The Châteauroux Version of the «Chanson de Roland»: A Fully Annotated Critical Text 9783110339741, 9783110339611

Citation preview

The Châteauroux Version of the «Chanson de Roland»

Beihefte zur Zeitschrift für romanische Philologie

Herausgegeben von Claudia Polzin-Haumann und Wolfgang Schweickard

Band 384

The Châteauroux Version of the «Chanson de Roland» A Fully Annotated Critical Text

Edited by Marjorie Moffat

DE GRUYTER

ISBN 978-3-11-033961-1 e-ISBN 978-3-11-033974-1 ISSN 0084-5396 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.dnb.de abrufbar. © 2014 Walter de Gruyter GmbH, Berlin/Boston Satz: Meta Systems Publishing & Printservices GmbH, Wustermark Druck: Hubert & Co. GmbH & Co. KG, Göttingen ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

Preface

The Châteauroux Manuscript «Ce manuscrit n’a point de valeur appréciable. Aux yeux de certaines personnes il peut valoir cent mille francs; aux yeux d’autres personnes vingt sous.» Jean-Louis Bourdillon

It is thanks to Antoine-Jean-Louis Bourdillon, a Swiss merchant and banker but also an ardent bibliophile, that the town of Châteauroux, administrative centre of the Indre département, has been since 1856 the possessor of the manuscript copy of the Chanson de Roland cited above, plus other treasures and the funds necessary for their proper preservation. Bourdillon (1782–1856) came of an Huguenot family who had fled to Geneva in the sixteenth century; a widower and childless, it was perhaps this which led him to bequeath his valuable collections, his property and his fortune to «Châteauroux, berceau de ma famille».1 The Châteauroux manuscript contains a single chanson de geste: one of only four complete texts of the Chanson de Roland in Old French still in existence and the only complete version to be found today in France. Constantly overshadowed by the greater literary merit of the Oxford MS (Digby 23) and lodged in a provincial library, it has not been highly rated. It has been dated by its gothic script, simple style of illumination and linguistic features to the late thirteenth century; as such, it reflects contemporary literary and popular taste, as well as historical events. Certain special features of this text also offer a unique insight into the whole process of remaniement. The place of this MS in the date-sequence of the major versions of the Chanson de Roland is today controversial, with scholars such as Segre, Beretta and Duggan arguing in favour of the anteriority of the Venice 7 manuscript, but the more traditional dating is given in the list below.2 All texts are written  Full biographical details may be found in Châteauroux et ses Livres: Histoire de la Bibliothèque, published by «Les Amis du Vieux Châteauroux» in 2000. Briefer details are given by Paul Thibaud (or Thibault) in his introductory article to the Mortier edition of the Châteauroux ms. Bourdillon was in fact mistaken: his family originated in the Nivernais, but it was from Bourges that Jehan II Bourdillon, persecuted as a Huguenot, fled in 1562 to Geneva. The family history is set out in fascinating detail (including the prestigious English branch) by Jacques Bourdillon, in the Bulletin de la Société nivernaise des Lettres, Sciences et Arts 51 (2002), as the text of a lecture given on 21 March 2002.  The dating of the ms. will be discussed later in the Introduction; as may be seen from the list given here, C, V7 and P are of approximately the same date; V4 is slightly later.

vi

Complete Texts and Incomplete Versions

in laisses monorimes but only O is assonanced throughout. Complete texts and acephalous versions are listed separately below.

Complete Texts (i)

The twelfth-century manuscript, Digby 23, in the Bodleian Library, Oxford (sigma O): the earliest known, assonanced version of the poem, comprising 4002 lines of verse; (ii) The Châteauroux manuscript (C), 8201 lines, late thirteenth century; mainly in rhyme; (iii) The late thirteenth-/very early fourteenth-century3 manuscript in the Biblioteca Nazionale Marciana, Venice, Gall.VII,4 known as V7: 8880 lines in the Duggan edition. V7 has smoothly rhymed laisses monorimes with very rare poetic licence; for the most part it closely resembles C; (iv) The early fourteenth-century5 manuscript Gall. IV (now 225) in the Biblioteca Nazionale Marciana, Venice, known today as V4: a strongly Italianised version with the first 3846 lines corresponding to O (but with assonance at times converted to somewhat imperfect rhyme), the remaining 2163 lines in fully rhymed laisses monorimes, but with a digression on the «Prise de Narbonne», inserted V4 3848–4417, which is not found in any other version of the Roland;

Incomplete Versions (v)

The late thirteenth-century manuscript 860 in the Bibliothèque Nationale in Paris (P): a mainly rhymed version with occasional assonance, at times very similar to C and V7, but lacking the first quire of text (line 1 matches C 1553, the first line of folio 24 verso in C); (vi) The fourteenth-century manuscript 984 in the Bibliothèque de la Ville de Lyon (L): a rhymed version sometimes very close to P, but often in abridged form, with an even larger section at the beginning of the poem missing (its first lines correspond approximately to lines 2150 seq. in C);  This date, generally accepted by scholars, is endorsed by D’Arcais in her analysis of V7’s simple illumination. Cf. D’Arcais 1984 for a detailed discussion of the style of decoration and its implications in regard to date.  Biblioteca Nazionale Marciana Cod. Fr. 2.7 (=251).  Cf. Introduction to Carlo Beretta’s 1995 edition of V4 and the article on its illumination by D’Arcais 1984, dating V4 to between 1320 and 1345. The Biblioteca Marciana reference is Cod. Fr. 2.4 (= 225).

Incomplete Versions

vii

(vii) The fifteenth-century manuscript R, 3, 32 at Trinity College, Cambridge (T): a rhymed version which also lacks the first part of the poem, starting at C 1093 (the third line of folio 17 verso in C). In addition, there are a number of fragmentary texts: the Lorraine fragments (F) of 347 lines and other 180-line fragments published by C Lavergne in Romania XXXV 445, plus the London «British Museum» fragment (B) corresponding to O 2776–2883 and dating from the end of the thirteenth century, but their brevity and lacunæ make them rarely useful in the editing of C. Similarly, several foreign versions – most notably the Middle High German Ruolandsliet (K) and the eighth branch of the Old Norse Karlamagnussaga (n), which are based on assonanced French sources – are not really relevant to the clarification of difficult readings in C, although they testify to the pan-European popularity of the Roland story and provide valuable insights into the evolution of the text and the dating of remaniements. The Pseudo-Turpin and the Provençal Roncsasvals have proved largely irrelevant. Of the above versions of La Chanson de Roland, the Oxford text stands preeminent. It has been recognized as the first great work of French literature,6 outstanding for its nobility of theme, psychological accuracy, effective use of a limited vocabulary, restraint and balance, a poem in which the desmesure of Roland is directly comparable to the hubris of the Homeric heroes. During the last sixty years, new editions of the V4 text, valuable because it contains the only other assonanced version of the poem, have been produced by Gasca Queirazza (1954), Robertson-Mellor (Salford, 1980) and Beretta (Pavia, 1995); and Cesare Segre has stressed the importance of V4 and the rhymed versions in his 1971 edition of the Roland, modified 1989, and in Segre 1974. A great many aspects of the other versions have been studied over recent years,7 but there has been no critical, fully annotated edition of the Châteauroux manuscript.8 This is still the situation today, despite the publication of all the Roland Corpus texts at the end of 2005 (see below). In the present edition, the focus on C rather than the CV7 model has given rise to an interesting change of perspective on its distinctive characteristics and thus on its position within any schema of the relationship between the different versions of the Roland. The theories and hypotheses propounded here  Cf. Bédier 1927, in respect of the précellence of O.  E.g. Rosellini 1960a; 1962, which compares V4 with O and the other versions; Rosellini 1960b; Brook 1990 and a wide range of books and monographs by Joseph J. Duggan, Cesare Segre, Karl Heinz Bender, Annalee C. Rejhon, etc.  Horrent comments: «Personne n’a étudié les versions rimées en elles-mêmes et pour ellesmêmes» (1951, 208 n. 7).

viii

Earlier Editions of the Manuscript

have arisen directly from a close study of the text; indeed, the consultation of other recent editions of the Roland and related (mainly post-1950) research of all kinds has been driven and guided by questions stemming from this study. Whilst there have been, therefore, no a priori assumptions deriving from the more recent studies by eminent researchers in this field, their findings and conclusions have been carefully examined and weighed against what may well be a rather naïve argument, but one solidly based on textual evidence.

Earlier Editions of the Manuscript The earliest «modern» transcription of the Châteauroux ms. (referred to henceforward simply as «C») was made in 1818 by Guyot des Herbiers, whose text (with corrections added in red ink circa 1870) is lodged in the Bibliothèque Nationale (mss. fr. 15–108, suppl. fr. 254–21). This text, together with the mss. from Venice (V4 and V7), Paris (P), Cambridge (T) and Lyon (L), appears to have formed the basis of a thesis presented by Monin at the Sorbonne at some time prior to 1840.9 In 1841, Jean-Louis Bourdillon published his own complete but idiosyncratic edition, Roncisvals mis en lumière, with supplements published in 1847, 1850 and 1851. Large extracts from C were quoted by Francisque Michel in his 1869 work on the Chanson de Roland. In 1883, Wendelin Foerster published a transcription of C with V7’s variations from C’s text printed below each line or, where there was greater divergence between the two texts, after the relevant laisse in C. Finally, Raoul Mortier published his lightly annotated transcription in 1943 as part of his complete edition of all the manuscripts of the Chanson de Roland; reference will be made later to the preamble to that edition, in which Mortier makes perceptive comments on the text and on its relationship with V7 and P. A comprehensive introduction to the MS was supplied by Paul Thibaud (or Thibault), secretary to the Châteauroux Library Committee, giving its history (insofar as this was known at the time) and a detailed palæographic study (composition and quality of parchment, style of script, punctuation, scribal errors). This was based in part on a lecture on Les Manuscrits de la Bibliothèque de Châteauroux given on the 16th May 1881 by M. Joseph Patureau, Membre de la Société de l’Histoire de France, the text of which is preserved in the archives of the library. Towards the end of 2005, a new critical edition of all the Old French manuscripts of the Roland was published under the title La Chanson de Roland: The French Corpus by a consortium of scholars under the general editorship of

 Cited in the Thibaud introduction to Mortier’s edition of 1940–1944, referred to below.

The Current Edition

ix

Joseph J. Duggan of the University of California. Each scholar was responsible for a particular manuscript: Ian Short for O, Robert F. Cook for V4, Annalee C. Rejhon for P, the late Wolfgang van Emden for T, William W. Kibler for L and the fragments. Professor Duggan was himself responsible for the edition of the Châteauroux-Venice 7 version, producing a fully annotated Critical Text based on V7 with the aim of providing a text approximating to the common ancestor of Châteauroux-Venice 7; he therefore included an unemended, non-annotated transcription of the Châteauroux manuscript on pages 529–807, as Appendix B. No fully annotated, critical edition of the Châteauroux manuscript has been published at the moment of writing. In the present edition, due note has been taken of the Duggan edition, particularly where opposing views are taken regarding the presence of clusters of assonanced laisses in C and the reasons for the associated repetitions. I am indebted to Professor Duggan for drawing my attention to Beretta’s Studio sui rapporti fra i manoscritti rimati della «Chanson de Roland» (Beretta 2001); but communication between us has been limited and confined to the period Autumn 2005–Spring 2006.

The Current Edition The current edition has been based on the original manuscript, originally using the microfilm in the Archives Nationales in Paris but later verified by close examination of the manuscript in Châteauroux, and checked against both the Foerster and the Mortier editions. Both Foerster’s transcription of V7 and Mortier’s photographic reproductions of V7’s complete text have proved invaluable sources of alternative readings, when emending scribal errors in C. For comparisons with the Digby 23 MS (sigma O), Whitehead’s edition has been used; indeed, the initial work on C for this edition was done under Dr Whitehead’s direction in 1951–1952 and influenced by his advice and comments in 1953. In respect of V4, Mortier’s edition was originally used, but this has been tempered by reference to Carlo Beretta’s recent edition. Mortier’s edition of the other rhymed versions has been the source of all other readings and comparisons, but the V7 readings have been checked against the Duggan edition and the Duggan V7 line-numbers used for ease of identification (since no other line-numbers exist). For a variety of reasons no work was done on the current edition between August 1953 and March 1998. Since then, the facilities available on computer have been fully utilized, as have also those of the British Library in London. This has been very much a piece of private and independent research; any errors are solely mine. However, I should like to record my thanks to Professor

x

The Current Edition

Bogdanow for her encouragement and advice in 1998, to Professor Mirjam Foot for very useful information about manuscripts, to Dr Michelle Brown for her comments on the marginal decoration in the ms., to Mme. Dominique Potard, Conservateur en chef de la Médiathèque Equinoxe de Châteauroux for her enthusiastic co-operation and to her and her colleagues for their generous assistance, to the librarians of the Biblioteca Nazionale Marciana in Venice for being so accommodating and helpful, with special thanks to Dottoressa Suzy Marcon for drawing my attention to her contribution to Dr. Carlos Alvar’s edition of La Entrada en España (2003). Her comments on my photographs of the Châteauroux decoration were particularly valuable. So too were those of Professor M. Canova, who kindly discussed them with me at very short notice, when I was in Padua. Professor Lorenzo Renzi also generously accorded me time to discuss my work. In both Padua and Mantua, librarians were unfailingly helpful and so too, in Paris, were the staff at the Bibliothèque Nationale, the Bibliothèque de l’Arsenal and the Bibliothèque de la Sorbonne; nearer home, the librarians of All Souls’, Balliol, Corpus Christi, Keble, Magdalen, Merton and New College, Oxford and Trinity College, Cambridge, accorded me generous access to early manuscripts, and those of the Bodleian and Taylorian Libraries offered valuable assistance in finding later reference books. I should also like to record my special thanks to the late dottoressa Ada di Nola, for her lively interest and practical help. To this list I must, most importantly, append Professor Philip Bennett of Edinburgh University, whom I first contacted in 2008. He has patiently read through all sections of this edition; and his comments on my Introduction have enabled me to revise and restructure my reasoning so as to produce greater clarity in my argument against the Segre theories. Most of all, this work is dedicated to my late husband, Edwin John Moffat, for his constant encouragement and, despite all his health problems, the practical assistance with everyday tasks which enabled me to devote time to this study. Marjorie Moffat Tonbridge, 11 August 2012

Contents I I.1 I.2 I.3 I.4

Introduction 1 The Relationship between C and the other Texts Provenance of the Manuscript 64 Orthography and Principal Linguistic Elements Editorial Policy 87 Appendix A 94 Appendix B 97 Appendix C 100

II

La Chanson de Roland (Critical Text and Notes)

III

Glossary

IV

Proper Names

V V.1 V.2 V.3

Bibliography 619 Editions (Roland) 619 Secondary Literature and Dictionaries Catalogues 625

479 583

619

3 71

103

I Introduction The Châteauroux manuscript of the Chanson de Roland is today the only one of the seven major Old French versions of that poem which has never been the subject of a full critical edition. It has become all the more necessary to remedy this, as a result of Joseph J. Duggan’s edition of The ChâteaurouxVenice 7 version as part 3 of La Chanson de Roland: The French Corpus, published in 2005. Whilst the Duggan edition studies the C manuscript on equal terms with that of V7 – its description, provenance, narrative features and their relationship with other versions, its linguistic characteristics and versification – Duggan’s declared aim (2005, Introduction, 33) is «to produce the text that served as the common model of C and V7» and he adopts V7 as the base of his Critical Text and provides only a non-annotated transcription of C’s text in Appendix B. This choice is based on the theories of Cesare Segre as to the C–V7 relationship, as set out in La Tradizione della «Chanson de Roland»,1 – theories whose validity is placed in serious doubt by a closer study of C. C and V7 were both in the Gonzaga library in Mantua for at least 300 years2 and there is general agreement that the major part of both texts has been copied from the same shared exemplar; this is clearly shown in Foerster’s edition of 1883, where C is printed as the dominant text, with any variant V7 words and graphies printed in smaller font beneath each line; Duggan adds (2005, 30–33) a list of shared errors in C and V7, plus innovations common to both, in support of this. Yet, even in the major (rhymed) part of the text, there are also occasional additional or variant lines in both C and V7, some clearly omissions by the one or the other scribe and necessary for the sense, others apparently representing «corrections» to metre or rhyme – or simply scribal innovation. In addition, C is unique in having two sizeable part-assonanced passages with noticeable variations from V7, each linked to a significant repetition for the most part omitted by V7. C has assonanced or quasi-rhymed laisses in these passages, whereas V7 consistently has rhyme, with very rare poetic licence.

 Cf. Segre (1974, 148–165) chap. 7: «Il problema delle lasse assonanzate nei codici rimati della Chanson de Roland».  They are listed in the 1407 inventory in Mantua and were sold by auction in Venice in 1707/ 08. Cf. Braghirolli/Meyer/Paris 1880. C is No. 52, V7 No. 43 and V4 No. 41 in this list. «Continet cart 59», cited in the Duggan edition (2005, 25) as an error in the C entry, in fact relates to Item 51, Guido de Nantoya (Marciana V10); C is correctly listed as having 125 folios, as verified by me in Mantua in 2005 and also, at my request, by the Mantuan archivist in 2007.

2

Introduction

Cesare Segre argues that, while both C and V7 derive from a shared, rhymed exemplar in the main body of their text, C in the two «assonanced passages» turns to a separate and authoritative assonanced source (lying between β and γ) in order to make good damage to the rhymed exemplar;3 that it sometimes modifies that assonanced source on the basis of the rhymed version, producing what Segre terms «pseudo-rhyme»;4 and that it is therefore slightly later in date than V7, which had earlier copied the unblemished rhymed exemplar. V7’s model thus becomes the hypothetical Roland Rimé, equivalent to δ′ in Segre’s stemma and assumed to be the earliest fully-rhymed version of the whole poem. Segre also sees C’s two major repetitions as resulting from scribal carelessness in having recourse for too long to the assonanced source. It is therefore of critical importance to look at (1) the more superficial features of both manuscripts which aid the dating of C and V7, plus the impact of historical evidence; (2) the structural evidence in respect of the whole CV7 text and its significance particularly to C’s two repetitions, but also to the two associated part-assonanced passages; and (3) the intrinsically textual evidence of the C–V7 relationship specific to those passages. Furthermore, it is essential to examine the relationship of CV7 to both OV4 and PTL, once the latter are available as potential control copies. In making all the above comparisons, the laisse- and line-numbers cited in respect of the Châteauroux manuscript are those of the annotated text which forms the second half of the present edition; in general, these correspond to those of the Mortier edition. For Venice 7, both the Foerster and the Duggan laisse-numbers are usually cited (differentiated by F and D), but the Duggan line-numbers are used throughout. Because the work of comparison was completed before the Roland Corpus edition was published, the Whitehead edition has been used for O, but the Mortier edition in respect of V4, P, T and L. C and V7 are so closely related as to be at times identical. They offer a unique perspective on the development of different versions of the same chanson de geste at roughly the same time; even in the major part of the text, the different aptitudes, attitudes and levels of comprehension of the copyists are discernable; and the presence in C of laisses deriving from a different but authoritative early assonanced model gives new insight into the whole question of remaniement. It is for these reasons, plus the fact that, compared

 Whilst accepting this argument, Duggan appositely comments (2005, Introduction, 34): «partial damage, in that case, because C still contains in those sections readings from the rhyming tradition».  Duggan (2005, 34) refers to a superficial transformation of assonance into rhyme by C.

The Relationship between C and the other Texts

3

with O, C’s text is of lesser literary merit, that the critical text of C in this edition has been modified to a minimal degree. In order to facilitate comparison, full footnotes have been inserted immediately beneath the text, rather than provided in a separate «Notes» section. A full Glossary and List of Proper Names are provided.

I. The Relationship between C and the other Texts The Dating of C and V7: External Characteristics and Historical Evidence Salient features of C, relevant to dating, are its size, quality of parchment and single-column lay-out;5 style of decoration is also a highly significant factor.6 The manuscript is a slightly narrow in-8o, 214 mm x 137 mm; this is somewhat smaller than V7, 228 mm x 154 mm, and the latter looks more spacious – perhaps because of its fewer lines per page (29 or 30 as against 33 in C) and larger, rounded script. The legibility of V7 is shown by the Mortier edition, which consists of reduced-size photographs, four to a page, but still quite readable. C’s angular gothic script is smaller and more worn by usage at times throughout the text, but generally legible. C uses a vertical line in red ink to enhance the (black) initial capital of each line and Proper Name, whereas V7 uses a gold line on red ink; and V7’s decorative «tail» to the capital letter of each laisse is longer and more ornate. The alternate red and blue «laissecapitals» in C always have red ink scrolling, whereas V7 (and also V4) ring the changes: red ink for a blue capital, but blue ink for the alternating red. C is written on parchment of variable, often very poor, quality, with signs of heavy wear; and holes within the writing block have caused the displacement of text, particularly on folios 12 verso, 27 recto and 30 recto (due to maximising the number of folios obtained from each skin, presumably on financial grounds). Thickness of parchment ranges from paper-thin fragile (folio 50) to coarse and the colour varies from ivory to deepest ochre. In contrast, V7 is written on pale ivory parchment of good, even quality throughout (as is V4); one folio (50) is cut slightly short, but there are no holes or flaws and only the slightest signs of use (damp on folio 34 recto? finger-marks or smears on folio 67? a blob of oil or candlewax on folio 121? occasional rubbing by a

 Cf. Hasenohr (1990, 243) for her comments on the archaic quality of this layout in a late thirteenth-century ms.  Detailed descriptions of both C and V7 may also be found in the Mortier edition of each manuscript and the Duggan edition (2005, 15–24).

4

Introduction

reader’s sleeve); like V4, it is in pristine condition. The texts of both C and V7 are set out in a single column (whereas V4, a later in-folio, has double column layout). The combination of small size, poor quality parchment and single-column lay-out (plus lack of gilding) is often seen as typical of early manuscripts, the so-called jongleur’s copies.7 It seems that the Crusades stimulated a fresh interest in the old chansons de geste, giving a new immediacy to their reassuring promise of immediate translation to Heaven of the «Holy Warrior» on his death, but they also gave rise to remaniements designed to satisfy contemporary taste and conform to more recent literary fashion: Section C (C 5888– 8201), which was probably written between 1180 and 1195,8 is an obvious example of this: its vast expansion of O’s 320 lines to 2316 lines introduces the Belle Aude episode into a male-dominated narrative and the «adventure story» element of Ganelon’s attempted escapes into the sober restraint of the earlier version. At the same time, in Italy the earliest Studia, the fore-runners of the Universities, were already established in Bologna and Padua, giving rise to that «explosion of learning» that produced not only the demand for manuscript copies amongst the nobility (and the thriving bourgeoisie) but also the pecia-trained copyists capable of meeting it. In the thirteenth century, the transition from the more fluid oral tradition to the more stable written form can be traced in the manuscripts produced.9 Apart from C and V7 there are only a few thirteenth-century manuscripts amongst the 67 French texts listed in the 1407 Mantuan inventory and known today: the small but modestly opulent Girart de Roussillon, item 4810 (now ms. Canonici Misc. 63 in the Bodleian Library), and the Partonopeus de Blois, ms. L, Gonzaga item 30 (now in the Bibliothèque Nationale in Paris ref. nouv. acq. fr. 7516); and in the Marciana, two in-folio manuscripts, V23, the Tristan, item 64 or 65 in the Gonzaga inventory and V20, item 45, the acephalous Foucon de Candie. The Girart de Roussillon is a strongly Provençal text written in a  Cf. Hasenohr (1990, 239–241 and particularly 249: «Ce n’est qu’à partir du 13e siècle que la mise par écrit des textes en langue française prit réellement son ressort»).  Cf. Palumbo 2003. His argument is based on textual elements in other chansons de geste. He specifies (2003, 410) that this dating relates to «la lunga coda finale del testo, quella piú perfezionata ed elaborata dal punto di vista formale e piú ricca di innovazioni narrative».  It may be seen in the punctus used from folio 5 verso on and particularly in the flamboyant right-slanting punctus elevatus with long hair-stroke which consistently marks the cæsura in C from folio 27 recto: a very useful aid to anyone reading aloud; in contrast, V7 marks the cæsura less consistently and less accurately, with a hair-stroke – perhaps looked upon as adequate for a «library copy». Cf. Destrez (1935) for details of the role of la pecia in thirteenth-century universities.  Cf. Hackett (Paris, 1953) edition.

The Relationship between C and the other Texts

5

single column on very thick, waxy, flawless parchment, small in-8o in size. Dated to the second quarter of the thirteenth century, it is, despite its small size, a valuable, high quality manuscript whose three greatly oversize capitals incorporate substantial areas of gold and tiny animals.11 In contrast, the Partonopeus is extremely plain with no gilding, a much less expensive product: it is written in a single column on poor quality parchment trimmed at some stage to 195 mm x 105 mm, with many holes within the writing block (and repaired large holes after folio 57). There are no historiated capitals but the drawing of a knight on horseback (with lance point down) fills the upper third of folio 1 recto. Even the simple, undecorated, large red capitals which introduce each laisse do not extend to the very end of the text (due to the illuminator’s lack of time or the purchaser’s lack of funds?). This has been dated to the second half of the thirteenth century.12 Finally, V23, the Tristanus in prose, extensively illustrated with lively, page-wide scenes in the readily identifiable Naples style, has been dated to 1280–1290 by D’Arcais 1984. This is in-folio in size (typical of the later date) but made of extremely flawed parchment with coarse texture and ochrous tint.13 The only other Marciana manuscript of approximately this date is the acephalous Foucon de Candie, V20, also an infolio; this has no illustrations (but lacks folio 1, the most likely location) and is written on paper produced in Bologna in or after 1300.14 Of these four manuscripts, the Partonopeus has the greatest similarity to C; it is a poor quality version, purchased when funds were low. C differs most strikingly from all the other manuscripts in having three pages with decoration in the margin. These are the pages bearing the oversize capitals which divide C’s text into three main sections of varying length: Section A, 4179 lines, contains all the events leading to the death of Roland; Section B, 1707 lines, deals with Charlemagne’s pursuit of Marsile and defeat  The initial capital B is written on a rectangular, gold background (54 mm x 37 mm) with a simple, abstract pattern inside the blue and pinkish-red B, but with two tiny animals hidden inside; the large uncial E which begins folio 1 verso is in plain green on a square gold background, 19 mm x 21 mm; below, in the margin, a figure (equivalent to 11 lines of text) drawn in red ink may well be a later addition. Finally, on folio 157 verso is a solid gold L 73 mm high with a short foot, bearing a delicate sketch in which a leafy branch at the top becomes a little Dachshund descending head-first round into the horizontal.  For all details of this ms., see the Internet site of the Arts and Humanities Research Institute, University of Sheffield, relating to Manuscript L: www.hrionline.ac.uk/partonopeus/Lma nuscriptnotes.htm (last accessed June 19th, 2013).  Could this possibly have been a «boys’ book», its lively illustrations designed to tempt the young reader and its parchment sturdy enough to withstand heavy handling?  Cf. Moreno (1997, 23–25) regarding the watermark and inscription linking this ms. to Bologna.

