis, The True Interpreter (Oxford: Blackwell, 1979) Kennedy, Edward, ‘Sir Thomas Malory’s (French) Romance and (English) Chronicle’, in Arthurian Studies in Honour of P. J. C. Field, ed. Bonnie Wheeler (Cambridge: D. S. Brewer, 2004), pp. 223–34

444

Bibliography

Ker, N. R., ‘Medieval Manuscripts from Norwich Cathedral Priory’, Transactions of the Cambridge Bibliographical Society 1 (1953), 1–28 Kittredge, G. L., Chaucer and his Poetry (Cambridge, MA: Harvard University Press, 1915) Kleinhenz, Christopher, and Keith Busby (eds), Medieval Multilingualism: The Francophone World (Turnhout: Brepols, 2010) Knott, Gordon, ‘The Historical Sources of Fierabras’, Modern Language Review 52 (1957), 504–9 Knutson, Roslyn L., and David McInnis (eds), Lost Plays Database (Melbourne: University of Melbourne, 2009), http://www. lostplays.org/ Konick, Marcus, ‘The Authorship of Sir Ferumbras’ (unpublished PhD thesis, University of Pennsylvania, 1953) Lambert, Sarah, ‘Heroines and Saracens’, Medieval World 1 (1991), 3–9 Lang, Andrew, The Book of Romance (London: Longmans, 1902) Langlois, Ernest, ‘Deux fragments épiques: Otinel, Aspremont’, Romania 12 (1883), 433–57 Latowsky, Anne A., Emperor of the World: Charlemagne and the Construction of Imperial Authority, 800–1229 (Ithaca, NY: Cornell University Press, 2013) Lauer, Philippe, ‘Le Poème de la Destruction de Rome et les origins de la cité léonine’, Mélanges d’archéologie et d’histoire publiés par l’école français de Rome 19 (1899), 307–65 Lawton, David (ed.), Middle English Alliterative Poetry and its Literary Background: Seven Essays (Cambridge: D. S. Brewer, 1982) Leedham, Suzanne, ‘The Impact of Charlemagne and Roncevaux in English Literature and Culture from the Fifteenth to the Twenty-First Centuries’ (unpublished PhD thesis, University of Reading, 2012) Le Gentil, Pierre, ‘Reflections sur la chanson d’Otinel’, Cultura Neolatina 321 (1961), 66–70 Legge, M. Dominica, Anglo-Norman in the Cloisters (Edinburgh: Edinburgh University Press, 1950) —— ‘Archaism and the Conquest’, Modern Language Review 51 (1956), 227–9

Bibliography

445

—— Anglo-Norman Literature and its Background (Oxford: Clarendon Press, 1963) Leitch, Megan, ‘Thinking Twice about Treason in Caxton’s Prose Romances: Proper Chivalric Conduct and the English Printing Press’, Medium Ævum 81 (2012), 41–69 Lejeune, Rita, and Jacques Stiennon, La Légende de Roland dans l’art du moyen âge, 2 vols (Brussels: Arcade, 1966) Leopold, Antony, How to Recover the Holy Land: The Crusade Proposals of the Late Thirteenth and Early Fourteenth Centuries (Aldershot: Ashgate, 2000) Le Person, Marc (ed.), Le Rayonnement de Fierabras dans la littérature européenne: Actes du colloque international [L’Université Jean Moulin, Lyon 3] (Lyon: Université Jean Moulin, 2003) Le Person, Marc, Fierabras: Traduction en français moderne (Paris: Champion, 2012) Libbon, Marisa, ‘The Invention of King Richard’, in The Auchinleck Manuscript: New Perspectives, ed. Susanna Fein (York: York Medieval Press, 2016), pp. 127–38 Linder, Amnon, Raising Arms: Liturgy in the Struggle to Liberate Jerusalem in the Late Middle Ages (Turnhout: Brepols, 2003) Liu, Yin, ‘Middle English Romance as Prototype Genre’, Chaucer Review 40 (2006), 335–53 Loomis, Laura Hibbard, ‘The Auchinleck MS and a Possible London Bookshop of 1330–1340’, PMLA 57 (1942), 595–627 —— ‘The Athelstan Gift Story: Its Influence on English Chronicles and Carolingian Romances’, PMLA 67 (1952), 521–37 Lucas, Henry S., ‘Edward III and the Poet-Chronicler John Boendale’, Speculum 12 (1937), 367–9 Machan, Tim William, ‘Manuscript Culture’, in The Oxford History of Literary Translation in English, I, ed. Ellis, pp. 29–44 Macmillan, Harold, Pointing the Way, 1959–61 (London: Macmillan, 1972) Malden, John, ‘The Double Tressure’, in Emblems of Scotland (Dunfermline: The Heraldry Society of Scotland, 1997), pp. 13–19 Manion, Lee, ‘The Loss of the Holy Land and Sir Isumbras: Literary Contributions to Fourteenth-Century Crusade Discourse’, Speculum 85 (2010), 65–90

446

Bibliography

—— Narrating the Crusades (Cambridge: Cambridge University Press, 2014) Marnette, Sophie, Narrateur et points de vue dans la littérature française médiévale: Une approche linguistique (Bern: Lang, 1998) Marshall, H. E., Stories of Roland Told to the Children (London: T. C. and E. C. Jack, 1907) Marx, C. W., The Devil’s Rights and the Redemption in the Literature of Medieval England (Cambridge: D. S. Brewer, 1995) Massie, Allan, The Evening of the World: A Romance of the Dark Ages (London: Wiedenfeld and Nicholson, 2001) —— Arthur the King (London: Wiedenfeld and Nicholson, 2003) —— Charlemagne and Roland: A Romance (London: Wiedenfeld and Nicholson, 2007) Mathew, Gervase, The Court of Richard II (London: Murray, 1962) Matthes, J. C., ‘Die Oxforder Renaushandschrift, MS Hatton 42, Bodl. 59, und ihre Bedeutung für die renaussage; nebst einem Worte über die übingen in England befindlichen Renausmss’, Jahrbuch für romanische und englische Sprache und Literatur, n.s. 3 (1876), 1–32 McAndrew, Bruce A., Scotland’s Historic Heraldry (Woodbridge: Boydell Press, 2006) McCarthy, Terence, ‘Malory and his Sources’, in A Companion to Malory, ed. Elizabeth Archibald and A. S. G. Edwards (Cambridge: D. S. Brewer, 1996), pp. 75–95 McDiarmid, M. P., ‘Rauf Colyear, Golagros and Gawane, Hary’s Wallace: Their Themes of Independence and Religion’, Studies in Scottish Literature 26 (1991), 328–33 McDonald, Nicola (ed.), Pulp Fictions of Medieval England: Essays in Popular Romance (Manchester: Manchester University Press, 2004) McFarlane, K. B., The Nobility of Later Medieval England (Oxford: Clarendon Press, 1973) McKitterick, Rosamond, Charlemagne: The Formation of a European Identity (Cambridge: Cambridge University Press, 2008) McLagan, Michael, The City of Constantinople (London: Thames and Hudson, 1968) Meale, Carol M., ‘Patrons, Buyers and Owners: Book Production and Social Status’, in Book Production and Publishing in

Bibliography

447

Britain, 1375–1475, ed. Jeremy Griffiths and Derek Pearsall (Cambridge: Cambridge University Press, 1989), pp. 201–38 —— ‘“…alle the bokes that I haue of latyn, English and frensch”: Laywomen and their Books in Late Medieval England’, in Women and Literature in Britain, 1150–1500, ed. Carol M. Meale, 2nd edn (Cambridge: Cambridge University Press, 1996), pp. 128–58 Mehl, Dieter, The Middle English Romances of the Thirteenth and Fourteenth Centuries (London: Routledge, 1967) Mehnert, R., ‘Alt und Neue Fierabras-Fragen’, Zeitschrift für romanische Philologie 60 (1940), 49–62 Ménard, Philippe, Le Rire et le sourire dans le roman courtois en France au moyen âge (1150–1250) (Geneva: Droz, 1969) —— ‘Le Rire et les mentalités médiévales dans Fierabras: Une réflexion sur la violence’, in Le Rayonnement de Fierabras dans la littérature européenne, ed. Marc Le Person (Lyon: Publication du CEDIC, 2003), pp. 97–108 Menichetti, Caterina, and Maria Teresa Rachetta, ‘The Gormond et Isembart Legend in the French Monastic Chronicles of the Thirteenth Century’, in The Proceedings of the Oxford Congress of the International Société Rencesvals, ed. Marianne J. Ailes, Anne E. Cobby and Philip E. Bennett (Edinburgh: British Rencesvals Publications, 2015), pp. 469–94 Metlitzki, Dorothee, The Matter of Araby in Medieval England (New Haven, CT: Yale University Press, 1977) Meyer, Paul, ‘Un récit en vers français de la première croisade fondé sur Baudri de Bourgueil’, Romania 5 (1890), 1–63 —— ‘Fragments de manuscrits français’, Romania 35 (1906), 22–67 Moisan, André, Répertoire des noms propres de personnes et de lieux cités dans les chansons de geste, 5 vols (Geneva: Droz, 1986) Mone, Franz Josef, ‘Die Haimonskinder’, Anzeiger für Kunde der deutschen Vorzeit 6 (1837), cols 189–205 Morris, John, et al., Scottish Book Trade Index, NLS online resource: http://www.nls.uk/catalogues/scottish-book-trade-index Morrissey, Robert, Charlemagne and France: A Thousand Years of Mythology, trans. Catherine Tihanyi (Notre Dame, IN: University of Notre Dame Press, 2003) Muret, Ernest, ‘Fragments de manuscrits français trouvés en Suisse’, Romania 44 (1915–17), 215–20

