Boundaries in Medieval Romance 184384155X, 9781843841555

Medieval romance frequently, and perhaps characteristically, capitalises on the dramatic and suggestive possibilities im

711 176 12MB

English Pages X+198 [214] Year 2008

Report DMCA / Copyright


Polecaj historie

Boundaries in Medieval Romance
 184384155X,  9781843841555

Table of contents :
Preface vii
Notes on Contributors viii
Abbreviations x
Introduction / Neil Cartlidge 1
When Romance Comes True / Helen Cooper 13
The Curious History of the Matter of England / Rosalind Field 29
How English are the English Charlemagne Romances? / Phillipa Hardman and Marianne Ailes 43
The "Sege of Melayne"; or, How the French Screwed Up and "Oure Brettons" Rescued Them / Elizabeth Berlings 57
Romance Society and its Discontents: Romance Motifs and Romance Consequence's in "The Song of Dermot and the Normans in Ireland" / Simon Meecham-Jones 71
England, Ireland and Iberia in "Olyuer of Castylle": The View from Burgundy / Elizabeth Williams 93
The Alliterative "Siege of Jerusalem": The Poetics of Destruction / Arlyn Diamond 103
The Peace of the Roads: Authority and "auctoritas" in Medieval Romance / Robert Rouse 115
The Hero and his Realm in Medieval English Romance / Laura Ashe 129
The Courteous Warrior: Epic, Romance and Comedy in "Boeve de Haumtone" / Judith Weiss 149
Rewriting Divine Favour / Ivana Djordjevic 161
Bodily Narratives: Illness, Medicine and Healing in Middle English Romance / Corinne Saunders 175
Index 191

Citation preview

Boundaries in Medieval Romance

Medieval romance frequently, and perhaps characteristically, capital­ ises on the dramatic and suggestive possibilities implicit in boundaries - not only the geographical, political and cultural frontiers that medi­ eval romances imagine and imply, but also more metaphorical demar­ cations. It is these boundaries, as they appear in insular romances circulating in English and French, which the essays in this volume address. They include the boundary between reality and fictionality; boundaries between different literary traditions, modes and cultures; and boundaries between different kinds o f experience or perception, especially the ‘altered states’ associated with sickness, magic, the supernatural, or the divine. Dr N eil C artlidge teaches in the Department o f English at the University o f Durham.

Studies in Medieval Romance ISSN 1 4 7 9 -9 3 0 8

Series Editors Corinne Saunders Roger Dalrym ple

This series aims to provide a forum for critical studies o f the medieval romance, a genre which plays a crucial role in literary history, clearly reveals medieval secular concerns, and raises complex questions regarding social structures, human relationships, and the psyche. Its scope extends from the early middle ages into the Renaissance period, and although its main focus is on English literature, comparative studies are welcomed. Proposals or queries should be sent in the first instance to one o f the addresses given below; all submissions w ill receive prompt and informed considera­ tion. Dr Corinne Saunders, Department o f English, University o f Durham, Durham, D H 1 3 AY Boydell & Brewer Lim ited, PO Box 9, Woodbridge, Suffolk, IP 12 3DF

Volumes already published I:

The Orient in Chaucer and Medieval Romance , Carol F. Heffernan, 2003


Cultural Encounters in the Romance o f Medieval England , edited by Corinne Saunders, 2005

III: The Idea o f Anglo-Saxon England in Middle English Romance , Robert A llen Rouse, 2005 IV: Guy o f Warwick: Icon and Ancestor , edited by Alison Wiggins and Rosalind Field, 20 07 V: The Sea and Medieval English Literature, Sebastian I. Sobecki, 20 07

Boundaries in Medieval Romance



© Contributors 2008

All Rights Reserved. Except as permitted under current legislation no part o f 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 o f the copyright owner

First published 2008 D. S. Brewer, Cambridge

ISBN 9 7 8 -1 -8 4 3 8 4 -1 5 5 -5

D. S. Brewer is an imprint o f Boydell & Brewer Ltd PO Box 9, Woodbridge, Suffolk IP 12 3DF, UK and o f Boydell & Brewer Inc. 668 Mount Hope Ave, Rochester, NY 14604, USA website:

A CIP catalogue record for this book is available from the British Library

This publication is printed on acid-free paper Printed in Great Britain by Antony Rowe Ltd., Chippenham, Wiltshire

Contents Preface Notes on Contributors Abbreviations Introduction Neil Cartlidge 1. When Romance Comes True Helen Cooper 2. The Curious History of the Matter of England Rosalind Field 3. How English Are the English Charlemagne Romances? Marianne Ailes and Phillipa Hardman 4.

The Sege o f Melayne - A Comic Romance; or, How the French Screwed Up and ‘Oure Bretonns’ Rescued Them Elizabeth Berlings 5.

Romance Society and its Discontents: Romance Motifs and Romance Consequences in The Song ofDermot and the Normans in Ireland Simon Meecham-Jones 6. England, Ireland and Iberia in Olyuer o f Castylk. The View from Burgundy Elisabeth Williams 7. The Alliterative Siege o f Jerusalem-. The Poetics of Destruction Arlyn Diamond 8. The Peace of the Roads: Authority and auctoritas in Medieval Romance Robert Rouse 9. The Hero and his Realm in Medieval English Romance Eaura Ashe 10. The Courteous Warrior’: Epic, Romance and Comedy in Boeve de Haumtone Judith Weiss 11. Rewriting Divine Favour Ivana Djordjevié 12. Bodily Narratives: Illness, Medicine and Healing in Middle English Romance Corinne Saunders Index

in memonam

W.R.J. Barron

Preface In April 2004, the 9th Biennial Medieval Romance Conference took place at Newman House, the historic heart of University College Dublin. Conferences in this series are traditionally focused on insular (as opposed to continental) romance, especially texts outside the Arthurian tradition, but they otherwise accommodate a broad range of approaches to medieval romance; they tend to define insular romance without regard to the the language in which it was written (whether Middle English, Old French or Latin); and they also foster consideration of a diverse set of intellectual contexts. The 2004 conference in Dublin was the first time that any conference in the series had been held outside Britain; and the crossing of the particular boundary presented by the Irish Sea seemed to create a distinct intellectual stimulus that was clearly evident in the papers presented in Dublin. Not only did the depiction of Ireland in medieval romance feature in several of the papers given: but also boundaries, and the crossing of boundaries, seemed to feature remarkably often in the approaches taken to the texts that were considered. As a result, it was relatively easy to define a theme for this Proceedings volume, which (as is traditional with volumes in this series) represents a developed version of the presentations and discussions that took place during the conference itself. Numerous debts were incurred in the organization of the conference. I am grateful to the Royal Irish Academy for agreeing to host the plenary lecture given by Helen Cooper (printed here as Chapter 1), in concurrence with one of the Academy’s own sessions; and to the School of English at University College Dublin for providing some financial support. I should also thank Margaret Robson, Frances McCormack, Veronika Hinterdorfer and Michelle Piazza for so cheerfully volunteering their assistance both before and during the conference itself. For assistance with the editing of this volume, I am grateful to the staff at Boydell & Brewer, particularly Caroline Palmer and Pru Harrison. I also owe an extensive debt to Corinne Saunders, both for her work as one of the general editors of the Studies in Medieval Romance series, and for all her help and advice since my move to Durham. I must also thank Laura Ashe for moral support and helpful criticism supplied at very short notice. The tolerance, good humour and email-companionship of all the contributors has made my work on this book a real pleasure. Closer to home, my partner Kate helped me with some perceptive proof-reading and equally perceptive advice; while Ian and Sue Thomas and Mrs Muriel Twiggs generously created time for me to work by giving up a considerable amount of time of their own. Finally, I should like to mention my daughters, Carensa and Imogen Thomas neither of whom has been any help whatsoever, but whose company I am very happy to have had.

Notes on Contributors Marianne Ailes is a lecturer at Bristol University, having recently moved from Wadham College, Oxford. She is the editor of the crusade chronicle, Ambroise’s Estoire de la guerre sainte and has published a book on the Chanson de Roland. She has also published on chronicle and chansons de geste in various books and journals including Medium JEvum, Romania,, French Studies and Reading Medieval Studies. Laura Ashe was a Research Fellow at Gonville & Caius College, Cambridge, and is now Lecturer in English at Queen Mary, University of London. She has published various articles on secular and religious literatures, focusing on the interplay of texts with historical, cultural and legal contexts. Her book Fiction and History in England, 1066—1200 came out with Cambridge University Press in 2007. Eli2abeth Berlings is an Associate Professor at St. John’s University, New York, with an M.Phil. from New York University. She has presented numerous papers on medieval romances, and also on Chaucer, Marlowe and Shakespeare. In 2006 she received a grant from the National Endowment for the Humanities and was a participant in a summer institute at the Oxford Centre for Hebrew and Jewish Studies. Neil Cartlidge is a lecturer at the University of Durham. He is the author of two books, Medieval Marriage: Literary Approaches 1100-1300 (1997) and The Owl and the Nightingale: Text and Translation (2001) and of articles in various academic journals including the Journal o f Medieval Latin, the Yearbook o f English Studies, Neuphilologische Mitteilungen, Medium Ævum, Viator and Studies in the Age o f Chaucer. Helen Cooper is Professor of Medieval and Renaissance English at the University of Cambridge. Her publications include The English Romance in Time: Transforming Motifs from Geoffrey o f Monmouth to the Death o f Shakespeare (2004). Arlyn Diamond is Professor Emerita at the University of Massachusetts, Amherst. She is the co-editor of Medieval Women: Texts and Contexts (2000). Her work is centered on the social meanings of romance. Ivana Djordjevic is an Assistant Professor in the Liberal Arts College at Concordia University, Montreal. She has published on Anglo-Norman and Middle English romance and on the poetics of rewriting, especially translation; she is the co-editor, with Jennifer Fellows, of a forthcoming volume of essays on medieval versions of the story of Bevis of Hampton.

Rosalind Field is Reader in Medieval Literature at Royal Holloway, University of London. Her publications include ‘Romance in England, 1066-1400’ in The Cambridge History o f Medieval English literature, ed. David Wallace (1999) and ‘Romance’ in the forthcoming Oxford History o f Uterary Translation in English, VoL /, ed. Roger Ellis (2007). She has recently co-edited Guy o f Warwick: Icon and Ancestor (2007). Phillipa Hardman is a Reader in Middle English Literature at the University of Reading. She has published on medieval romance, manuscript studies, Chaucer, Lydgate and the Gawain-pott. Simon Meecham-Jones is an affiliated lecturer for the English Faculty at Cambridge, and a research fellow for the Medieval and Early Modem Research Centre at University of Wales, Swansea. He has published on Chaucer, Gower, twelfth-century Latin lyrics and co-edited (with Ruth Kennedy) Writers o f the Reign ofHeniy II (2006). Robert Rouse has taught at the Universities of Bristol, Durham and Nottingham in the United Kingdom, and is currendy a Faculty member in the Department of English at the University of British Columbia in Canada. A medievalist whose interests range across both Old and Middle English literary culture, his recent research has addressed issues of historiography, nationalism and geography within the romance narratives of medieval England, resulting in the publication of The Idea o f Anglo-Saxon England in Middle English Romance (2005) and The Medieval Quest fo r Arthur (2005). Corinne Saunders is Professor of English at the University of Durham and English editor of the journal Medium Ævum. She is also the author of The Forest o f Medieval Romance (1993) and Rope and Ravishment in the Literature o f Medieval England (2001); and has edited Chaucer (Blackwell Guides to Criticism, 2001); A Companion to Romance: From Classical to Contemporary (2004); (with Françoise le Saux and Neil Thomas) Writing War. Medieval Uterary Responses (2004); Cultural Encounters in Medieval Romance (2005); (with Jane Macnaughton) Madness and Creativity in Literature and Culture (2005); and A Condse Companion to Chaucer (2006). Judith Weiss is an Emeritus Fellow of Robinson College, Cambridge. Her interests lie mainly in the field of Anglo-Norman romance and historiography. She has produced a parallel text and translation of Wace’s Roman de Brut, and her translation of Boeve de Haumtone and Gui de Warewic, for the French of England Translation Series, appeared in 2007. Elizabeth Williams lectured in Medieval English Language and Literature in the University of Leeds until she took early retirement. She has published articles on the Middle English romances and is particularly interested in the relation of medieval stories to their analogues in ballad and folktale.

Abbreviations Anglo-Norman Text Society British Library Bibliothèque nationale de France Classiques français du moyen âge Early English Text Society, Extra Series Early English Text Society, Original Series Exeter Medieval English Texts and Studies Feet o f Clay Guards, Guards! Journal o f English ford English Dictionary Patrologia cursus completus: series latina, ed. J.-P. Migne, 221 vols (Paris, 1844-64) Publications o f the Modem Language Association The Rolls Series: Rerum Britannicarum Medii Ævi Scriptores (Chronicles and Memorials o f Great Britain and Ireland during the Middle Ages), 99 works in 244 vols (London, 1858-1896) Studies in the Age o f Chaucer Studies in Medieval Culture The Consortium for the Teaching of the Middle Ages

Introduction N eil C artlid g e

The boundaries addressed in this volume are of various kinds. They include not only the geographical, political and cultural frontiers that medieval romances imagine and imply, but also more metaphorical demarcations, such as the boundary between reality and fictionality; boundaries between different literary traditions, modes and cultures; and boundaries between different kinds of experience or perception, especially the ‘altered states’ associated with sickness, magic, the supernatural or the divine. All of the essays in this volume demonstrate in one way or another how medieval romances frequently, and perhaps characteristically, capitaine on the dramatic or suggestive possibilities implicit in all such boundaries. Indeed, one conclusion that might be drawn from this collection is that medieval romances both define and challenge boundaries more self-consciously and more provocatively than is usually acknowledged. Yet romance is, as a genre, notorious for its indifference to limits - its apparent readiness to breach the rules both of literary decorum and of literary realism. As a result it has often been interpreted simply as an instrument of or an excuse for - a form of fantasy that is self-indulgently indisciplined, and therefore intellectually trivial. This has sometimes been seen, in turn, as a reflection of the relatively unsophisticated nature of medieval romance’s audiences (and authors). More recently, this has been translated into a more serious allegation, the suggestion that romance implicitly represents a profound form of intellectual failure, dishonesty or aggression. Such criticism perhaps says more about the preconceptions of medieval romance’s critics, the intellectual limits in which they choose to confine it as a genre, than about the specific effects or achievements of individual romancetexts - the way in which they themselves employ the various different kinds of boundaries mentioned above. By contrast, the premise that all the essays in this volume share is that romance is a complex and dynamic phenomenon, and considerably less amenable to generalization than it is often taken to be, so that understanding its boundaries is a process that both requires and repays detailed and sympathetic attention to individual texts. There is, though, some truth in the suggestion that the imagined worlds of medieval romance tend to be, by nature, rather unlocated and/or unlocalizable. One explanation for this is as a form of narrative efficiency, in the sense that it is the consequence of the subordination of setting to plot. So, for example, one of the more extraordinary boundaries in medieval romance is spanned by the famous Sword

Neil Cartlidge Bridge that Sir Lancelot encounters in Chrétien’s Chevalier de la Charrette, and which serves as a means of access to the mysterious Land of Gorre. This ‘bridge* —a sharp blade suspended over a torrent - ultimately provides Lancelot with the opportunity to demonstrate both extreme bravery and absolute devotion to his lady (Arthur’s queen, Guinevere). It is clearly this opportunity, rather than any attempt to provide a coherent geographical setting for Lancelot’s adventures, that explains its presence in the poem. Yet the Sword Bridge is also described with considerable attention to the physical details of its construction and its particular placing in the landscape: Au pié del pont, qui molt est max,1 Sont descendu de lor chevax, Et voient l’eve felenesse, Noire et bruiant, roide et espesse, Tant leide et tant espoantable Con se fust li fluns au deable, Et tant périlleuse et parfonde Qu’il n’est riens nule an tot le monde, S’ele i cheoit, ne fust alee Ausi com an la mer salee. Et li ponz qui est an travers Estoit de toz autres divers, Qu’ainz tex ne fil ne ja mes n’iert. Einz ne fu, qui voir m’an requiert, Si max ponz ne si male planche. D’une espee forbie et blanche Estoit li ponz sor l’eve froide, Mes l’espee estoit forz et roide Et avoit .II. lances de lonc. De chasque part ot un grant tronc, Ou l’espee estoit closfichiee. Ja nus ne dot que il i chiee Por ce que ele brist ne ploit, Que tant i avoit il d’esploit Qu’ele pooit grant fes porter. [At the end of that very dangerous bridge they get off their horses to see the treacherous water thundering swiftly past, black and turbid, as horrid and terrifying as if it were the Devil’s river, and so perilous and deep that there is nothing in the whole world which, having fallen into it, would not vanish into the salt sea. The bridge across it was quite different from all others: there never was nor will be one like it. If anyone asks me the truth, never did so terrible a bridge or foot-crossing exist. This bridge over the cold water consisted of a polished, gleaming sword; but it was a strong, stout sword as long as two lances. At each end was a tree-trunk to which the sword was