6

Introduction

of Baligant; Section C, 2315 lines, has the «new» version of the return to France, with the Belle Aude episode and Ganelon’s eventual punishment. There are no historiated capitals in C: the oversize C of Charlle in line 1 contains an abstract pattern, as do the M of Morz in C 4180 and the G of Granz in C 5887. On folio 1 recto two bands of decoration frame the text: these are a shaft or lance which runs from the top left-hand to the top right-hand corner, then via a small square down the page to the foot of the right-hand margin, ending in a single acanthus-leaf, and another shaft which runs down the lefthand margin to end in a spray of two acanthus-leaves stretching across the bottom margin to complete the «frame».15 There is a similar shaft or ribbon in the left-hand margin of folio 64 verso, beneath the capital M, and in the same margin of folio 90 verso, beneath the G. These shafts are lightly drawn, decorated with knobs, rings, a clumsy fleur-de-lis and a simple acanthus leaf at the end; the two on folio 1 recto each have a small greyhound or swan head at the end nearest to the capital C; that on folio 64 verso has a bird’s head with pointed beak; that on folio 90 verso has a mournful bear’s (or man’s?) head with cap at the shaft top. The colours are orange, blue, yellow, green, purple, palest pink and brown – all in rather subdued shades. The effect is attractive, but naïve, amateurish and rather primitive. V7 is distinctively different from C (and more opulent) in its style of decoration. It opens with a very large historiated capital depicting the seated figure of Charlemagne in pink and blue, very simple in form and primitive in facial featueres, alongside the capital K on a gold ground. The simple red initial letters beginning each line are also touched with gold. All V7’s oversize capitals are red in colour, with narrow bands of pink, white and brown following both the inner and outer contours of the letter, but after the initial K they are always set on a solid blue ground, faintly lined in white to frame the letter and decorated at the corners with small acanthus leaves in brown, red or pink. They cover 4 or even 5 lines in depth but, after the first page, they are not accompanied by any decoration in the margin. In folio 1 recto, V7 has a more formalized «shaft with acanthus-leaves» than those in C, with firmer outline, strong colours and golden balls or berries. The same style is found also in V4 and is typical of that seen in many very late thirteenth and early fourteenthcentury manuscripts produced (and still preserved in situ) in the Biblioteca  In Naissance et Développement de la Chanson de Geste en Europe (1961, 319s.), de Mandach suggested that these shafts represent two L (one reversed) to signify Luigi Gonzaga, who came to power in Mantua in 1328, but this is purely speculative. Similarly, any resemblance between the miniature of Charlemagne at the beginning of V7 and the portrait of Luigi in the ducal palace is extremely debatable. In either case, there seems no reason to assume a post-1328 date for the poor quality C: by then the Gonzaga were extremely wealthy.

The Relationship between C and the other Texts

7

Universitaria in Padua;16 the style seems generally typical of the mainland Veneto. In her article on Les illustrations des manuscrits français des Gonzague à la Bibliothèque de Saint-Marc, Francesca D’Arcais (1984) dates V7 to the end of the thirteenth century or beginning of the fourteenth century, with 1320 the latest date possible. This is based on its sequence of small, very simple initials and especially on the primitive style of the seated figure in the single historiated capital. She considers that both V7 and V4 (dated to 1320–1345 on similar grounds) were copied somewhere in the mainland Veneto, not in Bologna. Moreover, in the case of V4 (which has a range of historiated capitals throughout the Aspremont as well as that on the first page of the Roland), D’Arcais identifies a specific illuminator, the «Maître du B.18», whose work is to be found in the Biblioteca Universitaria in Padua.17 C, which has never been a Marciana manuscript, lies outside the scope of the D’Arcais article; it is the only manuscript to stress the «section divisions» with decoration in the margin of the pages bearing the oversize M and G. D’Arcais’ suggestion of Padua is interesting because, whilst there are many late thirteenth- and fourteenth-century manuscripts in the Paduan library with marginal decoration like that in V7 and V4, only one has been found to match that of C: ms. 0510: Petrus Lombardus: Sententiarum libri quatuor, produced in Padua in the thirteenth century. Like C, this is in-8o in size, but it is in prose, written in two columns with alternating blue and red capitals at the beginning of each sentence. It has a lightly-outlined shaft running down the left margin and across the top of the first page, turned to form a square (like  There are, in the Biblioteca Universitaria in Padua, 12 volumes of excellent photographs showing each individual item of decoration (grouped by manuscript) to be found in the mss. produced in thirteenth–eighteenth-century Padua; unfortunately, only the magnificent miniatures are reproduced in La miniatura a Padova (1999).  By 1163, Padua was already a free Commune, but it rose to new heights of prosperity, power and pre-eminence in both arts and sciences in the thirteenth and fourteenth centuries. The Palazzo della Ragione was built 1218/19, the University established by 1222, the Basilica of St. Anthony begun in 1232, the circular Baptistry in 1260; the Tomb of Antenore was erected 1283–1285, the Scrovegni Chapel with its magnificent Giotto frescoes dates from 1303 – and the list continues. The Biblioteca Universitaria holds many manuscripts, mainly in Latin, produced in Padua at this time; and there is evidence of both noble and plebian taste for chansons de geste (particularly the Charlemagne cycle). In the introduction to his edition of V4, Beretta (1995, xviii) argues in favour of Treviso, rather than Bologna, as the place of its production, on linguistic grounds, but Padua, Bologna and Treviso were all university cities in the thirteenth century. Padua and Bologna are equidistant from Mantua, both Padua and Mantua being to the north of the River Po, whereas Treviso is farther away, nearly 50 kilometres beyond Padua – but near enough, surely, for Padua to share linguistic characteristics with the Treviso area.

8

Introduction

that in C) in the top left-hand corner; this shaft has knobs, rings and other simple decoration, as in C, but foliage only at the terminals and particularly in a more formal acanthus design across the foot of the page. As in C, the effect is rather primitive and amateurish, when compared with V7 and V4; also, the dominant orange, palest pink and deep turquoise-blue shades, with touches of pale blue and a yellowy-olive green, echo those of C. The occasional large capitals usually contain a simple, formal pattern but there are three, on folios 2 verso, 75 verso and 135 verso, which conceal tiny animal-heads within the pattern, the first head being reminiscent of that on folio 90 verso in C. The capital C on folio 135 verso of the Paduan manuscript closely resembles the initial C of Charlle on the first line of C. There is no gilding. C contrasts strongly with the higher quality of production and pristine condition of V7 and V4, yet its careful lay-out produces a neat, attractive appearance, pleasing to the eye, and it has been well-used over the years. The impression one receives is of a well-loved manuscript, originally commissioned and purchased when funds were limited. V7 confirms this view, for the commissioner of V7 apparently wished to have the same text as C with only slight modifications, i.e. an improved and corrected copy taken from the same source – and C certainly contains meaningless lines caused by scribal incomprehension, unconcern or haste as well as by simple error, and a great many corrections to the text, both by the copyist and by a reviser. The better quality of the V7 parchment, the decoration on the first page and the skill of its copyist make it a more expensive product, implying production when funds were more readily available. The concept of an «improved copy» is also supported by the evidence of the two Foucon de Candie manuscripts V1 and V2, Nos. 46 and 45 in the Gonzaga inventory, V19 and V20 respectively in the Marciana. Foucon V1 (= V19) is in-folio size written in two columns of 36 lines on 91 folios of good quality parchment, richly decorated and gilded. D’Arcais dates it, via the heraldic symbols in the exuberant decoration on the first page, to 1360–1369 (although the black-and-gold horizontally barred shield was already in use in 1328 and possibly earlier, according to Giancarlo Malacarne).18 V2 (= V20), slightly smaller in size, lacks the first folio; the text is written in two columns of 61 lines each on 55 folios with single-line-space laisse capitals and no other decoration, on paper produced around 1300 (see footnote 14 above). Schultz-Gora (1966, 5)

 Cf. Malacarne (1993, 39s.); the Araldica Gonzaghesca contains large, clear illustrations of early pre-1328 Gonzaga shields with bars of black and gold.

The Relationship between C and the other Texts

9

suggested that V2 was an exact copy of V1, but Paola Moreno has more recently provided evidence that the reverse is true.19 Here again one apparently has an earlier, less expensive copy (with loose, detached or damaged folio 1) replaced by a much higher quality manuscript. There are also similarities between the Foucon V1–V2 relationship and that between V7 and C in the first repetition. It makes little sense to commission a cheap, poor quality version if one already has a better, richer copy, whereas the reverse is a logical procedure when one has more money available (and a certain style to maintain). There is no evidence in the Gonzaga library to support the idea of commissioning cheaper manuscripts for lending to other members of the North Italian nobility. The request for the loan of the Gonzaga «pulcrum Aspremontem» in 1371, cited by Francesco Novati 1890, may relate to the newly completed Venice 6 or the earlier and rather less lavish V4, but both would merit that description. Indeed, the quality of parchment and illumination increases commensurate with Gonzaga wealth.

The Historical Evidence There may, however, be other reasons for duplicate copies being found in the 1407 inventory, made on the death of Francesco I, the fourth Capitano del  Cf. Moreno 1992 (197–199) and 1997 (21–25, 152, 168–172 and 215). On pages 21 and 23 Moreno provides a detailed description of Foucon mss. V1 and V2; she stresses (1992, 197) the extremely close match between the two mss (except for 195 lines on V2’s missing folio 1). On 20 occasions, V1 and V2 (but not the other Foucon mss.) run two laisses into one, in 15 cases because of similarity of rhyme but without alteration (e.g. -eç and -aç in laisses 70 and 71, -anz and -auz in 319 and 320) – or due to scribal inattention; but on 3 occasions (laisses 14– 15, 269–270 and 422–423) justification for fusion is found solely in V2 and in the laisse-position on the page. Thus, in V1, laisse 15 (written as part of laisse 14) begins half-way down column b on folio 3 recto, whereas in V2 laisse 14 ends at the foot of column b of folio 1 recto, so that laisse 15 begins on the first line of folio 1 verso – but without the larger coloured capital needed to indicate a new laisse. This is a scribal error in V2, copied faithfully but uncritically by V1. On a further 15 occasions, V1 unites two laisses where V2 does not, whereas V2 quite independently runs together only laisses 777 and 778. There are 4 laisses (470–471, 556–557, 573–574 and 759–760) where laisse-fusion in V1 is due to incorrect marking in V2: a space being left but no indication of the lettera d’attesta in V2 (e.g. laisse 556 in-on immediately followed by 557 in -ier). – Moreover, Moreno cites (1997, 168 n. 24) 14 lines (519, 710–713, 2250, 5953, 7180, 9853, 10641, 12984, 13044, 13141, 14478) and 3 rhyme-words which V2 shares with the other Foucon mss. but which V1 lacks – and (1992, 199) draws attention to V2 lines 14478– 14479 (with line-inversion in error – cf. mss. S, B and Ch – and repetition of la mamelle where B and Ch read l’alemele) and to V1’s deliberate omission of line 14479 to eliminate V2’s

10

Introduction

Populo of the Gonzaga dynasty. The first Gonzaga to achieve this elective office was Luigi, who came to power in 1328 after leading an insurrection against Rinaldo Bonacolsi (known as Passerino), whose family had ruled Mantua as Capitani since 1272. In his introduction to the catalogue of the Mostra dei Codici Gonzagheschi 1328–1540, held at the Biblioteca comunale of Mantua, 15 September–10 October 1966, Ubaldo Meroni drew attention (1966, 43) to the Will of Filippone Bonacolsi, dated 25 September 1325, in which he bequeathed six manuscripts to his nephews: two copies of Brunetto Latini’s Tesoro, a Cesarianus, a Sidrac, a Tristano and a Trojano. With the death of the other legatee, these became the property of Passerino. Meroni suggests that they, with all the Mantuan archives and all other Bonacolsi property, eventually passed into the possession of the Gonzaga. However, if the D’Arcais datings are correct, of these six, only one manuscript found in the Marciana pre-dates 1325: the Tristan, V23, inventory item 64, dated to 1280–1290 (or, at the latest, to 1320 by Dezenhart-Schmidt and Perriccioli Saggese20). The two Tesoro, items 6 and 7, could also be former Bonacolsi mss., as could the Sidrac, item 18, but these are not in the Marciana and have not been firmly identified or dated. The Cesarianus, item 12, is V3, dated 1340–1350, so too late in date; but the Bonacolsi ms. could well be item 13, an unidentified copy of the same text. Finally, both Marciana copies of the Trojano, items 28 and 29 in the Gonzaga inventory, are too late in date: V17 has been dated to 1330–1340, V18 to 1360–1369. Thus, of the six manuscripts, one certainly could have belonged to the Bonacolsi, a further three may possibly have done so. Certainly, there is evidence that the Bonacolsi had contacts with the academic world in Padua: in 1295 Pietro d’Abano dedicated his treatise, Fisonomia, to Bardellone Bonacolsi, Capitano of Mantua.21 Nor should it be forgotten that twelfth–thirteenth-century Mantua was no cultural desert. This self-governing Commune22 encompassed within its boundaries the village of Andes (today known as Pietole), birthplace of Virgil in 70 BC, and that of Goito, birthplace of the poet Sordello (c.1200?–1270 or 1274). The monastery of San Benedetto di Polirone, founded in 1004, was a very active cultural centre; in the eleventh century, the law schools of Mantua equalled in repute those of Bologna and Padua, although they never developed

incomprehensible reading. On pages 169–172, she examines the degree of concordance between V (i.e. V1 = V2) and the other mss; and a stemma codicon is provided on page 215.  Cf. Degenhart/Schmitt (1977, 89 n. 8).  Cf. Marcon (2003, 299).  For details of the early development of the Commune of Mantua, cf. Coniglio (1958, 117– 142).

The Relationship between C and the other Texts

11

into a university and the number of students decreased in the thirteenth century.23 Other families may therefore have commissioned manuscripts. Within this context, there is intriguing evidence, sparse but significant, in relation to the Gonzaga family itself. This suggests that the earliest items of French literature found later in the 1407 inventory of its library were, from the first, Gonzaga manuscripts. The family took its name from the village or castle of Gonzaga, some 40 kilometres south of Mantua; but, according to Giuseppe Coniglio, it had no great status in that area, which in 1212 became the fief of the Casaloldi family by grant from the Emperor Otto IV. In his book, I Gonzaga, Coniglio refers to the family24 as i Corradi di Gonzaga, but the genealogical table25 given at the end of the book shows Corrado, the sixth and youngest son of Filippo, as having a son Gualtiero, so the name of Gualtiero dei Corradi di Gonzaga (1967, 8–11) may simply indicate that he was of this branch of the family. Coniglio stresses (1967, 7) the rich patrimony the Corradi di Gonzaga had accumulated at the beginning of the thirteenth century around the monastery of San Benedetto di Polirone (just south of Mantua), where Corrado’s son Gualtiero was the uomo di fiducia of the monastery. This monastery had earlier received rich gifts of lands from the Countess Matilda in particular (deeds of gift being signed as witnesses by earlier Gonzaga in 1146 and 1149); and a considerable amount of property had apparently been received – possibly as payment for services, rented or purchased from it by the Gonzaga. The family’s status in the Commune of Mantua may be inferred from the fact that the same Gualtiero was involved in 1199 with his cousin Corrado (son of Corbello, the third son of Filippo) in a Peace Treaty between Mantua and Padua, whilst, in the same year, the elder Corrado with his brother Abramino (Filippo’s second son) was signatory to an alliance between Mantua and Padua against Verona and Vicenza (Coniglio, 1967, 11); and Corrado also signed the Peace Treaty with Verona in 1217. The family tree traces the Gonzaga descent from the eleventh-century Filippo, who had six sons. At least two of these and four of his grandsons took an active part in public affairs: in addition to those mentioned above, Alberto, Abramino’s eldest son, signed a deed by which, in 1199, the Mantuan authorities granted concessions to the church of San Marco; Guido, Abramino’s second son, became ambassador to the Imperial court in 1209; and Corrado, son of Corbello, signed the Peace Treaty with Padua. Abramino’s grandson, Anto Cf. Faccioli (1959, 343s.).  Coniglio (1967, 7–15).  Tavola 1 of Albero Genealogico dei Gonzaga Signori di Mantova (dal Litta: Famiglie Celebre italiane)

12

Introduction

nio, was an Anziano (Elder) of Mantua in 1259 – and instrumental in the coming to power of Pinamonte Bonacolsi in 1272. Gualtiero dei Corradi’s close involvement in the stormy relations between Pope and Emperor put his estates in jeopardy on two occasions, but by 1260 his son Corrado was an Anziano of Mantua and another son, Ziliolo, was a member of the Council in 1285. The family also became increasingly wealthy, partly due to advantageous marriages. Gualtiero, a leading Guelph supporter in the area, had his lands declared forfeit by Henry VI’s legates, but was saved by the latter’s death in 1197; and Frederick II’s imperial decree of 1220 ordering their restitution to the San Benedetto monastery seems to have been implemented only in 1227, because of the frontier wars of 1220–1225 between Mantua and the neighbouring Communes of Reggio and Cremona. Gualtiero’s Will, drawn up before his departure in 1221 on Crusade, shows his considerable wealth. After the Emperor recognized the Casaloldi right to the fiefdom of Gonzaga, Gualtiero turned his attention to the Marmirolo area, north of Mantua, where by 1261 his sons had acquired extensive domains. The Abramino branch, particularly Guido and his son Antonio, had extended their estates into the neighbouring Commune of Reggio (although this caused them problems during the war); Antonio’s advantageous marriage brought further rich territorial possessions. Although occasional disputes led to temporary exile and dispossession during the thirteenth century, partial restitution was made by 1273 (after Pinamonte Bonacolsi, helped by Antonio, came to power as Capitano del Populo); and all their San Benedetto lands were finally returned to the family in 1287 (thanks to the opportune finding of documentary evidence by the Pope’s envoy, a Gonzaga Canon). Eventually, it would appear that, as the other branches faded out, their lands fell into the hands of the Abramino line whose scion, Luigi, added further substantial property in Brescia and Ferrara as his wife’s dowry – and became Capitano del Populo in 1328. When, in 1354,26 the former Bonacolsi properties which enclosed the huge square which is today the Piazza Sordello were confirmed as Gonzaga possessions, the Gonzaga family became quite fabulously wealthy. Although the extensive Gonzaga family tree at the end of I Gonzaga gives no dates of birth or death, it is possible to arrive at approximate dates from the historical evidence. Thus it is likely that Abramino and his youngest brother Corrado were at least 40–50 years old (Abramino perhaps nearer 60) when they signed the 1199 treaties, since their nephews were also signatories to treaties at this time. Corrado’s son Gualtiero, the confidential agent of the San Benedetto monastery, was already a man of property and active politically in 1197, when the Emperor Henry VI died, so probably at least 25 years old.  Cf. Paccagnini (1960, 31).

The Relationship between C and the other Texts

13

His cousin Guido, Abramino’s son, was probably in his thirties in 1209, when he was sent as Mantuan ambassador to the Imperial court on the coronation of the new (Guelph) Emperor, Otto IV. Mantua, a self-governing Commune trapped geographically and politically between Emperor and Pope, clearly took the opportunity to send an ambassador likely to find favour with the 35 year-old Emperor – and this paid off, for Otto was the guest of Guido Gonzaga during his subsequent visit to Italy (and Gualtiero retained his lands for several years longer). Guido must surely have been not only wealthy but also well-educated and cultured to be honoured by so prestigious a visitor. Unfortunately for the Gonzaga, Otto IV was deposed and excommunicated in 1212, then defeated in battle by Philip Augustus of France, the Pope’s ally, in 1214 and deposed. The next Emperor, the Hohenstaufen Frederick II, crowned Emperor in Aix-la-Chapelle in 1215 and in Rome in 1220, was eventually obliged to accede to the Pope’s demand for the San Benedetto properties of Gualtiero di Gonzaga to be returned to the monastery. Frederick was also forced to recognize the rights of the Casaloldi to the property around Gonzaga. It is not clear from Coniglio’s account whether the Abramino branch of the family (which produced the future rulers of Mantua from 1328 on) lost lands around Gonzaga at the same time as Gualtiero; certainly, all those lands were finally returned to the various branches of the Gonzaga in 1287. During the 1220–1225 war between Mantua and Reggio, Abramino’s son Guido and grandson Antonio were excluded from the considerable domains they had acquired in that Commune, but they were reinstated in 1257; subsequently, the exile of Guido and Antonio from Mantua in 1264 resulted in the temporary forfeit of their Marmirolo estates, but Antonio obtained some of the Casaloldi lands after his ally, Pinamonte Bonacolsi, came to power in Mantua in 1272 – and he had other extensive estates from his wife’s family. According to Coniglio, it was Antonio’s father Guido, not his son of the same name, who was exiled with him in 1264, but summoned back in 1265 to be made capitano di spedizione as ambassador to Charles d’Anjou, who was crowned King of the Two Sicilies in Rome on 6 January 1266. Charles was a major force in Italian politics during this period, defeating and killing both Manfredo, the illegitimate claimant to the Sicilian throne (in 1266) and his legitimate Hohenstaufen nephew, Conradin (in 1268) – but he also established a strongly French court in his new kingdom, being the youngest brother of Louis IX of France as well as the ruler of Provence. To return to Guido, it is rather surprising to find Coniglio identifying him as ambassador in 1265 and taking part in the battle of Benevento alongside Charles d’Anjou in 1266, when he was probably well into his eighties (even if he were only 20 years old when made ambassador in 1209, he would have been 76 in 1265). One is tempted to see this as the younger Guido, Antonio’s

14

Introduction

third son, and his appointment as a repetition of the earlier, successful Mantuan policy, since this Guido would be closer in age to the 39 year-old Charles d’Anjou. Antonio, his father, died in about 1283, so was probably born in the first decade of the thirteenth century, perhaps even in 1200; the younger Guido’s second son, Luigi, the future Capitano del Populo, was born circa 1267; so Guido himself could have been in his thirties in 1265. If the events of 1264/ 65 do indeed relate to the younger Guido, he must have enjoyed an exceptional reputation for intelligence, education and culture (as well as residual wealth despite the forfeit of estates) to be summoned back from exile to such high office. If it were the elder Guido, he must have been the Grand Old Man of Mantuan politics, with tremendous prestige – but perhaps, as a precaution, accompanied by his grandson, for surely it would be the younger who fought alongside Charles at Benevento in 1266? Thus, whilst the first Guido was subjected to the cosmospolitan influences of the Imperial Court in the first decades of the thirteenth century, it seems that he and/or his grandson Guido would feel the more specifically French cultural impact of the Neapolitan/Sicilian Court after 1265. Charles d’Anjou was a very strong, dynamic, proud and ruthless character – but also a poet, a friend of poets and an arbiter in poetry contests. At 20 years of age, he married the heiress to Provence and proceeded to impose his authority upon that County. A year later, in 1247, he took part in the 7th Crusade, led by Louis IX, and was imprisoned by the Saracens for a time with his brother before being ransomed and returning to France as co-Regent during Louis’ absence. In Charles d’Anjou, roi-trouvère,27 Maillard (1967, 20–30) examines his lively career, his complex character – and his links with poets and the world of epic poetry. He writes of «cet homme grand, aux traits accusés, avec un air grave et dur, qui parlait peu, ne riait pas… (qui) ébranla la moitié de l’Europe et fut célébré, ou maudit, dans toutes les langues…» who was extolled for his chastity, religious devotion, courage, taste for the arts (1967, 24). He refers to the «vie brillante et luxueuse» at the Court of Naples (1967, 20): «véritable radiant de tout un monde poétique et chantant de troubadours, trouvères, ménestrels et jongleurs»; and (1967, 27–30) he cites the tributes of poets such as Raoul de Soissons, Rutebeuf and Jean de Meung, the finesse of his judgements as arbiter in poetry contests (citing a whole sequence of these) and finally La Chanson du Roi de Sezile, the great panegyric of Adam de la Halle, the poet who followed Charles to Italy and eventually died there.

 Maillard’s assessment of Charles is based mainly on Langlois 1901 (Foreign Policy, 98– 102). De Joinville takes a slightly different, but complementary view in his Chronique des Croisades.

The Relationship between C and the other Texts

15

This view of Charles is endorsed and expanded by Perriccioli Saggese (1979, 26s.), with additional evidence cited by reference to Sabatini’s La cultura a Napoli nell’eta angioina (Storia di Napoli, vol. IV, tome II, 1974), Leonard, Les angevines de Naples (Paris, 1954), de Bouard, Documents en français des Archives angevines de Naples (1933 and 1935) and Loomis’ 1938 monograph on the Arthurian legends in mediæval art, amongst many other scholars. In particular, she quotes Sabatini (1974, 35) in stating that Charles d’Anjou, «cresciuto alla corte di Francia e fornito di un’educazione cavalleresca e letteraria, era in stretti rapporti d’amicizia con molti poeti, specie con quelli dell’attiva cerchia di Arras. Fu protettore particolare di rimatori celebri…»28 and that in Naples (1974, 37) «vi fu certo un andirivieni di poeti e giullari tra la Francia e Napoli». This was the man at whose court a Guido Gonzaga came into dramatic contact with French poetry – and surely, given Charles’ Imperial ambitions, with the Chanson de Roland, in which Charlemagne could be equated in flattering allusion to Charles? This was a court at which, despite its location in southern Italy, the references to «nostre François» and to «France l’alosee» would ring out proudly. In her 1958 thèse de doctorat at the Sorbonne, L’évolution du Roland,29 Jeanne Rosier made out a case for identifying each version of the Roland with a particular set of historical circumstances and thus a specific date. For Rosier, this involved linking the writing of C to the death of Charles d’Anjou in 1285. However, it seems far more probable that the model C shares with V7 was the «latest version», the «top of the pops» of the day, carried to Naples by the French courtiers and read aloud or recited there in honour of the newly-crowned Charles in 1266 or after his subsequent victories over Manfredo (1266) and Conradin (1268). It is hardly surprising that the Mantuan ambassador to that court should wish to acquire a copy, despite his somewhat straitened circumstances and the expenses of his office. Indeed, if he had already purchased the Partonopeus, whose single illustration of a knight on horseback may, as an allusion to the symbol of Charles d’Anjou, link it with a scriptorium in Naples30 and whose unfinished red lettering might well indicate  Amongst these was the Mantuan poet, Sordello (mentioned earlier, page viii), who was attached to Charles’ court in Provence from 1229 and whose presence in Naples (in prison) is recorded in 1266.  Bibliothèque de la Sorbonne, Réserve, réf. W 1958 (I09 (I–2), 4o; vol. 2, 266–541 and 529 for conclusions.  The University of Naples dates from 1224 and Charles d’Anjou built up a much-admired royal library in that city before his death in 1285. Both these developments, together with the highly literate royal courts of Federico II and Charles d’Anjou, were likely to stimulate the establishment of lay scriptoria.