448

Bibliography

Murray, Sarah-Jane K., ‘Plato’s Timaeus and the Song of Roland: Remarks on Oxford Bodleian MS Digby 23’, Philological Quarterly 83 (2004), 115–26 Negri, Antonella, L’episodio di Vaucouleurs nelle redazioni in versi del ‘Renaut de Montaubon’ (Bologna: Pàtron, 1996) —— ‘In limine du manuscrit Hatton 59 (Oxford, Bodleian Library)’, in La Geste de Doon de Mayence, ed. Boutet, pp. 97–111 Newstead, Helaine, ‘Arthurian Legends’, in A Manual of the Writings in Middle English, 1050–1500, I: Romances, ed. Severs, pp. 38–79 Newth, Michael A. H., Fierabras and Floripas: A French Epic Allegory (New York: Italica Press, 2010) O’Callaghan, Joseph F., Reconquest and Crusade in Medieval Spain (Philadelphia: University of Pennsylvania Press, 2003) Oliver, George, Lives of the Bishops of Exeter and a History of the Cathedral, 2 vols (Exeter: William Roberts, 1861) Opie, Iona, and Peter Opie, The Lore and Language of Schoolchildren (Oxford: Clarendon Press, 1959) Orr, Bridget, Empire on the English Stage, 1660–1714 (Cambridge: Cambridge University Press, 2001) Pagden, Anthony, Worlds at War: The 2,500-Year Struggle between East and West (Oxford: Oxford University Press, 2008) Palmer, J. J. N., England, France and Christendom, 1377–99 (Chapel Hill: University of North Carolina Press, 1972) Paris, Gaston, Histoire poétique de Charlemagne (Paris: Librairie Bouillon, 1905) Partner, Peter, The Lands of St Peter: The Papal State in the Middle Ages and the Early Renaissance (London: Methuen, 1972) Pastoureau, Michel, Les Chevaliers de la Table Ronde (Lathuile: Editions de Gui, 2006) Pauphilet, Albert, ‘Sur la Chanson de Roland’, Romania 59 (1933), 161–98 Payen, Jean-Charles, Le Motif du repentir au moyen âge (Geneva: Droz, 1967) Pearsall, Derek, ‘The Alliterative Revival: Origins and Social Backgrounds’, in Middle English Alliterative Poetry and its Literary Background, ed. Lawton, pp. 34–53 —— ‘The Pleasure of Popular Romance’, in Medieval Romance, Medieval Contexts, ed. Purdie and Cichon, pp. 9–18

Bibliography

449

—— ‘The Metre of The Tale of Gamelyn’, paper given at the Romance in Medieval Britain Conference, University of Bristol, 2014 Pensom, Roger, Literary Technique in the ‘Chanson de Roland’ (Geneva: Droz, 1972) Perkins, Nicholas, and Alison Wiggins, The Romance of the Middle Ages (Oxford: Bodleian Library, 2012) Perry, Curtis, Literature and Favoritism in Early Modern England (Cambridge: Cambridge University Press, 2006) Phillips, Jonathan, Holy Warriors: A Modern History of the Crusades (London: Bodley Head, 2009) Pieper, Josef, Introduction to Thomas Aquinas (London: Faber, 1962) Pope, Mildred K., From Latin to Modern French (Manchester: Manchester University Press, 1934) Popoff, Michel, and Michel Pastoureau (eds), Armorial de Gebre (Paris: Léopard d’or, 2012) Porcheddu, Frederick, ‘Edited Text and Medieval Artifact: The Auchinleck Bookshop and “Charlemagne and Roland” Theories, Fifty Years Later’, Philological Quarterly 80 (2001), 463–500 —— ‘The Cloning of the Karlamagnús Saga in Anglo-French Textual Criticism’, in Sagas and Societies, Proceedings of an International Conference at Borgarnes, Iceland, 5–9 September 2002, ed. Stefanie Würth, Tõnno Jonuks and Axel Kristinsson, online at https://publikationen.uni-tuebingen.de/xmlui/bitstream/ handle/10900/46205/pdf/13_fre~1.pdf;sequence=1 Pouzet, Jean-Pascal, ‘Mapping Insular French Texts? Ideas for Localisation and Correlated Dialectology in Manuscript Materials of Medieval England’, in The Anglo-Norman Language and its Contexts, ed. Richard Ingham (York: York Medieval Press, 2010), pp. 102–29 Pratt, Karen, ‘Medieval Attitudes to Translation and Adaptation: The Rhetorical Theory and the Poetic Practice’, in The Medieval Translator, ed. Roger Ellis (London: Westfield Medieval Publications, 1991), pp. 1–27 Pratt, Karen (ed.), Roland and Charlemagne in Europe: Essays on the Reception and Transformation of a Legend (London: King’s College Centre for Late Antique and Medieval Studies, 1996) Prestwich, Michael, Armies and Warfare in the Middle Ages: The

450

Bibliography

English Experience (New Haven, CT: Yale University Press, 1996) Purdie, Rhiannon, ‘The Search for Scottishness in Golagros and Gawane’, in The Scots and Medieval Arthurian Legend, ed. R. Purdie and N. Royan (Cambridge: D. S. Brewer, 2005), pp. 95–107 —— Anglicising Romance: Tail-Rhyme and Genre in Medieval English Literature (Cambridge: D. S. Brewer, 2008) Purdie, Rhiannon, and Michael Cichon (eds), Medieval Romance, Medieval Contexts (Cambridge: D. S. Brewer, 2011) Putter, Ad, An Introduction to the Gawain-Poet (London: Longman, 1996) —— ‘Story Line and Story Shape in Sir Percyvell of Gales and Chrétien de Troyes’s Conte du Graal’ in Pulp Fictions of Medieval England, ed. McDonald, pp. 171–96 —— ‘The Metres and Stanza Forms of Popular Romance’, in A Companion to Medieval Popular Romance, ed. Radulescu and Rushton, pp. 111–31 —— ‘Ralph the Collier’, in Heroes and Anti-Heroes in Medieval Romance, ed. Cartlidge, pp. 145–58 Putter, Ad, and Jane Gilbert (eds), The Spirit of Medieval English Popular Romance (Harlow: Longman, 2000) Radulescu, Raluca L., and Cory James Rushton (eds), A Companion to Medieval Popular Romance (Cambridge: D. S. Brewer, 2009) Reilly, Bernard F., The Contest of Christian and Muslim Spain, 1030– 1157 (Oxford: Blackwell, 1992) Rex, Richard, The Lollards (Basingstoke: Palgrave, 2002) Ribémont, Bernard, ‘Ius gentium, droit féodal et art du jongleur: Otinel et l’esthétique judiciaire’, in Chanter du geste: L’Art épique et son rayonnement: Hommage à Jean-Claude Vallecalle, ed. Marulène Possamaï-Perez and Jean-René Valette (Paris: Champion, 2013), pp. 401–14 Ricoeur, Paul, On Translation, trans. Eileen Brennan (London: Routledge, 2006) Riddy, Felicity, Sir Thomas Malory (Leiden: Brill, 1987) —— ‘The Auchinleck Manuscript: A Woman’s Book?’, paper given at the Romance in Medieval England conference, University of Bristol, 1992 —— ‘Middle English Romance: Family, Marriage, Intimacy’, in

Bibliography

451

The Cambridge Companion to Medieval Romance, ed. Roberta L. Krueger (Cambridge: Cambridge University Press, 2000), pp. 235–52 Rikhardsdóttir, Sif, Medieval Translations and Cultural Discourse: The Movement of Texts in England, France and Scandinavia (Cambridge: D. S. Brewer, 2012) Riley-Smith, Jonathan, The Crusades: A History, 2nd edn (London: Continuum, 2005) Rist, Rebecca, The Papacy and Crusading in Europe, 1198–1245 (London: Continuum, 2009) Robson, C. A., ‘The Character of Turpin in the Chanson de Roland’ , Medium Ævum 10 (1941), 97–100 Roques, M., ‘L’Élément historique dans Fierabras et dans la branche II du Couronnement Louis’, Romania 30 (1901), 161–83 Rouse, Robert Allen, ‘An Exemplary Life: Guy of Warwick as Medieval Culture-Hero’, in Guy of Warwick, ed. Wiggins and Field, pp. 94–109 —— ‘Crusaders’, in Heroes and Anti-Heroes in Medieval Romance, ed. Cartlidge, pp. 173–83 Rutter, Russell, ‘Printing, Prophecy and the Foundation of the Tudor Dynasty: Caxton’s Morte Darthur and Henry Tudor’s Road to Bosworth’, in Prophet Margins: The Medieval Vatic Impulse and Social Stability, ed. Edward L. Risden, Karen Rathmell Moranski and Stephen Yandell (New York: Lang, 2004), pp. 123–47 Rychner, Jean, La Chanson de geste: Essai sur l’art épique des jongleurs (Geneva: Droz, 1955) Sachier, Hermann, ‘Beschreibung der Cheltenhamer Handschrift 8345’, in Miscellanea di studi critici in onore di Vicenzo Crescini (Cividale: Stagni, 1927), pp. 315–25 Sachs, C., Beiträge zur Kunde altfranzösischer, englischer und provenzalischer Literatur aus französischen und englischen Bibliotheken (Berlin, 1875) St John Hope, William, Viscount Dillon, ‘Inventory of the Goods and Chattels Belonging to Thomas Duke of Gloucester and Seized in his Castle of Pleshy, Co Essex, 21 Richard II (1397)’, Archaeological Journal 54 (1897) 275–308 Sandler, Lucy Freeman, Gothic Manuscripts, 1285–1385, Survey of Manuscripts Illuminated in the British Isles, 2 vols (London