Introduction nailed. No one needed to fear falling off on account of its breaking or bending, for it was so well made that it could bear a great weight.]1 At the same time, Chrétien’s insistence on the materiality of the Sword Bridge and of the landscape in which it is so firmly fixed (‘closfichiee’) is a peculiarly self-defeating strategy, in the sense that all his gestures towards realism are inevitably undermined by the sheer improbability of anyone ever bothering to erect such a structure. Despite Chrétien’s absurd reassurance that no one need fear for its stability, it is evidendy not a convenient means of crossing the torrent; nor is it any more formidable as an obstacle than the torrent itself, despite the intimidating nature of its razor-sharp and slippery blade. The Sword Bridge thus hardly serves any purpose very effectively - not even as a symbolic disincentive to anyone trying to enter the Land of Gorre. Even when Chrétien draws attention to the solidity of the bridge’s engineering, the care (‘esploit’) implicit in its design and its capacity for heavy loads (‘grant fes porter’), the effect is only to emphasize, not the realistic qualities of the bridge, but rather its artificiality. This very artificiality, though, could be read as a pointer to the sheer meaningfulness of the bridge - its very ‘constructedness’ reflecting the constructedness of the narrative itself, the deliberate way in which Chrétien generates moral and psychological perspectives from the places through which the knight travels. One way of explaining the function of the Sword Bridge, and of the landscape implied by its presence, is as a means of enabling the ‘series of adventures’ that Chrétien’s romance tends to elevate ‘to the status of a fated and graduated test of election’, as Erich Auerbach puts it.12 As such, the topography of romance ‘becomes the basis of a doctrine of personal perfection’; and thus not just the setting for an assertion of the values of ‘courtly culture’, but the very means of making that assertion. In the light of this argument, Auerbach concluded that ‘courtly culture was decidedly unfavorable to the development of a literary art which should apprehend reality in its full breadth and depth’. This is a generalization that has been influential on criticism of medieval romance as a whole, including the insular romances that are specifically the subject of the essays in this book, and not just the continental tradition to which Chrétien himself belongs.3 Yet it is a fundamental premise of Auerbach’s study of the representation of reality in literature that no text can ever portray reality, to any extent, except by using particular techniques that can be subjected to literary-critical analysis. No more than any other form of fiction can medieval romance be read as an attempt at replicating reality. It can never be more than an imitation of reality (a ‘mimesis’); and from that point of view its tendency to endow landscape with psychological, moral and even allegorical dimensions is perhaps better seen as one aspect of the distinctive way in which it mimics reality, than as a proof of its indifference to, or unawareness of, reality itself. 1



Chrétien de Troyes, Le chevalier de la charrette: ou le roman de Lancelot, ed. Charles Mêla (Paris, 1992), lines 3007-27; trans. D.D.R. Owen, Chrétien de Troyes: Arthurian Romances (London, 1987), p. 225. Erich Auerbach, Mimesis: Dargestellte Wirklichkeit in der abendländischen Literatur (1946; repr. Tübingen and Basel, 2001), p. 132: ‘in den Rang einer schicksalsbestimmten, stufenweisen Bewährung eines Auserwähltseins’: trans. Willard R. Trask, as Mimesis: The Representation of Reality in Western Literature (Princeton, 1953; repr. 1991), p. 136. Auerbach, p. 138; trans. Trask, p. 142.


Neil Cartlidge Even if the Sword Bridge hardly serves in itself as a credible depiction of any kind of political or economic frontier, it certainly cannot be taken to illustrate Chrétien’s absolute lack of interest in the idea of boundaries. The Land of Gorre to which it provides access is itself a shadowy, undefined place, but this could be explained as a consequence, not so much of Gorre’s lack of borders, as of its apparent function as a kind of border. It is described, for example, as that land: .. .Don n’ist ne sers ne gentix horn Qui ne soit de la entor nez, N’ancor n’en est nus retornez. Les estranges prisons retienent Et cil del pais vont et vienent Et anz et fors a lor pleisir. (Unes 1904-09) [.. .which no serf or nobleman can leave unless he is a native of these parts: no one has as yet come back from there. Foreigners are held prisoner there, whilst the inhabitants come and go in and out as they please.]4 It is not impossible to imagine a country where such conditions could reign, but there is no suggestion here that the imposition of such conditions was motivated by any form of poUtical pragmatism. Rather, they seem somehow impUcit in the very nature of the land itself, and in this sense Gorre could be defined as an essentially liminal realm - a place fundamentaUy defined by the boundary that allows entry to non-citizens, but not egress. In that respect, it clearly resembles traditional conceptions of the underworld;56and this is a reading suggestively supported by the immediate context of the description that I have just quoted. The words are those of a hermit who helpfully explains to Lancelot that the people trapped in the Land of Gorre will only be released by the one knight who is capable of Hfting the Ud of a particularly large and heavy tomb in the hermit’s cemetery. It might be argued that if the prisoners can be released at aU, then they cannot be taken as dead in any Uteral sense, but the very fact that their release is presaged by the Hfting of the Ud of a tomb surely impUes that Chrétien intended it to be read at least as a symboUc resurrection. That possibiUty is perhaps stiU further reinforced by the hermit’s subsequent prediction that the man capable of Hfting the tomb’s Ud will eventuaUy be its occupant (Unes 1932-36). In effect, then, Chrétien’s depiction of Gorre serves to dramatize the boundary between Ufe and death, by suggesting that it is in some sense more permeable than we are accustomed to beUeve. Another text that apparently sets out to achieve much the same effect is the Middle EngUsh romance Sir OrfeoS In this case, it is the wilderness that serves as the liminal space between Orfeo’s kingdom and the strange otherworld into which his wife, Heurodis, is abducted. Driven distraught by his grief at her loss, he abandons his kingdom and disappears, barefoot ‘þurth wode & ouer heþ’ into the wilderness: Cf. also Chevalier de la charrette, Unes 640-43, where Gorre is described as ‘the kingdom from which no stranger returns, but is forced to stay in that land in servitude and exUe’ (Owen, Arthurian Romances, p. 193). 5 Cf. Je ff Rider, The other worlds o f romance’, in The Cambridge Companion to Medieval Romance, ed. Roberta L. Krueger (Cambridge, 2000), pp. 1 1 5 -3 1 , at pp. 115 -16 . 6 Sir Orfeo, ed. A.J. BUss (Oxford, 1971). 4


Introduction Noþing he fint þat him is ays, Bot euer he liueþ in gret malais*. He þat hadde y-werd þe fowe & griis, & on bed þe purper biis, - Now on hard heþe he liþ, Wiþ leues & gresse he him wriþ. He þat hadde had castels & tours, Riuer, forest, friþ wiþ flours - Now, þei it comenci to snewe & frese, Þis king mot make his bede in mese. (Sir Orfeo, lines 239-48) Again, the wilderness depicted here is more vividly apprehensible as a space in a psychological landscape - as the ‘objective correlative’ of Orfeo’s grief —than as a part of any kind of coherent geography; although it turns out that the gateway to the otherworld is actually very palpably located within this wilderness, for Orfeo eventually finds his way there by entering a cave in a cliff (‘in at a roche’, line 323). More clearly still, the wilderness serves as a contrast with the civilized world, the world of the living, to which Orfeo’s kingdom belongs, for the poet makes much of the pathos implicit in his protagonist’s loss of his high estate (‘He that hadde had castels and tours,/ River, forest, frith with flours...’). The otherworld to which Heurodis is abducted could even be seen as the mirror-image of Orfeo’s own kingdom - the only thing differentiating them being, in effect, the boundary between the two different kinds of reality that they represent. In describing the fairy-king’s realm, Heurodis uses much the same language as the poet subsequently uses to describe Orfeo’s own realm: [The fairy-king] made me wiþ him ride Opon a palfray by his side; & broußt me to his palays, Wele atird in ich ways, & shewed me castels & tours, Riuers, forestes, friþ with flours... (Sir Orfeo, lines 155-60)7 It could thus be said that both Chrétien’s Lancelot and the Middle English Sir Orfeo define boundaries in spaces that are imagined geographically, but that the significance of such boundaries in these two texts is ultimately moral, or even spiritual, rather than political. However, there is one particular moment in Sir Orfeo where the cultural and geographical boundaries of the real world do suddenly emerge, in a fashion that seems strikingly at odds with the more figurative nature of the boundaries that it otherwise draws. This is the moment when the poet rather disarmingly identifies the ancient Greek province of Thrace, which he describes as Tradens,/ Þat was a cité of noble defens’ (lines 47—48), with the English city of Winchester: ‘For Winchester was cleped þo/ Traciens wiþ-outen no’ (lines 49—50). This is often cited as a classic instance of medieval romance’s disregard for historical and geographical accuracy; there is no evidence that this idea could have been derived from any medieval 7

On the cultural contexts and implications o f Sir Orfeo’s imagining o f the otherworld, see Alan J. Fletcher, (Sir Orfeo and the Flight from the Enchanters’, SAC 22 (2000) 141—77; and Neil Cartlidge, ‘Sir Orfeo in the Otherworld: Courting Chaos?’ SA C 26 (2004), 195-226.


Neil Cartlidgt ‘authority’; and indeed the cultural distance between the ancient Greek province and the medieval English city is clearly so great that these lines can hardly be anything other than startling. It might be read as conspicuous proof of the poet’s lack of learning or as an illustration of romance’s cultural aggression, a bare-faced attempt by a member of western Christendom to appropriate the pre-Christian, pre-medieval world for his own imaginative purposes. More charitably, it could be read as a means of asserting a kind of continuity with the distant past, by reminding the audience of the status of Winchester as the ancient capital of England. It may also have had a local or personal significance now lost to us - one might speculate, for example, that so specific a reference to a particular city was designed to gesture at the Orfeo-poet’s own origins or affiliations. Little remarked, though, is that, far from simply exemplifying the readiness of medieval romance to commit such geographical and cultural solecisms, this passage in Sir Orfeo also illustrates the self-consciousness with which medieval romance characteristically disposes, or pretends to dispose, its boundaries. It is the author of these lines himself who emphasizes that this identification between the two places is a conscious exercise in historicization (‘For Winchester was cleped tho’) and insists on its incontrovertibility as a fact (‘was cleped... mthouten no*) - for all that the unhistorical, unfactual nature of what he says is precisely what strikes the modem reader. He also lays claim to a degree of judiciousness in his careful, and slighdy pompous, description of Traciens/Winchester as ‘a cite of noble defens’ (line 24). The posture he adopts, then, could be described as that of an antiquarian - someone deeply interested in the past and its places; and indeed one of his primary intentions in these few lines may have been to be seen to strike that posture. That may not entirely absolve him of the charge of intellectual naivety, but it does underline that his was not an unconscious naivety. Whatever his purposes were —and we probably do not have sufficient evidence to say with any certainty what they were (or even to be sure that these lines are not a scribal insertion)8 - his identification of Traciens as Winchester certainly does not prove his indifference to historical and geographical accuracy: indeed, quite the opposite. Moreover, this attempt to map the contemporary geography of the medieval world onto the ancient geography of the classical world is by no means unprecedented in insular romance. Towards the end of the late-twelfth-century romance Ipomedon by Hue de Rotelande, we encounter a character called Leander who is introduced as the duke and ruler of Thessaly (‘De Tessaille fut dux e sire’, line 8928) and as the brother of Ipomedon’s principal opponent at this point, the giant Léonin, who is himself described as the son of a rich emir from greater India (‘de Inde major,/ Ffilz a un riche almazor’, lines 7Ó97-98.9 Yet Hue also explains that: Si fist uns reis gualeis jadis, Jo quit ke il l’apelerent Ris, Il fut mut larges de Engleterre, A ses hirdmans parti la terre, Herefort e Glouecestre, Salopesbure e Wirecestre, 8 9

Bliss argues that lines 4 9 -5 0 ‘are unlikely to be genuine’ (p. 52; cf. p. xv).

Ipomedon, poème de Hue de Kotelande (fin du XHe siècle), ed. A J. Holden (Paris, 1979).


Introduction Mes il en lava ben ses mains; Il e li son ourent li meins Kar il fut vencuz e laidiz, Vilment chacez e descumfiz. (lines 8941-50) [He was once a Welsh king: I think they called him Rhys. He was very generous with England, dividing up the land - Hereford and Gloucester, Shropshire and Worcester - among his retainers. But he was pleased to wash his hands of it [in the end]: he and his men had the worst of it, for they were defeated and dishonoured, shamefully harried and routed.] For Hue, it seems, Leander exists simultaneously in two very different dimensions. He is, on the one hand, an exotic figure associated with both Hellenic and oriental culture; and, on the other, a Welsh war-leader operating in the Marches where Hue himself lived. Hue’s editor, A.J. Holden, tentatively identifies the Welsh prince alluded to here as Rhys ap Gruffudd, Prince of Deheubarth, but this is problematic for several reasons: Lord Rhys never campaigned in the English counties that Hue lists, nor was he ever defeated quite so comprehensively as Hue describes.101Yet even if Leander cannot be identified with any particular Welsh prince —even if he might also be read as ‘une créature composite, même symbolique’ as Holden also suggests the effect is to generate a topical resonance of local significance that is, in the context, really quite incongruous. As with Sir Orfeo, though, the incongruity is most plausibly explained as an index, not of any presumed ignorance or clumsiness on the part of the author, but of the deliberateness with which he chose to blur the boundaries between the fantastic and the familiar. The identification of Leander as a former Welsh war-leader could conceivably have been intended as a precise political allusion that is no longer readily explicable, but whether or not this is the case, the identification is clearly both self-conscious and highly provocative in the context. One might argue that its effect is to dignify the political turmoil in South Wales and the Marches in the last decades of the twelfth century by relocating it, at least imaginatively, onto a much grander and more glamorous stage. Alternatively it could be argued that, just as the romance of Ipomedon serves as an exposition and (intermittently ironic) analysis of the courtly ideology of the Anglo-Norman barons who fought the Welsh princes for control of South Wales, so Hue’s identification of Ipomedon’s opponents with the Welsh possibly serves to emphasize the extent of the cultural and political boundary between them. To quote Holden again: ‘Notre auteur, [u]n citoyen du royaume anglo-normand, regarde d’un œil méprisant et en même temps inquiet les régions à demi barbares qui s’étendent au-delà des frontières.’11 Yet this reading of the passage rests to some extent on Holden’s assumption that the word ‘hirdman’ is pejorative - on the basis that it denotes a herdsman, peasant or bumpkin (‘pâtre... paysan, individu fruste’);12 10 Holden, pp. 1 0 -11 . But see also Huw Pryce, ‘Rhys ap Gruffudd (1131/2—1197)’, Oxford Dictionary ofNational Biography, Oxford University Press, 2004 [ view/article/23464, accessed 12 April 2007]; R.R. Davies, The Age of Conquest: Wales 10631415 (Oxford, 1991), pp. 2 17 -2 4 ; J. Gillingham, ‘Henry II, Richard I, and the Lord Rhys’, Peritia 10 (1996), 225-36. 11 Holden, p. 566. 12 OED, s.v. ‘herdman’, sb.