16

Introduction

the purchaser’s sudden departure for the battle of Benevento, Guido could probably not afford to commission C until later – and therefore, perhaps, from an atelier de copistes nearer to Mantua (and Padua was the nearest university city). Even so, the younger Guido with his young family would only be able to afford a less expensive copy like C, in which a less experienced copyist transcribed a slightly earlier version of the text uncritically on poor quality parchment. Furthermore, in 1270 Louis IX of France led the 8th Crusade, directed (at Charles’ suggestion) against Tunis and including a contingent from England under the future Edward I and one from Flanders under Gui de Dammartin, as well as Charles’ Sicilians. On 25 August 1270, before the Sicilian contingent arrived in Africa, Louis died in Carthage, struck down by dysentery or plague; Charles then led a brief but successful campaign. The Crusader army then returned to Sicily and, crossing to Italy, headed north on their journey home. Within the Count of Flanders’ entourage was a famous poet, Adenet, whom the Flemish jongleurs had acclaimed as Roi des menestrels. Adenet (born c.1240) had been educated and trained (in France) under the patronage of Duke Henri III of Brabant, but had moved to the court of Flanders after the latter’s death; he wrote in French. The itinerary of the Flemish contingent can be traced in the account-books of Gui de Dammartin via a record of payments made to Adenet (for his services in providing entertainment perhaps) at each stop on the way back: after passing through Calabria to Naples and then Rome, they reached Florence in April 1271, then went on to Prato, Bologna, Modena, Reggio… and so to Milan and Aosta, in order to cross the St Bernard Pass in May, passing thereafter via Lausanne, Bar-sur-Aube (birthplace of Bertrand), Dôle and Provins to Paris31… and so eventually home. One thus had on the same Crusade the roi-trouvère and the roi des menestrels – and there is evidence elsewhere that works of French literature were taken on Crusade, the young nobles reading the great works of fiction (Tristan, Lancelot, le Roman de Troie, le Roman d’Alexandre, etc.) aloud to their fellowCrusaders during the long intervals between battles.32 Was this the case in 1270 and did it stimulate the poets in Charles d’Anjou’s entourage to the remaniement of the last few imperfectly rhymed laisses in the Chanson de Roland (as found in C’s assonanced or partially assonanced passages) so as to produce the fully rhymed version later found in V7? Were the rhymed versions of these

 The full list of the stops on this route is given by Holmes (1946, xl n. 83). These related to individual payments made to Adenet en route; he received his wages later, in 1275 and 1276.  Cf. Jacoby 1984 for evidence of this practice.

The Relationship between C and the other Texts

17

passages a communal effort, with Charles as arbiter setting standards as regards regular metre and rhyme? This can only be hypothesis, but the coincidences are extraordinary: the presence of a very literate French court in Italy at the appropriate time, with optimal reasons for celebrating with a «Charlemagne cycle» chanson de geste; the presence of ambassadors from the North Italian communes (surely not Mantua alone but, pertinently to both C and V7, certainly a Gonzaga) at that court; the strong Eastern and North-eastern French element due to the presence of poets from Arras and Flanders (and still visible in the linguistic features of C); the evidence of the Partonopeus, apparently copied in Naples; then, after the 1270/71 Crusade, the return journey of Adenet le Roi and the Flemish Crusaders through Italy, with a halt in Bologna (with its university and early evidence of independent copyists33), Modena and Reggio (so near to Mantua) and finally through Burgundy, southern Champagne and the Ile de France to Flanders. Whilst Crusaders often set out by sea from Aigues Mortes or Venice, this could well be the normal return route by land from southern Italy. It also offers an off-shoot branching North-East to Mantua, Padua, Venice, Treviso and North to the Germanic States beyond. Were these the routes by which the new, «improved» Roland, freed of alexandrines, assonance and faulty rhyme, travelled North? What is certain is the correlation visible between the thirteenth-century manuscripts in the Gonzaga library and that family’s known wealth or poverty. Thus there was enough money available between 1225 and 1250 to permit the purchase of the small, modestly opulent Girart de Roussillon; straightened circumstances between 1265 and 1273 forced a choice between quality and quantity, resolved by the purchase first of the Partonopeus, written on poor quality parchment with no decoration, then of C; and it was only in the final decades of the century that the restoration of the extensive Gonzaga domains resulted in the commissioning of V7. There is no comparable correlation with the Bonacolsi fortunes. Indeed, the potential identification of the Tristan in the 1407 inventory with that of the Bonacolsi Will of 1325 may be rendered dubious by the poor quality of the parchment, for during the years 1272–1328 the family enjoyed ever-increasing power, wealth and prestige. The poor quality of C and the Partenopeus, plus the extreme plainness of the latter, would accord ill with their status.

 Cf. Bosi (2000, 53s.) for evidence of the presence of 10 scribes/copyists and two illuminators amongst the witnesses signing Bologna’s Peace Treaty with Pistoia in 1219.

18

Introduction

The Structural Evidence a. C’s Two Repetitions and Associated Clusters of Assonance C’s three oversize capitals (C, M and G on folios 1 recto, 64 verso and 90 verso) divide the poem into three sections of disparate length: Section A (the sequence of events leading to Roland’s death) with 4179 lines, Section B (the pursuit of Marsille and the battle with Baligant) with 1707 lines and Section C (the dénouement) with 2315 lines. Thus the first half of the poem (Section A, 4179 lines) recounts the «Roland story» culminating in his death, while the second half (B and C together totalling 4022 lines) deal with the consequences of that death: the grief of Charlemagne and the justice he meted out to the Saracins (Marsile, Baligant and their armies) and then to Ganelon, with the tragic tale of Aude serving to show the passage of time between the Ganelon incidents as well as appealing to contemporary taste. This underlines the importance of balance between the incidents in the Roland. Palumbo demonstrates that Section C dates from the last two decades of the twelfth century, which implies a circa- or post-1200 date for any modifications to the earlier narrative to adapt its detail to that in Section C (for example, the earlier «grieving at Roncevaux» episode deliberately excised from C and V7). V7 has the same «chapter markers» (though with K, not C in line 1), but adds additional oversize capitals: L at V7 1052 (= C 1024), .M. for Marsillions at V7 4563 (= C 4449) and V in Vet, V7 7521 (= C 7338). Of these, the L marks the point at which, deceived by Ganelon’s false report of Marsile’s intentions, Charlemagne decides joyfully to return to France; the appointment of Roland to the rearguard then follows. The .M. introduces Marsile’s return to Saragossa, opening the whole Baligant episode (V7 4563–6006); it seems to imply that, despite the M of Morz (V7 4291), the V7 copyist saw the Death of Roland episode as ending with Charlemagne’s pursuit of Marsile’s army and subsequent prophetic dreams.34 The V marks a logical division in Section C’s 2,391 lines, inserted at the end of the Belle Aude episode and leaving a further 1,515 lines for Ganelon’s attempts to escape retribution. This V could possibly be an omission by the illuminator of C, occasioned by the fact that C 7338 is the first line of folio 112 verso as well as the first line of laisse 394. In V7, the length of «chapter» is usually a multiple of just over 500 lines, apart from that beginning with Morz (but see comment above and cf. .M. for Marsillions in V7 4563); V7’s oversize G of Granz occurs at line 6007. It will be seen later that, if one assumes a lay-out of about 32 lines per page in the CV7

 Taken with the testimony of L, which omits the whole «Battle with Baligant» episode, this may imply that that episode was treated as optional.

The Relationship between C and the other Texts

19

model (and thus 512 lines per quire), each quire of text corresponds to a separate sub-episode in the narrative. Such a lay-out is a credible possibility, since C and V7 have no common pattern of page formats: C has 33 lines per page, except for the 7th quire (32 lines per page) and the 32-line folios 77v, 78 r, 89r and 96v,35 whereas V7 normally has 29 or 30 lines per page. Indeed, the lack of standard page-length is emphasized by V7’s unusual sequence in quire 2: 33, 34, 35, 34, 35, 35, 35, 36, 34, 33, 33, 34, 34, 34, 34 and 27 lines per page. This gives an extra 60 lines in the quire, the equivalent of one extra folio. Perhaps the copyist suddenly realized that, if he continued with the 30-line pattern, he would need an extra quire of parchment to complete the poem (beyond what had been quoted for) and would use only a single folio of it (a great extravagance). The same problem, but a different solution, is seen in V7’s insertion of a 10-folio quire to accommodate additional (rhymed) material where C laisses 202–218 form a second «cluster of assonance» after the second repetition. V7 may also have wished the oversize L (V7 1052) to coincide with the first page of a new quire, to mark the beginning of a new sub-episode (the nomination of Roland and the arming of the Saracen peers), keeping to the quire-episode pattern. This modification brings the L to the 18th line of folio 17 recto and underlines the fact that the «Ganelon’s treachery» episode ends with laisse 60, not laisse 59,36 i.e. with Ganelon’s false report to Charlemagne, not his departure from Saragossa. C’s 32 lines per page in quire 7 may also be significant, for it opens with C 3168. the first line of laisse 197 and the beginning of the second repetition. When C found himself, by chance, starting a crucial episode on the first line of a new quire, it would be so much easier to adopt the lay-out of his model (at least for the quire in question). Looking at C, it is clear that Section A (4179 lines) is far too long for reciting without intervals in the «performance»,37 but C’s two repetitions serve to divide it neatly into three sections: 1533 lines, 1634 lines and 1012 lines, roughly equivalent to 3 quires, 3 quires and 2 quires of text on the basis of 32 or 33 lines per page. This provides convenient breaks for the performer’s rest

 Most of these have a line omitted or line-space left blank in the ms.  C and V7 do not coincide at this point, V7 having two laisses (58 and 59) where C has only one (58), each version clearly deriving from a different remaniement – implying confusion over where the «treachery» episode ends.  Various methods of recital, with or without musical accompaniment, and various speeds have been suggested, but Rychner (1955, 48s.) cites Murko’s suggestion of 1,000 lines per hour and suggests 200–500 ll. per half-hour as an average speed, depending on the singer or reciter and the tempo adopted. This is very close to the 512 lines per quire posited above and validates it on purely practical grounds.

20

Introduction

and refreshment.38 In contrast, V7 has 4290 lines in this section (in the Duggan edition) and its six oversize capitals split the text into episodes or chapters of 1051, 3239, 272, 1444, 1514 and 877 lines, the irregular sequence being immaterial in a «library copy». However, C is also destined for a library, so the «performance-based pattern» derives from the source – that is, either from the shared CV7 model or from the authoritative assonanced model posited by Segre as accessed by C to make good damage to the CV7 model. Examination of the texts reveals structural (and textual) evidence that, at the «division points», both C and V7 derive from the same CV7 model. b. The First «Cluster of Assonance»: C Laisses 73–100, V7 74D–91D (+ 92A, 108A) C contains two almost identical versions of the laisse beginning Dist Olivers, «Or ai paiens veüz»: C 91 (11 lines) in the assonanced passage and C 101 (9 lines) in the rhymed. V7 has just the one laisse, V7 91D (11 lines). In his 1883 edition, Foerster identifies the V7 laisse as based on the C 91 text; but Segre prefers C 101 as the base text. Both are to some extent correct, for a close textual examination (see Textual Evidence, page 31) reveals that V7 91 amalgamates the two laisses, mixing and matching the lines but ending with the final line of C 91, which is not found in OV4T or C 101. At this point P and L are still acephalous. This line, C 1421, is also the first line of folio 22 verso in C. Having written V7 1487 (= C 1421), the V7 scribe next proceeds (quite correctly) to copy V7 92D (identical to C 102, both rhyming in -er) from the shared source, thus eliminating C’s superfluous «repetition» laisses; but the V7 scribe copies the first 11 lines only: 1488–1498 in V7, 1543–1553 in C 102. C 1553 is the second line of C folio 24 verso. V7 then inserts into laisse 92D (after V7 1498) 11 lines rhyming in -ors closely corresponding to C laisse 92, which has 10 lines imperfectly rhymed in -or (C 1422–1431). The V7 scribe modifies the source text slightly and improves on the rhyme, but still does not perfect it, as may be seen in Appendix A to

 Duggan (2005, 38) draws attention to the length of the two assonanced passages where C differs from V7: 325 and 335 lines, contained within the third and the seventh quire respectively. (It seems preferable to make the first of these slightly longer, 336 lines, by incorporating laisse 73, but to change the second to 314 lines, ending with C 3481, or possibly 332 lines, ending with C 3499.) Since this close match in length is unlikely to be coincidental, Duggan suggests that C turned to an assonanced source with 21 lines to a single-column page (336 lines to an 8-folio quire) to make good material lacunæ in his rhymed model. One might add that any jongleur/trouvère converting assonance to rhyme would make sure that he kept to a sustainable recital-length.

The Relationship between C and the other Texts

21

the Duggan edition. This is V7 laisse 92A (11 lines), which Duggan correctly identifies as included through scribal error and removes from the Critical Text. The crucial evidence lies in the 2-folio backwards skip, from C 1553 (the first line on folio 24 verso) to C 1422 (the second line of folio 22 verso). In amalgamating two versions of the same laisse, the V7 scribe had apparently been consulting two different pages of his source; but this means that both versions existed in the CV7 model, which V7 was correcting because, in a library copy, the repetition became redundant. Some distraction had clearly occurred, for the transition is from the first line on one page back to the second line of the other. Moreover, V7 had no need to consult a different source at this point, since the rhymed version was in the CV7 model. After completing this laisse, the copyist returned to line 12 of the correct folio, copied out the final 6 lines of V7 laisse 92D (C 102) and then left several lines blank in acknowledgement of his error. However, before doing so, his eye must have been drawn to the next laisse (equivalent to C 93, the Sonnez l’olifant laisse) on C folio 22 verso, for he later inserted it after V7 108 (although it is found in neither C nor P at that point in the text). This is the laisse, 108A, which Duggan places in his Appendix A as a scribal error. Again, this indicates that the CV7 shared model contained laisses found in the repetition in C; and further evidence of «borrowings» by V7 is given under Textual Evidence. The aim of V7 was undoubtedly to effect a smooth transition (as approved by Segre) where the CV7 source had unnecessary repetition, but the resulting scribal error undermines the Segre thesis of C’s recourse to an assonanced source in order to make good damage to the CV7 model and also contradicts his basic tenet that C is a later manuscript than V7. If C laisses 91–100 were in the CV7 model, then this, like P, incorporated assonanced or pseudo-rhymed laisses; C’s Italian scribe, whose French (as Duggan comments) «cannot have been very secure»,39 therefore made some attempt to achieve good rhymes (traïsor for traïson, ciglator for ciglaton) but inevitably left much unmodified. Rather than C’s having to turn to an assonanced source to make good damage to the model, it was V7 who turned to a more recently rhymed redaction to eliminate assonance and imperfect rhyme in the CV7 source. Further evidence of this is provided by the C text in the second cluster of assonance and is cited later, under Textual Evidence; but first it is interesting to note that, precisely in the end-of-quire situation where V7’s error in copying occurred, at C 1553, the first line of C folio 24 verso, and V7 1498, 7 lines before the bottom of V7 folio 24 recto, the P manuscript opens. Since P is written on  Cf. Duggan (2005, 36). C’s scribal errors, confused handling of verb-subject inversion, misreading of words confirm Duggan’s judgement. Throughout the poem, the C scribe copies his model blindly and uncritically.

22

Introduction

large format parchment, in two 48-line columns per page with 8 folios per quire, it lacks simply the first quire: 1536 lines. This figure is slightly closer to C’s 1553 lines than to V7’s 1498 lines at the point where all the texts agree; but P laisses 1–7 match V7 1498–1504 rather more closely than C 1553–1558. When V7 inserts laisse 108A into the fully-rhymed part of the first repetition, V7 1840 (the third line of V7 laisse 109) reads: Ja le vos ai deus autres fois rovee, indicating that the redactor or scribe had deliberately inserted 108A as a triplication device – although this already existed, V7 110 providing the third part of the sequence. In contrast, C 1891 reads: Je le vos ai autre foiée rovee, P 349: Que je vos ai hui autre fois.. The CP reading autre fois shows that both were aware of the earlier «Sound the horn» passage found in their texts, that is, that the repetition was an integral (and deliberate) part of their indivual or shared model. Here, Duggan emends V7 1840 to read …ai autre foie… It is instructive to examine the above situation in the light of the general structure of the text. Even in O, the earliest text of the Roland known today, it is striking how much care is taken over the structural balance of incidents (and within incidents) in the narrative; and this is equally true of the CV7 version. When one takes an overall view of the structure of C and V7 up to the end of the second cluster of assonance (C laisse 218, V7 210F/208D), the balance between the episodes and the quires is plain to see. The first cluster of assonance, ending with C 1533, marks the first climax of the «olifant» episode and the culminating point of the initial development of the plot. First the introductory events set the action in motion: Blancandrin’s embassy to Charlemagne, Ganelon’s advice to treat with Marsile, his fury on being nominated by Roland as envoy, the ominous portents and his departure for Saragossa (ending at about C 511, V7 522, at the end of laisse 31, after roughly a quire of text); then the discussion between Ganelon and Blancandrin en route, leading to his treachery at Marsile’s court (with discrepancy between C 952–973 and V7 972–1000), his return (C 985, V7 1012) and false report (C 1023, V7 1051 ending the quire); and after this the nomination of Roland to the rearguard, the departure of Charlemagne and the main army and the arming of the Saracen peers, followed by Oliver’s urging that the olifant be sounded and Roland’s refusal. Up to here, in each succeeding quire (just over 500 lines) of text, there is a mounting dramatic tension – but the repetition in rhyme takes another quire (C 1534–2070) to rise from a less intense start with prayers and the arming of the French peers to the same peak of tension, which then leads in to the sequence of great battles, initially going in the French favour, then against them, until Roland is obliged to sound the horn. Thus quires 4–6 represent three main sub-episodes: C laisses 101–129 expand laisses 91–100, with the text of laisse 129 completing that of laisse 100 (a single laisse in O 91 and V4 87); quire 5 recounts the French success in battle until Margariz’

The Relationship between C and the other Texts

23

escape brings into play Marsile’s fresh armies;40 then quire 6 recounts the decimation of the French and, when all is already lost, the sounding of the olifant and Charles’ hasty return. The redactor of quires 4–6 no doubt thought that, in quire 4, he was improving on the earlier version (as found in O) by inserting some 300 lines on the arming of each French peer, to match the earlier arming of their Saracen counterparts. He did not appreciate the skill of the original author in maintaining the pace of the narrative by combining the description of each French peer with his individual success in killing the Saracen warriors – a success the redactor then recounts separately in quire 5. In quire 4, having lost the dramatic tension by this long description, the redactor producing the CV7P model therefore deliberately repeats in modified, smoothly rhymed laisses the «Sound the horn» urgings of Oliver. This leads straight into the battle sequence. c. The Second «Cluster of Assonance»: C Laisses 197–218, V7 187–208D The second «cluster of assonance» provides further structural evidence that C is the earlier version and firmly based on the CV7 exemplar, whereas V7, the slightly later manuscript, modifies the exemplar by reference to a different, fully rhymed redaction. The sounding of the horn is the culmination of the events in quires 4–6, just as the events in quires 1–3 culminate in C in Oliver’s vain pressure on Roland to sound the horn. C laisse 190 (V7 180D) recounts the first blast, provoking a minimal 2-line comment from Ganelon; C 191 (V7 181D) then records the second blast, whose force causes an internal hæmorrhage (ironically the only serious wound that Roland suffers, caused by his failure to warn Charlemagne before matters became desperate) – the description of the effects of this horn-blast being delayed to C 192 (V7 182D), 31 lines which end in a scathing 18-line speech by Ganelon; but it is only in C 193 (V7 183D) that the third great blast decides Charlemagne to turn back; and C 194–196 (V7 184– 186) describes this return. After this, a mainly assonanced passage which echoes V7 183–186D is inserted as C 197–201. The apparently extra laisse in C is due to C’s scribal error in separating laisse 198 (14 lines) from the 4-line 199 with the same assonance – but there is also discrepancy here, for C laisse 200 begins at the seventh line of V7 185. C 200 and V7 185 share the same -on rhyme.

 Another «end of episode» point at which there is discrepancy between the rhymed versions.

24

Introduction

The repetition begins with C 3168, the first line of laisse 197, and the partlyassonanced passage which follows comprises laisses 202–218, C 3217–3481, a total of 265 lines. If one counts in the five «repetition» laisses, then C 197–218 total 314 lines. In contrast, in the equivalent V7 laisses 187–208D, V7 3132–3583 total 452 lines. This requires nearly a full quaternion of parchment, since V7’s format of 30 lines per page gives a total of 480 lines per quire – but at line 3131, V7 has already filled 2 folios and 13 lines of his seventh quire: 133 lines already used. The V7 copyist may have become aware of this problem at the foot of folio 50 verso, where after V7 3118 he inserts (possibly from his new fullyrhymed model) the two lines on Alexander the Great which are found in no other version (and which interrupt the flow of the allusion to the assassins of Caesar). These two lines, excised by Duggan, form the last line of 50 verso and the first line of 51 recto (the fifth page of V7’s seventh quire). There are 452 lines to the new version to be inserted,41 but only 347 lines still available in this quire (480 minus 133). In order to make up the shortfall of 105 lines, V7 simply inserts an extra sheet of parchment, i.e. 2 folios, giving an extra 120 lines. For the scribe (and the scriptorium) it was easier and cheaper to insert a 2-folio spread into quire 7 (particularly if one could use a slightly cropped, perhaps previously rejected sheet, as found in V7) than to add an extra (eighteenth) quire at the end of the poem, where only four pages would be used – and presumably the price had already been agreed, based on 17 quires. Thanks to this insertion, V7 ends on folio 138 verso, the final page of the quire – and on the seventeenth line of that page. The extra pages permit V7 to insert 196D, 199D, 205D and 207D, 4 laisses unique to V7, plus the extra 8 lines into 188D on the Death of Blanzardin and later a further 33 lines in respect of V7 374– 375, not found in any other version. Just as V7 had gained an extra 60 lines in quire 2 by increasing the number of lines per page, here he gains over 100 lines by inserting one sheet of parchment. That V7 is here turning from the CV7 shared model used in most of the text to a «new» rhymed version is confirmed by the fact that, in laisses 197– 218, C has a partly assonanced passage (including C’s repetition) which totals 314 lines – a passage which would therefore have fitted easily into the 347 lines remaining of V7 quire 7, without any additional folios. That C here tallies with the CV7 model is also implied by the page layout adopted in quire 7. C’s

 V7’s «new» version had not only variant readings and expanded laisses, but also additional laisses (196, 198, 199, 205, 207) of which only 198 has an equivalent (in P 111, T 100, L 71 but not in COV4). The other 4 are unique to V7.

The Relationship between C and the other Texts

25

unusual 32 lines per page for this quire, in a passage which begins with the first line (3168) of laisse 197 and of folio 49 recto, suggests that, finding himself faced at this point with the CV7 model in a 32-line format, C adopted that format for one quire to facilitate copying. It is highly unlikely that, in either passage, the decision to insert repetitions was made by the C copyist, an Italian with imperfect knowledge of French42 who, even on folio 1 recto, was content to copy blindly the badly corrupted final lines of the page, lacking the skill, experience or confidence to excise the problem as V7 does and substitute a simpler line. It must stem from the redactor of quires 4–6, who was producing a standard length, fully rhymed episode for reading aloud or recital – a three-quire passage suitable for «slotting in» in place of the old assonanced episode. Segre and Duggan both suggest that a copy of an assonanced version was available in the scriptorium where the C and V7 manuscripts were produced. According to Segre, C’s repetitions occurred because the copyist, having turned to the assonanced source at laisse 73, persisted too long with it in error. This argument could apply in the first instance, where the partly-assonanced passage, laisses 73–100, is followed by ten laisses (91–100) which are expanded and repeated in rhyme in laisses 101–129; it cannot be valid in the second repetition, where the five assonanced laisses (197–201) immediately follow and repeat the rhymed version and where the partly-assonanced passage then runs from laisse 202 to 218. The insertion of laisses 197–201 must be deliberate. Duggan (2005, 36) refers to the laisses in both passages which «show signs of assonance» but sees the redactor as attempting to adjust these to the versification of his principal, rhymed, exemplar – by using «synonyms or equivalent phrases and words… to produce an adequate rhyme». It is very difficult to square the assessment of C’s poor knowledge of French with any argument that the copyist modified his assonanced model. The superficial modification of the last word of a line (producing Segre’s pseudo-rhyme classification) is much more his standard. In contrast V7, with a much better command of French, would either smooth the faulty rhyme himself or, given the chance, use a separate «newly improved» model (produced by a French redactor) to obtain a smooth rhyme. On one occasion (C laisse 76 and V7 laisses 76 and 77D) this involved a complete change of rhyme.  Duggan (2005, 36) comments that the Italian scribe’s French «cannot have been very secure». This is borne out by his scribal errors, confused handling of verb-subject inversion, misreading of words, etc. Throughout the poem, the C scribe copies his model uncritically; V7 emends most of C’s alexandrines to decasyllabic metre, corrects grammar and makes sure that the text is comprehensible.

26

Introduction

The Textual Evidence a. The First «Cluster of Assonance», C lines 1197–1533, laisses 73–100 inclusive For Segre, this «assonanced passage» begins with C laisse 74 – but in fact the first laisse with imperfect rhyme is C 73, and Beretta confirms this in defining the limits of his Sezioni in his comparison of the rhymed manuscripts of the Roland.43 C 73’s first line, 1197, has a slight problem with the line-ending, reading tenebros (the correct graphy for the adjective) despite the -or rhyme, whereas V7 1227 reads tenebror (with incorrect graphy) and T laisse 7, 89 solves the problem with de tenebrour, using the substantive. C 73 is then practically identical to V7 laisse 74D until the last line (C 1207) which V7 changes to 4 lines (1237–1240) so as to avoid the forced rhyme nevor. V7 1237 is the first line of folio 20 recto (a useful point at which to switch to a different source) and 2 of V7’s 4 lines are matched by T 98–99 (though T 100 is not found elsewhere). Throughout this passage, T (which frequently matches V7 closely) is the only other rhymed version available44 as a «control copy» (P and L being still acephalous) but its validity is questionable because T is so much later a manuscript and its readings potentially contaminated from other sources; its idiosyncratic version sometimes abridges the V7 text, sometimes expands it with new laisses. In the network of schemata developed by Beretta in his study of the rhymed versions of the Roland, T is consistently shown, despite its late (fifteenth century) date, as representing the first «branch» dividing off, subsequent to δ, from a version from which CV7PL (in Sezioni 1 and 2) and, via η, PL only (in Sezioni 3 and 4) derive. In the latter two sections, V7 and C are shown as representing a separate branch, in Sezione 3 via κ from δ and in Sezione 4 via κ from λ, but no account is taken of the close V7 = T correspondence found in specific circumstances in these passages and no link between them is shown in the stemma. Although the detailed Beretta analysis draws attention to individual lines and the discrepancies between the versions, a serious defect is its failure to see these lines in the context of the laisse (or take into account the effect of conversion of assonance to rhyme).  Cf. Beretta (2001, 42–134) for C’s second assonanced passage and the chapter on Le lasse assonanzate nei mss. C, V7, P (2001, 135–160) for both C 73–123 and C 200–214. Beretta’s Sezione 1a contains C laisses 99/128–165, Sezione 2a C167–199, Sezione 3a C 195–214 and Sezione 4a C 215–250 (using Beretta’s rather confusing laisse numbers). This study is based on the Segre thesis involving the earlier but «lost» sources δ, β, γ and λ.  T laisse 1, line 1 corresponds to C 1093, the first line of laisse 66 (O 61) though Duggan (2005, 476) comments in his note to V7 1121 that «T’s laisse is dissimilar in wording» to C’s; V7 lacks this laisse, but Duggan inserts C 66 as V7 laisse 67. T laisses 1–6 thus correspond to C 66–72, V7 67–73, since T 3 is equivalent to two laisses in the other versions (CV7OV4).