452

Bibliography

and Oxford: Harvey Miller and Oxford University Press, 1986) Sands, D. B. (ed.), Middle English Verse Romances (New York: Holt, Rinehart and Winston, 1966) Scattergood, John, ‘Literary Culture at the Court of Richard II’, in English Court Culture in the Later Middle Ages, ed. V. J. Scattergood and J. W. Sherborne (London: Duckworth, 1983), pp. 29–43 Schein, Sylvia, Fidelis Crucis: The Papacy, the West, and the Recovery of the Holy Land, 1274–1314 (Oxford: Clarendon Press, 1991) Scott, Margaret, ‘Previck and Leckprivick: Onomastic Connections in South-West Scotland’, Nomina 29 (2006), 115–28 Severs, J. Burke (ed.), A Manual of the Writings in Middle English, 1050–1500, vol. I: Romances (New Haven: Connecticut Academy of Arts and Sciences, 1967) Sharpe, R., et al. (eds), English Benedictine Libraries: The Shorter Catalogues, Corpus of British Medieval Library Catalogues 4 (London: British Library in association with the British Academy, 1996) Shenton, Caroline, ‘Edward III and the Symbol of the Leopard’, in Heraldry, Pageantry and Social Display, ed. Coss and Keen, pp. 69–81 Shepherd, Stephen H. A., ‘ “ I have gone for þi sak wonderfull wais”: The Middle English Fragment of the Song of Roland’, Olifant 11 (1986), 219–36 —— ‘Four Middle English Charlemagne Romances’ (unpublished DPhil thesis, University of Oxford, 1988) —— ‘The Ashmole Sir Ferumbras: Translation in Holograph’, in The Medieval Translator, ed. Ellis (1989), pp. 103–21 —— ‘Uninscribed Meaning: VRML Applications in Electronic Editions of Medieval Texts’, at http://myweb.lmu.edu/ sshepherd/vrml33.htm —— ‘“ This grete journee”: The Sege of Melayne’, in Romance in Medieval England, ed. Maldwyn Mills, Jennifer Fellows and Carol M. Meale (Cambridge: D. S. Brewer, 1993), pp. 113–31 —— ‘The Middle English Pseudo-Turpin Chronicle’, Medium Ævum 65 (1996), 19–34 —— ‘ “Heathenic Catechesis” and the Source of Awntyrs B’, Medium Ævum 81 (2012), 1–17

Bibliography

453

Shonk, Timothy A., ‘A Study of the Auchinleck Manuscript: Bookmen and Bookmaking in the Early Fourteenth Century’, Speculum 60 (1985), 71–91 Short, Ian, ‘The Anglo-Norman Translation of the Pseudo-Turpin Chronicle by William de Briane: An Edition and Study’ (unpublished PhD thesis, University College London, 1966) —— ‘The Pseudo-Turpin Chronicle: Some Unnoticed Versions and their Sources’, Medium Ævum 38 (1969), 1–22 —— review of The Old French Johannes Translation of the PseudoTurpin Chronicle, ed. Walpole, and La Traduction du PseudoTurpin du manuscrit Vatican Regina 624, ed. Buridant, Medium Ævum 47 (1978), 123–30 —— ‘Literary Culture at the Court of Henry II’, in Henry II: New Interpretations, ed. Christopher Harper-Bill and Nicholas Vincent (Woodbridge: Boydell Press, 2007), pp. 335–61 —— Manual of Anglo-Norman, 2nd edn (Oxford: ANTS, 2013) Sinclair, Finn E., Milk and Blood: Gender and Genealogy in the chanson de geste (Bern: Lang, 2003) Sinclair, Keith, ‘Fierabras in Anglo-Norman: Some Cultural Perspectives’, in Anglo-Norman Anniversary Essays, ed. Ian Short (London: ANTS, 1993), pp. 361–77 Smyser, H. M., ‘Charlemagne and Roland and the Auchinleck MS’, Speculum 21 (1946), 275–88 —— ‘Charlemagne Legends’, in A Manual of the Writings in Middle English, 1050–1500, I: Romances, ed. Severs, pp. 80–100 Snell, Rachel, ‘The Undercover King: Rauf Coilyear and its English Analogues’, in Medieval Insular Romance: Translation and Innovation, ed. Judith Weiss, Jennifer Fellows and Morgan Dickson (Cambridge: D. S. Brewer, 2000), pp. 133–54 Speed, Diane, Medieval English Romances, 2 vols (Durham: Durham Medieval Texts, 1993) —— ‘Chivalric Perspectives in the Middle English Otuel Romances’, in Medieval Cultural Studies: Essays in Honour of Stephen Knight, ed. Ruth Evans et al. (Cardiff: University of Wales Press, 2006), pp. 213–24 Spiegel, Gabrielle, ‘Pseudo-Turpin: The Crisis of the Aristocracy and the Beginnings of Vernacular Historiography in France’, Journal of Medieval History 12 (1986), 207–23

454

Bibliography

Stimming, A., ‘Die Entwicklungsgeschichte der Destruction de Rome’, Zeitschrift für romanische Philologie 40 (1919–20), 550–88 Stoneman, William P. (ed.), Dover Priory, Corpus of British Medieval Library Catalogues 5 (London: British Library in association with the British Academy, 1999) Stones, Alison, ‘The Egerton Brut and its Illustrations’, in Master Wace: A Celebration, ed. Glyn S. Burgess and Judith Weiss ( Jersey: Société jersiaise, 2006), pp. 167–76 Story, Joanna, ‘Charlemagne and the Anglo-Saxons’, in Charlemagne: Empire and Society, ed. Story, pp. 195–210 —— ‘Charlemagne’s Reputation’, in Charlemagne: Empire and Society, ed. Story, pp. 1–4 Story, Joanna (ed.), Charlemagne: Empire and Society (Manchester: Manchester University Press, 2005) Strickland, Debra Higgs, ‘Text, Image and Contradiction in the Devisement dou monde’, in Marco Polo and the Encounter of East and West, ed. Akbari and Iannucci, pp. 23–59 Strickland, Matthew, War and Chivalry: The Conduct and Perception of War in England and Normandy, 1066–1217 (Cambridge: Cambridge University Press, 1996) Strohm, Paul, ‘The Origin and Meaning of Middle English Romaunce’, Genre 10 (1977), 1–28 Stuckey, Jace, ‘Charlemagne as Crusader? Memory, Propaganda, and the Many Uses of Charlemagne’s Legendary Expedition to Spain’, in The Legend of Charlemagne in the Middle Ages, ed. Gabriele and Stuckey, pp. 153–68 Suard, François, La Chanson de geste (Paris: Presses Universitaires de France, 1993) Sunderland, Luke, Old French Narrative Cycles: Heroism between Ethics and Morality (Cambridge: D. S. Brewer, 2010) —— ‘Bueve d’Hantone / Bovo d’Antona: Exile, Translation and the History of the chanson de geste’, in Rethinking Medieval Translation, ed. Campbell and Mills, pp. 226–42 Tavormina, M. Teresa, ‘The Middle English Letter of Ypocras’, English Studies 88 (2007), 632–52 Taylor, Andrew, ‘Manual to Miscellany: Stages in the Commercial Copying of Vernacular Literature in England’, The Yearbook of English Studies 33, ed. Phillipa Hardman, Medieval and Early Modern Anthologies and Miscellanies (2003), 1–17

Bibliography

455

—— ‘The Self-Presentation of an English Mastiff: John Talbot’s Book of Chivalry’, in Language and Culture in Medieval Britain, ed. Wogan-Browne et al., pp. 444–56 Taylor, Larissa J., et al. (eds), The Encyclopedia of Medieval Pilgrimage (Leiden: Brill, 2010) Thompson, John J., Robert Thornton and the London Thornton Manuscript (Cambridge: D. S. Brewer, 1987) Thomson, John A. F., The Later Lollards, 1414–1520 (Oxford: Oxford University Press, 1965) Tolan, John V., Saracens: Islam in the Medieval European Imagination (New York: Columbia University Press, 2002) —— ‘Looking East before 1453: The Saracen in the Medieval European Imagination’, in Cultural Encounters between East and West, ed. Birchwood and Dimmock, pp. 13–28 —— Sons of Ishmael: Muslims through European Eyes in the Middle Ages (Gainesville: University Press of Florida, 2008) Trapp, J. B., ‘Literacy, Books and Readers’, in The Cambridge History of the Book in Britain, vol. III: 1400–1557, ed. Lotte Hellinga and J. B. Trapp (Cambridge: Cambridge University Press, 1999), pp. 31–43 Turville-Petre, Thorlac, ‘ “Summer Sunday”, “De Tribus Regibus Mortuis”, and “The Awntyrs off Arthure”: Three Poems in the Thirteen-Line Stanza’, Review of English Studies 25 (1974), 1–14, Appendix (pp. 12–14) —— England the Nation: Language, Literature, and National Identity, 1290–1340 (Oxford: Oxford University Press, 1996) Tyerman, Christopher, England and the Crusades, 1095–1588 (Chicago: University of Chicago Press, 1988) —— God’s War: A New History of the Crusades (London: Allen Lane, 2006) —— The Debate on the Crusades (Manchester: Manchester University Press, 2011) Tyler, Elizabeth M. (ed.), Conceptualizing Multilingualism in Medieval England, c. 800–c. 1250 (Turnhout: Brepols, 2011) Underwood, Vernon Philip, ‘An Anglo-Norman Metrical Brut of the XIVth Century’ (unpublished PhD thesis, University of London, 1937) Vale, Juliet, Edward III and Chivalry: Chivalric Society and its Context, 1270–1350 (Woodbridge: Boydell Press, 1982)