Neil Çartlidge but the word in question is much more likely to mean simply ‘retainer* —with no implicit suggestion of rusticity.13 What is striking, though, is that this Anglo-Norman writer should choose to adopt an English word when characterizing the war­ following of a Welsh prince - which illustrates, if nothing else, that, for Hue, political and cultural boundaries were not necessarily concurrent with linguistic ones. Sir Thomas Malory’s attempt to identify the fictional castle of Joyous Garde with either Alnwick Castle or Bamburgh Castle is also difficult to dismiss as a merely incidental or accidental detail, if only because - as with the instances that I have cited from Sir Orfeo and Ipomedon - it is so conspicuously gratuitous to the development of the narrative in hand. Moreover, the moment when Malory chooses to make this identification is a highly dramatic and momentous one, with his charismatic protagonist, Sir Lancelot, already at the point of death: So when he was howselyd and enelyd and had al that a Crysten man ought to have, he prayed the Bysshop that his felowes myght bere his body to Joyous Garde. (Somme men say it was Anwyk, and somme men say it was Bamborow.)14 Yet, as in the case of Sir Orfeo and Ipomedon, it is easier to argue that Malory’s suggestion was self-conscious and deliberate, than to define precisely what he might have expected to achieve by it. Both Alnwick and Bamburgh served as capitals of Lancastrian resistance (and Alnwick, of course, was the principal seat of the Percy earls of Northumberland - two of whom, the 2nd and 3rd, died for the Lancastrian cause during the 1450s and 1460s, along with three of the 3rd earl’s brothers).15 In 1462-63, Malory seems to have taken part in the Yorkist campaign that led to the siege and fall of the two castles, so he may have come to know them well - at least from the outside.16 In some ways that only makes it all the more surprising that he should have chosen to identify the two Northumbrian castles with Joyous Garde, a castle existing only in fiction, as if in denial of the existence of any gap between the glories of the Arthurian world and the realities of siege-warfare in a northern winter. Perhaps that is precisely the point: the identification of Joyous Garde as either Alnwick or Bamburgh being a calculated attempt to close that gap, to endow the brutal realities of the Wars of the Roses with some of the sheen of romance. It might also seem a little puzzling that Malory should have chosen to associate the castle of his hero, Sir Lancelot, with two castles that he knew as fortresses occupied by his political opponents; a possible solution for this difficulty is readily provided by the fact that, by the time he was writing the Morte Darthur; Malory was in prison as a consequence of falling out with the Yorkists. From this point of view Malory’s apparent glamorization of Alnwick and Bamburgh (and the Lancastrian defiance with which he would have associated it) could be read as a gesture of his own defiance against his gaolers. Alternatively, it might be argued that Malory’s association of 13 OED, s.v. ‘hirdman, hiredman’, sb. 14 Sir Thomas Malory, Morte Darthur, ed. Eugène Vinaver as Maloty: Works (Oxford, 1971), p. 724. 15 Alexander Rose, Kings of the North: The House of Percy in British History (London, 2003), pp. 495— 530. 16 P.J.C. Field, The Life and Times of Sir Thomas Mabry (Cambridge, 1993), pp. 129-30, 142; Vinaver, p. 778.


Introduction Alnwick and Bamburgh with Joyous Garde was not intended to be partisan; and that, in this sense, his imagination was perhaps more generous and less factionalized than his own behaviour during the Wars of the Roses might suggest. What I want to emphasize, though, is that such problems are not at all unusual in medieval romance: that they are, to some extent, characteristic of medieval romance. In other words, for all the unspecificity of some of the landscapes found in romance and for all the disorientations created by magic or madness or the supernatural, medieval romances do have a tendency to depict topographies through which the contours of the real can sometimes be perceived. More often than not, however, the real emerges in ways that are rather fitful and surprising. The problem, in other words, is not that the boundaries of romance simply do not correspond with those of reality, but that, where identifiable points of contact do exist, they are so often so troublesome to explain that they demand a high degree of tact and insight - if, indeed, it is possible to explain them at all. Consequendy, it could perhaps be argued that romance tends to draw boundaries in particularly provocative ways (relative to other genres of fiction) precisely because its correspondences with the real world are so very troubled, intermittent and complex. Nor is it always the case that such correspondences are explained by romance’s imitation of reality: there are certain respects in which the opposite is true, as Helen Cooper demonstrates in the first essay in this collection (which is based on the Conference’s plenary lecture, delivered at the Royal Irish Academy). She defines four different ways in which romance might be said to ‘come true’, including the possibility that certain historical events were modelled, to some extent, on narrative patterns typical of romance. In these cases, the process of ‘mimesis’, to use Auerbach’s term, is effectively reversed: it is reality that serves as the medium for an imitation of fiction, rather than the other way around. As Cooper shows, one of the mechanisms for this particular form of reverse-takeover was a shared preoccupation with the ideology of rightful inheritance. In the same way that the order of events in medieval romance is often subordinated to the vindication of the rights of the ‘true’ king, likewise kings in reality apparendy sought to play the part of the hero in a kind of ‘true’ romance. So too did the ‘imposters’ who attempted to compete with them for their thrones, even if their own attempts to impose romance on history by adopting a starring role in it were to be, by definition, frustrated. Implicit in such appropriations of romance-motifs for political effect is a deliberate blurring of the boundary between the real and the ideal - or, to put it another way, a deliberate attempt to supply romance with a specific localization, a specific set of borders: the literally generic realm of romance was interpreted as if it were a paradigm for the dynastic future of a particular realm: in this case, England. The sheer readiness with which the topography of medieval romance lent itself to being ‘realized’ in this way suggests that its apparent insubstantiality or abstractedness should be seen as a function, not of the genre’s historical irrelevance, but of its political adaptability. The next two essays both question the way in which specific groups of medieval romances might be said to demonstrate a coherently ‘English’ self-consciousness. Rosalind Field draws attention to the peculiar history of the term ‘Matter of England’, a term that has become widely used in romance-scholarship as a means of establishing a distinct identity for texts set in England (such as Guy o f Warwick and Havelok the Dane). As she shows, the origins of the term are much more obscure, and 9

NeiLCartlidge much less authentically medieval, than those of the other traditional ‘Matters’ of medieval romance (those of France, Britain and Rome). The boundaries implied by the term are thus highly artificial —and potentially misleading in that they suggest a much more rigid sense of national identity than romance-texts themselves support. Only Havelok the Dane, she suggests, comfortably inhabits the ‘Matter of England’ category; but even this romance has been more surprisingly and subversively influential than the traditional categorization would imply, as she shows in her provocative postscript, which observes the recent re-emergence of Havelokian motifs in some of the fantasy novels by Terry Pratchett. Meanwhile, Marianne Ailes and Phillipa Hardman joindy consider the extent to which Middle English romances deriving from chansons de geste depicting the deeds of King Charlemagne and his knights might be said to have given them a distincdy English spin. In this particular case, they suggest, English adaptations of the ‘Matter of France’ tended to involve a distinct shift of emphasis, towards a more consciously pious and moralistic reading of Charlemagne’s knights as Christian heroes in an implicitly universal fight against ‘heathenism’. In the next essay, Elizabeth Berlings tries to assess both the extent of the national partisanship expressed by the Middle English Sege o f Melayne and the extent to which the text is deliberately comical. The boundaries at issue in this essay, as in its two predecessors (by Field and Ailes/Hardman) might thus be said to be simultaneously political/cultural and generic; or, perhaps more accurately, all three essays might be said to address the question of how closely such boundaries might be said to be aligned. The same is true, to some extent, of Simon Meecham-Jones’s essay on the text traditionally known as the Song o f Dermot and the Earl\ but re-titled by its most recent editor as The Deeds o f the Normans in Ireland. These competing denominations reflect considerable disagreement about the genre to which this text actually belongs; and this disagreement in turn affects any assessment of its value as a witness to the Anglo-Norman conquest of Ireland. Since this is a work that presents a notably partisan view of this event, according to Meecham-Jones, the debate about its generic identity bears directly on any evaluation of its sophistication or success. The essay by Elizabeth Williams also discusses the depiction of Irish history in medieval romance (and she briefly considers the same Dermot as the one celebrated by the Song o f Dermot and the E arl); but in this case Ireland appears in a considerably longer perspective. The romance that she discusses, The Hystoiye o f Olyuer o f Castylle, originated in the ducal court of Burgundy during the fifteenth century. Leisurely and eclectic though this text is, it still alludes to historical events (in Ireland and elsewhere) in a fashion that is at times provocatively precise. The source of such allusions often seems to have been the chronicler Jean Froissart, as Williams demonstrates in this essay. Arlyn Diamond’s contribution to this collection addresses a text with a remarkable capacity to cause discomfort among modern critics: the alliterative Siege o f Jerusalem. She takes a bold approach to the difficulties that it poses, suggesting that it gratified its medieval audiences not because it celebrated violence against Jews, but because it celebrated violence itself. What makes the poem memorable, she suggests, is the ‘imaginative force’ of its descriptions of war, as much as the ‘figurai and historical exegesis’ that it might be taken to supply. Robert Rouse’s essay, by contrast, is much more concerned with peace - and specifically the motif of the 10

Introduction peace-creating king. As he demonstrates, this is a recurrent element in the idealization of England as an entity with a distinct national and legal identity to which insular romances tend to subscribe, but it is also a motif with parallels in a wide range of European texts. This he illustrates most strikingly with a brief excursus on its occurrence in the tradition of stories about the original Count Dracula - that is, the Prince of Transylvania, Vlad Dracul. ‘England’ also figures as a central idea in Laura Ashe’s study of the idea of the hero in medieval romance. She points out that neither Middle English nor Anglo-Norman possessed any word properly signifying the concept of the ‘hero’. Both languages lacked the term as it was elaborated in classical Latin, and as it was to be recoined in the sixteenth century, to designate ‘the man distinguished by supreme and superlative qualities of bravery’, ‘a person reverenced and idealised’. The absence of the word, however, evidendy did not preclude the writers of romances from inventing such characters, and she discusses the different strategies by which romancers went about constructing their idealized ‘heroes’. In discussing the insular romances of the twelfth and thirteenth centuries, she suggests that their rediscovery of the hero was concomitant with the emergence of a sense of the distinctness of England as a nation. Judith Weiss also touches on the question of what constituted heroism for the authors of insular romance in the course of her analysis of the generic affiliations of Boeve de Haumtone; and she discusses two other issues that are prominent elsewhere in this volume - the extent of comedy’s incursions into romance and the complexity of romance’s attitudes towards ‘Saracens’. She argues that ‘as a text that is part epic, part romance, Boeve de Haumtone could be said to exploit the tensions implicit in its occupation of the boundary between the two genres.’ Ivana Djordjevid’s essay is concerned with a different generic boundary: that between romance and hagiography. She demonstrates a shift in the way that medieval romances tend to depict the possibility of divine intervention, according to which the heroes of later medieval romances define and justify their heroism in terms of a much more intimate and demanding relationship with God. Paradoxically, she suggests that as the protagonists of pious romances came to rely increasingly ‘on God’s active help, the hero becomes in a sense less heroic, but God also becomes less divine’. The final essay in the collection, by Corinne Saunders, addresses a similar ambivalence: the uncomfortable relationship in medieval culture between different conceptions of medicine - on the one hand, as a sober science of the human body; on the other, as a more or less fantastic field of magic, miracle or superstition. Illness and healing have the capability of bringing us to the very frontiers of our rationality, to the very edges of our experience as rational individuals. As such, the process of sickness and recovery also readily serves as metaphor for moral and intellectual transformations of all kinds, but especially those transformations that take place at the limits of what we take to be ‘natural’. As Saunders herself puts it, the depiction of illness and healing allows romance writers ‘to create flash-points where human, otherworldly and divine may intersect, and where romance and realism meet’.


1 When Romance Comes True’ H elen C o o per

The romances that form the subject of this paper were not written by famous named authors, or by identifiable poetic masters. Almost all are anonymous, and so do not lend themselves to the kind of traditional criticism that one can apply to writers who have a known life and context; and since the dates of composition and the intended audiences for some are uncertain, and others are translations of works originally written within different political circumstances and a different social and linguistic culture, it is not at all easy to historicae them in the new or the old senses. They tend to be open about their meanings, avoiding subtexts and codes, so they are not amenable to the kind of hermeneutic of suspicion that fuels New Historicism. They are often talked about as ‘popular’ romances, though the term is somewhat misleading, since any text that was written down in the Middle Ages has by definition at least something elite about it. They are, however, written in English, not French or Anglo-Norman, and so mark themselves as linguistically accessible to all social classes. They do not generally carry the markers of high culture that characterize medieval French-language romance, though a number of them exist in continental or insular French versions as well as English; and several of the later ones were ‘popular’ almost by definition in the sense conveyed by the shift from individual manuscript copies to entire printed editions. It remains true, however, that all these romances overdy address the concerns of the gentry and the upper classes rather than peasants or townsmen; they emerge from an elite culture, first that of the Anglo-Norman romances written for aristocratic readers, later that of the ducal court of Burgundy, and throughout the Middle Ages their link with the aristocracy remains close - a fact that is of some importance for much of what follows. The tide ‘When Romance Comes True’ probably sounds like a paradox. ‘Romance’ has become the accepted antonym of ‘realism’, and we accordingly tend to define romance in terms of what is not true: much killing of dragons and giants by knights in shining armour. Those elements are of course there, but it is tempting to emphasise them to a degree that makes us overlook just how closely much Middle English romance connects with real life. Perhaps the very obviousness of those connections has something to do with the ease with which they are overlooked: This paper was originally delivered as a lecture at the Royal Irish Academy, Dublin, in April 2004, in association with the concurrent Medieval Romances Conference.

Helen Cooper dragons and giants make their presence more strongly visible, and are indeed what the authors seem most proud of. They are interesting just because they are not part of real life. There are, none the less, four major ways in which romance can ‘come true’. Two of those processes are deliberate and self-fulfilling. First is the writing of romances as a retrospective explanation of what is happening in the present; and second, the deliberate re-creation of romance in actual practice. The other two processes are more complex, in that the impact of immediate contemporary (or to us, historical) concerns on the writer serves to place romance at least partly outside his control: he is responding to a given context of historical or cultural incident that limits his freedom to invent or adapt or explain. Of these next two processes, the first has to do with how cultural practices, cultural changes, helped to create romance; and the second, with what happens when specific historical events appear to model themselves on romance structures, and how those events can be given a conscious extra boost by romance authors or patrons to make the parallels even closer. Across all four of those phenomena, there is a turning of history into romance, or romance into history. The first two of those processes, the deliberate and self-fulfilling connections, are comparatively straightforward, and require only a brief outline to indicate how they work. The use of romance to offer a retrospective explanation, or indeed a justification, of the known facts of the present, is something with which any scholar of romance will be familiar. Texts of this variety often develop romance into a kind of myth of origin; and in a Christian culture, such a mythic element carries with it a strong implication of endorsement by God. This usage emerges most often in genealogical romances, which tell the stories of the origins of countries, or towns, or aristocratic families. Geoffrey of Monmouth is in this sense writing a gigantic myth of origin, which runs from the foundation of Britain by Brut forwards; and if he wrote it genetically as history, many of his stories were given a later development as individual romances. Not all such stories are purely glorificatory, and romance can stretch itself to accommodate a degree of personal or political downfall alongside its celebratory function. Legends of origin can be invented or adapted to explain a present disaster, and therefore to shift the blame for that disaster back from the present onto the past. Melusine offers a particularly clear example. In its primary form, it tells of the foundation of the house of Lusignan by a woman who is half-fairy, and who is, in Donald Maddox’s term, the ‘mega-mother’ of the dynasty in all its numerous branches.1 In the late fourteenth century, however, a series of disasters that befell one particular branch of the family was given just such a retrospective explanation in the form of a curse imposed on one line of her descendants that was set to last for nine generations.12 The story does not obviously qualify as ‘true’ in any normal sense of the term; but to an age that lacked the techniques of historical 1


Donald Maddox, Fictions of Identity in Medieval France (Cambridge, 2001), pp. 172, 177-86. On the historical parallels for Melusine’s sons, see the introduction to the edition o f the metrical version by Eleanor Roach, Le Roman de Mélusine ou Histoire de Lusignan par Coudrette, Bibliothèque française et romane B. 18 (Paris, 1982), pp. 20—52. The prose version was composed in the 1390s in support o f Jean de Berry’s claims to the lordship. Both versions were translated into Middle English around 1500. The history is set out in the introduction to Laurence Harf-Lancner’s modernized edition, Coudrette: H Roman de Mélusine (Paris, 1993), pp. 26-35.