The Relationship between C and the other Texts

27

In terms of the Segre thesis, the question is whether C, the later manuscript in Segre’s view, turns to an assonanced source to make good damage to the CV7 model and then modifies some endings – or whether V7, slightly later in date than C, continues to copy the CV7 model (partly assonanced in this first and the later passage) but improves the rhyme and converts the assonance by reference to a newly rhymed source (attested to by T). The term «assonanced passage» is rather a misnomer: in each cluster a few laisses are fully assonanced but many are at least quasi-rhymed and some, depending on contemporary pronunciation and graphies, are rhymed. The latter is particularly true of laisses whose rhyme involves a final or penultimate r: -ee and -ere, -is and -irs. In the first cluster of assonance, C laisses 74 and 75, 86 and 87 have been divided in error by the copyist, misled by the 3-line -aigne rhyme in the first case, the 4-line -ent ending in the second. They are therefore re-united and treated as single laisses in this analysis (but not in the main text), although the graphies 74–75 and 86–87 are used. Thus there are just 6 indisputably assonanced laisses in this passage: C 74–75, 82, 83 and 86–87, plus 95 and 96 in the repeated section. The main part of C’s cluster of assonance, laisses 73– 90 equivalent to V7 74–90D, will be examined first. This examination is complicated by Segre’s inconsistency in the classification of pseudo-rhymed laisses. In his Table I (Segre 1974, 163), Segre classifies V7 74–90 all as smoothly rhymed, but C laisses 76–81, 84, 85 and 88–93 as pseudo-rhymed (the other four being unquestionably assonanced). In the main, this is accurate, but it is difficult to see why C 78 counts as pseudorhymed when V7 79D with the same endings is rhymed; and the same problem arises with C 80, V7 81D and later C 93, V7 108A. C’s passage, laisses 73–90, contains just four indisputably assonanced laisses. The main features of their comparison with O, V7 and T are as follows: (i) C 74–75, O 67 have –a.(.nasal)…e assonance and closely match; C 74 + 75 total 15 lines of which just 5 (already ending in -ance in O) are found in V7 and 4 in T; V7 75D (18 lines) and T 8 (19 lines) both rhyme in -ance, matching closely in 10 lines and approximately in a further 4 lines. (ii) C 82, with -o…e assonance, very closely matches O 74; but V7 83D is completely recast to obtain the -ez rhyme. T laisse 15, 11 lines ending in -ez, omits 5 of V7’s 16 lines and matches V7 closely in 9 lines. Nonetheless, V7 1353–1354, T 197–198 vary only in the rhyme-word from C 1300–1301: Plus vaut Mahons qe Seint Piere de Rome / S’a lui serviez, l’honor dou canp avrome and several first hemistichs match in CV7T. (iii) C 83, 8 lines with -e…e assonance, becomes 84D in V7, 10 lines with -is rhyme (again a complete change), and T 16, 10 lines rhyming in -is, has 6 lines matching V7 plus 4 an approximation. (iv) In C 86–87, 86 is

28

Introduction

assonanced in -ei…e / ein…e approximating to O 78’s -ei…e assonance, and ends with C87 (4 lines in -erent). In V7 laisse 87D, C’s 18 lines become 21 lines rhyming in -er (20 lines in T 19 with the same -er rhyme but matching V7 in only 10). In respect of C 74–75, where O has 9 different line-endings, V7 and T have used a rhyme occurring within the assonanced pattern, and V7 1246–1248, 1250, 1254–1255 have an almost exact match with -ance lines in C and O. In respect of the other assonanced laisses, all characterized by the great variety of line-endings in O, V7 and T turn to the same, completely different rhyme; and they particularly closely match in V7 83D, T15 and V7 84D, T 16, slightly less closely in the other two laisses. Whereas Segre sees this as C accessing an assonanced source to make good damage in the CV7 model, it is more logical to take this as V7 using a new, fully-rhymed version (known later to T) as a source when «correcting» the CV7 model’s assonanced laisses to rhyme. Where the CV7 source already had rhyme or quasi-rhyme (ie in C laisses 73, 77, 78, 80, 81, 84, 85 and 90, written in tandem with V7 by Foerster because the two versions so closely matched in rhyme, text-detail and length of laisse), V7’s readings were usually closer to C’s than to T’s; but, where C’s forced rhymes (or mis-readings) needed to be corrected, V7 turned again to his new source to obtain a rhyme or a more felicitous reading (often shared with T). This may be seen in V7 1237–1240, 1287, 1291 (approximately), 1305, 1310, 1343 and 1398–1399 amongst others. Thus, where C 1207 ends: C’as porz d’Espaigne a laisié son nevor, V7 1237 reads: Por son nevou, Rollant le poigneor and a further 3 lines, of which V7 1237–1238 match T 98–99 but T100 is not found in V7; when C 1242 reads: Son oncle apelle por mout bel contenant, V7 1287 and T both read: Son oncle apelle, si li dist en riant. Where C and V7 have minor variants, though the same rhyme, T matches them in rhyme but supports V7’s variants. For example, in C 78, C 1264: Tuit sunt jugié li .xii. conpegnon becomes V7 1310: Sera par nos misse a destruction, identical to T 160; in C 80, C 1274: Puis qe il est sor son chival corsier is replaced in V7 1325 = T 174: Puis que il doit sor son cheval monter; in C 81, C 1292 ends:…de mort li faz estregne where V7 1343 reads:…qu’il tienent a chastaigne, matched by T 190. This consistent pattern may also be seen in C84, 85 and 90. It is noticeable that, in C laisse 84, V7 86D, Duggan prefers some words and hemistichs in C as the base text, rather than V7’s variants (cf. V7 1377, 1380, 1381, 1382, 1386). This implies that C is the earlier version (based on the CV7 model) and V7 the modifier. Of the remaining four laisses within this passage, two (C 79 and 89) share the same rhyme as V7 but the others (76 and 88) have a different rhyme. C 79, 7 lines rhyming in -art with one mat (perhaps indicating a «silent r»), omits 2 vital lines found in O laisse 71 (9 lines): Devant Marsilie cil s’escriet mult halt: /

The Relationship between C and the other Texts

29

Jo cunduirai mun cors en Rencesvals, O 891–892. After C 1267, O 887, both C and O lack V7 1315: Je sui le tierz, or eslisez le quart, matched by T 165. C presents a confused and partly corrupt sequence of lines where V7 80D has 12 lines in -art (with jalant, line 1320 in error for galart: gaillart in T laisse 12, line 170). T here has 11 lines, identical to V7 except for variants in 3 hemistichs. The omissions and confusion in C 79 may well derive from the CV7 source and explain V7’s resort to a rhymed model. C laisse 89, 11 lines rhyming in -or, closely matches O laisse 80, 11 lines with –u- assonance, and V4 78, also 11 lines but with 8 in -or, 3 in -on (apparently an interim stage of conversion). C, however, omits O 1022 (= V4 957) with its reference to hauberks and helmets and inserts C 1396: Desoz paiens tant destrier coreor, almost identical to V7 1457, T 363 – probably a CV7 model reading. V7 laisse 89D (14 lines) and T 23 (17 lines) have the same -or rhyme. In the first line both V7 (1451) and T (354) reduce C’s alexandrine to a decasyllable; V7 and T add 2 matching lines after C 1393; and both substitute ..jel tieng a traitor (T: je le tien pour t.) for C 1398: en a fait traisor (cf. V4 959: oit faita la traisor, whereas O 1024 has Guenes le sout, li fel, li traitur). Here, V7 and T are once again clearly related, though a close match is found in only 7 lines – and V7 changes the order of COV4T in lines 1452–1454. This is the second time that C has used fait traisor in a forced rhyme (cf. C 1226). Where C and V7 have different rhymes, C laisse 76 has -or quasi-rhyme but V7 has 2 laisses: 76 rhyming smoothly in -ons, 77 in -on. C’s 17 lines become 1245 + 15 = 27 lines in V7, a considerable expansion; but T laisse 9 has 17 lines rhyming in -on, with 5 lines corresponding to V7 76 (but only 3 lines matching closely) and 12 lines matching V7 77 (closely in 10 lines): thus a close match in 13 out of 17 lines. O laisse 68 (19 lines and -u- assonance) has an even mix of -ur/-urs and -un/-uns endings, but opens with 3 lines in -urt, -ur, -ur – tempting a later scribe to adopt an -or rhyme he could not sustain. Faced with this unsatisfactory situation in his model, V7 or his rhymed source turned to the -uns/-un pattern, the easier alternative. Even so, two laisses were needed, one with each ending; it was only later that T’s model reduced these to a single laisse again. Segre cites this laisse as showing C turning to an assonanced model and converting it superficially to rhyme (traisun > traisor, ciglatuns > ciglator)46 – but it is in fact the only clear example of such a conversion in this passage. He argues that the innovation is C’s, not V7’s because (i) in this laisse (Segre, 1974, II, 150s.) C presenta una tradizione diversa, quella

 Duggan has an 11-line laisse here, excising the hypermetric alexandrine after V7 1265.  Segre (1974, 149). In the case of a fet traisor, C 1226, one wonders whether this could have been taken by the scribe to mean «acted the traitor»/«played traitor».

30

Introduction

di β-γ, cioè ha operato una contaminazione (V7 non poteva indivinare i limiti della parte additizia, ed eliminarla con taglio esattissimo); and that (ii) i margini dell’incastro in C non combacino perfettamente. V7’s use of lines and laisses from the C 91–100 passage in the rhymed version (C 101–129) gives good reason to believe that both passages were found in the CV7 model. If V7 is a slightly later manuscript than C, the scribe had no need to «guess» the size of the insertion – which V7 does not eliminate smoothly and accurately. V7’s two laisses totalling 27 lines represent the type of redaction produced when a tricky conversion demanded extra lines. T 9 much later pares this down to a single laisse with -on rhyme. C used the difficult -or rhyme either because the CV7 model here had an assonanced laisse which C was converting to rhyme, or because the CV7 model had already converted this laisse to (an imperfect) -or rhyme. In view of the C copyist’s poor knowledge of French, the second alternative is the more likely. If C had had the -on/-ons version available to V7, he would have used it (and it should not be forgotten that it is V7, not C, which modifies not only readings but also page-length and (in the later passage) number of folios to accommodate his text). Finally, C laisse 88, 21 lines, very closely matches O laisse 79, 23 lines assonanced in -ei-; V4 77 has -és rhyme. C rhymes in -is (with one -ins, two -eis) which is very similar to the -eis of V7 laisse 88D, 19 lines (where V7 has imperfect rhyme sors in line 1438, neis for nos in 1450). T laisse 22, 20 lines, rhymes in -ois, -oys with one -oirs, one -eys plus -aïs and -ais. V7 matches T in 10 lines, but C in 8 lines (T and C matching in 6 lines); but, apart from C 1385 = O 1011 = T351, neither C nor V7 has a full-line match with T in the final lines of its laisse, 347–353. V7 1445 changes the word-order in C 1382 so as to substitute avreis for Saracins at the end of the line, corresponding more closely to O (though T 346 has a much clumsier inversion ending in Francoys). In a line not found in C, Duggan substitutes T 338’s reading for escuz a or, V7 1435. In general, both C and V7 are very close to O, V7 sometimes closer than C: V7 1433 matches O 995 exactly, 1435 matches O 998 (omitted by C), but C 1374 matches O 999, which V7 lacks; V7 1439 paraphrases O 1002 but avoids C 1377’s hypermetric paraphrase. It is difficult to draw firm conclusions from this situation, but it reinforces the impression of V7 as the modifier producing the later version, whereas C copied his source faithfully but uncritically. Looking at the first cluster of assonance as a whole, there is strong evidence of a link between V7 and T in the laisses where C is assonanced – and also where C and V7 had a different rhyme or the same rhyme but very poor textual match. This suggests that V7 was accessing a different rhymed source (known later to T) in order to «improve on» the CV7 model’s readings. Where C is rhymed or quasi-rhymed and closely matches V7, it seems that the new

The Relationship between C and the other Texts

31

rhymed model was used by V7 only occasionally: to correct any forced or faulty rhymes and any infelicities of expression in the not-always-perfectlyrhymed CV7 source (caught at the interim stage of conversion to rhyme). When one looks at the detail of C’s first repeated passage alongside the V7 text, it is also difficult to reconcile the evidence with the Segre theory. If V7 (seen as the earlier manuscript) simply copied the CV7 rhymed model (still undamaged and fully legible) throughout laisses 74–90 D, he had every reason to proceed with the same exemplar in the subsequent laisses, which constitute a fully rhymed version, shared with C and P and running from C laisses 101 to 196 and P laisses 1 to 100. There would be no reason for the V7 scribe to turn to the partly-assonanced version of the next few laisses (C 91–100), which constitute the assonanced part of C’s first repetition. Nor would there be circumstances in which V7 could copy two C laisses out of context, deliberately moving one of them. In a passage where the CV7 reading is confirmed by P as «control copy», it makes no sense for V7 to select lines and hemistichs from C’s part-assonanced source and substitute them for the CP lesson. Yet V7 does precisely this – and the Duggan edition corrects it without comment. The first textual evidence of V7’s access to C’s part-assonanced source is found in V7 laisse 91D. As stated above, this is indeed based substantially on C 101, but it also incorporates elements of C 91: V F, D V  V –  V 

V  V  V  V  V 

V 

C  = C  = C – , = C  (droites > roides in C) NOT IN C  = C  = C 

C  = almost exactly C  (Je > Or in line ) paraphrased in C , ,  but C  has droites les astes… with change of word order and paraphrase

– but V = C  almost exactly (ot > vit) = C  = C  BUT both C lines change word-order in first hemistich = C  not in C : C  is quite different = in first hemi- but  = C  exactly stich only, C , NOT IN C  – but V = st hemistich C , nd hemistich C  (though V’s Respont Rollant in  really differs little from C : Ce dist Rollant).

The V7 copyist apparently seeks to amalgamate C’s two laisses, ending with C 1421, the final line of C 91, a line not found in O, V4, T or C 101. P and L are

32

Introduction

still acephalous at this point, but T laisse 25 (9 lines rhyming in -uz, -us) matches V7 91D more closely than C 91 or C 101. Having copied this line, the first of folio 22 verso, V7 next proceeds, quite correctly, to copy from the shared CV7 rhymed model laisse 92D (= C laisse 102 with very minor modifications) for the first 11 lines only (lines 1543–1553 in C, 1553 being the first line of C folio 24 verso). At this point (5 lines from the foot of folio 23 verso in V7), the V7 copyist’s work was apparently interrupted for a time for, when he returned to it, he continued with C laisse 92, whose first line (1422) is the second line of folio 22 verso. He had thus accidentally turned back two folios (132 lines) to the correct line but the wrong page of his exemplar, to a laisse whose rhyme (-ors in V7, but -or in C 92) is quite different from the -er rhyme in C 102. In the inserted 92A (11 lines where C has 10 lines) the slight variation in rhyme (C -or, V7 -ors) affects the case-ending of final nouns and thus the structure of some lines. V7’s first 2 lines match C 1422–1423 and its final 3 lines match C 1428–1430, although it omits C’s last line, 1431. V7’s line 3 almost matches C 1424, but modifies the second hemistich to obtain the -ors rhyme – and thus needs to expand and modify C 1425; but line 4 seems to be based on the 3rd line of C’s next laisse (C 1433, laisse 93) and line 5 (not found in C 92) on C 1434. V7’s 6th line is not found in C laisse 92 nor in 93, though lines beginning Secorra nos… with different second hemistichs are found in the later (rhymed) version (lines 1877 and 1893); Segre (1974, 171) points out that this line matches V4 990 (a later manuscript). It may thus have been found in the CV7 model, though it does not occur in O laisse 83. V7’s 7th line is initially identical to C 1435, but is then modified for the rhyme. The eighth line matches V4 993 and paraphrases C 1437. In this laisse, it is possible that V7 simply used standard formulæ suited to the -ors rhyme to replace C 1424–1427. Nevertheless, there are forced rhymes in lines 2 (avores), 9 (Durendors), 4 and 11 (both reading pors in the oblique case).47 The exact two-folio interval which is the source of this error in V7 points to the presence of both versions (first the partly-assonanced, then the fully rhymed) in the same exemplar (with 33 lines per page); if the assonanced source and rhymed model were separate entities, there would be no reason to leave the first open once the earlier laisses (C 91 and 101) had been combined in V7 – and less chance of a matching lay-out on the two pages, so no source of confusion. When both versions were found in one exemplar, V7 would need to refer back and forth, within the quire, in order to amalgamate them. The

 Intriguingly, the third, fourth and sixth lines of laisse 92A appear later in identical form as lines 2999–3001, the first lines of V7 laisse 179: 2999 is the final line of folio 48 verso in V7 – and matches C 3035, the first line of laisse 189 and last line of C folio 46 verso.

The Relationship between C and the other Texts

33

V7 scribe was led into error by his wish not only to eliminate the assonanced laisses, but also, where the «assonanced source» had rhymed or near-rhymed laisses, to incorporate the best lines of those laisses into his text. The precise 2-folio (132 line) interval could possibly imply that V7 was here using C itself as a model to be corrected – perhaps on the purchaser’s instructions – but it seems more probable that he had simply been told the position of the repetition to be excised. It is also possible that the CV7 source here had 33 lines per page.48 It is striking that V7 appears unaware of the change in rhyme between lines 1488–1498 (-er) and laisse 92A (-ors), but he had a limited focus – and his attention was apparently distracted for a moment. V7 had completed a sentence by copying line 1498 = C 1553, so a new laisse was quite acceptable; and he knew the position of the next line to be written: the second line of the source page. Moreover, when V7 returns to the correct laisse (92D, the equivalent of C 102) to complete the last six lines, his line 1500: Se Deus en France me laisse retorner is not found in C, although V7 1499 and 1501–1504 are identical to C 1554–1558. It is, however, exactly matched by P line 3 – proving that it was found in the CV7P source. The omission is therefore C’s scribal error. Duggan recognizes the insertion of V7’s version of C 92 into V7 laisse 92D, but simply excises it as having no part in the rhymed CV7 model he is recreating; he classifies its presence as V7’s scribal error and places it in his Appendix A, numbering it 92A. This is fully justifiable in terms of content, for this is the first «sound the horn» laisse and Oliver’s triple plea will not commence until C laisse 118 (V7 108) in the rhymed repetition. The relegation of laisse 92A to an appendix does, however, mask a highly significant feature of V7. This is also true of V7 laisse 108A, again placed in Appendix A by Duggan. Having copied C laisse 92, V7’s attention had been drawn to the smoothly  Both the 32- and the 33-line format would be a useful «scriptorium size», facilitating calculations and costing in respect of larger page-formats: the large in-quarto P manuscript with its 96 lines in two columns = 3 x 32 and V4, in-folio with two columns of 50 lines = 3 x 33 (+1). The acephalous P lacks its first quire, i.e. 8 folios with 2 columns of 48 lines per page, a total of 1536 lines. Its first line is Ganes li cuens nos a fait moult pener, identical to C 1553 but varying from V7 1498 in minor detail (cons > fels, tense changes to future). This is found in laisse 102 of C’s rhymed passage in the first repetition (C 101–129) and part of the longer passage (C 101–196) in which C, V7 and P (and later T and L) derive from the same source. There is no means of knowing whether P’s lost quire of text closely matched C and V7, but its 1536 lines approximate more closely to C’s 1552 (a difference of only 16 lines) than to V7 (1497 lines, a difference of 39 lines). Did P also insert repetitions of single laisses, as it did later? T laisses 25–27 give no help: 25 matches C 101 to some degree, but 26–27 correspond to C 118– 119, a considerable gap.

34

Introduction

rhymed C 93, the distinctive «car sonez l’olifant» laisse. It is found in almost rhe same form in O (12 lines with 6 x -ant ending, 3 x -ent and 1 each of -an, -anc and -enz) and may well have been well-known and popular. Except for C 1434’s tornera esrant becoming sera retornant in V7, the C and V7 versions (12 lines, -ant rhyme) are identical, even to zant and çant. V7 must have deliberately transferred this laisse from the first «assonanced passage» to the rhymed repetition, inserting it after C laisse 118. It is not found in T, which limits the «sound the horn» sequence to laisse 26 (22 lines) and abandons laisse 27 after 9 lines as an unnecessary repetition (leaving a line blank as acknowledgement of scribal error, then proceeding to the seventh line of C 130, V7 120D). Nor is it found in the rhymed version of the first repetition (C laisses 101–129, P 1– 28) although both they and V7 derive from the same shared model in this section of the text. It is noticeable that, in each case, the C laisse involved was rhymed/pseudo-rhymed, not assonanced. Both cases show that it is V7 which has modified the text (the first time through scribal error, the second by deliberate choice). V7 1840 in laisse 109 adds a further proof of this. Once again, Duggan modifies this line on the model of P 349, laisse 18: Que je vos ai hui autre fois rouvee, and…autre foiee rovee in C 1891, laisse 119. Having inserted C 93 immediately before this, in attempted stylised triplication, V7 stresses this fact by modifying his line to: ai deus autres foiz r. If laisse 108A is moved to the appendix, Oliver has not made the plea twice already. Duggan therefore corrects V7’s reading on the CP model, but removes vital evidence. However, the really significant factor is that V7 inserted these laisses at all; he was apparently still copying from the CV7 shared model here and had no need to refer to any other source. After laisse 108A = C 93, V7 continues quite correctly with laisse 93D = C 103 in the rhymed version of the repetition. In other words, it appears that here the «assonanced model» was not a separate entity available in the scriptorium but just as much an integral part of the CV7 model as was the subsequent «rhymed repetition» deriving from the CV7P model. V7 cannot be copying C, since V7 1500 exactly matches P 3 but is omitted by C. Aware that the repetition in his model now served no useful purpose, V7 tried to combine the two versions. There is also evidence in the text of V7 laisses 113–115D, where one finds quotations from C laisses 97–99, three quasi-rhymed laisses in the assonanced part of the repetition,49 replacing the agreed readings of the rhymed version, C 123–126 = P 22–24. Some of V7’s borrowings are of a minor nature: in V7

 No words or lines are selected by V7 from C 94 with its flawed mix of -ee and -ere endings or C 95–96 which, being assonanced, would not offer an acceptable variant.

The Relationship between C and the other Texts

35

laisse 113D, François line 1911 = C 1484 (and O 1112) where C 123 line 1963 reads A voiz… and P 420 A vois; and V7 1913 qe ci nos i lassa = C 1486 qi ici nos lassa (and V4 1030) where C 1965 has qi François nus laissa (= O 1114, P 422). More significant are V7 1912: amis, traiez vos ja = C 1485, where C 1964 reads e vos ne l direz ja and P 421 has mais ne le dire i a; and V7’s additional line, marked as [V+] in Foerster but excised and un-numbered in Duggan, reading Meillor vassal de moi ne la ara, which echoes meillor vasal ne l’avra, the variant second hemistich in C 1493, laisse 97, where the second hemistich of V7 1922 (dire poez i ja) roughly matches C 1973: dire poet qi l’avra and P 430: dire puet qui l’avra. Furthermore, V7 then adds 1923 D’un vassal noble ceste spee fu ja, exactly matching C 1494, where C 1974 reads Boen vasal ert icil qi l’avera and P 431: Iceste espee, vassaus hom la porta (both paraphrasing O 1123). In V7 114D, line 1929: As Crestiens devons estre aidiz exactly matches C 98, 1500, where C 124, 1980 reads Crestiente n’ait pas de nos mal diz and P 439: Crestienté n’ait de vos mauvais cris (cf. O 1129: Chrestientet aidez a sustenir); and 3 lines later V7 substitutes for C 1983: Garde chascun ses pechiez ait gehiz (matched by P 443) two lines, marked [V+] by Foerster but excised and unnumbered by Duggan: Clames vos coupes, si clamez Deu merciz / Asolvirai vos por vos armes garis, which exactly match C 98, 1503–1504: Clamez vos copes, si clamez Deu merciz / Asolvirai vos por vos armes gariz.50 Here, both V7 and C have the pseudo-rhyme garis/gariz, possibly representing the -irs form of the infinitive found elsewhere in C. The same pseudo-rhyme is found in V7 1939: Por penitance lor comande feriz, C 98, 1509: Por penitence lor comant a ferirs, a line not found in C laisse 124 but converted to two lines in P 449–450: Por penitence ferez sor Sarrazins / Qui ce ne croient que Deus fust surrexis. Here, O 1138 reads Par penitence les cumandet a ferir. Finally, there is just one borrowing by V7 in laisse 115D, again a line marked [V+] by Foerster and not found in C 125: V7 1941: Bien sunt ausols, quite de lor pechiez, identical to C laisse 99, 1511 and also O 1140 and P 452. In all these cases, Duggan simply replaces the V7 reading with the CP agreed version, without comment in the Notes. These are no doubt seen as occasions when V7 «refers to the model in assonance», but it is difficult to see why V7 should refer to a separate manuscript here, when he is copying the CV7P shared model. Only the presence of the assonanced passage within the CV7 exemplar, in the form of the repetition found in C, makes this a credible scenario. It also answers the Segre query as to how V7 could possibly know the exact size of the passage, so as to ensure an exact fit: the passage is there

 Duggan offers no discussion of the excision of these «additional lines» in the Notes.