456

Bibliography

Vallecalle, Jean-Claude, Messages et ambassades dans l’épopée française médiévale: L’Illusion du dialogue (Paris: Champion, 2006) —— ‘La Réception de la Chronique du Pseudo-Turpin en Europe’, Cahiers de recherches médiévales et humanistes 25 (2013), 465–9 Vallecalle, Jean-Claude (ed.), Le Livre de saint Jacques et la tradition du Pseudo-Turpin (Lyon: Presses universitiares de Lyon, 2011) van Cleve, Thomas Curtis, The Emperor Frederick II of Hohenstaufen (Oxford: Clarendon Press, 1972) van Emden, Wolfgang G., ‘“ La Bataille est aduree endementres”: Traditionalism and Individualism in chanson de geste Studies’, Nottingham Medieval Studies 13 (1969), 3–26 —— La Chanson de Roland (London: Grant and Cutler, 1995) —— ‘La Réception du personnage de Roland dans quelques oeuvres plus ou moins épiques des 12e, 13e et 14e siècles’, in Aspects de l’épopée romane, ed. van Dijk and Noomen, pp. 353–62 —— The Reception of Roland in Some Old French Epics’, in Roland and Charlemagne in Europe, ed. Pratt, pp. 1–30 Venuti, Lawrence, The Translator’s Invisibility: A History of Translation (London: Routledge, 2008) Vielliard, Françoise, Manuscrits français du moyen âge: Catalogue (Cologny-Genève: Fondation Martin Bodmer, 1975) Vincent, Diane, ‘Reading a Christian–Saracen Debate in FifteenthCentury Middle English Charlemagne Romance: The Case of Turpines Story’, in The Exploitations of Medieval Romance, ed. Ashe et al., pp. 90–107 Walden, Lord Howard de, Banners, Standards and Badges from a Tudor Manuscript in the College of Arms (The de Walden Library, 1904) Waley, Daniel, The Papal State in the Thirteenth Century (London: Macmillan, 1961) Wallace, David (ed.), The Cambridge History of Medieval English Literature (Cambridge: Cambridge University Press, 1999) Walpole, Ronald N., ‘The Source MS of Charlemagne and Roland and the Auchinleck Bookshop’, Modern Language Notes 60 (1945), 22–6 Ward, H. D. L., and J. A. Herbert, Catalogue of Romances in the Department of Manuscripts in the British Museum, 3 vols (London, 1883–1910) Warm, Robert, ‘Identity, Narrative and Participation: Defining a

Bibliography

457

Context for the Middle English Charlemagne Romances’, in Tradition and Transformation in Medieval Romance, ed. Field, pp. 87–100 Watson, Nicholas, ‘The Politics of Middle English Writing’, in The Idea of the Vernacular: An Anthology of Middle English Literary Theory, 1280–1520, ed. Jocelyn Wogan-Browne et al. (Exeter: University of Exeter Press, 1999), pp. 331–52 Weever, Jacqueline de, Sheba’s Daughters: Whitening and Demonizing the Saracen Woman in Medieval French Epic (New York and London: Garland, 2008) Weiss, Judith, ‘The Auchinleck MS and the Edwardes MSS’, Notes and Queries 16.12 (1969), 444–6 —— ‘Insular Beginnings: Anglo-Norman Romance’, in A Companion to Romance from Classical to Contemporary, ed. Corinne Saunders, Blackwell Companions to Literature and Culture (Oxford: Blackwell, 2004), pp. 26–44 —— ‘The Exploitation of Ideas of Pilgrimage and Sainthood in Gui de Warewic’, in The Exploitations of Medieval Romance, ed. Ashe et al., pp. 43–56 Westerhof, Danielle, Death and the Noble Body in Medieval England (Woodbridge: Boydell, 2008) Westwood, Jennifer, Stories of Charlemagne (London: Bodley Head, 1972) Wiggins, Alison, ‘Are Auchinleck Manuscript Scribes 1 and 6 the Same Scribe? Whole-Data Analysis and the Advantages of Electronic Texts’, Medium Ævum 73 (2004), 10–26 —— ‘The Manuscripts and Texts of the Middle English Guy of Warwick’, in Guy of Warwick, ed. Wiggins and Field, pp. 61–80 Wiggins, Alison, and Rosalind Field (eds), Guy of Warwick: Icon and Ancestor (Cambridge: D. S. Brewer, 2007) Wilcox, Rebecca, ‘Romancing the East: Greeks and Saracens in Guy of Warwick’, in Pulp Fictions of Medieval England, ed. McDonald, pp. 217–40 Williams, Hywel, Emperor of the West: Charlemagne and the Carolingian Empire (London: Quercus, 2010) Wilson, Derek, Charlemagne: The Great Adventure (London: Hutchinson, 2005) Wilson, Edward, ‘Sir Gawain and the Green Knight and the Stanley

458

Bibliography

Family of Stanley, Storeton, and Hooton’, Review of English Studies, n. s. 30 (1979), 308–16 Wilson, R. M., ‘The Medieval Library of Titchfield Abbey’, Proceedings of the Leeds Philosophical and Literary Society (Literary and Historical Section) 5 (1940), 150–77 —— The Lost Literature of Medieval England, 2nd edn (London: Methuen, 1970) Wingfield, Emily, ‘The Manuscript and Print Contexts of Older Scots Romance’ (unpublished DPhil thesis, University of Oxford, 2010) Wogan-Browne, Jocelyn, ‘ “Ceste livre lisez … chescyun jour”: Women and Reading, c. 1230–1430’, in Language and Culture in Medieval Britain, ed. Wogan-Browne et al., pp. 239–53 —— ‘General Introduction: What’s in a Name? The “French” of England’, in Language and Culture in Medieval Britain, ed. Wogan-Browne et al., pp. 1–13 Wogan-Browne, Jocelyn, Thelma Fenster and Delbert W. Russell (eds), Vernacular Literary Theory from the French of Medieval England (Cambridge: D. S. Brewer, 2016) Wogan-Browne, Jocelyn, et al. (eds), Language and Culture in Medieval Britain: The French of England, c. 1100–c. 1500 (York: York Medieval Press, 2009) Woodcock, Thomas, et al. (eds), Dictionary of British Arms: Medieval Ordinary, 4 vols (London: Society of Antiquaries, 1992–2014) Woodcock, Thomas, and John M. Robinson, The Oxford Guide to Heraldry (Oxford: Oxford University Press, 1990) Wright, Glenn, ‘Convention and Conversion: The Saracen Ending of the Taill of Rauf Coilyear’, Al-Masaq: Islam and the Medieval Mediterranean 14.2 (2002), 101–12 Yeager, R. F., ‘Politics and the French Language in England during the Hundred Years’ War’, in Inscribing the Hundred Years’ War, ed. Baker, pp. 126–57 Yeager, Suzanne, ‘The World Translated: Marco Polo’s Devisement dou monde, The Book of John Mandeville, and their Medieval Audiences’, in Marco Polo and the Encounter of East and West, ed. Akbari and Iannuci, pp. 156–81 Zumthor, Paul, ‘Étude typologique des planctus contenus dans la Chanson de Roland’, in La Technique littéraire des chansons

Bibliography

459

de geste: Actes du colloque de Liège (Paris: Les Belles Lettres, 1959), pp. 219–35 —— Essai de poétique médiévale (Paris: Seuil, 1972) —— ‘Intertextualité et mouvance’, Littérature 41 (1981), 8–16

Index Page numbers in bold refer to the main discussion of each text or manuscript. Acre, 72–3, 121 Advocates Library, 217 Aebischer, Paul, 120 n. 23, 346 Agincourt, 237 Aistulf, 308 Akbari, Suzanne Conklin, 7 Aksan, Virginia, 404 n. 9 Anglo–Saxon Chronicle, 54 Ambroise, Estoire de la Guerre Sainte, 114 n. 12 Anselm, Meditations, 206 Antioch, 41 Antioch Chamber, Westminster, 9 Aquinas, Thomas, 312–13, 315 Ariosto, Ludovico, Orlando Furioso, 402–3, 410 Arnold, Thomas, 48 Arthour and Merlin, 97 Arthur, King, 2–3, 8, 141–4, 146, 172, 211, 244–7, 402, 409 Arundel, Archbishop Thomas, 163, 200 Ashmolean Museum, 6 Ashton, John, 405 Asloan, John, 218, 400 Aspremont, 12 n. 41, 13 n. 45, 24, 38–9, 44, 46–9, 52, 58, 70, 72, 132–3, 152, 228 n. 18, 406 n. 15 Athelstan, 131 Audita tremendi, 332 Awntyrs of Arthur, 396–8 Babington Plot, 256 Babylon, 194, 336–40, 344

Bacon, Roger, 20 Bagnyon, Jean, Fierabras: see Fierabras Bajazet, the Great Turk, 383–4 Ball, John, 256 Barbara, Saint, 169 Barbour, John, Bruce, 131, 215 n. 161, 219, 265 n. 4 Barron, W. R. J., 36, 97, 235 n. 37, 288 n. 62, 394 Baudri de Bourgueil, Historia Ierosolimitana, 9 n. 33, 41–2, 73 Bayard’s Colts, 410–11 Beadle, Richard, 183–4 Beattie, William, 216 Beauchamp, Guy de, Earl of Warwick, 48–50 Beauchamp Pageant, 13 n. 43, 375 Beauchamp, Richard, Earl of Warwick, 375 Bédier, Joseph, 122, 124 Beer, Jeanette, 20, 23 Belgrade, 376 Bell, David, 46–7 Bellamy, John, 256 Bennett, Michael, 15 Berman, Antoine, 26 Bertrand de Bar sur Aube, 37–8 Bibbesworth, Walter de, 153–4 Blaess, Madeleine, 49 n. 63, 50 Blake, Norman, 17, 211 n. 150 Boeve de Haumtone, 10 n. 34, n. 36, 26 n. 92, 36 n. 16, 48 n. 58, 65, 67, 149, 151 n. 97 Book of John Mandeville: see Mandeville Bordesley Abbey, 48–51