When Romance Comes True investigation that we take for granted, it was perhaps the best that could be done (and we need to remember how frequently we do exactly parallel things even though we know better: the American myth of the colonizing of an empty and unpeopled land, for instance, is a close replication of what is found in Geoffrey's legend of Brut; and film habitually rewrites history in favour of the audience for whom the film is made, as with the U-571 version of the capture of the Enigma codes in the Second World War that turns it into an American rather than a British achievement). So far as the Middle Ages were concerned, the present state of affairs had to be reached somehow: how did the facts of the contemporary world come to be? Those facts, moreover, could themselves appear to ‘prove' the romance version of the past invented to explain them. Such versions of history were not always or altogether received without some degree of scepticism, even at the time; but at least they provided a kind of just-so story that was impossible to better. To borrow a term from the early development of science, such legends ‘saved the appearances', provided a working hypothesis that accounted for the observed phenomena, and so offered a functional stand-in for truth until such time as it was either proved to be true or replaced by a better hypothesis. Such a readiness to accord truth to a romance version of the past was confirmed by the deliberate recreation of romance in life, in a process of life imitating literature. The medieval social elites, particularly aristocratic and royal courts, had something of a genius for turning their lives into art, or ritual.3 This is what happened, for instance, with the creation of the Order of the Garter by Edward III in 1348. The Order was specifically and deliberately modelled on the fellowship of the Round Table; and indeed there was already a round table in existence that Edward could use if he so wished, which had probably been commissioned by his grandfather, and which is still preserved in Winchester Great Hall.4 The table seems to have been linked with Edward I’s revival of Arthurianism as courtly play,5 but it was play with a serious edge: the revival, and the ton-and-a-bit table as a physically massive endorsement of the point at issue, were above all a deliberate propaganda move, to show how the greatness of the imperial British past as embodied in King Arthur was recreated in himself, with particular reference to the dispute concerning the overlordship of Scodand. At some point, however, the origins of the table were forgotten, and it began to look as if it might be the real thing; and if it were, then, as Caxton noted in his Preface to Malory’s Morte Darthur, it constituted a proof of the historicity of


4 5

The adoption o f chivalric values and the rituals o f knighthood in both romance and aristocratic life is o f course a dominant feature o f medieval culture, and the processes o f imitation and symbiosis appear to have been mutual. The numerous studies include Maurice Keen, Chivalry (New Haven and London, 1984); Richard Barber, The Knight and Chivalry, rev. edn (Woodbridge, 1995); Michel Stanesco, Jeux d’Errance du chevalier médiéval: Aspects ludiques de lafonction guerriere dans la littérature du moyen âgeflamboyant (Leiden and New York, 1988); and, for a series o f case studies, the essays in Chivalric Literature: Essays on Relations between Literature and Life in the Later Middle Ages, ed. Larry D. Benson and John Leyerle (Kalamazoo, 1980). For more extensive modes o f performance, see for instance Susan Crane, The Performance of Self: Ritual, Clothing, and Identity during the Hundred Years War (Philadelphia, 2002). Not all the evidence is conclusive, but this is the best hypothesis reached in Martin Biddle et al., King Arthur’s Round Table: A n Archaeological Investigation (Woodbridge, 2000). The classic article is Roger Sherman Loomis, ‘Edward I: Arthurian Enthusiast’, Speculum 28 (1953), 114-2 7 .


Helen Cooper Arthur, just as the whole cult of chivalry seemed to promise a way back to a golden age of the past. There are strong arguments for regarding romance as always retrospective, always nostalgic, from the moment of its inception; the romans antiques describe the lost chivalry of Troy, Chrétien locates the chivalric Golden Age in the reign of Arthur. To set against that, however, is the fact that romance as we know it is the product of identifiable and specific changes in social practices, and therefore much more closely modelled on the immediate conditions of contemporary life than our association of the form with dragons allows. Far from being always exotic and implausible, romance would be almost unimaginable without those changes, which were happening just ahead of, or contemporary with, the emergence of romance itself in the mid-twelfth century. The simplest example is a purely technological one: the introduction of the stirrup in the early Middle Ages. That in turn enabled the mounted charge, impact combat, of the knight with the heavy lance couched under his right arm.6 Chivalric romance appears within a couple of generations of the introduction of such horseback combat (and of course the French terms chevalier,, chevalerie, literally ‘horseman’ and ‘horsemanship’, make the connection explicit, as the English ‘knight’ and ‘knighthood’ fail to do). Fighting of that kind in turn demanded heavier armour —plate armour. Knights in shining armour may look like fantasy figures to us, but shining armour developed out of the same practical considerations that enabled the emergence of chivalric romance; and its authors did not forget, as we tend to do, that armour needed to be kept shiny, to have the rust removed.7 Still more important to the emergence of romance, and indeed to the whole history of western Europe, were two more far-reaching social changes, both of them to do with those central concerns of the medieval secular world, inheritance and the family. One was the categorization of the principles of primogeniture in the twelfth and thirteenth centuries. It had long been normal practice for the eldest son to succeed to his father’s lands and tide; but if there were no suitable or obvious heir, then the dde had commonly passed to the most competent claimant - a system enshrined, for instance, in the election of the Holy Roman Emperor. Alternatively, a king could name his own successor, as Edward the Confessor named William of Normandy. Primogeniture as it developed in these centuries, however, insisted that there was only one right heir to a title, or a throne. That was in the first instance the eldest son and his issue (so that if the eldest son predeceased his father, his own eldest son was given precedence over the next living brother); if there were no son, then the inheritance passed to the eldest daughter; or if a direct line failed altogether, an elaborate series of rules was devised for working back up the generations and down again to establish the correct inheritance. What was initially set up as a legal principle rapidly came to be interpreted as ordained by God, a divine as well as a human law. On the death of a prince, you have to identify not just the legally correct heir, but the true heir in sight of God. 6 7

Discussed in e.g. Keen, Chivalry, pp. 23-25. A rust-removal process is part o f the service provided for Gawain at Bertilak’s casde (Sir Gawain and the Green Knight, ed. J.R.R. Tolkien and E.V. Gordon, 2nd edn rev. Norman Davis (Oxford, 1967), lines 2 0 17 -18 ); and Launfal returns from time to time from Olyroun to joust in order to ‘kepe his armes fro the rustus’ (Middle English Verse Romances, ed. D.B. Sands, EMETS (1969; repr. Exeter, 1987), line 1028).


When Romance Comes True The rules of primogeniture were presumably designed to prevent disputes over inheritance. In practice, they made them much worse. Historically and politically, they made the problem of a weak or tyrannical or mad king, or of an infant heir, or of an heir whose paternity was in doubt, impossible to resolve, since the replacing of a king or an heir meant, by definition, unrightful rule. It was that kind of situation that enabled successive English kings to lay claim to the throne of France in the Hundred Years’ War, when all the lines of inheritance except that of Edward III lay in an impossible tangle. It was the need for divine endorsement too that made Joan of Arc’s advent at the end of the war so important, not so much for military strategy, but because her appearance seemed like direct divine intervention on behalf of the man who therefore must be the true king, whatever the English claims or the doubts over his paternity. The other social change occurred in the mid-twelfth century, with the papal decision that what made a marriage valid was not a public ceremony nor parental arrangement, but simply the consent of the spouses.8 In everyday practice, this probably made very little difference; arranged marriages (as distinct from forced marriages) continued to be the norm. Combined, however, with those new principles that bestowed a father’s lands and titles on his daughter if he had no son, it potentially gave extraordinary political, economic and erotic patronage to the heiress. Her erotic patronage, moreover, was interpreted in romance not just as consent, but as free and faithful sexual choice. So if the invention of stirrup and armour and lance enabled chivalric romance, these other changes enabled all those romances about the dispossession and return of the true heir, or about the fair unknown who turns out to be the missing claimant; and they enabled too those other twinned romance plots, of the young man who makes good by marrying the titled heiress, and of the young woman who makes her own choice of husband and pursues that choice through all kinds of adversity - plots that constitute a high proportion of Middle English romance. Given the basis of such stories in actual inheritance practices, it becomes less surprising that history and romance can sometimes chime very closely: closely enough for poets to rewrite history into romance, to mythologize history, even as it happened, or for people caught up in political events to see themselves as participating in those quasi-mythic romance structures, structures that insisted that what was happening was providential, willed by God. There was a particular incentive to cast events in these terms if what you were doing (deposing the king, for instance) was driven by political ambition or desire for power, or if you knew that your claim to the title you held or desired was not as indisputable as you might have hoped. In such cases, there was all the more reason to present your claims and actions - to spin them - in just such patterns of divinely sanctioned romance. Spin is most typically thought of as antagonistic to truth; but events could also be spun to resemble romance motifs in ways that endorsed genuinely held beliefs rather than setting out to fabricate belief where none might otherwise exist. The rest of this paper will consider some instances of historical spin of all these kinds: romance as

On the edict and its context, see Neil Cartlidge, Medieval Marriage: Liferary Approaches, 11001300 (Cambridge, 1997), pp. 12 -19 .


Helen Cooper propaganda, whether employed idealistically or cynically, or, as is normal with human motivation, something of both at once. An early story of a dispossessed heir, Havelok, furnishes a familiar and transparent point from which to start. It was in origin an English legend that first appears in a chronicle setting, Gaimar’s Anglo-Norman Estoire des Engleis of about 1137. Over the course of the next century or so, it was reworked as an independent romance, first in Anglo-Norman, then in English. It re-entered chronicle history in the Prose Brut at about the same time as it was given its English romance treatment; and the story then cut between chronicle and romance for the next few centuries, becoming increasingly unrecognizable in the process, until it dropped decisively away from history with its conversion into a sentimental ballad in the eighteenth century.9 It is in fact a story about two dispossessed heirs, children who are disposed of after the deaths of their fathers by wicked guardians who want to keep power for themselves. Havelok, son to the king of Denmark, is ordered by his guardian to be killed; but he is saved when the wife of the fisherman Grim, his designated murderer, sees a light coming from the child’s mouth, and they recognize a bright birthmark on his shoulder as a ‘kynemark’,101a birthmark defining his kin as royal, a king-mark. Grim escapes with him to England, where his homestead becomes the origin of the future Grimsby: a major function of the legend, in fact, was to provide a foundation legend for the town, a legend recorded on its seal and still familiar in the early seventeenth century. Meantime Goldeburh, the orphan daughter of the king of England, is also being raised by a wicked guardian. (In the chronicle versions, he is her uncle, the male equivalent of the wicked mother-in-law, and for analogous reasons: both are cut off from potential or real power, one by the existence of the heir or heiress who prevents what would otherwise be his own inheritance, the other by the advent of the young wife who supplants her as the senior woman of the dynasty.11) In order to keep power for himself, he decides to interpret literally the promise he made to her dying father to marry her to the strongest and highest man in England, in the form of a heroically tall and athletic young scullion employed by the Bishop of Lincoln - a scullion who is, of course, Havelok. On their wedding night, Havelok, exhausted by his day’s labour, falls asleep; and she in her turn, grieving over her compelled fate, sees the light from his mouth, and a further sign of royalty, a king-mark on his shoulder in the form of a gold cross:

For its early history, see the edition by G.V. Smithers, Havelok (Oxford, 1987), pp. xvi-lvi. For its post-medieval history, see Helen Cooper, ‘The Elizabethan Havelok'. William Warner’s First V’ o f the English’, in Medieval Insular Romance: Translation and Innovation, ed. Judith Weiss (Cambridge, 2000), pp. 16 9-83, and, for a text overlooked there, a reworking o f Warner entitled ‘A Song o f the Strange Lives o f two Young Princes in England’, which relocates the story to Devonshire and entides the lovers Raymond and Maudlin (in A Collection of Old Ballads, vol. 3, possibly ed. Ambrose Philips (London, 1725), pp. 1-10). 10 Ed. Smithers, line 605. 11 The uncle in question is thus most often the father’s brother, the second son whose inheritance is foiled by the existence o f the child. In the Havelok stories, he is the dead mother’s brother: he therefore has no lineal claim to the throne, but his selection as guardian shows him to be the highest-ranking competent male, and therefore the kind o f man who under the older more flexible inheritance patterns that were in the process o f being displaced, or under a system o f election or acclamation, could have expected to succeed as ruler. 9


When Romance Comes True On hise shuldre, of gold red, She saw a swiþe noble croiz. « Of an angel she herde a uoyz: ‘Goldeborw, lat þi sorwe be! For Hauelok, þat haueþ spuset þe, He [is] kinges sone and kinges eyr Þat bikenneth þat croiz so fayr.’ (lines 1263—69) The angel's message not only interprets the physical symbol, but gives his royalty divine endorsement. Only at that point does Goldeburh make her own willed election of Havelok as her husband, with the implication that the consummation of their marriage, her full sexual choice, follows from that act of her will. In due course he wins back both his kingdom and hers, and rules them jointly; and the wicked guardians come to a nasty end. Whatever Gaimar or the compiler of the Prose Brut or the good folk of Grimsby or Lincoln thought about the story, there is no historical evidence that anything like this ever happened. What is likely is that the legend emerged in response to cultural pressures: in this instance, it has been argued, by the need retrospectively to legitimize Danish rule in England, especially in the eastern areas, not least Lincolnshire, that had embraced it so readily.12 The romance of Havelok addresses precisely that historical fact. It casts itself as predictive of the Danish rule of England that did indeed come about, even if it did not occur in anything like the way the romance represents it. If the Danes had been defeated, a romance might still have been produced at some point in the future, but it would not have been one that put a Dane on the throne of England, and that made his heirs legitimate linear English rulers through his marriage. To us, Havelok is a romance precisely in the sense that it is not true, and the element of miracle it contains, that divine symbol of true royalty, confirms that; but it is dangerous to make assumptions about its fantasy on that basis. Let me step aside into the historical record, to the year 1238, when the legend of Havelok was apparently already long established in oral tradition and familiar in written form in Anglo-Norman, though probably still before the Middle English romance had been composed. Here is another story, from the Greater Chronicle of Matthew Paris, about Henry III: In the same year, a great danger beset the king, such as astonished all those who heard it. On the morrow of the nativity of the blessed Virgin, a certain squire who was said to be educated came to the royal court at Woodstock, and pretending to be mad, he said to the king, ‘Resign to me the kingdom which you have usurped unrightfully and held for yourself too long.' He also added that he bore a king-mark on his shoulder (signum regale in humero). The king’s servants seized him and wanted to beat him out of the royal presence, but the king stopped them as they ran on him, saying, ‘Leave the lunatic alone, as it’s natural for someone like that to play the fool; such men’s words carry no weight of truth.’ But in the middle of the night, that same man got in 12 See Thorlac Turville-Petre, ‘Havelok and the History o f the Nation’, in Readings in Medieval English Romance, ed. Carol Meale (Cambridge, 1994), pp. 121-34.


Helçn Cooper through the window of the king’s chamber, carrying an unsheathed dagger, and rushed in a frenzy on the king’s bed. He was baffled at not finding him there, and hurriedly searched for him in all the comers of the chamber. By the Lord’s providence, however, the king was lying with the queen. A certain maid of the queen’s was by chance awake, reading her psalter by candlelight; she led a holy life in devotion to God, and was called Margaret Biseth.13 So Margaret screams and raises the alarm, and the king is saved. The squire who had wanted to assassinate him (to kill him in the manner of the assassins, more Assessinorum) is tortured until he names his • co-conspirators, whereupon he is condemned as a traitor and executed in appropriately nasty ways. It is a story about a claimant with a king-mark who did not succeed; and it remains as history, not romance. It is however worth pausing on the evidence that the madman, or the feigned madman, cites for his demands: to paraphrase just a little, ‘Resign the kingdom to me, for you have usurped it unrightfully, and I bear a kingmark upon my shoulder.’ How do you know who is the rightful king? The laws of primogeniture insisted that there was one, and one only. Henry’s father, King John, was not such a rightful heir, having overridden the claims of Prince Arthur, the young son of his elder brother Geoffrey: John was, in fact, the conventional wicked uncle. And even if a man plausibly claims to be the son of the rightful king, how can you be sure that he is what he asserts, in an era before DNA testing? Another unknown young squire, named Arthur, drew a sword from a stone to prove his right to the throne; the scullion Havelok had his king-mark, the gold cross inscribed on his body, and the light from his mouth that became visible in the dark, which marked him indelibly even in the most adverse of circumstances - providentially endowed and endorsed signs. So the squire of Woodstock who demanded Henry’s throne from him, claiming a similar king-mark, was much more dangerous than we might at first glance think. Henry’s dismissal of his words as the ravings of a madman may have been humane, but it was also politically astute, since it disarmed the force of his demands. The rest of the story, however, recasts the lunatic as only pretending to be mad, as his naming of a further group of conspirators confirmed. If the man were indeed sane, it none the less seems a crazy way to go about mounting a conspiracy; but the claim he makes about his king-mark was presumably thought by the other malcontents, if they indeed existed, to carry real weight - for if a king is faced with a man who makes such a claim, how can he prove his rival is not what he says he is? 13 Eodem anno accidit regi periculum, omnes audientes nimis reddens attonitos. In crastino enim nativitatis beatae Mariae, venit quidam armiger literatus, ut dicitur, ad curiam regis apud Wodestok, se fingens infatuatum, dicens regi: ‘Resigna mihi regnum, quod injuste tibi usurpasti, et diu detinuisti.’ Addidit quoque, quod signum regale gestabat in humero. Quem cum ministri regales arripuissent, volentes eum baculatum a praesentia regis propellere, rex irruentium in eum impetum compescuit, dicens: ‘Sinite infatuatum ut talem decet desipere; verba enim talium carent pondere veritatis.’ Media autem nocte, ecce ille idem per fenestram regii thalami introgressus, cultellum portans extractum, lectum regis adiit furibundus; quem cum non invenisset, confusus est; sed festinus quaesivit eum per plura thalami diverticula. Erat autem tunc temporis, Domino providente, rex quiescens cum regina. Quaedam autem puella reginae, cum forte vigilaret, psalterium psallebat ad candelam; erit enim sancta et Deo devota, nomine Margareta Biseth (Matthew Paris, Chronica Maiora, ed. H.R. Luard, Rolls Series 57 (London, 1872-83), 3.497; my translation).