36

Introduction

in his model – and the fit is not smooth and exact, but rather confused at times. The first repetition offers one final piece of evidence, showing how new episode-based redactions were dovetailed into the text. In O, laisse 91 has 18 lines whose -an- assonance comprises 11 x -ant, 4 x -ent, 1 each of -anz, -anc and -ains, greatly facilitating conversion to rhyme. In C laisse 100, rhymed throughout in -ant, this is reduced to 12 lines, of which eight match O very closely: after the first two lines, C 1522–1523, C inserts in error and out of context a group of four lines relating to Turpin and foreign to O, which interrupt the description of Roland. These begin Li arcivesqe de Deu les va seignant… so clearly they should follow Si compeignon apres li vont siguant, which provides the antecedent for les. They have therefore been moved to that position where they form the final lines of C laisse 100. This laisse is repeated as C 129, almost identical to V7 119D, both with 22 lines, also as P 28 with 23 lines and L laisse 1 with 30 lines – and in each case without the four «Turpin» lines. (The laisse is lacking in T, which resumes its narrative in the next laisse.) All four versions are rhymed, C smoothly in -anz, V7 smoothly in -ant, P less perfectly in -ans, -ant and L in -anz, -ant. It was therefore not any problem of versification which caused C’s source to abandon the assonanced model, but the fact that O’s (and V4’s) final lines lead directly into the battle, whereas in C’s model there is first the long hiatus of the first rhymed repetition. The four additional lines, found only in C, stand in place of O’s final 9 lines (1161–1169) which end with A cez paroles vunt les oz ajustant). They serve to round off the laisse, curtailed here to allow for laisses 101– 128. Thus, when the laisse is repeated, C 129 closely matches V7 119 D and P 28, all three inserting an additional introductory line. L laisse 1 is a paraphrase with occasional matching lines. Taking C 129 as representative of CV7P, after the initial line, C 2050–2052 match O 1152–1154 quite closely, but C 2053–2057 replace O 1155–1156 with five lines of very different description; then C 2058– 2070 correspond quite closely to O 1157–1169. There is thus an overlap in the correspondence between «assonanced» and «rhymed» versions, equivalent to O 1157 + 1159–1160 in the first, O 1157–1160 in the second. The CV7P version aims to avoid repeating the description of Roland’s arms already given in C 100 and O, but to complete the laisse truncated by C 100 in accordance with O. This means that the writer of the CV7P source was aware of the earlier version of the laisse in the «assonanced passage», just as the scribe of that earlier version was aware of the imminent «rhymed passage». The division of O laisse 91 between the two supposedly «duplicate laisses» was deliberate. This makes sense only if both assonanced and rhymed passages were in the

The Relationship between C and the other Texts

37

CV7 model. It thus reinforces the earlier evidence provided by V7 – and overturns the Segre theory of the «insertion» of the assonanced passage by C into the rhymed text.51 C’s assonanced passage in the first repetition is crucial to this argument because, solely at this point, two versions of C are available, plus P as a «control copy» of the «rhymed C» as well as the sometimes variant V7. In the preceding passage, C 73–90, there is only the one C version and V7’s intermittent but persistent match with (the much later) T seems to indicate reference to another (rhymed) source;52 there is no «control copy» to provide proof as to which copyist is modifying his model. The same is true, despite the multiplicity of versions available, in the cluster of assonanced laisses which follow C’s second repetition; and the repetition itself, C laisses 197–201, contains no signs of modification of the assonanced version (which closely corresponds to O) on the basis of the rhymed. It is C laisses 91–100 which provide compelling evidence that the repetitions and their associated clusters of assonanced laisses were found in the CV7 source. The rhymed version of C’s first repeated passage is a complete, integrated redaction contained in laisses 101–129 and filling a single quire. It is found in identical form in V7 and C (apart from the variants already noted) and also in P laisses 1–28. T laisse 25 corresponds to V7 91D but T lacks the ensuing 17 laisses, inserting laisse 26 = V7 108D and the first 9 lines of laisse 27, approximating to V7 109D; after leaving a gap, T completes laisse 27, T 422 corresponding to V7 2028 in V7 120. There is no need for the V7 copyist to refer to a source other than the rhymed model common to CV7 (and P). That he does so arises partly from a desire to effect a smooth transition (by amalgamating C 91 and C 101) whilst eliminating the assonanced passage of the first repetition, partly through scribal error (in inserting laisse 92A) and partly through deliberate selection of a laisse (108A) and individual lines which he judges superior. In every case, the laisse involved as source is rhymed or quasi-rhymed in C – and in every case, it is V7 which modifies his base text.  It also counters one of Segre’s proofs of his theory, namely that V7 non poteva indovinare i limiti della parte additizia (since there would be no need for guess-work) as well as his subsequent words, ed eliminarla con taglio esattissimo. Cf. Segre (1974, 150s. n. 1).  In this passage, the close correspondence between C and V7 in C laisses 73, 76, 77, 79, 80, 83, 84 and 89 is striking, but V7 has occasional variant lines or hemistichs which are matched in T. In fact, the fifteenth-century T consistently has a slightly closer match with V7 than with C and a particularly close match in rhyme with V7 where C’s rhyme differs; T and V7 also match closely in laisses where C is assonanced or differs greatly from V7. Detailed study of the relationship of C and V7 with O, V4 and T in the first cluster and with O and V4, on the one hand, and P T and L, on the other, in the second cluster, reveals that V7 used a separate, fully rhymed source in both cases – but not that V7 or that source was earlier in date than C.

38

Introduction

Looking at the first cluster of assonance as a whole, one finds that: C accords with V7 in two-thirds of the laisses, with only minor discrepancies, similar to the variants in the preceding 72 laisses; that, where C has assonanced laisses, V7’s variant readings are usually supported by T; that V7 and T always have the same rhyme and close concordance; and that C’s retention of full-blown assonance (the exception rather than the rule) arises from the difficulty of converting a difficult mix of endings to smooth rhyme, not from access to a different source. Where C 76 has 17 lines and a few glaringly imperfect -or rhymes, V7 needs two laisses with different rhymes (-ons and -on) and 27 lines, but closely matches O 68 where its assonanced lines end in -un/-uns and also in 4 first hemistichs. b. The Second «Cluster of Assonance», C laisses 197–218, lines 3168–3481, V7 laisses 189F–210F (187–208D), lines 3132–3583 In this case the repeated passage, which comes first, is much briefer: C 3168– 3216, laisses 197–201, repeat (mainly in assonance) the contents of the smoothly-rhymed C 3102–3167, laisses 193–196, and the matching V7 3066– 3131, laisses 183–186D. Thus 49 mainly assonanced lines in C repeat in 5 laisses the 66 preceding lines in 4 matching laisses in C and V7 (199 being separated from 198 in error but sharing the same assonance).53 C 200 is rhymed in -on like C195 and V7 185, but loses the visual rhyme through the graphy poing (although C often writes this pon). This laisse is paraphrased in CV7, with only C 3133, 3141, 3143 and 3145 found in the CV7 rhymed version; C’s Sanson becomes Bovon in CV7; but both C versions of this laisse and V7’s read cons or quns in error for cous. C laisses 197–199 and 201 are all assonanced. Once again, the repetition in C coincides with a crucial event in the narrative, occurring at the end of a major 3-quire episode – just where an interval in performance would have been necessary in earlier times. It also conforms to the pattern of consecutive assonanced and rhymed laisses found several times in P later in this passage, demonstrating the acceptability of such a mix. Another factor was that the older assonanced laisses were a piece of very effective writing, appreciated by the scribe and probably generally popular. Compared with C 193–196 (= V7 185–188F/183–186D), the C 197–201 text is briefer, pithier, clear and more dynamic.

 It has been noted under Structural Evidence (page 24) that V7 inserts 2 extra lines relating to Alexander the Great into laisse 186, after line 3118. These are not found in any other version; they interrupt the sense and the flow of narrative and are excised by Duggan, but have significance to the structure.

The Relationship between C and the other Texts

39

P, T and L have poor correlation with CV7 in the laisses preceding C’s repetition; in fact, P laisse 100 is imperfectly rhymed (-amp, -anc, where C 196 has -ant, -anz rhyme) and matches in at most 8 of its 25 lines. T 85–89 match C 193–196 in roughly half the lines – but L 55–58 is generally a paraphrase. There is no reliable «control copy». After the repetition a more complex situation obtains in laisses 202–218: (i) O lacks laisses equivalent to C 203, 206 and 216, but otherwise closely matches C, particularly where C has assonance but also where it has rhyme or pseudo-rhyme; (ii) V4 generally has a less close match and, as a later Gonzaga manuscript, may be contaminated; (iii) P, T and L are all now available as potential control copies; but P has a mix of rhyme and assonance, with consecutive repetition of individual laisses – and PTL do not present a consistent picture; (iv) V7 has 4 laisses (198F/196D, 201F/199D, 207F/205D and 209F/ 207D) not found in any other version;54 (v) C laisse 210 is equivalent to two laisses in all the other versions. C has just 5 assonanced laisses (C 204, 207, 211, 213, 218), 6 laisses where C and V7 have the same rhyme and closely match (C 202, 208, 212, 214–216), another 4 with identical rhyme but paraphrase (C 203, 205, 206, 217) and 2 with different rhymes (C 209, 210). It is at this point that V7 inserts a 10-folio quire55 because, although it has gained 49 line-spaces by excising C laisses 197–201, it has already reached the fourteenth line of folio 51 recto and still has to insert a 452-line rhymed passage (laisses 187– 208D) where C has only 265 lines (laisses 202–218); this is due partly to the 100 lines in V7’s extra laisses, partly to its expansion of the other laisses. C’s seventh quire of text does not end until C 3679, 6 lines before the end of the 48-line C laisse 226, but C is practically identical to V7 from C 219 on. In this passage, the presence of 4 laisses unique to V7 serves to underline the role of V7 as «agent of change», contrasting with C as the more passive copier; the themes and purpose of these laisses reinforce this idea. V7 laisse 198F/196D, 14 lines with -ent rhyme, relates the dying Oliver’s thoughts of Charlemagne and Aude, his prayer to God. This is an innovative laisse, deliberately inserted by V7 between two laisses corresponding closely to C 210, 20 lines with -uz rhyme: 197F/195D, 13 lines in -u, is almost identical to the first 9 lines of C 210, whilst 197D, 11 lines in -er, matches C’s first hemistichs in 6 lines. V7 198F/196D also foreshadows, in a minor key, V7 201F/199D (26 lines

 The rhymed V7 194D and 198D are marked as corresponding to P 110, 114, T 97, 100 and L 68, 71 only, but they also correspond to C 209 with -oi rhyme and C 211, assonanced in -a…e and to assonanced versions in P, O and V4. In this passage P has duplication via 4 pairs of rhymed and assonanced laisses.  As discussed under Structural Evidence, c, pages 23–25.

40

Introduction

with -ez rhyme) in which Roland thinks of Charlemagne, Aude and all he has lost; these two «unique» laisses thus prepare the ground for the events in Section C, the most recent and expansive redaction. In contrast, V7 207F/205D (27 lines) and 209F/207D (22 lines) recount the battles of Turpin, Gautier and Roland with the Saracens but are really non-essential; perhaps they were inserted simply because they were available. There is considerable discrepancy here between C and V7: just 6 rhymed or pseudo-rhymed laisses (C 202, 208, 212, 214, 215 and 216) where Foerster prints V7 as a subscript to C and these contain a considerable number of variant forms, with addition/omission of lines; of the remaining 6 laisses (discounting V7’s 4 «unique» laisses) just 3 (C 205, 206, 217) have the same rhyme as V7, C 203 has pseudo-rhyme, 209 has a completely different rhyme and 210 represents two differently-rhymed laisses in all the other versions. C has assonance (and closely matches O) in 5 laisses and poor match with V7; but V7’s relationship with PTL varies as set out below: (i) V7 191F/189D, 14 lines with -ez/-iez rhyme, paraphrases C 204, 16 lines with -e- assonance, matching C in only one line; but in C the first 5 lines end in -ee, the rest in -ien/-en, and the correspondence with O 141 (17 lines, -e- assonance) varies from exact match to paraphrase. P 103 (15 lines) is assonanced but has -iez ending in 7 lines; it paraphrases C and matches V7 fully in 4 lines, partially in most of the rest; L 61 shares V7’s rhyme and matches V7 in 3 out of 8 lines. Only T laisse 92 (12 lines, -ez/ -iez rhyme) matches V7 closely in almost every line. (ii) C 207 (18 lines, -i- assonance) is reasonably close to O 143 (19 lines) and V4 156 (17 lines); V7 194F/192D (25 lines, -ie rhyme) opens with 3 extra lines, matches C for 4 lines, then paraphrases to obtain a smooth rhyme. P 106 (35 lines) has -ie rhyme, matching V7 in 13 of its first 21 lines, then assonance in -i…e and partial match with C, amalgamating 2 different versions in a single laisse. L 64 (23 lines, -ie rhyme) has close match with V7 in 9 lines, partial match in 8. T 95 (18 lines in -ie) opens like V7 and matches V7 in 11 of its first 15 lines. (iii) C 211 (10 lines and -a…e assonance) matches O 148 in 7 lines and also P 113 (10 lines with the same assonance) closely in 5 lines, approximately in 3. It is paraphrased throughout by V7 200F/198D (13 lines, -ee rhyme). Both P 112 (15 lines) and L 69 (10 lines) match V7 in part, but T 100 is almost identical to V7; all rhyme in -ee. (iv) C 213 (14 lines, -o- assonance) is practically identical to O 150. It is paraphrased in V7 201D (18 lines, -ant rhyme) with line-to-line correspondence and occasional change of word-order. PTL all have -ant rhyme: P 115 matches V7 in 13 lines, L 72 has limited match, but T 102 (36 lines) has the closest match in 17 of its lines.

The Relationship between C and the other Texts

41

(v)

C 218 (15 lines, assonanced in -e-) closely matches O 154; it is much expanded in V7 210F/208D (31 lines, -iez rhyme). P 121 (21 lines in -ier/ -iers) is closely matched by T 107 (15 lines in -ier/-er), L 78 (18 lines in -ier/-iez). They roughly correspond to C (often line-to-line) but completely paraphrase V7, which derives from a different redaction. (vi) C’s semi-rhymed laisse 204 seems to represent a intermediate stage in the conversion-to-rhyme process – and this is underlined by the slightly different assonance in P 103 and P’s paraphrase of C’s line-content (plus P’s lack of match with V7 after the first 5 lines). The availability of different redactions visible here is also shown in P’s duplicate versions, rhymed then assonanced, in P 106, the variant readings of C 211 and P 113 and the accord of P 121, T 107 and L 78 with C whilst differing so much from V7 210F/208D. Against this must be set the close correspondence between T and V7 in the first 4 comparisons set out above. (vii) However, C 218 (V7 210F/208D) is a special case. This is the final laisse of the cluster of assonance; after this, PTL derive from a redaction different from that found in the CV7 model, right through the Death of Roland, the grief at Roncevaux, the battle with Baligan, and only return to a common model at C 5887, laisse 331, for Section C of the poem. At this point, V7 had already inserted laisse 209F/207D (22 lines, -age rhyme), one of the 4 laisses unique to V7, and simply took V7 210F/208D from the same source. In a further 6 laisses, C and V7 have the same rhyme and are printed in tandem by Foerster. They match closely in most lines, but also have significant variants, emendations, occasional omissions and insertions. Where V7’s variant constitutes an emendation/improvement of C’s imperfect or ungrammatical rhyme or the insertion of a vital line, it is difficult not to conclude that C’s was the earlier (CV7) reading – and indeed Duggan does insert lines from C into V7 on occasion, as representing the CV7 model. (i) V7 189F (17 lines, but 18 in 187D through insertion of C 3223 as V7 3138) rhymes in -is, as do C 202 (18 lines), T 90 (16 lines); but P 101 (16 lines) has an -is/-iz/-ir mix and L 59 (14 lines) is assonanced in -i- like O 140 (18 lines). V7 emends the ungrammatical morz jesis in C 3218 (where OPL have gesir) to tant gesir ocis (T 1385: gesir tant d’ocis), inserts into V7 3137 a reference to flors (cf. O 1856, P 1823) and avoids repeating ocis in V7 3145 and 3148 (with V7 adding [V+] 3149 = P 1832, T 1399 and L 932). V7 lacks C 3223, 3231 and 3234 (which closely match O); PTL also lack C 3231. (ii) V7 195F/193D (11 lines) and C 208 (8 lines) rhyme in -ent; both are close to O 144 (8 lines). P 107 (14 lines) and T 96 (8 lines) have -ent rhyme, but

42

(iii)

(iv)

(v)

(vi)

Introduction

L65 (12 lines) has -ant. Of V7’s 3 additional lines, line 3257 is unique to V7, 3262 approximates to P 1994 but closely matches T 1484 and 3264 matches P 1996, T 1486 and L 1034. V7’s variant line 3261 conforms to the PTL pattern (though with variety in the verb). Once again, V7 turns to a source known to PTL and particularly T. V7 202F/200D (30 lines) and C 212 (20 lines) rhyme in -é, as does T 101 (24 lines), but P 114 (45 lines) has predominantly -é and -ez endings, and L 71 (33 lines) has -ez. PTL have some match with C and V7 in the first few lines, but then generally a paraphrase. V7 adds 12 lines not in C or O 149 (21 lines, -e- assonance), with V7 3364 and 3369 matched by P (2106 and 2120), 3380 by T 1559; the rest are not found in CPTL. Few of V7’s variants from C find a match in PTL, but V7 2263 accords with P and T: de mer né where C 3367 reads carné (mortel in O 1993). Out of 7 instances where C and V7 vary from O to obtain the -é rhyme, they are identical in 4 lines (though both have parlé, pardoné and V7 3362 has esgardé where the infinitive is required). The V7 scribe seems to turn to an alternative source (with limited evidence of a link with PTL) or to create his own variant readings. V7 204F/202D (14 lines) and C 214 (12 lines) have -is rhyme and so do T 108 (16 lines) and L 73 (10 lines, but P 117 (13 lines) is assonanced in -ilike O 151 (11 lines). Here V7 changes C’s two opening lines (3397–3398) to 3 (3406–3408) which exactly match PTL, inserts V7 3412 (matched only by T), and corrects C’s poor or incomprehensible rhymes in V7 3416 (= T), 3418 (= PT) and 3419 (no match in PTL). C 3403 presents a strange reading: le cons pasmez s’est ils, with ils perhaps standing for i, where OPL read se pasmet li marchis but V7 and T have chaï pasmis, changing the -er to an -ir verb. Evidence of V7’s derivation from the same source as PTL must therefore be set against evidence of textual variation. V7 205F/203 D (22 lines) and C 215 (21 lines) both rhyme in -uz, as do V4 165 (22 lines), P 117 (23 lines), T 104 (20 lines) and L 74 (18 lines). CV7 match O 152 (21 lines, -u- assonance) and also PT extremely closely, L slightly less; PTL also correspond closely to O. V7 3429–3430 and 3441 (all matching PTL) modify C 3418, 3419 and 3429 to avoid the forced rhymes tuz and luz; V7 3433 (approximating to P 2190), 3434 (matching PT), 3437 (matching L and almost P) and 3441 (found in PTL) replace C’s readings with variants: so a more consistent match with PTL. V7 206F (37 lines in -antz, -ants, but 39 in 204D through insertion of C 3440, 3447) adds 13 lines (in groups of 2, 1, 3, 4, 2, 1) to C 216 (26 lines in -anz with one -ans, one ants), but [V+] 3464 approximates to C 3445. None of these [V+] lines is found in PTL. P 119 (32 lines in -ant, -ans) paraph-

The Relationship between C and the other Texts

43

rases CV7, but L 76 (27 lines in -anz) matches C and/or V7 in about 9 lines; and T 105 (19 lines in -ant) matches V7 in 9 lines, C in 8 – but it omits the [V+] lines and abridges the laisse. In a further 9 lines, V7 has minor variations from C, but these rarely coincide with PTL. In general, in the above six laisses there is consistent match of C and V7 with O, but very slight and intermittent correspondence between PTL and what in V7 are variant or additional lines, compared with C, and the variations are only small details with no consistent pattern from laisse to laisse. There is no sign of C’s modifying an assonanced source reading by reference to any rhymed version; but V7 expands two laisses considerably as well as emending the C readings in others. There remain six laisses where C and V7 are too dissimilar to be printed together: four (C 203, 205, 206, 217) with a rhyme identical to V7; one (C 209) where C and V7 have completely different rhymes; plus the single laisse (C 210) equivalent to 2 laisses in V7 (and all other versions) which has approximately the same rhyme in the first laisse, but a completely different one in the second. Looking first at the laisses with identical rhyme, V7 190F/188D (20 lines, -ee rhyme) and C 203 (11 lines rhymed in -ee, with 2 x -ere) are a special case. In V7, the first few lines (3150–3157) roughly match 3235–3241 although C lacks V7 3152 and 3157 is a free paraphrase of C 3241. After this, V7 has a completely different version, recounting the death of Blancandrin. O lacks this laisse, but C 203 roughly corresponds to the first 9 lines of V4 153 (25 lines, 10 rhymed in -é then 15 in -er); the rest of V4’s laisse approximates to C 204. P 102 (34 lines) has 21 lines with -ee rhyme, followed by a 13-line version of the same laisse with a mix of -ee and -ere (i.e. either rhyme with effacement of intervocalic -ror pseudo-rhyme). P’s first part has a very occasional match with C; the second is closer; very few lines match V7. L 60 (14 lines) and T 91 (11 lines) both rhyme in -ee; L paraphrases C, with no match; T has 4 lines matching C, 4 different lines matching V7. V7 has a very different, individual version of this laisse, where the Blancandrin passage is unique to V7. This recalls the earlier mention of Blancandrin in V7 181D, C 191 (unique to CV7) and the first 2 lines of V7 182D, C 192 (found in no other version); but C copies the CV7 model untroubled by the fact that the «Blanandrin story» remain unfinished. It is typical of the V7 scribe that he inter-relates different parts of the text and ties up loose ends – as is shown in his insertion of [V7+] 196D and 199D in anticipation of the «Belle Aude» episode in the final section of the poem. He therefore turns to a different source – possibly consulted in the previous laisse (the first in the cluster of assonance) where errors and pseudo-rhyme or assonance in the CV7 model (as found in C 3217, 3221, 3226, 3231, 3234) need emending.

44

Introduction

The next 3 laisses present a more straightforward situation. V7 192F/190D (33 lines in -on) matches C 205 (26 lines in -on) in 20 lines, of which 11 are identical. L 67 (34 lines) and T 93 (28 lines) rhyme in -on, but the vastly expanded P 104 (55 lines) is semi-assonanced (43 lines in -on, the rest variable) and matches C and V7 in only 15 lines. In L, at most 9 lines match V7, rather fewer C. Only T has close match with both C and V7. C omits 2 essential lines, 3264a (V7 3189) and 3279a (V7 3208) through scribal error and both C and V7 omit the simile of the lion found in O 1888, P 1887, T 1426 and L 958. V7 also inserts additional lines: 3185, 3187–3189, 3194, 3203, 3204 – but there is little correlation with PTL. C shows no sign of substitution based on any of the rhymed versions. V7 193F/191D (12 lines) and C 206 (7 lines) both rhyme in -u but only 2 lines match exactly. P 105 (12 lines), T 94 (10 lines) and L 63 (11 lines) have the same rhyme. O lacks this laisse but V4 matches C in 6 of its 7 lines. V7 has 6 lines not found in C, of which 3218 has no equivalent in PTL, 3219 is found in PTL, 3224 is identical to T 1457, with part-match in PL. P and L match V7 in 5 lines, but T matches in 7. V7 208F/206D (23 lines) rhymes in -is like C 217 (11 lines) but 4 at most of V7’s first 11 lines match C and V7 then adds 12 lines not in C. C is almost identical to O 153 (8 lines, -i- assonance). P 120 (19 lines) is also assonanced in -i-, but T 106 (10 lines) and L 77 (14 lines) have -is rhyme; there is little if any match between PTL and V7. V7 apparently has its own unique version of this laisse, which lies between 2 laisses (207F/205D and 209F/207D) peculiar to V7 and forms part of the «Gautier» sub-episode which ends with V7 210F/ 208D, another very individual laisse. Here once again V7 is turning to a rhymed source unknown to the other versions. The remaining 2 laisses are more disparate. V7 196F/194D (16 lines with -iez rhyme) simply paraphrases C 209 (12 lines rhyming in -oi with omission of many final consonants, i.e. pseudo-rhymed) and there is no match; but C closely matches the -o- assonanced O 145 (12 lines) and V4 158 (10 lines). V7 retains many of OV4C’s first hemistichs and stays close in meaning. P has two laisses: P108 (22 lines with mainly -iez endings but 5 x -ier, 2 x -ies) matches V7 closely in 11 out of its first 17 lines; but P 109 (12 lines, -o- assonance) is very close to COV4 – and P’s retention of assonance shows it has no cognizance of C’s -oi rhymed version. L 66 (18 lines with -iez, -ier like P and final 5 lines matching P) has an 8-line match with V7; but T 97 (20 lines with -ez rhyme) closely matches V7 in 10 lines, P in 13. Because V7’s lines closely correlate with COV4, it is interesting to see where it matches P, T and L. Thus V7 3269 = P 2003, T 1490, L 1038; V7 3270–3271 = P 2004–2005, T 1491–1492; V7 3272 = P 2007, T 1493; V7 3273 = P 2008, T 1494, L 1041; V7 3274 = P 2009, T 1495;

The Relationship between C and the other Texts

45

V7 3275 = P 2010, T 1496; V7 3275 = P 2010, T 1496; V7 3276 = P 2011, L 1043 (but becomes 2 lines in T)… and the pattern continues for V7 3277, 3278, 3279. This is a very close match, particularly with P and T. In C one finds 3 occasions when O’s second hemistich has to be changed by more than a single word: O 1944: des esperuns a or > C 3325: ce vos creant per foi, O 1947: li mist fors > C 3328: a besloi, O 1950: nen est dreiz qu’il s’en lot > C 3331: si com je cuit et croi; but C’s modifications show no sign of contamination by reference to the other versions. V7 usually retains O’s first hemistichs, but its second hemistichs match P and T. Finally, V7 197F/195D (13 lines, -u rhyme) and 199F/197D (11 lines, -er rhyme) correspond to a single laisse in C: 210 (20 lines, -uz rhyme but several pseudo-rhymes). The intervening laisse, V7 198F/196D, is unique to V7; it has 14 lines and rhymes in -ent. C’s first 9 lines, 3333–3341, correspond to V7 197, C matching V7 in 7 lines. V7’s -u rhyme permits correction of C’s tuz, luz. V7 has 2 [V+] lines: 3282 (lacking in O) after C 3333, and 3287 (with second hemistich matching O 1957) after C 3337; and C 3340 becomes V7 3290–3292, quite different from O. C 3333–3341 are practically identical to O 146 (13 lines, -uassonance) but C omits 3 lines and modifies O 1964. P 110 (19 lines rhymed in -u, but 3 x -us, 1 x -ui) matches C closely in 5 lines, partially in 3. T 98 (13 lines) has -uz in the first line, then -u rhyme in the rest. T matches C in 6 lines (plus T 1519 = C 3350), but V7 in 9. L 67 (13 lines) has 4 lines in -uz, with -u thereafter, but minimal match with C and V7. There are a number of minor variant readings in all versions. C is closest to O; it has limited match with P and T, even less with L; but V7 is slightly closer to PT, especially T. C 3342, duplicating 3333: Voit Olivers qe a mort est feruz, opens the second half of C 210 (11 lines in -uz). C then inserts V7 3282: De lui venger mout entalentez fu, the second line of V7 195D, not found in C earlier, but = P 2055 in P 111. Next C turns to V7 3310–3318, inserting [C+] 3348 but omitting V7 3315, 3317 and paraphrasing 3316; but C changes V7’s -er rhyme to -uz, so matches only the 6 first hemistichs in V7. The other versions have a new laisse: O 147 (13 lines, -e- assonance) and V4 160 (13 lines), V7 197D (11 lines), P 111 (21 lines), T 99 (11 lines) and L 68 (14 lines), all rhyming in -er. V7 matches P in 7 lines, L in 5 lines, but is practically identical to T in 9 out of 11 lines. C matches the first hemistich in 9 of O’s lines, 8 of P’s, 7 of T’s and approximately 6 of L’s lines. There is no good reason for C to continue in -uz. C’s error can arise only from duplicate first hemistichs (Voit Oliver in C) beginning two consecutive laisses (cf. Olivers sent./ Sent Oliviers… etc. in V7PTL). Thus the CV7 model copied by C did not contain V7198F/196D; since OV4PTL all lacked this and the other 3 [V7+] laisses in this passage, their source was more likely to be a later addition to the text.