461

462

Index

Boulogne, 131, 335 Bowers, John M., 29 Bradbury, Nancy Mason, 395 Brandin, Louis, 146 Brault, Gérard, 42, 224 n. 8, 229 n. 21 Bretigny, Treaty of, 139 Bristol, 197, 201–2 Brut: see Wace Brut (prose), 238 Brutus, 4, 140 Burgundy, 133, 279, 379 Busby, Keith, 44 n. 40, 45 n. 44, 149 n. 90, n. 92, 150 n. 95, 151 n. 99 Butterfield, Ardis, 15, 27 n. 96 Byzantium, 119, 123 n. 35 Calin, William, 16, 34 n. 8 Calixtus III, 332 Calkin, Siobhain Bly, 98 n. 212, 99 n. 214, 182 Campbell, Mary, 341, 343 Camps, Jean–Baptiste, 39 n. 24, 150 nn. 94–5, 406 n. 16, 415 Canterbury, St Augustine’s, 48, 74 n. 149 Carlisle, 246–7, 397 Caxton, William, 2, 3, 8 n. 28, 17, 21–3, 26, 28–9, 43, 248, 294, 409 Four Sons of Aymon, 21–2, 42, 80–1, 86, 209–14, 402, 406 n. 15 Godeffroy of Boloyne, 72, 214 Lyf of Charles the Grete, 21–2, 78 n. 162, 81, 86, 106, 209–14, 244, 247 n. 68, 250, 261–2, 293, 389 n. 92 Morte Darthur, 214 Chanson d’Antioche, 9, 304 n. 86 Chansons de geste, 9–11, 17, 21, 26–8, 32–4, 36–8, 40, 46, 49 n. 64, 50–2, 54, 57–8, 64–73, 81–2, 85–6, 88, 96–8, 102, 108–10, 122–3, 126 n. 40, 137, 141, 150–1, 159 n. 6, 184, 209,

222–4, 227–8, 260, 272, 286 n. 59, 314, 317, 325, 330, 339, 346–7, 354 n. 33, 356, 370, 388, 393, 397, 406, 409 Chanson de Guillaume, 33, 38–40, 58, 64, 67, 72, 224, 249 n. 78, 317 n. 121 Chanson de Jérusalem, 9 Chanson de la première croisade, 41 Chanson de la reine Sibille, 45 Chanson de Roland, 3, 9, 33, 37, 38–9, 53, 58, 64, 67, 69, 70, 99–100, 110–11, 113–15, 120–2, 135–7, 155, 173, 221–47, 250, 256, 264–5, 276, 312–13, 317–18, 323, 338, 365, 412 Rhymed Roland, 70, 110–11, 222–3, 226, 236, 238 n. 48, 246 n. 65, 373 n. 52 Charlemagne, passim; see particularly as Emperor of Christendom, 1, 30, 63, 99, 114–5, 133, 138, 146–8, 153, 170–2, 176, 221, 228, 231–4, 274, 323, 363–4 as Father of Europe, 1, 5 as King of France, 30, 101, 104, 146–7, 170, 366, 390, 403 as ‘rex pacificus’, 173 his Journey to the East, 5, 63–4, 123, 165, 252, 391–2; see also Descriptio qualiter; Pèlerinage de Charlemagne ‘Charlemagne and Roland’, 76–8 Charlemagne, or The Distracted Emperor, 403 Charles V, 141 Charles VIII, 217 n. 165 Charteris, Henry, 215 n. 161, 219 Chartier, Alain, 339 Chartres, 5, 259 Chaucer, Geoffrey, 29, 219 n. 170, 343 Book of the Duchess, 234 Canterbury Tales, 279–80, 342, 345 Troilus and Criseyde, 280–1 Chepman, Walter, 216, 218

Index Chertsey tiles, 11–12 Chevalerie Ogier de Danemarche, 44, 47 Children, Charlemagne stories retold for, 405–6 Chrétien de Troyes, 98 n. 209, 246, 286 n. 59 Christendom, 4–5, 84, 115, 117, 119, 122, 133, 139, 155, 157, 170–1, 176, 181, 195, 207, 209, 232, 277, 297, 310–11, 333–5, 364–7, 377–8, 380, 385–6, 388, 392, 402, 407, 409, 411 Clarendon palace, 11 Clement V, 332 Cobby, Anne, 64, 133 n. 51 Coke, John, The Debate between the Heralds of England and France, 408 Coldiron, Anne, 28 Complaynt of Scotland, 81 Constantinople, 63, 119, 121, 123, 133, 165, 195, 296, 336, 391, 404 n. 11 Conversion, 12, 72, 100, 104, 106, 108, 111–15, 118, 120, 135, 151 n. 98, 157–64, 171, 173, 176, 180–1, 185, 192, 198, 201, 245, 250, 252–3, 264, 271, 276, 278–9, 283, 286–7, 302, 314, 317, 319, 332, 335, 340, 344, 347–8, 351, 358, 361–2, 364, 367, 369, 371, 374–6, 382, 384–6, 391–5, 399 Cooper, Helen, 209–10, 244, 261, 284, 288, 294, 402–3 Copeland, Rita, 18–19 Copland, Robert, 210 Courtenay family, Earls of Devon, 280 Cowen, Janet, 31 n. 113, 169, 382 n. 76 Crécy, 15, 237 Crofts, Thomas, 99, 104, 390 n. 94, 397 Crosland, Jessie, 322 Crowland Abbey, 54 Crusade and Death of Richard I, 10 n. 36

463

Crusades, 5–6, 8–11, 31, 41–2, 72–3, 84, 113–22, 132, 139, 144, 154, 189, 207–8, 276–7, 279–80, 309–11, 333, 336, 408 Cruz, Jo Ann Hoeppner Moran, 340 n. 154 Cunningham, I. C., 218 Cursor Mundi, 74 n. 150, 163, 187–8 Cycle de Mayence, 38, 40, 42–3, 49–50, 52, 69, 132, 134 Cycle de Guillaume (Monglane), 37, 48–9, 52 Cycle des croisades, 9, 37, 41 n. 30, 42, 72 Cycle du roi, 38, 46–7, 52, 58, 68–9, 71, 98, 234, 347 Dafydd ap Gruffydd, 256 Daubeny, William, 212–13 Daurel et Beton, 51 Davenant, Sir William, The Siege of Rhodes, 404–5 Dean, Ruth and Maureen Boulton, 38–40, 54 n. 87, 120 n. 23 Débat des hérauts d’armes de France et d’Angleterre, 408 De Mandach, André, 265 n. 2, 267–9, 303–5 Descriptio qualiter Karolus Magnus clavum et coronam Domini a Constantinopoli Aquisgrani detulerit, 12 n. 42, 123 n. 35 Despenser, Henry, 153–4, 311 n. 107 Destruction de Rome, 13, 36 n. 16, 39–40, 57, 59, 61, 63, 67–8, 73, 87, 94, 117–19, 122, 127–8, 131–2, 138, 144, 156, 174, 193, 265 n. 2, 266, 281, 301–16, 317, 330–1, 333–4, 336, 350 n. 20, 410 n. 25, 414–15 Devils’ Parliament, The, 198–9 DeVries, Kelly, 379 Dover Priory, 46, 55

464

Index

Duggan, Joseph, 235, 246 n. 65, 412 Duke Rowland and Sir Otuell: see Roland and Otuel Edinburgh, 214–16, 218–19, 400 Edward I, 65, 73, 141, 256, 396 Edward II, 139 n. 65, 256 Edward III, 15, 16, 44 n. 41, 51, 138, 139–42, 144, 237, 247 n. 68, 255, 276, 279, 390 Edward IV, 211–13, 262 Edward of Woodstock (Black Prince), 140 Edward, Saint, 3 Einhard, Vita Karoli Magni, 52, 409 Eleanor of Aquitaine, 63 Eleanor of Castile, 9 Elizabeth I, 256 Elizabeth of York, 212–13 Ellis, George, 198 n. 111, 405 Ellis, Roger, 401 Enfances Ogier, 147 n. 86 English Charlemagne Romances (EETS), 4, 83, 412–17 Eremyte and the Owtelawe, The, 198–9 Estorie del Euangelie, 204, 206 Fantosme, Jordan, 65 Field, Rosalind, 11, 16, 34, 69, 83, 88, 151 Fierabras, 13, 14, 37, 39, 40, 44, 49–51, 57–61, 64, 67, 69, 72–3, 78, 81, 87, 90, 93 n. 199, 94–6, 101, 103–6, 110–13, 115–18, 120–1, 123–4, 126–31, 133–5, 138–9, 144–5, 149, 155, 158–9, 165–7, 172, 177, 194, 200, 209–10, 222, 226, 233–4, 262–3, 264–345, 346–8, 351–2, 362, 364, 377, 381 n. 72, 383, 387–9, 394, 397, 399, 403, 406–7, 410, 413–14 Bagnyon’s Fierabras, 247 n. 68, 250 Fierenbras (Anglo–Norman), 24,