When Romance Comes True At one level, Matthew Paris, like the king himself, seems to want to dismiss the affair as the actions of a madman; but he is also anxious to prove the claim untrue, and therefore treasonable not just in men’s sight but God’s. Havelok being a romance, the king-mark is a true, and therefore also divine, signifier, and the story records its hero’s restoration; there, the man who holds the throne unrightfully is the one who comes to a sticky end. In Matthew Paris’s story, Henry’s legitimacy as king is confirmed by enlisting God on his side. It is by God’s providence, Domino providente, that the king is not in own bed; and the devout Margaret Biseth, reciting the psalms and therefore with a hotline already open to God, serves as the divine agent in raising the alarm. That may indeed have been what happened; but if it was not, something of the kind would have had to be invented something that demonstrated that the man who occupied the throne was indeed the true king in the sight of God. The contemporary stories of the scullion of Lincoln and the squire of Woodstock invite reading against each other. Jump forward two and a half centuries, and you find another set of contemporary romances that invite similar parallel readings between their own texts and the sequence of children, men and one woman who in the years following 1483 all claimed to be the true heir to the throne of England. Two of these are fifteenth-century prose works emanating originally from Burgundy that were translated into English on either side of 1500: Blanchardyn and Eglantine, translated by Caxton around 1489; and Olyuer o f Castylie, printed in 1518. Probably dating from slightly later is a third text, a ballad-style romance entitled Lady Bessy, much more demotic in style and dissemination,14 that fictionalizes history more directly: most of its characters are historical, but their actions, as in a historical novel, are rewritten to produce a version of events that is close enough to fact to be credible but that reaches its final outcome (here, the Tudor takeover) by imaginative means. The background to the late fifteenth-century struggle for the English throne went back almost a century, and demonstrated all the problems consequent on the equation of the rightful monarch with the true heir as defined by the system of primogeniture. The trouble had started in 1399 when Henry Bolingbroke deposed Richard II; both were grandsons of Edward III, Richard through his eldest son, the Black Prince, Bolingbroke through the third son, John of Gaunt, duke of Lancaster. Since Richard was so evidently the true king in linear terms, this necessitated the fiction that Richard had not been deposed, but rather had freely resigned the crown to Bolingbroke as his designated heir, somewhat as Edward the Confessor had designated William the Conqueror. Bolingbroke’s reign was, however, haunted by stories that Richard was still alive, that he was a king in waiting, like the dispossessed heir of romance, for the moment of his return.15 In addition, and less spectrally, Bolingbroke was beset by the descendants of Edward’s second son, whose claims he 14 It now survives in two manuscripts, one among the papers o f John Stowe in London, British Library, Harley MS 367, and also in the collection o f popular literature assembled in the Percy Folio Manuscript. It is printed as The Most Pleasant Song of Lady Besy, ed. J.O. Halliwell, Percy Society 20 (1847), and in Bishop Percy’s Folio Manuscript: Ballads and Romances, ed. John W. Hales and FrederickJ. Furnivall (London, 1868), 3.319-63. 15 On these issues see in particular Paul Strohm, England’s Empty Throne: Usurpation and Textual Legitimation, 1399-1422 (New Haven and London, 1998).


Helen Cooper had simply overridden - and done so all the more easily since the grandchild initially in question was female. The importance of the equation of rightfulness with the rules of primogeniture necessitated the further fiction, accepted by Parliament as underlying his claim, that he was ‘desendit be right hyne of the Blode comyng fro the gude lorde Kyng Henry therde’, on the grounds that Edmund Crouchback, earl of Lancaster and younger brother of Edward I, was in fact the elder son of Henry III, and so carried a superior right to the throne. Bolingbroke accordingly claimed the crown not by virtue of his descent from Edward III, but through his mother, Blanche of Lancaster.16 Perhaps not surprisingly in view of its implausibility, this rewriting of history proved no deterrent to the displaced descendants of the senior line from Edward III. It was their claims that haunted the whole Lancastrian dynasty and finally overturned it, even though it took five generations of Yorkists before they succeeded; for a claim based on primogeniture never goes away so long as the line continues. The Yorkists could and did represent themselves as the equivalent of Haveloks, true heirs emerging from the shadows to claim the throne that was rightfully theirs. While a strong man held the crown, rival claimants stood litde chance, as the Lancastrian Henry V disposed of the earl of Cambridge, and as the Yorkist Edward IV could keep the last Lancastrian claimants at bay; but Edward died when his sons were still children, and the linear system promptly broke down. There followed two successful usurpations by men who had no valid claims from primogeniture; and a third, unsuccessful, attempt by a pretender who did make such a claim, but who failed to impress it on history. The first usurpation was Richard duke of Gloucester’s seizure of the throne from the young heirs of Edward IV. He justified his action by claiming that they were illegitimate, on the grounds that an earlier contract of marriage entered into by Edward rendered his marriage to their mother bigamous or adulterous, or indeed both; but that still left the child of an intervening brother, the young earl of Warwick, surviving, just as Bolingbroke had ignored the line intervening between the Black Prince and John of Gaunt. In the late fifteenth-century case, the boy in question may well have been feeble-minded; but in genealogical terms, that made no difference to the linear strength of his claim. Richard might pragmatically be the man best equipped to rule, but in no way was he the rightful heir. The disappearance of the princes from the Tower, whatever in fact happened to them, did not help; for it was all too familiar as an act of a usurping tyrant or a wicked uncle, like the ones in the Havelok story, to try to kill the child heir. Whether or not Richard was actually guilty counted for nothing beside the fact that he was believed to be guilty. The next usurpation followed from the first both chronologically and logically: Richard’s failure to impress his legitimacy on his subjects made Henry Tudor’s takeover all too easy. Henry had an even less plausible claim than Richard: his accession indeed marked the biggest disruption to the linear descent of the crown since the Norman conquest. He too was descended from John of Gaunt, founder of the Lancastrian line, but illegitimately, and even though the duke had eventually married their mother, the Beauforts had been explicidy excluded by Act of 16 John Ashdown Hill, ‘The Lancastrian Claim to the Throne’, in Tant d’Emprises —So Many Undertakings: Essays in Honour of Anne Sutton, ed. Livia Visser-Fuchs, The Ricardian 13 (2003), 27-38; quotation from p. 30 (my italics), citing Rota Parliamentorum (1832), 3.422-23.


When Romance Comes True Parliament from any claim to the throne. Richard, however, was killed in battle against him at Bosworth, and Henry had Parlianient declare him king by virtue of the indisputable fact that he occupied the throne; and he proceeded to liquidate every possible rival claimant over the next few years, the feeble-minded earl of Warwick among them. It was all the more necessary, therefore, for Henry to mythologize his seizure of throne on the romance model, to claim a status as the divinely identified true heir. How well he succeeded can be measured by the fact that we never describe him as a usurper: we still buy into the Tudor myth of rightful kingship. It was not, however, an easy myth to create. There were indeed Welsh prophecies ascribed to Merlin of the advent of a ‘son of prophecy’, which Henry could apply to himself; and prophecy, the foreseeing of the present in the past, was a way of guaranteeing that what was happening in the present was right, was divinely foreordained. Writing the Faerie Queene a century later, Edmund Spenser similarly found Merlin useful for prophesying just such a providential advent of the Tudors.17 Henry also claimed that God had made His own views clear, not by a king-mark, but through trial by batde on Bosworth Field. He made some claim to being in the line of descent from Arthur, though he did not press that too hard, as it lacked plausibility as grounds for asserting a contemporary right to the English throne even in the age of Sir Thomas Malory;18 he famously called his eldest son Arthur, but the ploy died with the child. In addition, Henry’s mother, Lady Margaret Beaufort (who, if his linear claim had anything in it, should herself have been the one to be occupying the throne), commissioned a translation of a romance from Caxton that offered a story analogous to Henry’s. She was renowned for her piety, and this was the only secular work in which she ever showed any interest. Blanchardyn and Eglantine describes how a young prince leaves home to test himself in chivalry; in his absence, his father is overthrown by pagan enemies, and he himself in due course returns to claim his own tide and to marry a neighbouring heiress. The story offers a series of parallels to the overthrow of Henry VI, Henry Tudor’s sojourning on the continent to keep himself safe from any Yorkist attempts to harm him, and his return from over the sea (a distincdy English motif, as Rosalind Field has pointed out) to recover his throne and to marry the heiress to the Yorkist line, Elizabeth of York, Edward IV’s eldest daughter and, since the presumed death of her younger brothers, his linear heir.19 Blanchardyn and Eglantine thus provided the romance patterning that the Tudor takeover so singularly lacked. It suggests that what happened was not usurpation but the return of the rightful heir, and so offered a way to assimilate the deeply disturbing historical and genealogical upsets of the Tudor accession as right and proper. Blanchardyn was not the most obvious, nor at first glance the most appropriate, choice of romance for Lady Margaret to have selected for translation. She might 17 Edmund Spenser, The Faerie Queene, ed. A.C. Hamilton, 2nd edn (Harlow, 2001), III.iii.48. In II.X.75, he invents an elfin genealogy for the Tudors, so bypassing the problematic nature o f their lineal claim. 18 See Sydney Anglo, ‘The British History in Early Tudor Propaganda’, Bulletin of the John Bylands Library 44 (1961), 17 -4 8 , and his revisions to those views in his Images of Early Tudor Kingship (London, 1992), pp. 40-60. Direct descent from Arthur was o f course impossible, as he died without legitimate issue. 19 Rosalind Field, ‘The King over the Water: Exile-and-Retum Revisited’, in Cultural Encounters in the Romance ofMedieval England, ed. Corinne Saunders (Cambridge, 2005), pp. 4 1-53.


Hekn Cooper more logically have chosen the mid-fifteenth-century French Olyuer o f Castylle (which is also discussed elsewhere in this volume by Elizabeth Williams), for this romance already fortuitously contained a wicked duke of Gloucester who usurped the throne of England. He is overcome in batde and killed by a stronger claimant named Arthur, whose claim derives from his wife’s status as the true heiress; and so the rightful line is restored. Olyuer does, however, have one decisive drawback, evident even in that short plot summary: it makes it very clear that the husband’s claim of kingship lies solely in his wife. That was a step too far for Henry Tudor. He claimed the throne in his own right, not in that of Elizabeth of York; and he was careful to establish his own hold on the crown before he married her, so that there could be no question that his kingship was in any way dependent on her. It is perhaps, therefore, not surprising that Olyuer o f Castylle was translated into English only after his death, in the reign of his son Henry VIII, who inherited the claims of both his mother and his father to the English throne.20 At first glance, Olyuer o f Castylle might seem like the perfect example of romance that comes true; but for all its coincidences with history, it was not a model that Henry VII himself could have tolerated. That did not mean, however, that there was no contemporary awareness of the possibility that both Henry and Lady Margaret rejected, that the crown righdy descended through the Yorkist line as embodied in Elizabeth. Her own historical story was given a romance treatment some time in the next few decades, to bring it into line with that social change mentioned earlier in this paper: the location of political and erotic patronage not just in the passive consent of heiress, but in her own active personal choice. Cady Bessy turns the story of Elizabeth of York, the Lady Bessy of the tide, into a romance of the dispossessed heiress who herself instigates the wooing of the man she loves. It might sound as if the true heir and the true heiress should have analogous biographies, but in fact there are interesting differences between them. Typically, the true heir is lost from sight: he becomes a foundling, a fair unknown, who may himself not know his true identity, and others certainly do not. He is brought up away from the court, out in the world at large. A woman, by contrast, is oppressed or imprisoned rather than lost. She typically remains within the land that constitutes her inheritance, as if she were a metonym for the territories she owns. The process of restoring her to her rightful position and power is a matter of rescuing her from a tyrannical father who forbids her choice of marriage partner, or from rival suitors of a highly undesirable sort (such as pagans), or from a usurper. Thus Havelok’s wife, Goldeburh, the heiress to England, is never ‘lost’ in the same sense as he is. He is brought up as a fair unknown in an alien country; she remains under her guardian uncle’s control. She does not need to be found, but to be rescued and married to the right man. Goldeburh is an unusual romance heroine in that her active choice of husband comes after her forced marriage. Most heroines make their own choice much more positively, as do Lavine in the Eneas, Rimenhild in Horn, Josiane in Bern o f Hamtoune, or the eponymous Melusine. Willed choice of this kind is especially common in genealogical romances, with their concern with the founding of a family, as if the future of a dynasty must lie in the active choice of the founding mother 20 The nearest thing to a modem edition is by Gail Orgelfinger, The Hystotye of Olyuer of Castylle (New York and London, 1988).


When Romance Comes True (Maddox’s ‘mega-mother7) more than the founding father; and Lady Bessy accordingly rewrites history to fit. The poem opens when Richard III is already on the throne. Bessy, who is consistently presented as the true heiress to her father, herself decides to woo the exiled Henry Tudor, whom she loves despite never having met him; and she sends messengers and money to bring him back over the sea. She possesses a book of prophecy foretelling that she will be queen, and there is no mention of any Tudor claim to the throne at all; Henry is merely the means to her own declared end. Bessy is accordingly present at the batde of Bosworth, as a kind of spirit of rightful victory, and she marries Henry on the field of batde. This gives a decisively Yorkist spin to the Tudor takeover - indeed it turns the Tudor myth into a Yorkist myth. Here, the Plantagenet princess legitimizes the Tudor gendeman, as if he were a squire of low degree winning the hand of a superior lady; which is perhaps precisely why Henry in fact made so sure that he established himself on the throne before he did marry her. There is no evidence whatsoever as to what the historical Elizabeth thought about the marriage, whether she was enthusiastic or reluctant, though she certainly had no choice in the matter. She is unlikely to have objected to becoming queen, especially in view of what the alternatives would have been: all the evidence we have indicates that Henry was much more ruthless than Richard III. A romance should end at that point, with the ‘true’ heir restored to the throne; that was how Henry and his mother and the Yorkist author of Lady Bessy aimed to structure their propaganda. Another claimant, however, soon made his appearance on the field of history, and this one had a still more compelling claim to a biography modelled on romance. There is something of the Havelok about him, and also something of a more up-to-date romance hero, the Valentine of Valentine and Orson, for Valentine becomes a foundling in consequence of a charge of adultery brought against his mother, just as this new pretender has lost his status as a result of a comparable charge. Valentine has no idea who he is; he too, however, bears ‘a crosse upon [his] shoulder, the whiche is also yelowe as the fyne golde’, a mark that makes him suspect that there is more to his lineage than he knows and that impels him to seek his true parentage; and in due course he recovers his status as heir to the emperor of Greece.21 His real-life counterpart was the young man we know as Perkin Warbeck; but to most of the crowned heads of late fifteenth-century Europe, he was Richard of England, a name he was accorded by virtue of his claim to be Richard duke of York, the second son of Edward IV and the younger of the princes in Tower.22 He had, he said, been spared (like Havelok) by the man who had been 21 Valentine and Orson, ed. Arthur Dickson, EETS OS 204 (1937), quotation from p. 85; the English translation was made by Henry Watson. Its first edition dates from some time in the first decade o f the sixteenth century - interestingly, after the threat represented by Warbeck had been eliminated (both he and the earl o f Warwick were executed in 1499); but it may date from as late as 1510, by which time any coincidence between the stories would have ceased to resonate. 22 For a double biography o f Warbeck/Richard, see Ann Wroe, Perkin: A Story of Deception (London, 2003). Warbeck did declare himself to be an impostor on the scaffold, a moment when it would be very unlikely indeed that he would not tell the truth; but by that time he had an infant son, and therefore also a strong motive to try to protect the child from the consequences o f royal birth. Nothing further is known about the child, or the circumstances o f its death.