46

Introduction

In the situation set out above, certain features stand out: persistence of assonance in C and P; V7’s strong links with T where C has assonance and with P and/or T (less often L) in its [V+] lines; plus the significance of V7’s additional laisses. The greatest variation in detail between C and V7 occurs in the laisses where, at a basic level, they most closely correspond (i.e. in C 202, 208, 212 and 214–216) since V7 constantly «improves on» and expands the CV7 model. V7’s laisses seem to fall into groups based on proximity to its additional laisses. This situation is particularly striking in the V7 laisses 205D–207D group. Both 205 and 207 are unique to V7 and presumably copied from another (rhymed) source; they are longer than the norm at this point (27 lines and 22 lines); and they enclose laisse 206D (23 lines rhyming in -is) which, after the first line (3508), has only lines 3523–3524 matching C (laisse 217, 11 lines rhyming in -is) and hardly any match with PTL. Similarly, V7 laisses 194–199D form a sequence in which two are unique to V7 and two have match in rhyme and text with PTL only, but the other two (195D and 197D) are equivalent to C 3333– 3341 and 3342–3352 in C 210 – with V7 196D inserted between them. This is V7 copying from the source of his unique laisses in order to avoid C’s -oi rhyme and later -uz rhyme with their forced endings deriving from the CV7 model. V7 then continues with 198 and -ee rhyme where C 211 has -a…e assonance, followed by 199 (26 lines and unique to V7), so that it is hardly surprising that, after so much recourse to the other source, V7 has in laisse 200 (written in tandem with C 212 by Foerster) an unusually large number of variants from C, including 12 additional lines. Moreover, as C 213 is assonanced (and -o…e becomes -ant rhyme), V7 201 once again has needed to turn to the alternative source; and V7 203D and 204D modify C’s forced rhymes in laisses 215, 216. All these instances point to V7’s turning to and then continuing with a different model, either replacing or modifying the CV7 model. The variable degree of match between V7 and PTL may be explicable via the «French court» in Naples. In a gathering of French poets, trouvères and jongleurs to celebrate Charles d’Anjou’s coronation in 1265 and later to go on the eighth Crusade, several could well have known variant redactions of the Roland whose versions of individual laisses were approved there by their peers. The selection of texts would depend on the individual copyist or the scriptorium. However, the most striking feature of C’s quire 7 is that it is a mirror-image of the earlier quire 3. Indeed, C 3102–3684 present the structure of 1024–2070 in reverse. Thus quire 3 begins (C 1024–1196) with a short, fully rhymed passage (173 lines) practically identical to V7; then, 1197–1533, there is the partlyassonanced passage in which (i) just 6 laisses are fully assonanced, (ii) two

The Relationship between C and the other Texts

47

laisses are rhymed and (iii) the remainder are quasi-rhymed with forced rhyme in some two or three lines per laisse; also, 1534–2070 repeat in a rhymed version the part-assonanced text in 1410–1533 (i.e. the first repetition, in which the rhymed element occupies the whole of quire 4). Similarly, but taken in reverse order in C’s second «cluster of assonance», 3102–3167 are the rhymed version (at the end of quire 6) which C 3168–3216 then repeats in mainly assonanced laisses at the beginning of quire 7; C then continues with 3217–3481, with assonance in 5 laisses, rhyme or quasi-rhyme in the rest (as detailed on pages 39–45 above); and finally completes the quire in a version identical to V7, in C 3482–3679 (ending five lines before the foot of laisse 226). This structure originates in the early assonanced version of the Roland, but these two episodes are emphasized in C by the use of assonance and imperfect rhyme, in the first case, in a sequence of ominous signs leading to Oliver’s three-fold urgings to sound the horn and Roland’s fatal refusal, then in the second case, after the too-tardy horn blasts, in a disastrous sequence of deaths until only the dying Turpin and Roland remain on the field. In the first passage, C laisses 73–100 contain a closely linked, pithily-expressed chain of evemts, of which 91–100 is the culmination; when these last ten laisses are omitted (as in V7), C laisses 101–117 are longer and slow-moving, causing delay in reaching the crucial «sound the horn» sequence. In the second passage, the effectiveness of the mix of dramatic action with assonance and imperfect rhyme is shown by the amount of reader-produced wear on the pages of quire 7. In both cases, the sequence of laisses derives from that in the assonanced versions, but the retention of assonance and quasi-rhyme at these specific junctures is too precise to be coincidental. This deliberate stylistic feature is not within the capacity of the C copyist, so must have been found in the CV7 model. It cannot simply be attributed to ‘C turning to an assonanced source’ when each passage extends into the «CV7 model» area where there is close Cto-V7 match; and the large proportion of rhymed and quasi-rhymed laisses (with no more than two or three forced rhymes per laisse) provide significant evidence of a shared model. This puts the use of assonance in a totally different light: the originator of the CV7 model chose to use assonance and quasi-rhyme in the major part (two thirds or three fifths) of two discrete quires, spaced at precise intervals within that section of his epic which chronicles the events leading to Roland’s death, i.e. Section A, C 1–4179. These two passages are crucial to the narrative: the first ends with Roland’s triple refusal to sound the horn; the second begins with his sounding the horn, too late to save himself, the twelve Peers or any of the rearguard. There are good reasons for the survival of assonance and quasi-rhyme in these two passages. First is the sheer difficulty of converting certain assonan-

48

Introduction

ces to rhyme; in V7, this is achieved only by a complete change of rhyme and therefore paraphrase of the text. Thus in C one finds the -a…e assonance with variants -aingne, -ance, -ange, etc. and -age, -aile, -alle, -ace in seven laisses; -o…e with a variety of consonants in four laisses; and a variety of endings involving the vowels e (plus -en) and i in a further five laisses. Second, the nature of the laisse-content sometimes contributes to the difficulty: for example, the three laisses in the first passage relating to Torchis de Tortolose, Estormis de Valterne and Cornuble de Mont Nigre – six-syllable names which fit easily into a second hemistich, but do not provide an easy rhyme, nor a choice of vocabulary to complete the line – and there is no dominant rhyme within the assonance. As five other Saracen peers had already been introduced, few options remained for rhyme or variety in speech. The names (and lines) were probably well known, so popular appeal protected these laisses. In the second passage, the same -a…e assonance is found in vital laisses describing Roland blowing the horn, Charlemagne hastening back and later the mortally wounded Oliver. Assonance, whether in -a-, -e-, -o- or -i-, is used in significant laisses, for it has a certain rude vigour compared with the more expansive versions with their standard formulæ. It also serves to stimulate audience-attention, dulled by listening to smooth laisses monorimes in which one could second-guess the second hemistich on the basis of standard epic formulæ and the rhyme required. Since assonance offers a wide choice of endings (as the Torchis de Tortolose laisse shows) it makes the audience pay full attention – as do the forced rhymes with their «surprise» element. Much as a modern poet might change metre as well as rhyme-pattern for dramatic effect, the CV7 model originator uses assonance and less-than-perfect rhyme to enliven his text and heighten the impact of two crucial passages. There is cumulative evidence (i) that C was earlier in date than V7, (ii) that C’s assonanced passages and repetitions were found in the CV7 model and (iii) that V7 modified that model’s assonance, imperfect rhyme and metre by reference to a more recent redaction later known to T. V7 was an «agent of modification», inserting «new» laisses found in no other versions and eliminating C’s assonance, imperfect rhymes and other errors, «borrowing» lines from C’s pseudo-rhymed laisses and modifying the page-format in quire 2 and the number of folios in quire 7. Further Evidence in the Main Body of the Text, outside the Clusters of Assonance When one examines the first 34 lines of both texts, it is tempting to see V7 as a «corrected version» of C. In line 3, V7 adds the final t to conquist, distinguish-

The Relationship between C and the other Texts

49

ing the perfect 3 from the past participle; in line 11, he corrects C’s mis-reading of f as s; in the hypometric line 12 he changes tint to trahi. In this last case, much depends on whether the symbol over tit in C indicates -re (giving treit/ trait, an alternative graphy of trahit) or a nasal (tint in error for atint, ataint which fits better with por engigne). Only the fact that V7 inserts line 7, omitted in error by C, indicates a shared model, rather than V7 copying and correcting C. In laisse 2, the line-order in C lines 16, 17 makes better sense than that in V7’s 17, 18 when one treats C’s 17, 18 as forming a new sentence; in line 25, V7 reduces C’s alexandrine to ten syllables; and V7 omits C 30 with its erroneous plural verb. V7 deals summarily with the hopelessly corrupt last two lines of C laisse 2 (C 32, 33) – lines which must relate to the nomination of Blançardin, as is shown by C 34: V7 simply replaces them with a line (33) containing clumsy reference to Blançardin’s domicile in the first hemistich and slight modification of C’s text in the second. As a result of this «correction», C 34 is no longer relevant since there is no speech in V7 33 for Blançardin to hear, so V7 has to replace it with a line of banal description. This first page sets the pattern of the CV7 relationship. C’s is the uncritical copy, the scribe hampered by his lack of expertise in French when he tries to copy faithfully the model he sees before him, but misreads and omits elements – and lacks the skill and confidence to cut or modify when corruption occurs. It is also very much the norm for C’s occasional single alexandrines to be converted to decasyllabic metre in V7: this may be seen in lines 53, 66, 124, 369, 458, 531, etc.; some 70 individual alexandrines are found throughout C’s text, as shown in Appendix A (which lists the equivalent line or hemistich in V7). In addition, C has in Section C nearly 300 consecutive alexandrines between C 7841 and C 8133, with extremely rare lapses into decasyllabic metre. This long passage covers the single combat between Thierry and Pinabel (the extra syllables being extremely useful when 3-syllable names constantly occur) and the counsellors’ advice on the method of execution of Ganelon (the alexandrines giving weight and solemnity to the occasion). At this point, L 2753–3032 has alexandrines interspersed with the occasional 10 or 11 syllable line; T uses alexandrines for the judicial combat and for the first suggestions regarding Ganelon’s punishment, T 5358–5620, but again with decasyllables interspersed; P uses alexandrines for most of the Thierry-Pinabel fight, P 6537–6670, but has 23 decasyllabic lines within that passage and then reverts to decasyllabic metre; V4 has a much shorter version of the combat between Thierry and Pinabal, with irregular metre (5:6; 6:8; 4:7) in many lines; it is only from V4 5848, as Pinabel falls and is killed, that the V4 text corresponds to C and has frequent alexandrines. In contrast, V7 8035–8328 (= C 7841–8133) has only 29 alexandrines, but 77 lines with 5:6 metre – evidence of a vain attempt to reduce 12-syllable to 10-syllable metre.

50

Introduction

C’s dodecasyllabic metre in Section C clearly does not derive from any early assonanced version: the extremely close match between the rhymed texts in this section (where even L’s abridged version is composed of selected essential lines and not a paraphrase) indicate a fully rhymed source, requiring no «updating» to satisfy contemporary taste. Section C was a late development, written between 1180 and 1195.56 Moreover, O is decasyllabic in lines 3873–3933, with only occasional 5:5, 5:6 or 4:7 variation within the combat sequence; only O 3940, rounding off laisse 294 with a solemn note, is an alexandrine; but the execution of Ganelon’s hostages and of the traitor himself is related in decasyllabic verse with only the final line of each laisse in 5:6 metre. There is, of course, no discussion of possible methods of execution in O 3934–3973 – and it is noticeable in C that, when the «final solution» (as found in O) is proposed by Otes, the author of the «new» version reverts to decasyllabic verse. It is also relevant to the second «assonanced passage» and the CV7 relationship therein to PTL, to consider both the preceding «inter-repetition» passage (C 130–196) and the final part of Section A (C 219–247) plus Section B (248–330) which frame C laisses 202–218. In C’s 3-quire passage, laisses 101– 196, it is noticeable that, despite V7’s early borrowings from C’s imperfectlyrhymed laisses, C and V7 are practically identical and that P for its first 55 laisses (= C 101–157) matches them closely, certainly deriving from the same source. The «Margariz episode» occurs at this point, involving extra laisses in P (matched by L) and confusion over laisse-order in all the other versions; and thereafter, in laisses 56–102, P has an increasingly uneven match with CV7. Nevertheless, a persistent common source (however much modified by P) is shown by the shared error in the final lines of C laisse 165, 2679–2682, V7 laisse 155D, 2627–2630 and P laisse 67, 1277–1280 + 1281, lines according with the laisse in rhyme but completely out of context and not found elsewhere. From this point L, in which laisse 1 corresponds to C 129, generally matches only half the lines in CV7; it echoes P more closely, but often in abridged form and with variant lines. After a gap roughly equivalent to C laisses 101–129, T’s match with CV7P fluctuates from almost identical to minimal (some laisses being completely omitted, as in OV4). P, T and L vary increasingly from CV7 in the laisses leading into the second repetition; P and L have little match with C laisses 192–196, V7 laisses 183D–186D, and only T laisse 89 has a fairly good match (in roughly 14 out of 25 lines) with C196, V7 186D. After C laisse 218, V7 208D, there is once again no close match between PTL and CV7, despite the fact that this is the «Death of Roland» episode: PL and T apparently derive from different redactions, T often varying from the  Cf. Palumbo 2003.

The Relationship between C and the other Texts

51

other rhymed versions. The same situation obtains in Section B (C laisses 248– 330). Thus P and T both correspond to OV4 in placing Charlemagne’s visit to Roncevaux immediately before the battle with Baligant, then P alone inserts the vain search for Roland’s body and its discovery after prayer – an incident not found in CV7 at this point. P and T also have the OV4 interment of the French dead, though only T contains the embalming of Roland, Oliver and the French peers (not Turpin); in P laisse 184, arrangements for transporting the bodies of Roland and Oliver to Blaye are interrupted by news of Baligan’s approach. Even in the «Baligant episode» (which L lacks) P and T have slightly variant, though parallel, versions – and neither has a close match with that in CV7. Indeed, P and T each have assonanced and quasi-rhymed laisses from time to time. Thus Beretta justifiably sees T as a separate «branch» from PL at this point, though the chronological element cannot be ignored when considering this late manuscript. The «clusters of assonance» occurring in C and their significance to the C–V7 relationship have to be seen within this context. On the one hand, it is not a matter of sudden change in this relationship, but of the same pattern persisting: V7 seeking constantly to improve on the CV7 model (and there was every incentive for V7 to turn to a newly available fully rhymed model in order to meet the specification set for his copy). On the other hand, there was no call to change the CV7 model where it was already smoothly rhymed and made sense; the major part of laisses could be retained unchanged where the CV7 model’s imperfect rhymes numbered only two or three and were readily correctable by reference to the new version. Only in the assonanced laisses and those recalcitant to full conversion would a complete change of ending be necessary; and it is T rather more than P which V7 matches in detail, wherever it differs from C. This is particularly noticeable in T laisses 99 and 100, slightly less in 98 and 104 (V7 laisses 197D, 198D, 195D and 203D). The Episodic Structure of C (and CV7) and its Implications Examination of the episodes in the Roland narrative, as found in C and V7, reveals a pattern in the structure of the CV7 model, fluctuation in the dating and degree of redaction and variation in the relationship between the different versions, assonanced and rhymed. The first episode relates to the prime causes of the action: Blancandrin’s deceitful embassy to Charlemagne, Roland’s nomination of Ganelon as envoy and the latter’s fury. It ends with C 511 in laisse 31 as Ganelon goes to his tent to prepare for departure; in C with its 33 lines per page, this is half-way down folio 8 verso, on the last page of the first quire. Laisse 32 seems to be a laisse

52

Introduction

de transition leading into the «treachery» episode which follows. V7’s laisse 31 ends with line 522 (but the Duggan edition includes 8 lines found only in C, reducing the V7 total to 514 lines); and the final line of the V7 episode coincides with the first line of folio 9 verso, i.e. because of V7’s 30 lines-per-page format, after a full page of the second quire has been filled. The V7 copyist must have become aware of this as he turned to the second quire of parchment, for it is at this point that he carefully marked the blank parchment to allow for an increased (and variable) format: 33, 34, 35, 34, 35, 35, 35, 36, 34, 33, 33, 34, 34, 34, 34 and 27 lines per page so as to obtain an extra 60 lines (a full folio in his normal format) in the second quire. V7 was clearly not copying a model with 30-line page, for no adjustment would have been necessary; and the 60 extra lines would have no relevance to a 33-lines-per-page format (with an extra 48 lines compared with V7’s 60); only a 32-line format approximates to the 60 lines, particularly if the unusually short final page is taken into account. C’s second quire (theme: «treachery») contains lines 512–1023, laisses 32 to 60, one line short of 512; V7 laisses 32–61, 523–1051, has 529 lines minus 7 lines from C inserted by Duggan = 522 lines, of which 8 are in V7’s additional laisse, 58. The situation in C and V7 at the end of the second quire is extremely interesting. At first, it seems more logical to end the quire with Ganelon’s departure from Saragossa, that is, with C 973 at the end of laisse 58, V7 1000 in laisse 59, since there is a change of scene to Charlemagne at Cordres; however, the texts prefer to see Ganelon’s false report (in C 60, V7 61) as the culmination of the «treachery» quire – and this is confirmed by V7’s oversize L at the beginning of laisse 62, V7 1052, marking the beginning of the third episode/third quire. If one posits a 32-line-per-page CV7 model, the final laisses of this episode (C 58–60, V7 58–61) would be found on the final pages of the quire, where rubbing or other damage might easily have occurred. Since O laisse 51 (5 lines) and V4 46 (12 lines) both endorse and clarify V7 58, such a scenario would explain the latter’s incomprehensibility and corruption – and also its excision by C. Although C 58 and V7 59 have approximately the same number of lines, it is clear they derive from different sources; they have the same general theme as OV4 but only C has the slightest match – and that in a few first hemistichs only. Both O 52 and V4 47 are assonanced in -a…e and considerably shorter than C 58 and V7 59: C has -é rhyme, V7 -ez. The third quire thus begins with C 1024 in laisse 61 and runs to 1533 in C 100, just 510 lines; in V7, this is laisses 62–91D in the Duggan edition, V7 1052–1487 (only 436 lines because V7 omits (most of) the assonanced passage of the first repetition). This quire contains first Charlemagne’s premonitory dreams, then the nomination of Roland to the rearguard, where he is joined

The Relationship between C and the other Texts

53

by the twelve peers – laisses 61–72 in C, 62–73 in V7; then comes the first «assonanced passage» with the introduction of the Saracen peers and the approach of the great Saracen armies, followed by the first «horn passage». The roughly 170-line length of each of these three sub-episodes (a Roland – Saracens – Roland sandwich) shows the care taken over balance within the quire. Duggan inserts C laisse 66 (14 lines), which V7 lacks, and also two of C’s additional lines into the «pre-assonance» passage, V7 62–73. The acephalous T opens at C 66 but has only.intermittent match with C in T 1 and with CV7 in T 2–6, with frequent paraphrase (whereas C 66–72 and V7 67–73D remain almost identical). From laisse 7 to the first few lines of laisse 27 T’s fully rhymed text has a intermittent match with C, but matches V7 closely in the laisses where C is assonanced and in individual variant lines. The intra-repetition passage, C laisses 101–196, lines 1534–3167, falls into three quires: laisses 101–129, C 1534–2070 (537 lines); the rhymed repetition, much expanded to include speeches, prayers and the 11-laisse introduction of the French peers (only 11, because Gautier has already left to fight Amauriz and his army); 130–161, C 2071–2587 (517 lines), the French defeat the first Saracen armies but Margariz escapes to warn Marsile; then 162–196, C 2588–3167 (580 lines) in which, attacked by fresh Saracen armies, the rearguard is decimated and Roland is finally forced to sound his horn. P opens at C laisse 101, line 1553; L laisse 1 does not begin until C 129; but T begins again at C 130 with the remainder of T 27 corresponding to that laisse. C and V7 are generally identical, although V7 inserts into the first of these quires Duggan’s laisses 92A and 108A and also individual lines taken from C’s assonanced passage.57 P laisses 1–100 closely match CV7 for most (but not all) of this three-quire passage, showing that all three derive from the same rhymed source; in CV7 quires 5 and 6, T laisses 27–89 and L laisses 1–58 have a more variable match with CV7P: both omit C laisses 131, 135 and 147–149 (as do OV4); otherwise, L corresponds quite closely to P. However, there is discrepancy after C 2518, for C laisse 158 opens with 6 lines not found in V7, but matching PTL. This laisse is equivalent to P laisse 55, after which P inserts 5 extra laisses, matched by L. It opens the «Margariz episode» (C 158–161) in which Marsile learns of the Saracens’ defeat; this is the turning-point in the battle, as his fresh armies then slaughter the French peers and their followers. Once again, this is an end-of-quire situation, where P matches O, V4 and L, but not CV7 and only partly T – and again a separate episode. So P is probably combining two different sources at a point where their variants overlap.  Cf. Introduction, Textual Evidence, above, 31–37.

54

Introduction

Quire 6, the final quire in this group, contains C laisses 162–196, with the first sounding of the horn reported in laisse 190. The match between CV7 and PTL (and OV4) becomes very variable in this long passage (579 lines). Although the shared error in CV7P in the final lines of C laisse 165 confirms their common source, this laisse is not found in OV4TL and nor are the first 14 lines of C 166; L lacks C 176, O and L lack C 179–181, O lacks C 182–183, PL lack C 184, L lacks C 186, O lacks C 189, and C 191 is unique to CV7. Within each laisse, the match of PTL with CV7 ranges from close to practically nonexistent in this quire, but it decreases considerably in the latter part of the quire. Did the CV7 model access and meld in another source here? Certainly in C 191 and 192 (V7 181, 182D), CV7 are unique in mentioning Blanzardin (whom Roland will kill in V7 188D, but not in the truncated C 203); and the intrusion of C 191 (V7 181) increases the «horn sounded» laisses to four, making nonsense of V7’s tiercha foiz in V7 3070 – and possibly justifies C 3106’s hypometric tres foiz (either a mis-spelling of terce, matching V7, or a scribal error for autre fois, as in P 1750, T 1302). After the second «assonanced passage» (which occupies 314 lines in C, 452 lines in V7), C laisses 219–247, 3482–4179, (V7 209–237, 3583–4290), lead steadily up to the Morz est Rollant… which opens Section B. In both manuscripts this passage fills approximately 700 lines and the CV7 texts are practically identical. They contain laisses not found in OV4 or PTL and vary considerably from them; PTL appear to derive from different redactions,58 so the situation is entirely different from that in quires 4–6 earlier. This lack of match continues through the whole of Section B; concordance is re-established only in the final section of the poem. It should be noted that, whereas V7 3591 (the fifth line of laisse 209D) marks the end of V7’s (10-folio) quire, C’s seventh quire does not end until line 3678, six lines before the end of laisse 226, a laisse in which the Saracens

 Even in C 219 (V7 209), found in all versions, there is a poor match between CV7 and PTL, though both groups are similar to O laisse 155; and there is no equivalent to C laisses 220, 223, 227, 231, 235 or 238 and their V7 equivalents in OV4 or PTL. C 237 (V7 227) has the same first line as P 137, T 121 and L 91 but no further match, although all have -ir rhyme; but C 3927– 3929 echo P 2545–2548, T 1862–1865, and L 1432–1435 in the preceding PTL laisse rhyming in -ant. In C 239, 3958–3959, 3961 and 3963 (and the V7 equivalents) with -ent rhyme correspond in sense only to the -ir rhymed T 121, 1878–1880 and 1882 and, apart from 3961, P 137, 2556– 2557 and 2559, and L 91, 1443–1445. Additional laisses, paraphrase and expansion are a feature of CV7 throughout this passage; differences in rhyme between CV7 and PTL show that they derive from different conversions of an assonanced source. Beretta (2001, 135–160) provides a useful list of laisses with non-matching rhymes in CV7 and PTL in this section.