39, 57, 61, 64 n. 127, 67–8, 116–17, 128 n. 41, 130–2, 134–5, 163 n. 10, 174, 193, 265–6, 272 n. 31, 305–7, 316–17, 319–21, 323–4, 325–30, 334–5, 337, 414–15 Finlayson, John, 83, 188 n. 75 First Crusade (1096–9), 9, 41–2, 72–3, 189 Firumbras, 83 n. 177, 87, 89, 94–5, 101, 157, 159, 165, 168–9, 172–3, 176, 197–8, 200–1, 267, 287–301, 325 n. 130, 336 n. 146, 352, 361, 373 n. 53, 375, 387–8, 414 Fischer, Steven, 204–5 Fischer, Walther, 90 Fitzgerold, Warin, 54, 153–4 Flodden, 401 Florence of Rome, Saint (Fierabras), 158 Formisano, Luciano, 62 Fouques de Candie, 40, 50 n. 72, 72 Foxe, John, 203, 207–9 Actes and Monuments, 203, 207, 208 n. 144 Frederick II, 309, 311, 409 Freeman, Charles, 166 n. 17 Froissart, Jean, 237, 383–4, 388 Frolle, 141–2, 144 Fulcher of Chartres, Historia Hierosolimitana, 54 Furrow, Melissa, 3 n. 11, 34, 47–9, 51 n. 75, 58–9, 69, 82–3, 88, 113 Gaimar, Estoire des Engleis, 35, 231 n. 25 Gaunt, Simon, 222 n. 4, 225 n. 11, 226 Gautier, Léon, 14 n. 46, 221 n. 2, 267 Geoffrey of Anjou, 231–2 Geoffrey of Monmouth, Historia Regum Britanniae, 35, 55–6, 138, 152 George, Saint, 11 Gilte Legende, 169

Index Girart de Vienne, 37–8, 49–50 Godeffroy of Boloyne, 8 n. 28, 21 n. 73, 213 n. 155, 214 Golagrus and Gawain, 218–19, 396–7 Golden Legend (Legenda Aurea), 158, 259 n. 95, 340 Gormont et Isembart, 33, 35, 39, 48, 58 Gower, John, Confessio Amantis, 7 n. 23, 276–9 ‘In Praise of Peace’, 277–8 Granada, 376–7, 393 Gratian, Decretals, 308 Gray, Douglas, 395 Greenblatt, Stephen, 26 Greene, Robert, 403 Gregory I (the Great), 308 Gregory VIII, 315 Gregory IX, 309–10 Gröber, Gustav, 61 Guard, Timothy, 139 n. 65, 280 Gui de Bourgogne, 45, 48, 74 n. 149 Gui de Warewic, 46 n. 47, 48 n. 58, 150–1 Gunpowder Plot, 259 Guy of Warwick, 10 n. 34, n. 36, 48, 85, 92, 180, 210, 267 n. 5, 402 Haidu, Peter, 232–3 Halliwell, J. O., 405 Harington, Sir John, 402 Hastings, 231, 311, 406 n. 14 Hattin, 115 Hausknecht, Emil, 4–5, 193 Heinemann, Edward, 328 Heinermann, Theodor, 63 Hellinga, Lotte, 22, 210 n. 146 Heng, Geraldine, 10, 392 Hengist, 141 Henry II, 232 Henry III, 11, 137, 310 Henry V, 16, 237, 390 Henry VI 3, 40, 140, 196, 211, 390

465

Henry VII, 43, 211–13, 262, 376 n. 63, 400 Henry VIII, 43, 211, 401 Henryson, Robert, Moral Fables, 219 Heraldry, 42–3, 139–48, 177, 195, 197 n. 105, 302, 306, 330, 336, 354, 390, 400 Heralds’ Roll, 42 Heresy, 163, 201–2, 257, 375–6, 409 Herrtage, Sidney J. H., 4, 96 n. 206, 212, 281 n. 50, 377 Higden, Ranulph, Polychronicon, 56, 195–6 Higgins, Iain, 340–1, 343, 345 n. 163 Hilka, Alfons, 268 Holland, Richard, The Buke of the Howlat, 219 Holy Land, 5–6, 9, 13, 72–3, 85, 114 n. 12, 115, 121, 123 n. 35, 165 n. 16, 172, 189, 207–8, 214, 219, 332, 336, 375, 392, 397 Holy Roman Empire, 146, 365 Horrent, Jules, 63–4, 229 n. 20 Hundred Years’ War, 2 n. 9, 14, 16, 31, 73, 133, 138, 173, 390 Hungary, 208 n. 144, 377, 383–4 Hutchings, Mark, 404 n. 11 Indagine, Joannes ab, 215 Innocent III, 309–10, Innocent IV, 207, 309–10 Isabella of France, 51 Islam, 7 n. 24, 84, 138, 278 n. 42, 339–40, 343, 345, 366–7, 376, 407 James IV of Scotland, 400–1 James, Saint, 28, 123, 164 Jerome, Saint, 194 Jerusalem, 5, 9, 63, 73, 123, 165, 188, 252, 276, 309 n. 99, 332, 391–2 Johannes Translation: see Pseudo– Turpin Chronicle

466

Index

John the Reeve, 396 Karlamagnús saga, 74, 78 n. 163 Keen, Maurice, 279–80 Keiser, George R., 188, n. 77, n. 79, 189 n. 82 Kennedy, Edward, 238 King Horn, 256 n. 89 Konick, Marcus, 272 n. 31, 273 Kosovo, 379 Kyng Alisaunder, 97 Laisse form, 32–3, 36 n. 16, 59, 65–6, 68, 81, 86–91, 93, 95–7, 108, 226, 242, 317, 326–9, 365 Lancastrian affiliation, 196, 211, 261 Laneham, Robert, 81 n. 169 Langley, Henry de, 137 Langlois, Ernest, 353 Langtoft, Peter of, 65 Lateran Council, Fourth (1215), 72, 163, 199, 333 n. 141 Latowsky, Anne, 63 Lay le Freine, 79 Lee, Sir Christopher, 410 Leedham, Suzanne, 403, nn. 5–6, 410, n. 24 Legge, Dominica, 34, 57, 65 Leitch, Megan, 257 Lejeune, Rita, 259 Lekpreuik, Robert, 81, 214–20, 417 Le Morte Arthur (stanzaic), 239 Lendit fair, 124, 126, 130–1, 155, 266–7, 351 Leo IV, 308 Lepanto, 404 n. 9 Le Person, Marc, 112, 146 n. 84, 149 n. 92, 265–6, 270, 316 Libbon, Marisa, 15, 99 n. 212 Liturgy, 227, 331–2, 340 Livingston, Michael, 189, 277 ‘Livre de Charlemagne’, 44, 78 n. 162

Lollards, 200–2, 300, 376 Lombardy, 13–14, 190, 366, 376–8, 383, 385, 386 n. 86 Loomis, Laura Hibbard, 77, 131 Louis (son of Charlemagne), 33, 35, 48 Louis VI, 305 Louis VII, 63 Louis IX, 207, 311 Louis, Saint, 3 Love, Nicholas, 163 n. 11 Lupack, Alan, 395, 407 Lydgate, John, 3, 193, 196 Macaire: see Chanson de la reine Sibille Machaut, Guillaume de, La Prise d’Alexandrie, 279 Macmillan, Harold, 1 Malory, Sir Thomas, Morte Darthur, 181 n. 48, 182 n. 49, 238–9, 244, 246–7, 257 n. 90, 293, 370 Mamluks, 208, 279 Mandeville, John, 332, 338–41, 343, 345, 375 Manion, Lee, 72 n. 145, 84–5, 103 n. 220 Manuscripts Cologny Geneva Bodmer Library MS 168, 39, 149–51, 179, 346 n. 1, 354 n. 32, 365, 415 Edinburgh NLS MS 19.2.1 (Auchinleck), 15, 75–6, 87–8, 95, 98, 177 n. 32, 178–83, 350 n. 16, 352, 394, 413, 416 Edinburgh NLS MS 16,500 (Asloan), 218–20 Hanover Niedersächsische Landesbibliothek MS IV 578, 39, 61, 116, 120, 127–30, 144–6, 265, 269, 271, 307, 316, 413–14 San Marino, Huntington Library MS HM 28,561, 83 n. 177,

Index 177 n. 33, 195–7, 413 London BL MS Add. 31042 (Thornton), 87–8, 100, 187–92, 352, 377, 383, 390, 416 London BL MS Add. 34114, 41 n. 30 London BL MS Add. 37492 (Fillingham), 74, 83 n. 177, 87, 100, 107 n. 229, 159, 197–202, 257, 287, 294 n. 70, 349 n. 15, 352, 361, 374–6, 414, 416 London BL MS Add. 38663, 33, 40 London BL MS Arundel 220, 56, 152–4, 413 London BL MS Egerton 1994, 403 London BL MS Egerton 3028, 24 n. 86, 34, 39, 61, 87, 120, 128, 138–44, 145, 151, 154, 158, 197, 247 n. 68, 265–6, 270, 304–5, 307, 316, 325, 330, 390, 400 n. 123, 414–15 London BL MS Lansdowne 388, 88, 202–9, 412 London BL MS Royal 2 B vii (Queen Mary Psalter), 179 London BL MS Royal 15 D ii (Welles Apocalypse), 78 n. 162 London BL MS Royal 15 E vi (Shrewsbury Book), 15, 40, 43, 78 n. 162, 136, 146–8, 196, 271, 282 n. 53, 286 n. 60, 370 n. 49 London BL MS Royal 16 G ii, 35 n. 12, 43–4, 132 n. 50 London, College of Arms MS Vincent 170 (Kings of Britain Roll), 140 Louvain University Library MS G 171 (Didot), 149, 267–8 Oxford Bodleian Library MS Ashmole 33, 87, 92 n. 198, 158 n. 3, 183–7, 273–4, 280 n. 46 Oxford Bodleian Library MS