Helen Cooper ordered to murder him; he had then (like Havelok) escaped over the sea; and he now came back from the dead, returned to claim his throne. He further declared that he bore marks on his body that proved his identity - in effect, king-marks, like Havelok’s and Valentine’s, though the reference is presumably to distinctive birthmarks (also widespread in romance: Cymbeline’s lost eldest son and heir is identified by just such a birthmark).23 He furthermore carried himself with a natural grace and authority such as all those other fair unknowns had possessed who turned out to be indeed the heirs to great fathers. His supposed aunt, Margaret of Burgundy, seems to have believed he was genuine, as did Charles VIII of France. James IV of Scotland gave him a close kinswoman as his wife; and Margaret’s son, the emperor Maximilian I, recognized him as Richard IV. If he were indeed the son of Flemish parents, as Henry VII claimed, it was odd that he spoke English perfecdy, with no accent; and it was odd too that after he had captured him, Henry absolutely refused to have him brought face to face with those supposed parents. Was he afraid that they might confess that they were not his true parents, just as Grim was not Havelok’s father, nor Sir Ector Arthur’s? Francis Bacon, writing his history of Henry VII a century later, confessed himself baffled as to just what the truth of the matter was.24John Ford, in his play with the double-edged title The Chronicle Historie o f Perkin Warheck: A Strange Truths has Henry (of course) insist that Perkin is no more than that, but neither Warbeck nor the play offers support for that insistence: it leaves the question of his real identity, the ‘truth’, unresolved and ‘strange’, though the play’s sympathies clearly lie with the pretender.25 That we speak of him now as Perkin Warbeck simply echoes the verdict of history. If he had succeeded - if, when he invaded England with a pitifully small force, the people had risen in his support then we too would know him as Richard IV. Henry Tudor would be no more than a brief interlude in the royal line of the Plantagenets, an adventurer who had seized the throne and forced the heiress into marriage, only for the foundling prince who had escaped death to return and claim his crown, to assert his own right above his sister’s; for if Henry is cast as the wrongful king, then the shape of his life becomes not romance wish-fulfilment, but nightmare. The point at which we traditionally mark the end of the Middle Ages, when the dynasty of the Plantagenets was replaced by the Tudors, thus offered itself as four different romance plots: as a story of the fulfilment of supernatural prophecies; as a story of the dispossessed heir, Henry Tudor; as a story of the dispossessed heiress, Elizabeth of York; and very nearly, as a story of the true heir spared as an infant from murder by his wicked uncle, and who returns as a fair unknown to claim the throne that is rightfully his - though we hear of Richard of England now only as Warbeck, a mere impostor on the edge of more significant political events. The only 23 Cymheline, 5.6.365-70, in William Shakespeare: The Complete Works, ed. Stanley Wells and Gary Taylor (Oxford, 1986). Warbeck is unlikely to have had Havelok itself in mind, as the romance is not known to have been copied after c. 1400 (Smithers, p. xv) and the chronicle versions omit the king-mark; he could well have known the French Valentin, but that itself witnesses to the continuing currency o f the motif. The return o f an exiled hero from over the sea was also widespread: e.g. in Blanchardyn, or in the prose reworking o f the Horn romance entitled

Ponthus and Sidoine. 24 Francis Bacon, History of the Reign of King Henry VII, ed. Brian Vickers (Cambridge, 1998), p. 96. 25 In John Ford: Three Plays, ed. Keith Sturgess (Harmondsworth, 1970).


When Romance Comes True figure of all these ambitious climbers whose story comprehensively resists any such romance shaping is Richard III, the one among them most demonized by history; for history, as we know it, is romance written by the victors.


The Curious History o f the Matter o f England R o sa lin d F ield

The discussion of several of the texts central to our understanding of insular romance habitually employs the familiar term ‘Matter of England’. The familiarity has encouraged a comfortable sense of the meaning and value of the term; this paper aims to investigate whether such confidence is well-founded and to trace the development and use of the term, and its effect on the perception of the romances associated with it. The term ‘Matter’ is modelled on Bodel’s famous classification of the subjects of narrative into the three Matters of France, Rome and Britain, a genuine, and rare, glimpse into the medieval sense of narrative.1 However, Bodel’s twelfth-century classification is itself not as straightforward as it is often assumed. The recent study of the origins of romance by D.H. Green draws attention to Bodel’s assessment of the differing status of the three Matters: that of France is voir (true), that of Rome is sage (wise, knowledgable), while the Matter of Britain is vain et plaisant (frivolous or false entertainment).12 It is in Bodel’s interest thus to privilege the Matter of France, to which he is contributing, but this also provides a gesture of support for the political hegemony of France in competition with the Anglo-Norman exploitation of Arthurian material, and that of the Matter of Rome by the German Empire.3 In terms of modern romance criticism we can also see that Bodel’s perceptions have as much to do with modality as with matter - that in his hierarchy of truth Arthurian romance is more quintessentially faithful to the fictionality of the genre as it has developed, a finding with which most modern readers would concur. But how does this give rise to the ‘Matter of England’? This is not, of course, a medieval term used by Bodel or anyone else, and there was neither the literary material nor the contemporary political interest to support it. Nor does the term come from the nineteenth-century ‘inventors’ of Middle English. Although there was plenty of interest in the origins and nature of English romance amongst English and continental scholars in the nineteenth century, their concerns are more with the perceived Germanic origins of that romance. For early scholars such as Ritson and Ellis and the early editors, Madden, Furnivall and Skeat, as John Ganim has shown, 1 Jean Bodel, La Chanson des Saisnes, ed. A. Brasseur, 2 vols (Geneva, 1989), lines 6-7. 2 D.H. Green, The Beginnings ofMedieval Romance: Fact and Fiction, 1150-1220 (Cambridge, 2002), pp. 138-9. 3 For the interests o f the German Empire, see Green, Beginnings, p. 235, n. 57, cidng Heinz Thomas, ‘Madère de Rome - Matière de Bretagne. Zu den politischen Implikationen von Veldekes “Eneide” und Hartmanns “Erec’”, Zeitschriftßr deutsche Philologie 108 (1989), 65-104.

Rosalind Field romance was a central genre, as important to understanding the development of Middle English literature as Chaucer and Langland.4 But they were not in the business of constructing the ‘Matter of England’.5 It was evidendy the 1906 publication of Bodel’s Chanson des Saisnes, with its famous lines about the three matters of medieval literature6 that encouraged literary scholars to adopt the term ‘Matter of England* —of whom the first seems to have been W.H. Schofield in the same year: These different “matters** [Bodel’s of France, Britain and Rome] we shall discuss in the order given by Bodel, developing, however, two more by separating from the Matter of Britain stories of Germanic origin, and from that of Rome those that have their source in the Orient.*7 Schofield is here treating Bodel’s classification as simply one of matière, ignoring the issues of status and veracity and he is of his time in extending this to a theory of Germanic origin. It becomes obvious that his concept of ‘Matter of England* is not only a means of describing content, but also of tracing a native narrative - a ‘saga* - beneath the foreign veneer of Frenchness: It seemed almost necessary at first for a saga to assume the exterior semblance of French poetry in order to maintain its dignity [ that,] reading about their own heroes, Englishmen learned foreign ways.8 Having thus invented a new Matter he draws up an ambitiously comprehensive list of texts to represent it. His Matter of England includes Havelok, Waldef (assuming a lost translation into English), Bern o f Hamtoun, Guy o f Warwick,, all the versions of Horn (the Anglo-Norman Romance o f Horn as well as King Horn and Horn Childe) Atheisten, the Robin Hood ballads, Gamelyn and other outlaw tales. He wants to include Tristan but for the ‘superadded Celtic tone’ (p. 260). If Schofield’s influential grouping chimes with pro-Germanic cultural constructions of the early twentieth century, its terminology is adaptable enough to dodge the association when it becomes awkward as the century develops. The ‘Matter of England* can shake off its assumed Germanic origins as necessary while still acting as a definer of Englishness.9 The other purpose here is that of detaching medieval English literature, especially narrative literature, from any French associations. At this date there is little use of the concept of ‘insular’ with regard to English French-language literature, although the 4 John Ganim, ‘The Myth o f Medieval Romance’, in Medievalism and the Modernist Temper, ed. R. Howard Bloch and Stephen G. Nichols (Baltimore, Md, 1996), pp. 148-66. 5 Joanne Charbonneau attributes the first use o f ‘Matter o f England’ to George Ellis in 1805 following Bodel’s model. However, Ellis has his own idiosyncratic terms ‘Anglo-Saxon romances’ for Guy and Bevis and ‘Anglo-Norman romances’ for Richard Coeur de Lion. Moreover, Bodel’s work was not available to him. See George Ellis, Specimens of Early English Metrical Romances, 3 vols (London, 1805), 2 :3-174; and Joanne A. Charbonneau, ‘Romances, Middle English’, in Medieval England, Garland Encyclopedias o f the Middle Ages (Vol 3), ed. Paul E. Szarmach et al. (New York, 1998), pp. 646-49, at p. 647. This is a telling instance o f the widespread impression o f the antiquity o f the classification. 6 Jean Bodel, Saxonlied [Les Saisnes], ed. F. Menzel and E. Stengel, 2 vols (Marburg, 1906, 1909). 7 W.H. Schofield, English Literature from the Norman Conquest to Chaucer (London, 1906; repr. 1925), pp. 258-82. 8 Schofield, English Literature, p. 259. 9 For the situation in England after the First World War see, for example, David Matthews, The Making ofMiddle English, 1765—1910 (Minneapolis and London, 1999), pp. 188-89: he cites the 1921 Newbolt Report: ‘English studies is to be purged o f its Germanisms.’


The Curious History o f the Matter o f England nineteenth century English scholars had been comfortable with the notion if not the term, Madden for example accepting that there must have been many native English writers writing in French.1012However, Schofield’s list, including as it does the extant French-language version of Horn, does raise the issue of whether Matter of England romance can be such if not written in English, and scholars since have differed on this. The next influential use of the term is another American scholar, H.L. Creek, writing on ‘Character in the Matter of England Romances’ in 1911,11 for whom Schofield’s new category is convenient. He does not acknowledge Schofield, nor define the term; his texts are King Horn, Havelok the Dane, Bevis and Guy - ‘the four most important of the “matter of England” romances’ (his quotadon-marks). This unexplained and unexamined use of the term ‘Matter of England’ may be the cause of its ubiquity; it seems, lacking explanation or definition, to have an established authority. So for U.S. editors French and Hale in their important 1930 anthology it requires no introduction, but is now represented by Havelok, King Horn, Athelston and Gamelyn}2 It is significant that it is an invention not of English literary historians, but of those from the United States. That the Matter of England might link medieval English literature to precisely that founding body of Anglo-Saxon texts that Frantzen has discussed as contributing to the nineteenth-century American sense of national, linguistic and legal identity is particularly attractive to those constructing Middle English as a subject for serious study in American universities.13 The Matter of England becomes necessary and visible not as evidence of a late medieval sense of national identity with which it has since been associated, but rather when ‘England’ becomes a multi-national, international, concept - a shared past, reaching across the Empire of the nineteenth century and North America. It is a concept of England and English literature that requires a substantial body of legendary material - its own Matter, to equal those of European romance. In this respect it is a development comparable to Bodel’s original three Matters, an ideological constuct, not simply a classifying label. It is not a neutral term. However, it must be recognised that for most scholars and readers of romance, the primary attraction of the Matter of England classification is as a means of bringing some order into the unwieldy taxonomy of Middle English romance, reducing the number consigned to the bin of ‘Miscellaneous’. But even as a classification there are problems, as has long been recognised. One problem is that of which texts to include. As we have seen, Schofield, followed by Creek, concentrates on Havelok, King Horn, Guy and Bevis, while Schofield further includes Athelston, the Anglo-Norman Horn and outlaw material. French and Hale select their own small group, presumably a selection from Schofield’s original list. Derek Pearsall in the 1974 Cambridge Bibliography o f English Uterature, lists the Matters of Britain, France and 10 F.M. Madden, Havelok the Dane (Roxburghe Club, 1828), pp. xlv-xlvi, cited by Matthews, p. 123. See also Ellis, Specimens, vol. I, section II. 11 H.L. Creek, ‘Character in the Matter o f England Romances’, JEGP 10 (1911), 429-52, 5 85609. 12 W.H. French and C.B. Hale, eds., Middle English Metrical Bornâmes (New York, 1930). 13 Allen J. Frantzen, The Desire for Origins: New Language, Old English and Teaching the Tradition (New Brunswick and London, 1990), pp. 2 0 1-25.


Rosalind Field Rome, and then has a section headed - in quotation-marks - ‘Matter of England’ which includes Horn, Havelok, Atheisten, Gamelyn, Guy, Bevis, Richard Coeur de U on and William ofPaleme .14W.R.J. Barron’s important study of 1987 has a Matter of England group predominantly made up of Havelok, King Horn, Guy and Greenwood romances, including Gamelyn but his discussion also touches on Bevis and Atheisten .15 Richard Coeur de Uon appears in a later section as a ‘derivative of the Matter of England’ (pp. 179—80). Diane Speed has perhaps the most generous collection, including the Auchinleck Arthour Merlin, Sir Tristrem and Sir Otfeo by virtue of that manuscript’s setting of their action in England.16 As well as differing opinions as to which texts to include in the category, there is an increasing note of unease as the twentieth century progresses about the value of using the classification by Matter at all. By 1965, Derek Pearsall in his ground­ breaking re-examination of the corpus, can write of the ‘traditional classification according to the Matters (of Britain, France, Greece and Rome, and England)’ and notes that ‘this classification is used in virtually all the standard textbooks and bibliographies’ - and there is an implication here that the Matter of England is as ‘traditional’ as the other three.17 He dismisses the classification by Matter as ‘misleading where it lumps together quite dissimilar romances because of some superficial coincidence of plot material’, in favour of an analysis through form and style.18 A similar impatience is voiced by John Finlayson in 1980:


Nor is the classification by matières much more useful [...for] to know that a poem is about Arthur, Charlemagne or antiquity is to know only the subject which is no more useful for critical purposes than to know that Hamlet is about a Danish prince of that name.19 But such impatience with the blunt classification by matière only serves to link the fortunes of the Matter of England with those of Bodel’s original three; they sink or swim together. However, the Matter of England continued to provide a rationale for editors of anthologies, the most blatant of which is probably that offered by Donald Sands: ‘An additional category which Bodel does not mention is of great importance to English literature - the “Matter of England’”.20 His representative texts —King Horn, Havelok, Atheisten, Gamelyn — are described as ‘the four genuinely English romances’, by contrast with other associated texts, William o f Pakme, Bevis, Guy. This seems a 14 Derek Pearsall, ‘Matter o f England’, in The New Cambridge Bibliography ofEnglish Literature, Vol. 1:600-1660 , ed. George Watson (Cambridge, 1974), pp. 429-36. 15 W.R.J. Barron, English Medieval Romance (Harlow, 1987). 16 Diane Speed, T he Construction o f the Nation in Medieval English Romance’, in Readings in Medieval English Romance, ed. Carol M. Meale (Cambridge, 1994), pp. 13 5 -15 7 , at pp. 145—46. 17 Derek Pearsall, ‘The Development o f Middle English Romance’, Mediceval Studies 27 (1965), 9 1 -1 1 6 ; repr. in Studies in Medieval English Romances, ed. Derek Brewer (Cambridge, 1988), pp. 11-3 6 , at p. 16. 18 Dieter Mehl takes a similar line in The Middle English Romances of the Thirteenth and Fourteenth Centuries (London, 1968), pp. 31-3 3 . 19 John Finlayson, ‘Definitions o f Middle English Romances’, Chaucer Review 15 (1980-81), 4 4 6 2 ,1 6 8 -1 8 1 , at p. 45. There is a side glance here at the motif-based school o f criticism which would see Havelok as an analogue o f Hamlet. 20 Donald B. Sands, ed., Middle Engish Verse Romances, EMETS (Exeter, 1986), p. 4; my emphasis.