The Relationship between C and the other Texts

55

see Charlemagne’s army approaching and throw their darts from a safe distance before taking flight. C 226, V7 216D have minimal match with OV4PTL; C 227, V7 217D are not found in OV4PTL. The close C-V7 match is re-established from the beginning of C laisse 219, V7 209D, but even in this laisse C inserts line 3486, not found in V7: Per grant vigor sor piez est revenu. V7 may have omitted this because he had (in error) used the same first hemistich in V7 3586, where C 3484 begins Isnelement; Duggan substitutes the C reading in V7 3586, but does not insert the [C+] line 3660: Il se dist voir, ja mar lo mescrerer. V7 3593 was apparently omitted by C through scribal error. This situation has other interesting implications in respect of C and V7. First, C 3678 (or 3686 if one continues to the end of the laisse) tallies neatly with 3673, the half-way point between C 3168 (opening the «sounding the horn» repetition) and Morz in 4180: thus C’s quire 7 relates the events from the sounding of the horn to the flight of the Saracens from the battlefield, then quire 8 contains Roland’s gathering of the peers’ bodies for blessing, the death of Turpin and finally that of Roland. This explains and gives impact to the oversize M for Morz in C 4180 (with its decorated margin for further emphasis). In contrast, in V7 4291 this M occurs part-way through a quire. In V7, 3584– 4085 (the eighth quire) end with the death of Turpin and 4086–4562 (approximately another quire) deal with Roland’s death, plus Charlemagne’s grief at Roncevaux and his premonitory dreams – leading straight into the change of scene, with the oversize M of Marsillions in V7 4563, the first line of laisse 249 and the beginning of quire 10. The weak mid-quire position reduces the impact and significance of Morz considerably; but it makes the «Baligant episode», C 4449–5886 and V7 4563–6006, into a three-quire passage – in a 30 lines per page format. In contrast, C’s Section B, 4180–5886, has 1707 lines, equivalent to 3 exceptionally long quires of 569 lines (roughly 35 or 36 lines per page) or 3 quires (at 32 lines per page) plus an additional 172 lines. The short final passage may result from CV7’s excision of the «grief at Roncevaux» passage with its search for Roland’s body and interment of the French dead – a passage found in CV7 and L only at the beginning of Section C. If one turns again to the «mirror-image» situation, it is clear that, just as quire 3 opened with fully rhymed laisses based on the CV7 model (C 61–72, 1024–1196 = 173 lines and V7 62–73, 1052–1226 = 175 lines), so quire 7 in C ends with laisses 219–226, 3482–3679 (3684 to the end of the laisse), a total of 198 or 203 lines based on the CV7 model. The difference lies in the fact that V7, with its extra laisses and expansions, had already reached the end of quire 7 at 3591 on folio 58 verso, so that quire 8 began on the ninth line of laisse 209D and quire 9 with V7 4547 in laisse 248D (or, having completed that laisse, with 4563, the first line of laisse 249D). When the V7 scribe reached folio 75

56

Introduction

recto, the first page of quire 10, he recognized that laisse 249, beginning 4563, marked a change of scene and of topic, so inserted an oversize, decorated M for Marsillions (whereas C 4449, the first line of laisse 259, begins Li rois Marsille… and has no oversize capital). The fact that the whole Baligant episode, from this point to the end of Section B at line 6006 on folio 99 recto, then falls neatly into three episodes/quires (Marsile calls on Baligant for aid and the latter arrives with his army; Charlemagne organizes his forces into divisions; so too does Baligant and the ensuing battle ends in single combat in which Charles kills Baligant) may well indicate that this was always a separate, optional unit, to be inserted or omitted as required – and its omission by L reinforces that impression. It is therefore less significant that C does not offer a pattern of quires and episodes in the narration after the M of Morz; Section B derives entirely from the CV7 model, with C and V7 practically identical; it should be seen as a single «performance» contained in 1,707 lines. The fact that both C and V7 have the oversize M of Morz, although it has much reduced significance in V7’s case, implies that this was found in the CV7 source and that, as usual,.V7 was the modifying agent in treating the M of Marsillions in the same way on structural grounds. In Section B (where L has only 10 laisses, omitting the Baligant episode entirely) P and T have the same sequence of incidents as OV4, with Charlemagne’s initial grief quickly interrupted by his pursuit and extermination of Marsile’s army and his return to Roncevaux (postponed in the text until after the Marsile-and-Baligant interaction) culminating in the latter’s arrival with his army. In contrast, CV7 have their own version of Charlemagne’s grief on his arrival at Roncevaux (C 4180–4238, the first laisses in Section B) and no further return there until the beginning of Section C – whereas PT match OV4 in the return to Roncevaux prior to the battle with Baligan. P 3282–3290 (laisse 178 in the Rejhon «Corpus» edition) then match O in Charlemagne’s vain search, prayer for Divine help and miraculous discovery of Roland’s body; P then continues for several laisses, as does O, with Charlemagne’s lamentations over Roland and the interment of the French dead, but omits the embalming of Roland, Oliver and Turpin for transport to France, since P laisse 184 is interrupted after P 3353–3355 by the Saracen announcing Baligan’s approach. T (like V4) has no mention of any difficulty in finding Roland’s body (cf. laisse 158, 2424–2437), but it continues (like OV4) with the lamentations and interment of the French dead (though T says «burial», not in a charnel-house) and in the embalming of Roland and Oliver (laisses 159–166) – although T (laisse 166 in the van Emden «Corpus» edition) substitutes the twelve peers for Turpin. PT often have different rhymes from CV7 in the «battle with Baligant»

The Relationship between C and the other Texts

57

and, even where the same rhymes are used, tend to paraphrase the CV7 text. P and T each retain assonance in a number of laisses throughout this Section; but they derive from a different redaction from CV7. Section C: A closer match between the rhymed versions is not found again until this final section of the poem, beginning Granz fu li deus… These words take one straight into Charlemagne’s return to Roncevaux; it is here that CV7PTL all accord – whereas V4, which (like O) has already recounted the events at Roncevaux earlier in Section B, inserts a new and unique episode, La Prise de Narbonne. For C, V7 and L, the Roncevaux events constitute new material, not found earlier in the poem; they are recounted in C laisse 332, 86 lines: in lines 5960–5961 (vain search), 5962–5984 (invocation and prayer), 5991–5992 (discovery); then, after the 20-line laisse 333, the miracle of the aubépines in 334–335, the burial of the French dead (laisse 336) and the miraculous growth of trees to furnish biers for transporting Roland and Oliver to France (337–338). The same events are recorded in V7, L and also P and T. Thus there is considerable duplication and indeed contradiction in P and T in respect of the events listed in Section B above.59 The late twelfth-century date of this section, the use (or non-use) of alexandrines, the oversize capital inserted by V7 to mark the division of its 2,391 lines into two «chapters» have already been discussed. Judging by its extensive use of alexandrines, C 5887–8201 may well provide the earliest version extant of this late, hugely expanded and innovative dénouement to the poem; but the derivation from a common model, here discernable in all the rhymed versions (including V4), is unique to this final section of the poem. Section C is a vast expansion of O’s final passage, with many incidents – such as the «flights of Ganelon» and «Belle Aude» episodes – added to the original storyline and its 2,050 lines have minimal correspondence to O’s 320 lines; but V4, after inserting the «Narbonne episode» to replace the «Roncevaux» passage

 It is difficult to understand why Duggan (2005, 47s.) affirms that, in the PTL version, «the French simply return to Roncevaux; it is not a question of searching for Roland’s body, which was already found before the Episode of Baligant». The problem does not arise with CV7L, which do not have the earlier version; but in Section C, P 4503–4504 read Entre les mors va son neveu querant /Quant il nel trouve, moult a le cuer dolant (practically identical to C 5960– 5961, V7 6080–6081) followed by a prayer and discovery of the body, P 4538–4539 – thus duplicating the events in P 3282–3292 (which match OV4). T is rather ambiguous: it omits nel from the hypometric 3478 (Quant il trouve, moult a le cueur dolant) but contains Charlemagne’s prayer and the discovery of the body, 4538–4539. In Section C, CV7PTL all contain the «miracle of the aubépines» (C 6080) and that of the biers (C 6122–6127), so P 4714 and T 3623 have a second burial of the dead (though in tombes, not the fosses found in C 6123, V7 6243).

58

Introduction

already recounted earlier as in O, corresponds – sometimes roughly, with partly assonanced laisses – to the «new» version. The dating of Section C to between 1180 and approximately 1190–1195 is based on the arguments of Palumbo (2003). Palumbo makes it clear that this dating relates solely to Section C60 and that he finds the supposed evidence based on Daire in the Roman de Thèbes unsatisfactory – in particular, because it depends on which version of the Thèbes is consulted. Duggan (2005, 54) apparently applies the dating to the whole text and makes no mention of the rider when he refers to the Palumbo article. In fact, apart from the «Daire» reference (in C 3559, V7 3662), all of Duggan’s evidence dating the whole poem to 1180–1195 is contained within Section C. There seem to be no good grounds for identifying Daire (or daire/d’aire) with Daire le Roux in the Roman de Thèbes. Whilst his case may be relevant to the Ganelon theme and the «treachery» argument, it is difficult to see how this allusion relates to the «loss» of Roland. If one reads jugement daire as «sentence / judgement on / punishment of Daire», this could well be an allusion to Darius, king of Persia, whose huge losses at Marathon would be well known to a classically-educated readership and whose fall is documented not only in the twelfth-century Roman d’Alexandre but also in the Book of Daniel – and therefore familiar to less educated audiences. Alternatively, daire may be deliberately written without initial capital in C and V7. This is an occasional but very infrequent feature in both texts in respect of Proper Names. Read as d’aire, variant of d’aive/d’aïeul, this could be a reference to the Judgement of God on Adam for Original Sin and the loss of Eden – again a familiar concept in a religious setting and a much more potent image. One other discrepancy in Section C must be recorded: that C has the single laisse 383, 7 lines rhyming in -ie, where V7 376F/373D has 15 lines with the same rhyme, then adds 377, 378F/374, 375D which are not found in any other version. C 7082, the final line of C 383 reads: N’est pas mervelle s’ele est descolorie, which V7’s final line (7228) matches exactly; but V7’s seventh line, 7220, reads: N’est pas merveille car grant soif ot soffrie. C’s saut de même en même is confirmed by V4 352, P 312 and L 159, but T here lacks a complete folio. However, V7’s additional 2 laisses (19 lines in -age and 14 lines in -is rhyme = 33 lines) are not found in any other version. This is further evidence of V7’s reference to a different model. The question of the «Roland Rimé» must be considered in the light of all the foregoing evidence. Whilst it is indisputable that C and V7 derive for the  Cf. Palumbo (2003, 410): «gli elementi a nostra disposizione invitano a fissare la datazione del ‹Roland› rimato – o almeno la datazione del capostipite che comprende la lunga coda finale del testo… nell’ultimo ventennio del XII secolo… tra il 1180 e il 1190–95 circa.»

The Relationship between C and the other Texts

59

most part from the same model, there is nothing to suggest that V7 derives entirely from that model – and no proof that that model was fully rhymed. In contrast, there is evidence that V7 modified the model shared with C: by inserting additional laisses culled from another (rhymed) source, by modifying readings by reference to that source, by substituting readings from C’s assonanced or quasi-rhymed laisses for lines common to C and P – and in general by modifying the lay-out of his text to accommodate these changes. Both Duggan and Segre accept that V7 had access to (and occasionally used) an assonanced source – but, if V7’s model (i.e. the CV7 shared model) were the Roland Rimé (and therefore fully rhymed) there was no good reason for the V7 copyist to turn to that source. However, if the CV7 source contained the assonanced and quasi-rhymed laisses found in C, there were good grounds for V7 to turn to a newly-rhymed source in order to produce a fully-rhymed version. Segre’s theories hold good only if V7 is an earlier version than C; but there is no proof of this. Indeed, the evidence points to C as the earlier version – and the predominantly close match between C and V7 in the quasi-rhymed laisses in those passages where Segre posits a damaged model as the cause of C’s imperfect rhymes puts Segre’s explanation into question. The evidence of T in the first passage and of T and P in the second reveals the existence of an alternative rhymed version and V7 as the agent of change, modifying C’s infelicitous readings where C is assonanced or where there are forced rhymes. On the first page (148) of Chapter 7 in La tradizione della «Chanson de Roland», Segre’s third paragraph reads: Se la presenza di queste lasse non fosse razionalizzabile, se cioè queste lasse dovessero risultare come dei relitti della vecchia redazione assonanzata lasciati indenni, o riaffiorati, in seno alla nuova, rimata, potrebbe riacquistare prestigio quella specie di Wellentheorie che considera ogni codice come risultato casuale di una rete memoriale d’incroci e di sovrapposizioni. Se viceversa la presenza di queste lasse risultasse razionalizzabile, non solo essa non lederebbe la validità dello stemma, ma fornirebbe forse degli indici preziosi sul metodo di lavoro dei copisti della ChR.

The argument set out in the present edition of C shows that there is a logical explanation for the existence of C’s two partly-assonanced passages, other than that of «making good partial damage» put forward by Segre. It also reveals, in its examination of the quire-episode relationship in both C and V7, a very different view of how both remanieur and copyist worked and how the different versions evolved. The evidence of both C and V7, as set out in the past 59 pages, raises serious doubts as to whether the «Roland Rimé» existed as a complete, fully rhymed text prior to V7. V7 itself masks, by its consistency of rhyme, rare

60

Introduction

forced rhymes and regular decasyllabic metre, its derivation from at least two different models: predominantly the CV7 shared model, but also a later rhymed version eventually known to T and related at times to P. V7’s use of a third model is implied by the presence in V7, in the second cluster of assonance, of laisses which COV4PTL lack. The above details are confirmed by Beretta’s comparison61 of the rhymed versions, in which he distinguishes four different Sezioni between C laisse 129 and laisse 251 (i.e. C 2049–4278): laisses 129–166, 167–201, 196–216 and 217– 251, when one uses laisse-numbers as in the present edition of C. In fact, Beretta’s examples from C begin at laisse 130, line 2071. Whilst the limits of the Sezioni are often debatable, taking insufficient account of the laisse-content,62 the evidence Beretta adduces generally underlines that set out above. Thus Sezione 1a reflects the rhymed source common to CV7 and P in C quire 5, but the greater variation of L and T from that source; Sezione 2a shows an increasing polarization of first T, then PL, from the CV7 version in quire 6; but the third section, covering C’s partly-assonanced passage, links V7 more closely at times with PLT, though Beretta does not set the lines he cites in the context of the whole laisse; finally, Sezione 4a clearly shows the separate source of CV7 compared with PTL. Throughout the poem, there is a manifest concern with the balance of episodes, so that the work of redactors was restricted by considerations of space, as well as dramatic tension. The space element also had a practical impact, the quire-episode correlation setting an approximate limit to the degree of re-working: substitute pages allowed for different numbers of lines. It was only in the open-ended Section C that a vast flow of new incidents occurred, although earlier C’s ten partly assonanced laisses were expanded to fill a whole quire. The balance between V7’s additional laisses in quire 7 and CV7’s expanded «death of Roland» in quire 8 also prevented a change of emphasis in the «new» version, compared with what was to follow – so that the fatal wounding of Oliver, his accidental striking of Roland and subsequent death did not overshadow by too great expansion that of Roland in the next quire – and it encouraged a lively tempo of events, expressed in short laisses. Whilst the use of a limited vocabulary and standard epic formulæ in the remaniements produced an overall consistency and harmony, the degree of  Cf. Beretta (2001, 43–134).  For example, the division between Sezioni 1a and 2a takes no account of the textual significance of the Margariz episode; Sezione 3 overlaps with the end of Sezione 2a, beginning with C’s final assonanced «horn» laisse, 196, instead of 197 and ending with 216 (although C and V7 return to the shared model in C 219). These are arbitrary decisions, for they need have no effect on Beretta’s stemmata.

The Relationship between C and the other Texts

61

expansion during conversion to rhyme fluctuated from quire to quire in a rather haphazard sequence. Thus, comparing O with C, one finds a 33 per cent increase in the predominantly «Baligant episode», from O’s 1268 lines to 1707 lines in C 248–330, a 50 per cent increase in C 1–72 (813 lines in O become 1196 lines in C) and in C 130–196 (where O’s 611 lines become 1097 lines in C), then a doubling of the number of lines in C laisses 219–247 (where O’s 314 lines become C’s 698 lines). In contrast, in C’s first repetition, laisses 101–129, O’s 131 lines and C’s 123 lines in the assonanced text expand four-fold to 537 lines in rhyme; and the final 320 lines in O increase to 2315 lines in Section C, a more than seven-fold expansion. The variations in the length of episode are reflected in the number of lines per laisse, the largest being 86 lines in C laisse 332 – in the final section, of course). The degree of redaction visible in C’s non-assonanced passages varies greatly; it does not seem to have been done as a consecutive piece of work by a single redactor. Whilst the above analysis gives only a very rough, oversimplified view of C’s version compared to the earliest assonanced text known today, it does offer insights into the redaction process which accord with the limited evidence regarding dating. The main feature is the lack of continuity, in terms of re-working, between the different episodes in the narrative. Even where C and V7 have a consistently close match, between C 219 and the end of the poem, there are three very different degrees of expansion. Since V7 is practically identical to C throughout the passages identified above, this also implies that the CV7 model was not the result of any regular redaction, such as would have produced a Roland Rimé. The situation is underlined by the discrepancies between the different sections of the poem in the various versions. The most glaring example is found in the «Roncevaux incident», examined above: neither P nor T takes account of their earlier version of this event when copying Section C. As Duggan points out, Richard of Normandy, killed in C 5586 in Section B, is very palpably alive in 6908, 7470, 7640, etc. in Section C. Similarly, confused readings at «intersection points» where different redactions meet (for example, the end of the «treason episode» and the Margariz episode) show where different versions are being melded. Also, the fact that P and T have at times a variegated mix of assonance, pseudo-rhyme and rhyme, with few common features, indicates that they have no knowledge of a full-text Roland Rimé. Beretta 2001 relates the readings of these assonanced laisses to those of OV4 and, where relevant, T and L. On page 155, he acknowledges that, in trying to explain why the C copyist took the trouble to substitute assonanced laisses for those already rhymed, «l’ipotesi della lacuna da colmare è… insostenibile» in the light of his own findings and that the fact that C laisses 90–99

62

Introduction

(91–100 in the present edition) are then duplicated in rhyme makes this even more incredible; similarly, «Lo stesso dicasi per la seconda serie». Beretta concludes that «l’inserzione delle lasse assonanzate in C non dipende da cause accidentali; deve invece attribuirsi ad una scelta cosciente del copista, qualunque siano le ragioni che l’hanno ispirata». However, belief in a stemma based on the Roland Rimé (δ) causes Beretta to argue that, whilst C revises his assonanced laisses by reference to κ, V7 obtains some corrections from γ: «con molta probabilità, lo stesso che gli fornì le sue due uniche lasse assonanzate» – i.e. laisses 92A and 108A in Duggan’s edition. This is carrying the difficilior argument to extremes: once it is accepted that no fully rhymed text of the Roland existed prior to V7 and that C was the earlier manuscript, then the CV7 model with its mix of rhyme and assonance in two passages (as found in C, a close but uncritical copy made by an Italian scribe with somewhat limited French) offers a simple source of the discrepancies. Since V7 acquires his two «assonanced laisses» from the CV7 model, through scribal error in copying 92A from the wrong passage and through deliberate choice in inserting 108A, there remains only his access to one newly-rhymed source to explain the match with readings in the other rhymed versions. This eliminates many of the variant sources which abound in the stemmata based on the Segre argument.

Conclusions It is on the grounds of the evidence set out in the last 61 pages that C is held to be the earlier manuscript and treated as such in this edition. The small size, single-column lay-out on poor quality of parchment and amateurist decoration of C all imply an early date. The historical evidence suggests that C was a Gonzaga manuscript from the first, produced soon after Charles d’Anjou’s coronation as King of the Two Sicilies in 1266, possibly between the French-led Crusade in 1270 and 1286. Guido Gonzaga as Mantuan ambassador to the court in Naples had experienced the strongly French culture of that court, but had limited resources available for commissioning copies of French literature. V7 was a later copy of the same shared model, made shortly after extensive rich domains were restored to the Gonzaga in 1287. The structural evidence shows how V7’s role as the «agent of change» is reflected in the constructional details of the V7 manuscript: lines-per-page layout in quire 2 and 10-folio size of quire 7; V7’s 2-folio backward skip when copying the shared CV7 model. This provides textual evidence that some of C’s laisses in the first repetition were in the CV7 model, since V7 copied them

The Relationship between C and the other Texts

63

in error and took some hemistichs from them, in preference to the PC rhymed exemplar. It also identifies the changes made by V7 to the CV7 model as exemplified in C and their links with an exemplar from which (the much later) T derives, an exemplar known also to P on the evidence of the second cluster of assonance. In the latter, the insertion of the 4 additional laisses peculiar to V7 provides not only strong evidence of V7’s role as agent of change but also of C as representative of the CV7 model and an earlier copy than V7 of that model. In both clusters, there is consistently close match between C and V7 where they have the same rhyme, showing that they share the same model; but there are also many closely matching first hemistichs in laisses where the rhyme varies or C is assonanced, indicating the same shared model modified by V7 to perfect the rhyme. The features identified here in the two clusters of assonance are also found occasionally but consistently in the rest of the text: in modifications made by V7 to the earlier version (as found in C and the CV7 model), in «correcting» alexandrines to 10-syllable metre, in adding extra lines and laisses not found elsewhere and in generally «improving on» the CV7 model. Nonetheless, after the second cluster the CV7 text is clearly based on a different redaction from that of PTL and close correspondence is re-established only in Section C of the text, which opens in C with line 5887: Granz fu li deus la nuit en Rencesvaus. The episodic nature of CV’s correspondence with the acephalous PTL becomes clear only after C’s first repetition, i.e. from C 130 on, when the first sequence of battles commences. The discrepancies between the CV7 and PTL readings in the «end of episode» laisses cast fresh light on the reasons for V7’s insertions of ornate oversize capital letters additional to those found in C – and also underline the later date of V7, compared with C. Comparison of the different episodes in the various versions reveals no sign of a single, continuous, rhymed redaction having existed until one reaches Section C. There is no evidence that a complete Roland Rimé existed prior to V7. On the contrary, all the evidence points to C’s being the earlier manuscript, so that C’s text represents the CV7 model, with V7 the innovator and «improver» of that model. The historical evidence of French trouvères at the court of Charles d’Anjou, King of the Two Sicilies, gives a possible source of the rhymed exemplar used later by V7; the Gonzaga evidence supports this. The implications are far reaching: not only a re-evaluation of the authority of C’s readings, whether single words, hemistichs or lines, and a re-examination of the reasons for C’s assonanced laisses and its repetitions, but also the need for a revision of the Segre thesis. This throws into question the validity of Duggan’s edition of CV7, since it is based on the reconstruction of the CV7 model in accordance with the Segre theory. Finally, it demonstrates the need for the present annotated edition of C.

64

Introduction

I. Provenance of the Manuscript Documentary evidence post-1407 The history of C is for the most part well-documented and well-known, from the 1407 inventory of the Gonzaga Library in Mantua, as detailed in Braghirolli/Meyer/Paris 1880, to its purchase with other Gonzaga manuscripts by Giambattista Recanati in 1707/08, on the death of the last Gonzaga Duke of Mantua. Recanati, a wealthy Venetian bibliophile, bequeathed 216 volumes to the Republic of Venice63 on his death in 1734. The Venice 4 and Venice 7 manuscripts, plus other Old French manuscripts with Gonzaga provenance, thus moved from the Gonzaga library to the Recanati library and so to the Marciana with very little disturbance, enjoying ideal conditions. C, however, was bequeathed by Recanati to his friend and fellow-bibliophile, Jacopo Soranzo. It is (only just) identifiable from the laconic one-or-two line entries in the Recanati catalogues of 1722 and 1734, but fully described as item CCXLV in Vol.III, Section II of the Soranzo catalogue, in a volume entitled Catalogo de’ manoscritti in-4o: Comincia dal CCXI fino al CCCC. A’ di 14 settembre 1746. Fatto da me Francesco Melchiori da Oderza.64 The detail given leaves no doubt that this is the Châteauroux ms. In the transcription of this entry given below, the continuous text has been spaced for ease of reading and the crossings-out inserted, with the replacement text shown in superscript. Romanzo in lingua provenzale antica francese in versi intitolato Roncesvalle: questo è un Poema Romanzesco intorno le guerre de francia e la gran battaglia seguita a Roncisvalle, dove per frode di Gano de Maganza perì il fiore de’ Palladini di Francia. Com:senza alcun titolo Charlle li rois a la barbe grifaigne Sit anz toz plenz a este en espaigne Conquis… f.(= finis) Et au terz jour de mort resuscitez. Deo gratias. Amen. Explicit Roncisvali e de .R. e doliver e de aude.,

 Cf. Zorzi (1987, 250–252) and footnotes (489–490). For other useful details of the Recanati mss. later known as V4 and V7, cf. Bisson (2008, 17–19 and 31–33).  This catalogue entry (245) was transcribed and printed by Rossi (1930, 284), but as a continuous text with some of the deleted words given in footnotes and without any mention of the Châteauroux inscription. The lay-out and corrected detail are critical to recognition that the two entries were made at the same time, by Melchiori, his assistant Antonio di Santo or Soranzo himself. A photocopy of the Châteauroux end-paper has been compared with the actual Soranzo catalogue entry: to a non-expert eye, the writing seems identical.

Provenance of the Manuscript

65

vide che si spiegano probabilmente sic fortasse integra scriba cìsè fortasse nobili fra magiorentes (?): explicit Roncisvallis et de Rolando (id est Orlando) et Doliver (id est Oliviero) et de Aude, qui i quali tres insigniores Duces Galli in hoc poema fortasse laudantur. Codex membranaceus perantiquus seculo scilicet XIII, aut paulo post exaratus, ut videbitur ex maximis initialibus in tres Libros divisus, quorum quisque in strophas nunc longiores, nunc breviores. Inter optimos nostros Codices ennumerandus. Altri due Poemi Romanzeschi in lingua antica francese abbiamo notato neli Codici infoglio 763.806.65

This is exactly matched, in abridged form, by the inscription on the third page of the folded end-paper inserted into C, cf. Duggan (2005, III/18–19, n. 4). This reads: anzi in Lingua antica francese

Romanzo Provenzale intitolato Roncisvalle, cisè Poema Romanzesco intorno la guerra di Francia e la battaglia seguita a Roncisvalle dove perì il fiore de’ Paladini di Francia – per frode di Gano di Maganza. In fine leggesi: explicit Roncisvali e de .R. e Doliver e de Aude. forma quellà .R. significa Roland, cìsè Orlando, ed Oliviéro, due comes de’ Paladini qui nominati.

This is identical to the Soranzo catalogue entry; it is in Italian in the same distinctive flowing hand, with the same insertions and corrections, so it is likely to have been written by Melchiori or his assistant, Antonio di Santo, on the same date – and possibly corrected by Soranzo, who must have known the contents well from his discussions with Recanati. Since there is no record of how or when C entered the French royal library, and since C and V7 were not the only copies of this text, the inscription is a valuable confirmation of the identity of the ms. and of the evidence of the binding. Soranzo’s habit of  This catalogue entry (245) was transcribed and printed by Rossi (1930, 284) but as a continuous text with some of the deleted words given in footnotes and without any mention of the Châteauroux inscription. As may be seen above, the lay-out and corrected detail are critical to the recognition that the two entries were made at the same time. A photocopy of the Châteauroux inscription has been compared with the actual Soranzo catalogue entry: to a non-expert eye, the writing seems identical.