467

Bodley 264, 339 Oxford Bodleian Library MS Digby 23, 2 n. 8, 33 n. 4, 136–7, 222–3, 235, 412 Oxford Bodleian Library MS Hatton 59, 40 Oxford Bodleian Library MS Hatton 77, 41 n. 30, 42 n. 31 Paris BnF MS fr. 25408 (fragment), 152 Paris BnF MS nouv. acq. fr. 5094 (Mende fragment), 39, 120 n. 23, 152, 346, 348, 353–4 Princeton University Library MS Garrett 140, 87, 192–5, 415 Vatican City, Biblioteca Apostolica Vaticana MS Reg. lat. 1616, 112, 149, 346 n. 1, 349 n. 13, 415 Marco Polo, Description of the World, 338–42 Margaret of Anjou, 15, 40, 43, 196, 211 Margaret of England, 400 Marie de France, 45 Markward of Anweiler, 310 Marx, C. W., 198–9, 202 Massie, Alan, 409–10 Matilda, Empress, 232 Maugis d’Aigremont, 43 McKitterick, Rosamund, 8 McSparran, Frances, 199 Meale, Carol M., 192–3, 195 Mehl, Dieter, 83, 236, 288, 351, 371, 395 Mehnert, Rudolf, 131 Mézières, Philippe de, 279 Milan, 14, 189, 383 Miracles of the Virgin, 167 n. 19 Morte Arthure, 96, 172 Mouskès, Philippe, Chronique rimée, 35 n. 14, 60, 302 n. 81, 305 Mull family (Thomas and William), 177 n. 33, 195–7, 261 Muslims, 6, 8–11, 121–2, 278, 297, 310,

468

Index

331–2, 336, 339–40, 377, 384, 392–3 Myrrour of Mankind, The, 198–9 Neckham, Alexander, De naturis rerum, 42 Negri, Antonella, 41 Neville, Cecily, 43 n. 37, 132 n. 50 Newstead, Helaine, 239 Nicopolis, 296, 379–80, 383–4 Nine Worthies: see Worthies Normandy, 232, 268–9, 271, 364–5 Northern Passion, 187, 190–1 Norwich, 153–4, 311 n. 107 Notre Dame, Paris, 124–5, 129–31, 152, 335 Octavian, 46, 102, 103 n. 220, n. 222 Octovian Imperator, 102–3, 338 Odo of Bayeux, 311 Offa of Mercia, 228 Oldcastle revolt, 375 Olive and Landres, 74 Orr, Bridget, 404 n. 10 Osney Abbey, 137 O’Sullivan, Mary I., 288, 373 n. 53 Otinel, 13, 14, 39, 45, 52, 58, 68–9, 71–3, 76–9, 87, 90, 91 n. 197, 95–6, 101–4, 106, 110–13, 115, 118, 120–1, 133, 135, 149, 152, 155–7, 165, 170–1, 179 n. 36, 190, 201, 209, 233–4, 250, 253, 257 n. 91, 262–3, 346–401, 406, 410, 415–16 Otto IV, 309 Ottoman Empire, 208, 296, 377, 380, 404 Otuel, 76–7, 87, 89, 95–6, 98, 100, 108, 157, 170, 181–2, 349–50, 352, 364 n. 41, 366–71, 380 nn. 70–1, 393–4, 405, 416 Otuel and Roland, 56 n. 98, 74–9, 83 n. 177, 87, 89, 95–6, 100, 107, 157, 159–60, 171, 198, 201, 243–4, 250,

253–9, 294 n. 70, 348–50, 352, 361, 371–7, 380 nn. 70–1, 391, 405, 416 Papal States, 308–9, 316 Pardons (indulgences), 121, 190, 200–1, 299–300, 309 Paris, 13, 104, 123–6, 129, 131, 133, 155, 167 n. 18, 219, 299 n. 77, 335, 363, 380, 396 n. 115, 397 Paris, Gaston, 64 n. 125, 124 Paris, Matthew, 65, 207 Vie de saint Auban, 65 Pascal III, anti–Pope, 168 Passion relics, 12 n. 42, 13, 59, 63, 73, 115–16, 123–4, 126–7, 129–31, 164–6, 168, 193, 200–1, 252, 264, 266–7, 297–8, 300–2, 305, 333–4, 337, 341 Paul, Saint, 158, 310 Pearsall, Derek, 79 n. 166, 83, 86, 95 n. 202, 184–5 Pecham, Archbishop John, 163 Pèlerinage de Charlemagne, 34, 58, 63–4, 123, 130, 133 Peter I of Cyprus, 276, 279 Peter, Saint, 117, 231, 310 Peterborough Abbey, 45–6, 55 Peter’s Pence, 228, 231, 233 Philip II (Philip Augustus), 124 Philip VI, 139 Philippa of Hainault, 51 Phillips, Jonathan, 9, 11, 296, 376, 408 Pigouchet, Pierre, 216 Piramus, Denis, 66 Plato, Timaeus, 137 Porcheddu, Frederick, 77, 78 n. 163 Pratt, Karen, 18 Pseudo–Turpin Chronicle, 3, 12, 25, 28–9, 32, 46–8, 52–7, 69–70, 73, 77–9, 111, 113–15, 120, 122, 138, 144, 153–5, 161, 169–71, 176, 222, 238, 243–4, 247–63, 264, 332, 371–5, 391– 2, 405, 410, 413

Index Anglo–Norman Pseudo–Turpin, 27–8, 53, 54, 73, 114, 120, 153, 247–50, 260, 413 Johannes Translation, 12, 53, 87, 88, 99, 101, 165, 170, 251–3, 391 Middle English Pseudo–Turpin (Turpines Story), 86, 162, 164, 170, 176, 195–7, 244, 247, 251, 259–61, 373, 376, 413 Purdie, Rhiannon, 77–9, 86–7, 89–90, 95, 101 n. 216, 350 Purvey, John, 203 Putter, Ad, 94 n. 201, 186 n. 71, 395–6, 399, 409 Quatre Fils Aymon, 22, 40, 42, 48–9, 79–80, 210 Rauf Coilyear, 71, 84, 88, 96–8, 104, 158, 161–2, 176, 214–220, 352, 381–2, 394–401, 406, 417 Relics: see Passion relics Renaud, Count of Boulogne, 53, 248 Renaud de Montauban, 24, 40–4, 46, 48, 52, 68, 71–2, 79–80, 132, 210, 410 Riccoldo da Montecroce, 393 Richard Coeur de Lion, 10, 12, 14, 15, 72, 85, 172, 189, 296 n. 72 Richard I (the Lionheart), 10–13, 114 n. 12 Richard II, 279–80 Richard III, 212–13, 262 Richard of Cornwall (King of the Romans), 310–11 Riddy, Felicity, 179 n. 40, 195 n. 98 Rigord, Gesta Philippi Augusti, 130 Riley–Smith, Jonathan, 121, 332 Rist, Rebecca, 310 Robert the Bruce, 131, 220 Rochester Priory, 54 Roland and Otuel, 87, 95, 100, 102, 157,

469

169–70, 172, 187, 189–91, 348 n. 10, 349–50, 352, 377–81, 383, 385–6, 390, 416 Roland and Vernagu, 6 n. 20, 12, 13, 53, 74–8, 87–9, 96, 98–100, 102, 114–15, 161–3, 165, 169–171, 178–9, 181–2, 228 n. 18, 250, 252–4, 264, 352, 371, 381, 391–4, 405, 413 Roman de Horn, 36 n. 16, 65, 67, 151 n. 97 Rome, 2, 13, 59–60, 80, 91, 99, 103, 115, 117, 122–8, 130 n. 45, 131, 155, 158, 173, 190, 194–5, 208, 228, 264, 265 n. 2, 273, 276–7, 301–2, 305–9, 311–12, 314, 316, 331, 333–5, 337, 339, 342, 344, 347, 377, 383, 409 Roncevaux, 53, 79 n. 164, 110–11, 155, 190, 221–63, 312, 323, 333 n. 144, 366, 371–2, 374, 405, 410 Rouse, Robert, 72–3, 99, 104, 386 n. 87, 390 n. 94, 397 Rutter, Russell, 213 Rychner, Jean, 68 n. 138, 328 Saints: see under the saint’s name St Albans Abbey, 228 St Andrews, 214–15 Saint–Denis, 124–5, 129–30, 267, 303–4, 330, 335, 396 n. 115 Saladin, 10–13, 115, 310 Santiago de Compostela, 123, 253 Saracens, passim; see particularly ‘belle Sarrasine’, motif of, 60, 112, 265 n. 2, 291, 347, 375, 407 n. 18 ‘noble Saracen’, motif of, 108, 182 n. 50, 344, 347, 369, 375, 385–6, 405 Schein, Sylvia, 121 Scotland (Escoce), 4, 211, 229–30, 246 n. 66; see also Rauf Coilyear Seege of Troye, 97 Sege of Melayne, 14, 71, 81, 84, 88, 89, 100, 158, 161, 165–7, 172, 187, 190–1,