The Curious History o f the Matter o f England complex and misleading occlusion of the practical motivation of the editor of a student anthology - to choose short texts *that fit comfortably into teaching programmes. More recendy W.R.J. Barron adopted classification by Matter as a fundamental structure for his survey of English romance, despite the misgivings of Pearsall and others. His Matter of England section makes it quite explicit that the basis for such a classification differs in important ways from the Bodel-derived Matters. The Matter of England is seen as ‘a body of national literature’ (p. 63), some of which leaves traces in oral tradition, represented by the romances of ‘those heroes who passed from history to folklore entered English literature only through the intermediary of French romance’. The suppositions underlying this are the familiar and long established ones of Germanic, saga and folk origins and the intervening activity of ‘French’ romance, a term which does not distinguish between insular and continental, although his discussion does consider the Anglo-Norman originals of the Middle English texts. The ‘body’ of literature that results is cohesive enough to be generalised about in a final section on ‘the character of the Matter of England romances’ (p. 85) in which the Middle English romances are distinguished from their continental counterparts by their wide appeal to a socially diverse audience. The TEAMS edition of Four Romances o f England has a title that suggests a different terminology, but the introduction sounds a familiar note (my emphasis): Both Horn and Havelok the Dane belong to a group of poems known as the Matter of England, late medieval romances based in part on the oral folk culture that survived the Norman Conquest. This category also usually includes Athelston and Bern o f Hampton?' As with Barron, the essential difference between this concept of Matter of England and Bodel’s Matters is that between a hypothetical oral folk culture giving rise to the Matter of England and a strong literary tradition, supporting named authors and creative inter-textuality represented by Bodel’s original three Matters. A timely note of caution is sounded by John Frankis who remarks that the term ‘Matter of England’ does not distinguish between origins and setting, the prime example being King Horn, always included in the Matter of England category on the grounds of its apparent origin, but which is not set in England.212223This difficulty tends to be avoided in the construction of the Matter of England by shifting attention to the more historicised and geographically authentic Horn Childe despite general agreement that it is of inferior literary quality. Another example from an authoritative reference-volume entry, by Sharon Stevenson, includes King Horn, Havelok the Dane, Guy o f Warwick, Bevis and Richard Coeur du Lion?3 She attributes the Matter of England classification to the imitation of Bodel’s model by ‘modern bibliographers’ (unnamed). She provides a definition that 21 Introduction to King Horn, ed. Ronald B. Herzman, Graham Drake, and Eve Salisbury; originally published in Four Romances ofEngland (Kalamazoo, MI, 1999), p. 1. 22 John Frankis, ‘Views o f Anglo-Saxon England in Post-Conquest Vernacular Writing’, in OraHty and Literacy in Early Middle English, ed. Herbert Pilch (Tübingen, 1996), pp. 228-247, at pp. 238-39. 23 ‘The Matter o f England (1240-1400)’, Dictionary of Literary Biography: vol 146: Old and Middle English, ed. Jeffrey Helterman and Jerome Mitchell (Detroit, 1994), pp. 371-82.


Rosalind Field goes beyond that of material, one in which all the texts ‘are metrical, use rhyme rather than alliteration, use chivalric conventions to varying degrees, deal with a historical or legendary English personage, and, with the exception of Richard, utilize an exile and return motif (p. 372). In the same volume, Jane Dick Zatta writes about the ‘so-called Matter of England romances’ (p. 474) to group together the AngloNorman Romance o f Horn, Boeve, Gui, Waldef, Fouke Fit^warin.2AThis is a more recent development, in which the taxonomy of Anglo-Norman romance uses a term usually seen as essentially English in both content and language.2425 As is often the case with these encyclopaedic volumes, there is no attempt to reconcile the contradictions between entries, but these can stand for the deep differences that run through the usages of the term: one language or two, the use or not of ‘so-called’ and quotationmarks, the inclusion of texts in Anglo-Norman that do not have extant Middle English equivalents (Waldef and Fouke Fityvariri), the exclusion of Anglo-Norman versions of the Havelok tale, the exclusion of the wider range of Middle English romance. At the same time, from the early twentieth century, important scholars such as Laura Hibbard Loomis were working with a classification of ‘Romances of English Heroes’ - a less portentous phrase retained by others including Renwick and Orton and a number of more recent studies including that of Susan Crane.26 Charles Dunn in the Manual also avoids the term Matter in favour of ‘Romances Derived from English Legends’;27 he is also avoiding the term used for the same group of romances in Wells’s 1916 Manual —‘English and Germanic Legends’.2829But despite some variants on Laura Hibbard Loomis’s alternative category, the ubiquitous usefulness of the term continues: Helen Cooper gives a clear account of some of the complexities involved in the ‘matter of England’ group of romances (her quotationmarks), focussing on those romances that have sixteenth-century versions: Fing Horn and Havelok, Guy and Bevis.29 Robert Rouse discusses Middle English Matter of England romance, taking it to include ‘those romances that are explicitly set prior to

24 Jane Dick Zatta, ‘Anglo-Norman Literature in the Development o f Middle English Literature’, pp. 472-79. 25 Compare A.C. Gibbs, Middle English Romances (London, 1966) in discussing the AngloNorman romances: ‘the most distinctively Anglo-Norman romances are those which use for their subject-matter tales from “The Matter o f England”, that is, English traditions and legends which the Normans had adopted’ (p. 19). 26 W.L. Renwick and Harold Orton, The Beginnings of English Literature to Skelton (London, 1939; 2nd edn, 1952; 3rd edn (revised), 1966). Crane does describe the ‘Matter o f England’ as ‘omitted from Jean Bodel’s list’ and continuing strongly into Middle English, but her own discussion consistently adopts the term ‘romances o f English heroes’: see Susan Crane, Insular Romance (Berkeley, CA, 1986), p. 13. 27 J. Burke Severs, ed., A Manual of the Writings in Middle English, 1050-1400 , Vol. 1: Romances (New Haven, CT, 1967), pp. 17-37. His list consists o f the versions o f the Horn story, Havelok the Dane, Bevis of Hampton, Guy of Warwick, Gamelyn, Athelston and William of Paleme (‘included in the present section only for editorial convenience’, p. 17). 28 John Edwin Wells, A Manual of the Writings in Middle English, 1050-1400 (New Haven, 1916), pp. 763-64. 29 Helen Cooper, The English Romance in Time: Transforming Motifs from Geoffrey of Monmouth to the Death ofShakespeare (Oxford, 2004), pp. 29 -3 1.


The Curious History o f the Matter o f England the Norman Conquest’, in effect, the Auchinleck Bevis, Guy and Horn Childe, and Havelok.30 We seem to have reached a situation in which whenever the term is used - as it is frequendy - everyone confidendy assumes that we know which works are referred to and what the term means in relation to literary and cultural history, while at the same time feeling free to re-define it to fit whatever argument is being developed and to apply it to a shifting range of texts. Frequendy a sense of unease is marked by the use of quotation-marks or ‘so-called’, but there is no clear discussion of what causes this unease.31 Such unease is, I suggest, well-founded in respect of both elements in the term, ‘Matter’ and ‘England’. The problem with the use of the term ‘Matter’ to classify these Middle English romances concerns both reputation and reception. The three Matters have a pan-European dimension beyond their national boundaries and are an official and recognisable part of European culture. The term ‘Matter of England’ lacks international status - individual heroes, namely Guy and Bevis, will have their tales told in a number of languages, but there is no instandy recognisable body of material evident in heavy and exuberant tapestries or in the traditions of the Nine Worthies across Europe.32 To envisage Havelok or Bevis as finding their way into Dante’s Divine Comedy as do Ganelon, Ulysses and, indirecdy, Lancelot - is to show how ludicrous this is. However, if Bodel’s Matters belong to official high-status literature, they operate under the restrictions that go with it, at times labouring under the weight of their tradition. The other problem with ‘Matter’ lies in its implication of a coherent corpus of related texts. Despite the inadequacy of the term to describe romances fully, it is the case that each of the matters of Rome, France and Britain has a recognisable setting and cast of characters, commonly available to all texts within the group. The romance writer needs only to sketch in a setting and drop a few well-known names and can rely on an audience to respond accordingly. This happens in texts as different as Sir Eaunfal, The Sege o f Melayne and Troilus and Criseyde. Each writer of these romances of English heroes has to start afresh, introducing place, people and issues; because the material is fresh, there is no familiar setting, no continuum of inter-textual reference. A comparison of the openings of some minor branches of the cycles of the Matters proper with those of the Middle English romances in question shows how bereft these romances are of stock opening ingredients. For Sir Eaunfal two lines suffice to place the action historically and thematically: Be doughty Artours dawes That helde Engelond yn good lawes.. .33 30 Robert Rouse, The Idea ofAnglo-Saxon England in Middle English Romance (Cambridge, 2005), pp. 52-69 (p. 52, n. 1). 31 As used intermittendy by the present author to signify undefined unease: in ‘Romance as history, history as romance’, in Romance in Medieval England, ed. Maldwyn Mills et al. (Cambridge, 1991), pp. 163-74. 32 Admittedly later English tradition would make Guy a Worthy (replacing Godfrey): see Helen Cooper, ‘Guy as Early Modem English Hero’, in Guy of Warwick: Ancestor and Icon, ed. Alison Wiggins and Rosalind Field (Cambridge, 2007), pp. 18 5 -9 9 , at p. 193. 33 Sir Eiunfal, ed. A.J. Bliss (London, 1960), lines 1-2 . Marie de France’s Eanval introduces the Arthurian world equally briefly, but adds wars against the Piets and Scots (lines 5-10).


Rosalind. Field The more courtly Y wain and Gawain, drawing on Chrétien, encompasses the chivalric, political and literary implications of its subject matter in a few introductory gestures: Ywayne and Gawayne; þai war knightes of the tabyl rownde, þarfore listens a lytel stownde. Arthure, þe Kyng of Yngland, þat wan al Wales with his hand And al Scotland, als sayes þe buke.. .M The Sege o f Melayne similarly sums up the essential themes of the Matter of France in its first stanza: All werthy men that luffes to here Off chevallry that byfore us were, That doughty weren of dede; Off Charlies of Fraunce, the heghe kinge of alle, That ofte sythes made hethyn men forto falle.. .3435 And Kyng Alisaunder provides an authenticated synopsis: 3ee shullen heren noble geste, Of Alisaundre þe rich[e k]yng, þat dude by his mais[t]res teching, And ouercom, so J fynde, Darrye of Perce and Pore of Ynde, And many oþere, wi3th and hende, Jn to þe est werldes end.. .36 There is no such ease of reference for our romances, even when, as in the case of Atheisten, England is established as the locus of action: Off foure weddyd breþeryn I wole 3ow tel, þat wolden yn Yngelond go dwel, That sybbe were nou3t of kyn.37 Most often, kingdoms are indeterminate and their rulers obscure, as in King Horn: A sang ihc schal 3ou singe Of Murry þe kinge. King he was biweste.. .38 There is a concern to establish family relationships, often the key to the action, at the expense of clear geographical or temporal contexts, as in Bevis where the place is more unspecific than often realised: 34 Y wain and Gawain, ed. A.B. Friedman and N.T. Harrington, EETS OS 254 (London, 1964), lines 4—9. 35 Maldwyn Mills, ed., Six Middle English Romances (London, 1973), lines 1-5 . 36 Kyng Alisaunder, vol. 1, ed. G.V. Smithers, EETS OS 227 (London, 1952): Oxford, Bodleian Library MS Laud Mise. 622, lines 30-36. 37 Atheisten, ed. A.McI. Trounce, EETS OS 224 (London, 1951), lines 10 -12 . 38 King Horn ed. George H. McKnight, EETS OS 14 (London, 1901), Cambridge, University Library MS Gg. 4.27 (part 2), lines 3-5.


The Curious History o f the Matter o f England Ich wile yow teilen al togadre Of that knight and of is fadre, Sire Gii. Of Hamtoun he was sire And of al that ilche schire.. .39 These openings resemble the scene-setting of independent romances rather than the inter-textual referencing of cyclic ones - as in Amis andAmiloum What sumtyme fel beyond the see Of two Barons of grete bounté And men of grete honoure.. -40 There is no need to labour the point: these romances are not drawing on a body of material familiar to author and audience. Place in these romances is often unspecific, at best local and only rarely national. Characters need to be introduced afresh, they come with no established reputation - there is no English equivalent of the figure of ‘Gawain’ whose identity carries over between several texts. The only figures who occur in more than one text are the historical kings, Athelston41 and Richard I, each of whom has his own romance as well as a walk-on role in Guy and Fouke F i t a r m (and eventually the Robin Hood tradition) respectively. The city of Winchester has some status as a significant place, but this is limited by comparison with Troy or Camelot. It is not until the Robin Hood ballads that a corpus of texts set in England develop an introductory motif that links one text to another. The few examples given above are enough to indicate that there is a further problem with ‘England’. For recent romances scholarship the grouping of the Matter of England has provided a useful shortcut into considerations of nationalism and awareness of national identity. So there has been an expansion of meaning as the label ‘Matter of England’ shifts its emphasis from the grouping of narratives which can be seen to have similar material or origins, to a group of narratives concerned with the concept o f ‘England’. Implicit in this is a perception of national qualities and virtues - and more recent criticism has kept implicit assumptions and bias that nineteenth-century critics were comfortable about voicing. ‘England’ in the ‘Matter of England’ is alleged to encode structures of political and cultural complexity that the texts themselves barely support. Even where ‘England’ may supply authentic and familiar topography, there is little agreement as to historical moment. It is usually, but not invariably, vaguely Anglo-Saxon, or at any lightly detailed time up to and including the audience’s present. This is not as reliable or recognisable as the Arthurian moment of the Matter of Britain, or the France of Charlemagne and his sons. 39 Bern ofHamptoun, ed. Herman et al, in Four Romances ofEngland, lines 7 -1 1 . In her paper ‘Light Entertainment and Upward Mobility: the Anglo-Norman Boeve and Its Patrons’ (given in Southampton, May 2004), Ivana Djordjevid questioned the identification o f the original ‘Hamtoun’ with Southampton. 40 Amis andAmiloun, ed. MacEdward Leach, EETS OS 203 (London, 1937), lines 4-6. 41 In fact, though, the Athelston in the romance o f that name, a foreigner and something o f an anti-hero, seems only nominally related to the Anglo-Saxon king. Frankis tentatively suggests the author may have wished to disassociate his Athelston from the historic Athelston, in ‘Views o f Anglo-Saxon England’, p. 242.


Rosalind Field But the question remains —if the ‘Matter of England’ never existed, have events proved that it was necessary to invent it? There are alternatives; if we adopt Laura Hibbard’s formulation - Romances of English Heroes - we gain the recognition of pluralities at work in these texts as distinct from the authoritatively monolithic ‘Matter’. If we escape the heavy shadow of nationalism only to be caught by gender specificity that may not be too high a price. We are released from the requirement to find valid generalisations to group these texts together. There is no exile-and-retum in Guy for example, but rather a pilgrimage. The majority of English heroes may be positive figures, but Athelston is not. Law may be a major theme, but it is skewed somewhat in Gamelyn and other outlaw texts. England may be saved by Guy, but it destroys Waldef and ejects Bevis. The constructed past may often be ‘Anglo-Saxon’, but in Havelok it is Anglo-Danish, and in Richard, Gamelyn and Fouke Fit^warin it is the recent past, or even the present. This is not a literary grouping, there is no evidence, textual or extra-textual of a contemporary consciousness that any of these Middle English romances belong to an established body of material.42 If we were to generalise from this group of texts, it might be more along the lines of modality - if the canonical Matters are expected to function as true, wise or entertaining, then perhaps these texts display an exploratory freedom unavailable to them. There is little sense (outside the Havelok tradition) of a known destiny in the form of a historical outcome which haunts and encumbers those who write about Troy or Arthur, as Chaucer and Malory bear witness. The reader or audience has little warning of some unexpected turn in the narrative - such as Guy’s conversion, Ascopart’s treachery or Waldef s death. There is, perhaps, opportunity for genuine inventiveness on the part of the authors in their selection and combination of traditional motifs. The concept ‘England’ is not present in all of them, and is not monolithic when it is - not surprising when we take into account that the AngloNorman originals pre-date, or even contribute to, a growing awareness of national identity. This is to say that the space offered by the English past —that is the insular past differentiated from the Arthurian tradition — is one of constructive fantasy, of exploring a world similar to but not under the same constraints as, the known present. It has not (yet) been colonised by literary auctores, nor adopted by international political factions. There is further recent evidence of the lack of fixity in the term ‘Matter of England’ to be found in an important inaugural lecture by the historian Rees Davies at Oxford in 1996.43 His title The Matter of Britain and the Matter of England’ is in literary-historical terms, somewhat misleading. For Davies the ‘Matter of Britain’ is the historiographical movement initiated by Geoffrey of Monmouth and the ‘Matter of England’ is the counter-movement, led by William of Malmesbury, Henry of Huntingdon and Geffrei Gaimar. It is concerned to displace Britain by England in the perception of the past - a movement which proved successful and dominant until the recent challenge from historians of the other regions of Britain, including Davies himself. The value of Davies’s version of the Matter of England is that it 42 For a contrary view to this see Diane Speed, ‘Construction’, pp. 144—5. 43 The Matter o f Britain and the Matter o f England’: an inaugural lecture delivered before the University o f Oxford on 29 February 1996 by Rees Davies (Oxford, 1996).