470

Index

280 n. 47, 310 n. 101, 352, 377, 379–81, 382–90, 391, 394, 401, 406, 416 Segré, Césare, 223 Shepherd, Stephen H. A., 90–1, 93 n. 199, 171 n. 25, 183, 186 n. 73, 195–7, 235, 250–1, 260–1, 397 n. 116 Short, Ian, 3, 248–9 Shrewsbury Book: see Manuscripts, BL Royal 15 E vi Sicily, 310–11 Siege of Jerusalem, 187–9, 191 Siège/Prise/Estoire d’Antioche, 41 Simon de Puille, 44 Sinclair, Keith, 58 n. 104 Single combat, motif of, 11–13, 31, 102–4, 113–15, 133, 157–8, 162, 180–1, 233, 243, 254, 264, 270–1, 275, 281–3, 308 n. 91, 314, 352, 369–70, 374, 386, 397–9, 403 Sir Ferumbras, 26, 70, 87, 89–91, 93–5, 100–2, 157–9, 168, 170, 174 n. 30, 176, 183–7, 272–87, 293, 295 n. 71, 296 n. 73, 325 n. 130, 336 n. 146, 367 n. 45, 388 n. 90, 414 Sir Gawain and the Green Knight, 186, 284, 399–400 Sir Isumbras, 210 Sir Orfeo, 79 Skelton, John, 80–1, 266 n. 4 Smyser, H. M., 74, 77, 80, 89, 98, 165, 235 n. 37, 273, 330 Somnia Danielis, 204 Song of Roland (Middle English), 70, 88, 96, 101, 110 n. 1, 156, 173, 202–3, 205–7, 209, 234–47, 250–1, 262, 266, 321, 333 n. 144, 391, 412–13 Sowdone of Babylone, 13, 87, 89, 94, 95 n. 203, 101–2, 156, 158, 162, 168, 172, 174, 192–4, 265, 267, 272 n. 31, 282 n. 51, 283, 285–7, 291, 306, 315–16, 318 n. 122, 322 n. 127,

330–45, 366, 374 n. 56, 387 n. 87, 405–6, 415 Spain, 5, 13, 79, 113, 115, 121, 138–9, 164, 169, 171–3, 194, 222, 233, 246, 249, 252, 337, 344, 349, 371, 376, 391–3, 407 Speed, Diane, 377 n. 64, 394–6, 400 Spiegel, Gabrielle, 54 Stanley, Sir William, 213 Stimming, Albert, 62 Stones, Alison, 138–40, 144–6 Strickland, Debra, 339 Strickland, Matthew, 311 n. 107 Strype, John, 202–3, 209 Sunderland, Luke, 26 n. 92 Talbot, John, Earl of Shrewsbury, 43, 69, 146–8, 196 Taylor, Andrew, 15, 179 n. 38 Third Crusade (1189–92), 10, 11, 114 n. 12 Thomas, Duke of Gloucester, 44 n. 41 Thompson, John, 187–9 Thomson, J. A. F., 201–2 Thoresby, Archbishop John, 163, 199, 201–2 Thornton, Robert, 172, 187–9, 202, 237 Titchfield Abbey, 47 Tolan, John V., 278 n. 42, 331, 340, 367 n. 44, 376 n. 61, 393 Translation, passim; see particularly 15–30, 54, 69–71, 98, 185–6, 238–9, 247–50, 259–61, 272–3, 287–93, 401 Treason, 211, 213, 243, 254–9, 261–2 Trevisa, John, 177 n. 33, 195–6 Trial by combat: see Single combat Tristan legend, 45, 234–5 Turks, 31, 73, 106, 108, 119, 121, 195, 207–9, 220, 277–9, 297, 331–2, 359, 366, 376, 379–80, 383–4, 389, 404 Turpines Story: see Pseudo–Turpin

Index Chronicle Turville–Petre, Thorlac, 98 n. 212, 179 n. 41 Tyerman, Christopher, 8, 10, 73, 139 n. 69, 208, 379, 392 Underwood, Vernon, 138 Usk, Thomas, 256 Vale, Juliet, 51 Venuti, Lawrence, 18 n. 60, 26, 28 Vere, John de, Earl of Oxford, 29 n. 105, 43, 210–11 Verse forms, 23, 32, 82, 86–97, 237, 288, 290–1, 371, 397; see also laisse Vie de Seint Clement, 27, 30, 66 n. 133 Vielliard, Françoise, 151 Vienna, 208, 404 n. 9 Vincent of Beauvais, 250, 339 Vincent, Diane, 376, 394 Vitkus, Daniel, 404 Voragine, Jacobus de, Legenda Aurea: see Golden Legend Wace, Roman de Brut, 35, 54, 138, 151, 231 n. 25 Waldef, 150–1 Wales, 4, 229

471

Wallace, Sir William, 219–20, 256 Walsall, 410–11 Warkworth, John, Chronicle, 238 n. 50 Warm, Robert, 2 n. 9, 139, 315 Wars of the Roses, 173, 211 Watson, Nicholas, 29 Weiss, Judith, 77 Westerhof, Danielle, 258 Westminster palace, 9, 11 Wiggins, Alison, 178, 181 n. 47, 267 n. 5 William de Briane, 28–9, 53, 73, 114, 154, 247–50, 260, 413 William of Jumièges, Historia Normannorum ducum, 54–6, 138 William I (the Conqueror), 3, 4, 228, 232, 311 Wilson, R. M., 74 Windeatt, B. A., 280 Wingfield, Emily, 216, 218 n. 169 Worde, Wynkyn de, 210, 217 Worthies (the Nine), 51 n. 76, 148 n. 86 (the three Christian), 2, 84 n. 182, 211, 409 Yeager, Suzanne, 341 Ypocras, 204

Bristol Studies in Medieval Cultures previously published The Madonna of Humility: Development, Dissemination and Reception, c. 1340–1400 Beth Williamson The Medieval Art, Architecture and History of Bristol Cathedral: An Enigma Explored Edited by Jon Cannon and Beth Williamson Chaucer and the Cultures of Love and Marriage Cathy Hume Translators and their Prologues in Medieval England Elizabeth Dearnley Cross and Culture in Anglo-Norman England: Theology, Image, Devotion John Munns

charlemagne: a european icon Charlemagne and his Legend in Early Spanish Literature and Historiography Edited by Matthew Bailey and Ryan D. Giles The Charlemagne Legend in Medieval Latin Texts Edited by William J. Purkis and Matthew Gabriele

CHARLEMAGNE: A EUROPEAN ICON Previous volumes in the series Charlemagne and his Legend in Early Spanish Literature and Historiography Edited by Matthew Bailey and Ryan D. Giles

The historical point of departure for this volume is Charlemagne’s ill-fated incursion into Spain in 778. After an unsuccessful siege of Zaragoza, the king of the Franks directed his army north and on his passage through the Pyrenees, he turned his wrath on Pamplona, destroying the Basque city and its walls. The Basques subsequently ambushed the rearguard of Charlemagne’s army on the heights of the Pyrenees, killing numerous officers of the palace, plundering the baggage, and then vanishing into the forested hills, leaving the Franks to grieve without the satisfaction of revenge. In Spain, popular narratives eventually diverted their attention away from the Franks to the Spaniards responsible for their slaughter. This volume explores those legendary narratives of the Spaniards who defeated Charlemagne’s army and the larger textual and cultural context of his presence in Spain, from before their careful elaboration in Latin and vernacular chronicles into the early modern period. It shares with previous studies a focus on the narration of historical and imaginary events across genres, but is unique in its emphasis on the reception and evolution of the legendary figure of Charlemagne in Spain. Overall, its purpose is to address the diversity and importance of the Carolingian legends in the literary, historical, and imaginative spheres during the Middle Ages, Renaissance, and into the seventeenth century. The Charlemagne Legend in Medieval Latin Texts Edited by William J. Purkis and Matthew Gabriele

This book explores the multiplicity of ways in which the Charlemagne legend was recorded in Latin texts of the central and later Middle Ages, moving beyond some of the earlier canonical “raw materials”, such as Einhard’s Vita Karoli Magni, to focus on productions of the eleventh to fifteenth centuries. A distinctive feature of the volume’s coverage is the

diversity of Latin textual environments and genres that the contributors examine in their work, including chronicles, liturgy and pseudohistories, as well as apologetical treatises and works of hagiography and literature. Perhaps most importantly, the book examines the “many lives” that Charlemagne was believed to have lived by successive generations of medieval Latin writers, for whom he was not only a king and an emperor but also a saint, a crusader, and, indeed, a necrophiliac.

The Matter of France, the legendary history of Charlemagne, had a central but now largely unrecognised place in the multilingual culture of medieval England. From the early claim in the Chanson de Roland that Charlemagne held England as his personal domain, to the later proliferation of Middle English romances of Charlemagne, the materials are woven into the insular political and cultural imagination. However, unlike the wide range of continental French romances, the insular tradition concentrates on stories of a few heroic characters: Roland, Fierabras, Otinel. Why did writers and audiences in England turn again and again to these narratives, rewriting and reinterpreting them for more than two hundred years? This book is the first full-length study of the tradition. It investigates the currency and impact of the Matter of France with equal attention to English and French-language texts, setting each individual manuscript or early printed text in its contemporary cultural and political context. The narratives are revealed to be extraordinarily adaptable, using the iconic opposition between Carolingian and Saracen heroes to reflect concerns with national politics, religious identity, the future of Christendom, chivalry and ethics, and monarchy and treason. PHILLIPA HARDMAN is Reader in Medieval English Literature (retired) at the University of Reading. MARIANNE AILES is Senior Lecturer in French at the University of Bristol.

Cover image: Charlemagne kneels before an altar, praying for Oliver's success in his combat with Fierabras. British Library, Egerton MS 3028, f. 89r. © The British Library Board.

Bristol Studies in Medieval Cultures