The Curious History o f the Matter o f England gives an intellectual and cultural context to the earliest of the romances of English heroes, those of the twelfth-century Anglo-Norman writers from Gaimar onwards. But in seeing the Matters as conflicting histories, and the Matter of England as re­ shaping the identity of the British Isles with the emphasis on England, he both gives the term ‘Matter of England’ a precise and recognisable function and renders it less useful as a term to describe fiction. We now have competing uses of the term, one historical, one literary-historical. We can therefore sum up the problems with The Matter of England’ thus: it is a hybrid term, both elements of which are problematic and ill-defined. It is a classification of texts, about which there is no clear agreement, which has no medieval justification although it sounds plausibly medieval and which has been appropriated very effectively to describe the historiographical debate of the twelfth century. It is a term used to yoke together romances that may have more significance in terms of differences rather than similarities, and it carries an ideological weight in its assumptions about national literature and language which render the texts themselves rigid, official and univocal. It is a cause of confusion rather than clarification. And as this paper has argued, it belongs to the discourse of history — not the history of literature, but rather the history of scholarship and the history of the perception of Englishness in literature, whether Rees Davies’s twelfth-century Matter of England, or the literary ancestry constructed by the early twentieth-century English-speaking world.

Postscript: Havelok, the returning king It seems clear that the central figure in any conceptualising of English romances is Havelok - the most apparendy authentic figure, the one instantly recognisable across several versions, most rooted in local tradition44 and the figure that emphasises the key concerns of good rule, law and social order. Most readers would still agree with Creek that: in Havelok we seem to have a truly popular hero, not entirely created in the image of crude or chivalrous knighthood. But he is the exception that proves the rule.45 Where other romances may cause difficulties in classification, Havelok the Dane alone represents everything that has been claimed for the Matter of England. It contains exile-and-retum, a popular hero (in both senses of the word), evidence of oral and local cultural connexions and probably origins, a positive, even good-humoured view of the past as a locus of creative fantasy for the present, a view of a cohesive society, of physical strength used in support of legal rule, of personal relationships contributing to social stability. It could be argued that the ‘Havelok-type’ even if it is only a group of one, has had a disproportionate effect on the modern awareness of English medieval narrative.46 Furthermore, in various rewritings, Havelok continues 44 See Nancy Mason Bradbury, ‘The Traditional Origins o f Havelok the Dane’, Studies in Philology 90 (1993), 115 -14 2 . 45 Creek, ‘Character’, p. 609. 46 See David Matthews on the importance o f Havelok in early scholarship on Middle English (Making ofMiddle English, pp. 118—27).


Rosalind Field as a presence beyond the medieval, into Elizabethan versions and nineteenth-century children’s literature.47 So it is pleasing and important that HaVelok as a traditional figure is still alive and well in the work of a major popular writer. The Discworld novels of Terry Pratchett have attracted litde academic attention, but provide one of the most potent routes by which the informed and innovatory handling and reception of traditional tale is kept alive in contemporary culture. With huge international sales and a continual presence at the top of all best-seller lists, Pratchett reaches an audience well beyond the range of the most ambitious re-packaging of medieval romance. His work is parodie of traditional tale types, but arguably moves beyonid parody to expose and question the assumptions underlying the modem appetite for medieval fantasy. Of the - to date - twenty-six novels, one strand deals with the city of AnkhMorpork, combining the qualities of New York, Rome, Calcutta and all other archetypal cities, but increasingly across the 1990s used to explore the problems of a multi-cultural and ancient city, not unlike London.48 Ancient enough to have once had a king, Ankh-Morpork is now run as a reasonably benign dictatorship by the machiavellian Patrician and its law and order upheld by a motley company of guards, led by Vimes the fiercely republican descendant of the Cromwell figure who assassinated the last of the kings. Into this strolls the figure of Carrot, a six-and-a-half-foot tall dwarf (a dwarf by culture, that is, if not physique), who comes to the city to join the guards. Immediately, of course, nature and nurture clash - as an orphan raised by dwarf foster parents, he both is and is not, a member of that species (in the Discworld species stands for both race and class). As the lost descendant of the ancient line of kings he is shadowed by the association with Arthur, and Tolkien’s equivalent, Aragorn from The Lord o f the Rings. But Pratchett’s fiction has litde patience with Arthurian romance or with Tolkien and it is the Havelok legend that supplies the alternative and the challenge to such figures.49 In the unlikely setting of the fantasy city of Ankh-Morpork the clash between the romantic Matter of Britain and the more pragmatic historical romance of English heroes is enacted with a new cast of characters. For readers of Havelok the Dane, Carrot has many familiar characteristics. He is physically larger and stronger than anyone else, on one occasion fighting with a roof beam: It was quite a large and heavy roof timber and it scythed quite slowly through the air, but when it hit people they rolled backwards and stayed hit... Carrot turned slowly, the roof beam held like a staff. His gaze was like a lighthouse 47 See Helen Cooper, T he Elizabethan Havelok: William Warner’s First o f the English’, in Medieval Insular Romance, ed. Judith Weiss, Jennifer Fellows and Morgan Dickson (Cambridge, 2000), pp. 16 9 -18 4 , and Cooper, English Romance in Time, pp. 343—44. Velma Bourgois Richmond gives an account o f Edwardian rewritings o f heroic romances, including Havelok, in ‘Historical Novels to Teach Anglo-Saxonism’, in A nglo-Saxonism and the Construction of Social Identity, ed. Allen J. Frantzen and John D. Niles (Gainesville, 1997), pp. 17 3 -2 0 1. 48 The present discussion is confined to the Guards Trilogy: Guards, Guards! (London, 1989) [henceforth GG\; Men at Arms (London, 1993) [MA\\ Feet of Clay (London, 1996) [FCj. 49 The most explicit reference to Havelok is the revelation in MA that this is the Patrician’s first name.


The Curious History o f the Matter o f 'England beam. Where it fell, the crowd lowered their weapons and looked merely sullen and uncomfortable. (GG, p. 286) He has a kingmark, of which he takes litde notice, is sexually naïve, but forms a politically significant partnership with the feisty Angua. He walks through the city as if he owns it and has a hypernatural ability to get on with all the varied sections of society. He also has a devotion to the law which overrides all other considerations: Vîmes turned his head slightly. ‘He killed Angua. Doesn’t that mean anything to you?’ Carrot nodded. *Yes. But personal isn’t the same as important.’ [...] Vimes looked down at Angua’s corpse and felt a train of thought derail itself... Personal isn't the same as important. What sort of person could think like that? And it dawned on him that while Ankh in the past had had its share of evil rulers, and simply bad rulers, it had never yet come under the heel of a good ruler. That might be the most terrifying prospect of all. [MA, pp. 35865]50 He is an opaque character, always viewed from the outside and an object of some mystification - only seen by his effect on others, apparently lacking a strong sense of self-identity. Havelok seems throughout most of his career to think like a peasant, Carrot like a dwarf. As with the medieval Havelok it is unclear how or when he realises the truth of his identity.51 The Havelok figure draws attention to the gap - at one and the same time both comic and politically forensic - between the king, marked out by destiny and equipped with kingmark and ancient sword, and the man, simple, unambitious, yet charismatic. As a creative re-reading of Havelok, Pratchett’s development of Carrot engages with that text’s concern with the nature of charismatic leadership, a term Pratchett analyses, but which medieval narrative did not have a name for: ‘Everyone likes Captain Carrot, and, well... rumour’s got about that he’s the hair to the throne, sir.’ There’s no proof of that, Sergeant.’ ‘Dunno what is proof, said Colon... ‘But he’s got that sword of his, and the birthmark shaped like a crown, and... well, everyone knows he’s king. It’s his krisma.’

50 Compare Diane Speed’s reading o f the ideology o f Havelok: ‘If a protagonist places selfinterest before the common good, the nation suffers’ (‘Construction’, p. 147). 51 The Middle English Havelok is informed by Grim ’s daughter. However, although in this version he was eight years old when exiled from Denmark, there is no explanation as to his failure to remember who he is (a consequence o f changing the infant’s age from the two years o f the Anglo-Norman tradition). See the detailed analysis o f the inconsistencies in the revelation o f Havelok’s identity in the three main versions o f Havelok the Dane, by G.V. Smithers, in the introduction to his edition o f the poem (Oxford, 1987), pp. xxxvii-xlvii.


Rosalind Field Charisma, thought Vîmes, Oh yes. Carrot has charisma, He makes something happen in people’s heads. He can talk a charging leopard into giving up and handing over its teeth and doing good work in the community... Vimes distrusted charisma. ‘No more kings, Fred.’ Royalty was like dandelions. No matter how many heads you chopped off, the roots were still there underground, waiting to spring up again. [FC, pp. 95—96] Carrot is also used to challenge the power of narrative tradition in his refusal to accept his destiny, or even to explicidy notice it. But as a lost heir he attracts the attention of sinister aristocratic plotters for whom a king represents an opportunity to seize power, only to be thwarted by the nature of the genuine article: ‘And then young Carrot turns up with charisma writ all over him, and he’s got a sword and a birthmark and everyone gets a funny feeling and dozens of buggers start going through the records and say ‘Hey, looks like the king’s come back.’ And then they watch him for a while and say, “Shit, he really is decent and honest and fair and just, just like in all the stories. Whoops! If this lad gets on the throne we could be in serious trouble! He might turn out to be one of them inconvenient kings from long ago who wanders around talking to the common people...’ [FC, pp. 386] The tension between the good nature of the man and the suspicion with which Vimes - and it seems his creator - holds the office is still running through the series and seems irreconcilable. This, I would suggest, is not simply a feature of its contemporaneity, for the ambivalent attitude to kings is familiar from the romances. The Havelok figure thus provides a version of the lost king which is free of Messianic, Arthurian associations52 (and an alternative to Tolkien, as is much of Pratchett). Free of such baggage, he is more adaptable to an enquiry into types of rule, the nature of law, and an examination of how power operates in the changing, contingent and, above all, urban world. Finally, as far as these romances are concerned, the modem mutation of Havelok is a useful demonstration of how they have always operated, not as a labelled, official, predictable Matter, but as a series of one-offs, unpredictable, creative and exploratory forays into the narratives of romance, legend and fantasy.

52 There is a typically wry reversal o f the sword-in-the-stone m otif in MA.


How English Are the English Charlemagne Romances?

M aria n n e A iles

an d

P h illipa H ard m an

In 1879 the Early English Text Society published its edition of Sir Ferumbras, the first volume in a series collectively entitled The English Charlemagne Romances'. Ten more texts appeared in the series, including both verse romances and prose translations.1 This collective treatment was not given to any other body of works related purely by subject matter - there was no series entided The English Romances of Antiquity', or ‘The English King Arthur Romances’, for example, though plenty of texts were available in the Society's publications for either grouping. Intrigued by this unique treatment, we propose to investigate the extent to which the Charlemagne romances may or may not be thought to have a specifically English character. Very few of the hundred-odd Old French chansons de geste were translated into Middle English. The few Middle English translations made were usually preceded by Anglo-Norman texts, and this pre-selection gives useful hints on what interested audiences in England.12 Boeve de Hamtoune is obviously insular in content and was indeed originally written in Anglo-Norman rather than continental French, while the didactic chansons Ami et Amile and Florence de Rome could be thought to have universal exemplary appeal, but the third type, chansons of Charlemagne, the French national 1


The series appeared between 1879 and 1887 in the Early English Text Society Extra Series, as follows: ES 34, The English Charlemagne Romances I: Sir Ferumbras, ed. Sidney J.H. Heritage (1879); ES 35, II: The Sege offMelayne, The Romance of Duke Rowland and Sir Otuell of Spayne, with a fragment o f The Song of Roland, ed. Herrtage (1880); ES 36 & 37, III & IV: The Lyf ofthe Noble and Ctysten Prynce Charles the Grete translated by William Caxton, ed. Herrtage (1880-81); ES 38, V: The Romance of the Sowdone of Babylone, ed. Emil Hausknecht (1881); ES 39, VI: The Taill of Rauf Coilyear with the fragments o f Roland and Vemagu and Otuel, ed. Herrtage (1882); ES 40, 41, 43 & 50, VII, VIII, IX & XII: The Boke of Duke Huon ofBordeux, ed. S.L. Lee (1882-87); ES 44 & 45, X & XI: The Foure Sons ofAymon, ed. O. Richardson (1884—85). Two further volumes o f English Charlemagne texts were published separately: EETS OS 198, Firumbras and Otuel and Roland, ed. Mary I. O ’Sullivan (1935); EETS OS 322, Turpine's Story: A Middle English Translation ofthe Pseudo-Turpin Chronicle, ed. Stephen H.A. Shepherd (2004). See Ruth J. Dean and Maureen B.M. Boulton, Anglo-Norman Literature: A Guide to Texts and Manuscripts, Anglo-Norman Text Society, Occasional Publications Series 3 (London, 1999).

Marianne Aiks and Phillipa Hardman hero, has been considered a strange choice for Middle English adaptations since they glorify French victories and were mosdy translated and copied during the Hundred Years War.3 However, any claim made by the French Capetian dynasty to be the legitimate successors of Charlemagne was of course available to the English royal house too, given Edward I ll’s claim to the French crown through his maternal grandfather, Philip IV.4 Therefore, translating the texts into English could be seen as a form of cultural appropriation, echoing the English kings’ claim to France. Furthermore, Charlemagne provides an imperial model for fifteenth-century English kings who were keen to present themselves as .Christian Emperors.5 In this respect it is perhaps a significant observation that the Old French chansons of revolt against Charlemagne are not among the texts translated into Middle English.6789By contrast, both Fierabras1 and Otinel? each represented by three separate Middle English romances, are notable narratives of incorporation, focusing on the conversion and assimilation of the eponymous protagonists.

Formal Characteristics The terminology adopted in discussing the translation or adaptation of the Old French and Anglo-Norman texts into Middle English usually assumes that the change of language is accompanied by a change of genre, from chanson de geste to romance - indeed, W.R.J. Barron argues that the change had already all but happened in the case of ‘romanticized epics’ such as Fierabras and Otinel? It is interesting, therefore, to investigate how the exemplary formal characteristics of the French chansons de geste are treated in the Middle English adaptations. The English poems certainly show an awareness of the importance of prologues in the epic tradition prologues are present in all three Otuel poems, the Sege o f Melayne, and vestigially in Poland and Vemagu. Although two of the three Ferumbras poems begin imperfectly, the multiple prologue in the Sowdone o f Babylone points to the importance of this feature. These elaborate prologues, like the French originals, set up the oral-delivery mode, appeal to heroic expectations (with much mention of victorious battles), and establish the historical context for the narrative. But all (by contrast with the French) stress the conflict between heathen and Christian as their central concern. In the 3

4 5 6

7 8 9

See, for example, 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, 1999), pp. 87-100. Warm explains the apparent paradox by stressing the Christian status o f the knights: ‘Christian heroes who happen to be French, rather than French heroes who happen to be Christian’ (p. 87). The crown used in the coronation ceremonies o f medieval French kings was known as the Crown o f Charlemagne. See further, Phillipa Hardman, ‘The Sege of Melayne-. A Fifteenth-Century Reading’, in Tradition and Transformation, ed. Field, pp. 7 1-8 6 , at pp. 75-76. The exceptions are Caxton’s translation o f the French prose Renaud de Montauban, The Right Plesaunt and Goodly Histone of the Foure Sonnes ofAymon (1488/9), and Lord Bemers’s translation from a French prose version o f Huon de Bordeaux, The Boke of Duke Huon ofBurdeux (1534). Fierabras, chanson degeste du Xlîe siècle, ed. Marc Le Person, CFMA 142 (Paris, 2003). Otinel, ed. F. Guessard and H. Michelant (Paris, 1859). W.R.J. Barron, English Medieval Romance (London, 1987), pp. 98—99. For discussion o f the terminology in relation to English Charlemagne romances, see John Finlayson, ‘Definitions o f English Romance’, Chaucer Review 15 (1980), 4 4-62, 16 8-81.


How English are the FLnglish Charlemagne Romances? French Otinel, the prologue ends with the promise of a hitherto untold adventure of Charlemagne, ‘whom God loved so much that He did miracles for him in his lifetime’, while the English poems promise ‘batailles [...] bitwene/ Cristine men & sarrazins kene’ {Otuel, lines 4—6), Roland’s defeat of infidel opponents {Otuel