66

Introduction

having his manuscripts rebound, with the detailed catalogue entry summarized in similar wording on a fly-leaf is confirmed by Mitchell (1969, 129). This provided an easy means of identification. These papers pre-date the purchase of the MS by Louis XVI and each folded sheet is firmly attached within the «new» binding for the French royal library.66 In the Recanati catalogues, C and V7 are classed as in-quarto, despite their smaller in-octavo size; and the Recanati catalogue of 1733/34 refers to C, V7 and V4 as fourteenth-century manuscripts. However, Melchiori suggests a thirteenth-century date for C; and this was adopted as correct when C was rebound for the library of Louis XVI of France. Melchiori assessed C as being «worthy of enumeration amongst our best manuscripts», so it seems unlikely that an avid collector such as Jacopo Soranzo would have parted with C in his lifetime. When Melchiori retired in 1748, he noted that the library contained some 4,370 manuscripts, about 2,000 of them uncatalogued. Moreover, the Soranzo catalogue, though detailed, was not divided according to language or subject-matter (unlike the earlier Gonzaga and Recanati inventories); it seems to have been a chaotic mass of manuscripts on a wide range of subjects, grouped only by size. After 1748, there is no record of any further cataloguing being done. Zorzi (1987, 344) confirms that, on Soranzo’s death in 1761, ereditarono la sua sostanza i nipoti Corner della Ca’ Grande a San Maurizio e Zorzi a San Severo. I Corner vendettero la loro parte all’abate Canonici; Marin Zorzi… vendette 220 codici a Teodoro Correr e gli altri, non si sa quando, probabilmente allo stesso Canonici, che finì con l’avere la maggior parte della raccolta Soranzo.67

C (No.245 in the catalogue) would have gone to Corner, who was the first to sell his manuscripts to Canonici. Matteo Luigi Canonici, born in Venice 1727, studied in Bologna and became a Jesuit priest. After teaching in Ferrara, then further study in Bologna, in 1761 he became a lecturer in Parma. He had a high reputation as a Latin scholar

 Duggan (2005, 18 n. 4) draws attention to the slip of paper glued to the verso of the second guard folio and bearing a translation of the summary in Old French on the opposite page. Did the impression at the top in fact read ‘Q.–G…’ (not Q.–C), i.e. Quartier-Général. Dépôt. Réserve (or Réservé) *Royal- : possibly signifying «Headquarters, Registration of incoming items intended for the private royal library (at Versailles)»?  For further details of events subsequent to Soranzo’s death, see Zorzi (1987, 290, 299, 320, 360, 385 and also 344 and 524 n. 269). These, plus the authoritative book on l’Abate Matteo Luigi Canonici and his library by Merolle Tondi (1958, 34s.), correct the under-estimation of the number of mss. made by Rossi (1907, 131s.) based on the figures stated by Moschini (1806, 60).

Provenance of the Manuscript

67

and antiquarian, amassing an interesting collection of medals and manuscripts which greatly impressed the abbé Barthélemy, the royal Garde des Monnaies et Médailles in France and also a noted antiquarian, when he visited Parma in 1757. This collection had to be abandoned when the Jesuits were expelled from Parma in 1767. Canonici eventually returned to Venice just before the Soranzo manuscripts became available. Soranzo’s library contained all manner of manuscripts: reports made by Venetian ambassadors or Papal nuncios, accounts of conclaves and Papal elections, archives relating to regional history, besides works of literature and biblical and theological texts of all kinds. These were categorized according to size, but otherwise unsorted. In contrast, Canonici was a connoisseur, a noted Latinist and antiquarian, and his library had a double focus, theological and literary: Merolle Tondi (1958, 49) divides his manuscripts into some 1,800 sacred and l,730 profane, the sacred including Bibles and Commentaries, early theological works, ecclesiastical writings, and the profane comprising classical Greek and Latin authors, Italian fourteenth–eighteenth-century literature, Venice-related documents, plus a few Arabic, Turkish, Persian, Chinese and Indian manuscripts. As a very cultured, highly educated expert in the field of manuscripts, he was also an experienced dealer with a wide range of contacts (in Constantinople, Flanders, Poland, Vienna, France and England). Most of the Canonici letters in the British Library, and particularly those to the Abate Carlo Maria Masnago,68 relate to the sale and purchase of manuscripts. In a letter dated 5 September 1778, Canonici announced with quiet but tangible satisfaction that sono in un mare dove può pescarsi assai and followed this on 18 November 1780, with ho scoperto molti tesoretti che saranno miei fra pochi giorni.69 On 31 March 1781 (Merolle Tondi 1958, 33 n. 52) he was writing to his friend Paciaudi: questo è un paese nato fatto per un pazzo del genio mio; rari sono i giorni in cui non se offrano incontri di buoni acquisti, ma il denaro manca. As Merolle Tondi suggests, this must surely relate to the appearance on the market of the Soranzo library, probably the Corner inheritance, but possibly also with the potential Zorzi share in view. The only problem was money.

 Lettere Bibliografiche Autografe deli Abate M L Canonici, 1788–1793, British Library shelfmark Add. 26.059. See also Merolle Tondi (1958, 33) for an illuminating analysis of Canonici’s sophisticated expertise as a dealer, as shown in his letters to emissaries.  Cf. Merolle Tondi (1958, 33 n. 51), relating to the Soranzo mss. coming or about to come on the market. This is followed (Merolle Tondi, 1958, 33 n. 50) by what must surely be the initial purchase from Corner some 2 years later.

68

Introduction

Canonici therefore set to work to evaluate each item he had bought from Corner and to decide which he should sell (and to whom). One finds him writing in a letter dated 20 September 178370 that Il prezzo in contanti di mia raccolta potrà valutarsi; le fatiche, gli affanni, i maneggi, il carteggio, la schiavitù sofferte per essa nessuno la potrà valutar mai and, on 23 June 1784,71 regretting that il mio dolore è di non poterne pubblicare un catalogo ragionato perchè non ho aiuto è sono troppo distratto in mille cose. Merolle Tondi (1958, 36) relates this to the evaluation of some 2,000 Soranzo manuscripts. On 5 September 177872 Canonici had earlier written to Paciaudi of i miei codici, i quali sormontano il numero di 3 mila and later letters record his library as having increased by over a thousand volumes by 1784 (after purchase and «sifting» of the Corner volumes). In another letter to Paciaudi, dated 17 January 1783 (Merolle Tondi 1958, 39 and n. 77), he affirmed, L’acquisto di un solo pezzo mi costa, oltre il dinaro, industrie, raggiri, lettere, brighe infinite. However, in recompense he recorded in a letter of 13 November 1783 to De Rossi (Merolle Tondi 1958, 39 n. 81) the beautiful Greek and Latin mss. he had acquired. More specifically, if Soranzo’s Rencesvals (i.e. C) formed part of the earlier purchase, then Canonici would be likely to sell it, however reluctantly, because it was extraneous to his theological and classical themes and so that he could purchase other manuscripts of greater interest and value to him. Moreover, the dates would fit. In a letter dated 20 December 1788 to the Abate Masnago,73 Canonici gives a short list of sixteenth- and seventeenth-century books, ending with Lorenzo Medici’s Stanze alla Contadinessa (1612). He then continues: Io la stima più che Carlo in Francia, ma le saro ugualmente obbligato se l’esito non corrispondera alle ricerche. Io con quei pezzi acquistarsi un codice prezioso. There is no Medici poem relating to Carlo in Francia, so one wonders whether this might possibly be a veiled reference to the Châteauroux manuscript (opening Charlle li rois..), sold at a very good price to the King of France. (Canonici rarely speaks plainly of his dealings.) There is otherwise no mention of C in any of the Canonici letters74 up to 1790 – which must surely be the final date possible, allowing

 Cf. Merolle Tondi (1958, 36) with letter quoted in footnote 61.  Cf. Merolle Tondi (1958, 36) quotation in footnote 62.  A letter to be found in Parma, Biblioteca Palatina, carteggio Paciaudi.  Lettere Bibliografiche Autografe deli Abate M L Canonici, 1788–1793 (British Library ref. Add. 26.059).  The Letters to J Strange, British Consul in Venice, British Library ref. Add. 23730. f. 349 and e.g. 2002. f. 149 contain just two letters from Canonici, dated 1780 and 1782, the first thanking Strange for his help in the acquisition of a manuscript, the second on a quite different subject.

Provenance of the Manuscript

69

for the French Revolution and time for the manuscript to be re-bound in Paris. Nor is C in the list of manuscripts attached to the sale catalogue75 of 1780 (unless it can by any chance be identified as Rime Diverse senza nome di Autore con alcune Latino nel fine). With revolution brewing in France, the transition of the C manuscript from Venice to Versailles appears to have passed unrecorded – unless one counts the embossed sheet of paper bearing the translation into French of the summary description on the third page of the end-papers (see Duggan’s footnote, 2005, 19). Otherwise, only the care taken to rebind it and place it in the King’s personal library at Versailles implies that its significance was recognized. This might possibly suggest a continuing link between Canonici and the abbé JeanJacques Barthélemy, the royal Garde des Monnaies et Médailles, whom he had met in 1757; but no evidence of this exists. It is striking that, in France and particularly Paris at that time, there was particular interest in the stories of the Charlemagne cycle, as shown by the success of the monthly publication, La Biblothèque universelle des Romans, founded in the 1770s by the Marquis de Paulmy. In the November and December issues of 1777, the latter recounts these stories as related in the Chronique de Turpin and the wildly fantastic poems of Italian poets such as Boiardo and Ariosto;76 in March 1779, he bemoans the disappearance of all trace of the Chanson de Roland which must have existed previously (and prints his own «song»). Paulmy was Minister of War under Louis XVI and, a noted bibliophile with a passion for Old French literature, sold his library of manuscripts (housed in the Arsenal) in 1785 to the Comte d’Artois, the future Charles X, who allowed it to remain in situ with Paulmy until the latter’s death in 1787. (This collection was sealed off during the Revolution and formed the basis of the Bibliothèque de l’Arsenal.) It can have been only between about 1780 and 1790 that the Châteauroux manuscript was purchased for the private library of Louis XVI at Versailles. It is not recorded in the catalogue of 1775/76,77 made shortly after the king’s  This is the 3-volume Catalogo de’ Libri che saranno posti in vendita nella città de Padova l’anno 1780, Marciana reference 118 C.171–173. The 8 pages of manuscripts represent nowhere near the 2,370 in the 3-volume Soranzo catalogue and far less than the 4,000+ in the Soranzo collection.  These are the sources Paulmy himself cites, but a more complete list of post-1350 Italian poets with an evaluation of their poems, plus translations into French and transpositions into opera or plays, is given by Faral (1932, 286–297). He also indicates (1932, 296) additional texts used by Paulmy, adding: De la vraie «Chanson de Roland», point de trace.  This hand-written catalogue is to be found in the Bibliothèque Nationale de l’Arsenal in Paris under the reference: nouvelles acquisitions françaises 2622; Arsenal ref. nos. 5381–5399 for the private libraries of the royal family; vol. IX (5389) relates to Louis Capet (Louis XVI).

70

Introduction

accession; and there was sufficient interest in its subject in the 1770s to ensure it would have been noticed. Given his links with the royal family, Paulmy would surely have recognized – or at least been intrigued enough to investigate – the title Roman de Roncevaux placed on the spine of the Châteauroux manuscript when it was re-bound for Louis XVI’s library. Taken with the evidence in the Canonici letters, this implies that the ms. was bought from Canonici in 1787 or 1788. If so, the time taken in binding and gilding could have delayed its appearance in the library to 1789, when all eyes were on political events. The manuscript certainly reached the king’s private library, for it is listed in Citizen Buffy’s 1791–1795 inventory «des bibliothèques nationales établies à Versailles dans les dépôts de la Commission des arts du département de Seineet-Oise…», in vol. IX: Catalogue des livres provenant de chez Louis Capet, condamné, district et comune de Versailles, Dépôt C.78 Under Manuscrits. Belles Lettres, page 468, one finds against No. 2038: Romanzezio di Roncisvalle, 1 Vol., Grand.in-8o. V.E. Vel. Poésie. 14e Siècle. Manuscrit Gothique. The term Romanzezio is borrowed from Melchior’s description (cf. Introduction, 64, 65) but corresponds to the title Roman de Roncevaux on the spine of C; the size is correct according to standard measurements; the leather of the cover is indeed Veau Ecaille, the pages are of parchment and the script is gothic; only the date is surprising, unless the over-worked Buffy mis-read XIIIe SIECLE on the spine as either XIIIIe or XIVe SIECLE. Certainly the eighteenth-century binding with the royal monogram on the spine confirms that C was once in Louis XVI’s cabinet privé at Versailles. The fact that Buffy recorded it so briefly may indicate the pressure under which he was working (for he catalogued over 120,000 volumes amongst those amassed at Versailles during the Revolution) or it may reflect a personal appreciation of incunabula or other works, rather than mediæval manuscripts; but this terse entry may reflect a dismissive reaction to the amount of wear and damage visible on the parchment.79 What happened next is not clear,80 but on 14 March 1822, it was listed as No. 571 in the Sale Catalogue of the effects of Count Germain Garnier, Prefect  The condamné in the title implies a date after the Court’s verdict in December 1792 and before 21 January 1793, but the entry itself may have been written later in 1793.  The mottled calf of the cover, brand new at the time, also shows signs of harsh treatment. On 26 vendémiaire, an IV (17 October 1795), the architect Leroy noted that the books still in place in the two libraries of Louis XVI at Versailles were affected by damp during foggy weather and during thaws. This is reported by Guilleminot-Chrétien (2003, 112). She comments that this shows that the books were still in situ.  According to the Mortier edition, C was sold to Garnier during the Revolution as a bien d’Etat, but there seems to be no documentary proof of this. Guilleminot-Chrétien (2003) mentions that Lucien Bonaparte as Ministre de l’Intérieur gave away some volumes by way of

Orthography and Principal Linguistic Elements

71

of Seine-et-Oise (chef-lieu Versailles) from 1800 under Napoleon I, and bought by the Swiss bibliophile Jean-Louis Bourdillon, whose evaluation of the MS is quoted at the beginning of the Preface. Bequeathed by Bourdillon to the town of Châteauroux (Indre et Berry), it has been carefully preserved there since 1856. What is indeed particularly striking about the history of this manuscript is how carefully it, like V4 and V7, has been preserved for some 700 years in a sequence of libraries. Even during the French Revolution, it remained peacefully (if not always comfortably) at Versailles.

I. Orthography and Principal Linguistic Elements Orthography The text is written in miniscule cursive gothic script, square in shape, with a neatness that initially masks the errors of content and the frequency of scribal (and revisor’s) corrections. In the Introduction to his (1944) edition, Mortier endorsed Foerster’s opinion that the angular shape of the letters indicated a French copist, since a more rounded form is found in Italian MSS, but closer study of the regional characteristics of the language disproves this. It is the general opinion that C was the work of a single copyist and Foerster (1883, vii) quotes Delisle’s opinion: «L’écriture ne paraît pas être celle d’un copiste habitué à copier des manuscrits». The ms.’s typical thirteenth-century fluidity in the graphies of words is particularly evident where C 2669–2673 repeat 2663–2666 + 2668 verbatim, but with minor variations: Olivers/Oliver, feron/ferons, dolerose/dolerouse, barguegne/bargaigne, toz/tuit, mosrons/morons, recovreigne/recovregne, genz/gent. There are a small number of special orthographical features worthy of note: From C 3965 to the end, the copyist very frequently writes the combination iu as if it were ui, marking the final vertical stroke of the 2-vowel combination

grants or payments to individuals and also that in 1801 there still remained books at Versailles which were considered readily marketable. It is not clear how Garnier acquired his manuscripts, but the damage suffered by the new eighteenth-century calfskin covers on both C and the Partonopeus suggests that he may have rescued both mss. from destruction. Since Germain Garnier was in self-imposed exile in Switzerland from August 1792 to 1795 (and his brother Charles-Georges-Thomas had withdrawn from Paris to be commissaire du pouvoir exécutif in Auxerre from 1793 to his death in 1795), it seems most likely that his Roncesvals was acquired after his appointment as Prefet in 1800.

72

Introduction

with the same light, right-slanting upward slash he uses to indicate an i. Thus, lui = liu is found in line 5856 and luis = lius in 5668, 5878, 6096, 6199, 7060 and 7408; sauies (savies) = saiues (saives) in 6787, 6801, 6827; Blavie(s) = Blaive(s) in C 6482, 6483, 6522, 6826, 6835, 6985; suiront = sivront in 3965 and suianz = sivanz in 4283; glauies = glaives in 7933. In some cases (e.g. lui) it is the sense which makes emendation essential, in many other cases (especially Blavies and savies) it is the metre. This carelessness in positioning the indicator may well be the result of fatigue. Its occurrence in roughly the second half of the text may either reinforce the view that C is the work of a single (and by then weary) copyist or indicate a change of copyist. (Indeed, Blavie could be read as Blauie or Blauje, an Italian copyist’s rendering of an unfamiliar name.) In the case of iut = jut, the correct marking is used in almost every case: C 1064, 3834, 3888, 3958, 3998, 4025, 4139, 4338, 4395, 6066, 6152, 8019, so uit in C 4018 and 7083 are probably due to scribal error. Since ui for iu is so prevalent, this graphy has been consistently emended (with footnote) in all the cases cited above. What can only be described as an extremely large cedilla is attached to both c and t upon occasion: in respect of c, usually in the middle of a word, but occasionally as the final letter; in respect of t, usually at the end of the word. Both these graphies appear to represent z, the ç being the soft fricative and the t + cedilla a harder, dental ts/tz; they have both been transcribed simply as z. However, the graphy -tz has been retained. This is found as -antz at the end of C 832, 3442, 3606, 3609, 3610, 3613, 3615, 4285 and 4286, in laisses rhyming in -anz, but also in 832 and 3464 earlier in the line, as a singular subject form, a variant of -ants; it also occurs in agutz, at the end of C 2728. (A very much lighter, even «whispier», «tail» attached occasionally to a t or even a final r has been ignored as purely ornamental.) According to Pope (1934, 290, § 739) the writing of c with z under it to denote the value ts was initiated in Spain, then spread to Italy and only in the late thirteenth century reached the south of France. If this chronology is accurate, the above graphy must be accounted a FrancoItalian feature inserted by the Italian copyist. Whereas g was regularly combined with n to indicate palatalisation, when the sound was final (and occasionally elsewhere), it was appended or written alone. Thus the graphies poing, puing, poig and puig found in C may be seen as valid alternatives to puign in the thirteenth century; and the value of g where -gn or -ng is expected must remain in doubt, as indicated in the relevant footnotes.

Orthography and Principal Linguistic Elements

73

There is also some confusion of rs and ss within the MS and specific instances of the assimilation of r to s where this is needed for the rhyme. Cf. Editorial Policy, below, p. 90. Laisse 331 (C 5887–5925 inclusive) is written with a rhyme in -ax (= -aus). This graphy is a stylistic feature of C and the laisse itself represents a poetic tour de force, so the graphy may possibly be a surviving feature (like the alexandrines) of the original Section C redaction; but -x quickly became a standard abbreviation for -us and is treated as such here.81 In C’s script, it is at times difficult to distinguish c from t, s lunga from f, and the d with its left-curving upper stroke is also easily mis-read. There is considerable variety in the capital letters: E is found in the normal squared-off angular form immediately below a «round-backed» E with simple central bar; two different styles of S and M are found; and very occasionally the abbreviations for et and con are used as capital letters to begin a line. The capital letter Y has posed a special problem: the names Ysaïe (C 3811) and Ysorez (6326) are found, but also Yhesu Crist (C 657, 3973, 7758) and simply Yhesu (C 1633, 2031, 4178, 5774, 7247, 8028), Yhesus (C 2121) alongside Ihesu = Jhesu (C 3684, 3696, 3967) and Ihesus = Jhesus (C 2585, 2659, 3809, 5039). Should Y then be read as an alternative graphy of I, especially where Yh = Jh? (The letter y is not otherwise found as an initial letter, although it occurs occasionally in lower case throughout the text: jamay, moy, loy, etc., plus the various graphies of Naymes). Finally, the Yh- graphies have been retained as a distinctive feature of C.

Principal Linguistic Elements: Regional Characteristics The manuscript has certain distinctive, but generally infrequent and randomly scattered, phonological and morphological characteristics. Some of these point to a source document originating in Northern and North-Eastern France, whereas others indicate a text copied in northern Italy. There are also a few elements typical of Eastern and particularly South-Eastern France, a logical geographical link between the two areas. The occurrence of these characteristics apparently has no correlation with the different episodes of the poem or with the different degrees of redaction identified in it. The following analysis is based on Pope (1934, 486–497, §§ 1320–1324) and on the Introduction to Bogdanow’s edition of La Folie Lancelot (1965, XXXIX–  The use of -x (= -us) began in the late twelfth century in vernacular words in which the vowel preceding -us still retained its value, but in the thirteenth century its use was rapidly extended to all words ending in -us.

74

Introduction

XLVIII) with its detailed enumeration of Franco-Italian (and other) characteristics. This has been updated and enhanced by reference to the analysis of language by Zarker Morgan (2009, 19–49), to Holtus (1985, XLVIII and LI–LXVIII for analysis of examples) and to his more recent analysis of the four phases of Franco-Italian (Holtus 1994, 149), plus a list of 65 ex-Gonzaga mss. (Holtus 1994, 150–160) and examination (161–168) of 7 of these (some earlier, some later in date) with varying degrees of Franco-Italian.

A Features peculiar to Northern and North-Eastern French Phonology (i) Reduction of triphthong ieu to iu in lius, 2679, 2770, 5596, 5668, 5878, 6096, 7060, 7408. (ii) Differentiation of the triphthong eau to iau in the graphies castiax, C 278 and 1025, chastiaus 4517, biaux 5910, piaus 5917 and Biauliant 6296. (iii) The use of ceaus 6 times, ceus once only, may show late retention of -eau diphthong. (iv) Reduction of unstressed feminine possessive adjective, ma > me, is not found in C. The possess.adj.masc., men, is found solely in men esciant C 348, 826, 845, 1113, 1487. (v) A small number of examples of retention of intervocalic group -nr- without denasalisation or development of interconsonantal glide are found in C: avenra 1968, prenra 372, remanra 3 times, tenrai C 448, tenra 7 times, tanra C 4343, tenrez C 50, venra 10 times, vinrent C 6497. (vi) There are 68 examples of the Northern French future forms arai, aras, ara, arons, arez, aront in C, compared with 40 examples of the standard Old French avrai, avras, avra, etc. Both occur in a random distribution (cf. Glossary for details). In contrast, the future of savoir has the standard Old French savr- stem except for the following cases: sarai C 6677; sara C 3312, 3516; sarez C 8070 and saront C 6845. Palatalisation of countertonic e to i is found in chival C 1274, 2040, 2781, 8151, compared with 54 instances of cheval. There are thus over 100 examples of phonological features peculiar to Northern and North-eastern French, but 68 of these relate to the future of aver, an extremely common verb.

Morphology (vii) The indirect object pronoun, mi (< mihi) is found in C 378, 380, 3682 and possibly 435 and 7623.

Orthography and Principal Linguistic Elements

75

(viii) The use of fisent C 1533, 2865, 7289, 7358, 7853, misent 1510, 1990, 4936, 7736, prisent 1943, 3735, 3746, quisent 7149 instead of standard Old French fistrent/firent, mistrent/mirent, pristrent/prirent, quistrent/quirent in the past definite 6. (C 1990 repeats C 1510 almost exactly.) (ix) The form celison C 4668 and past participle cellie C 246, both relating to the verb celer/ celler/ celeir/cellir, attest to the reduction of the first conjugation termination -ier to -ir, but this is rare in C. (x) The present indicative 1, morir, moert C 1793 and muert 1973, shows the Northern -ts ending (graphy ch or c), plus a (Franco-Italian?) palæographic confusion of c and t. There are just 19 definite examples of morphological features peculiar to Northern French – but a considerable number of phonological and morphological features in C which are common to both Northern and North-Eastern French and to Franco-Italian.

B Features common to Northern and North-Eastern French and to FrancoItalian Phonology (i) The diphthong ai is found reduced to a in: masnie C 244; masnee 1042, 3200, 4558; and or n’a je mas 7704; remanez 416; remanra 859, 1284, 8102; remandra 5323, 5376; remanrons 1480, 1958; remandrez 7186; remanront 2540; repara (past def. 3) 1211; lasse (pres. 3) 1498, 2170, 2396; (imperative) 6392; lassera (fut. 3) 4615; lassa (past def. 3) 1486, 2626, 2948, 4379; lasserent (past def.6) 4485; lassee (p. part.) 6686; lassié (p. part.) 7528; lasser (infin.) 7327; mantenir 2144; mantenuz 2906; mantenoient 4187; plast 3455, 5120, 7209, 7404. (ii) Open e, blocked, is diphthongised to ie in the adjective fier/fiers 56 times, compared with fer/fers 15 times (possibly also in siens 2632 and 2712, although scribal error is possible). This is characteristic of North-eastern French but not unknown in Franco-Italian. (iii) The use of e for tonic o is found only in the phrase men esciant, but men, ten, sen are Northern French forms (cf. § A (iv) above). (iv) The only evidence of absence of breaking is: orgoillos(e) 38, 1153, 3607, 4302, 4419; orgoil 2028; orgoillosement 5129 – but orgueillos 4882 and recoillent 3201. (v) Reduction of oi to o is found in C as follows: angoiser 3 times, angoissant twice, but the stem angoss- or angos- 4 times; cosir, infinitive, is found in line 6087 (but coisi, past participle, 4115, 4429, 7163); scosie 742 (but

76

Introduction

scoisi 4121); crostre 1358, 6080 and the stem crost- used in the present and future (but croistre in 6516, crestre in 6534 and croissant in 6123); crot 1912 for croit, levot 6741 for levoit. (vi) Retention of velar k before Gallo-Roman a and i (very common in both Northern French and Franco-Italian) is found in: cange 1460, 3953 and cangier 2015 but changee 5484, changer 7095; similarly cassé 580 but chasez 444, 668 and chasé 6991; canz 1068 but chant 4909, canzons 1388 but chanzon 3628, 4577, 5198; carpentiers 1162 but cherpenter 3825; catheigne 4237 but cheveteigne 4228; caroler 6858 but charoler 6868; car («dear») 2142 and qier 475 but chier in all other cases; escaufé 477; caut (= chaud) 3523, 7930; caucha 908; encaucer 3739; encauchent 2472; encaucent 2810; escanpez 457; escanpa 1056; escanpe 1101; escanp 6024, 6110, 6430; escaper 1279; escapez 2510, 6322; escanperont 3027 – alongside the stem eschap- in 3135, 4274, 4811, etc.; descauz 5163; escaigne 1222. (vii) There are 6 examples of metathesis of r and a vowel: sen trage for se targe (probably) 547; grepis for gerpis (guerpis), 715; estromie for estormie, 747; orie flor/orieflor = oireflor = oreflor 1237; sevreron + metathesis produces severron > sevesro 6043; entres for enters 8102. (viii) Intrusive inorganic n, a common feature of Northern and North-Eastern French and Franco-Italian, is found occasionally in C: escanpez 457, 2510; escanpa 1056, escanpe 1101, escanperont 3027, escanp 6024; sen trage (= se targe) 547; deslongent (= deslogent) 1023; canploier (= chaploier) 2016; ratranzon (= ratrazon) 2139; ensir (= essir) 2150; tanlent 3500, garnir (= garir) 3585–3586; forfaint (= forfait) 4113; seint (= seit) 6367; porine (= porrie) 7192. (ix) The s impura without prosthetic e, very common in North-eastern French and Franco-Italian, is found about 18 times: scherni 702; schine 2328; scoisi 742, 1843, 1886, 2323, 4081, 4121; smarie 7177; spee 1360, 1449, 1636, 1843, 1886; speronee 1771, 4247; spiez (possibly in 2522, definitely in 3479); spines 6088; spoisse 6763. There are thus at least 170 examples of phonological features common to Northern and North-eastern French and to Franco-Italian.

Morphology The only features common to both areas are: (x) The dropping of final -t of a verb ending after n, r, s, found in Conquis line 3, chevaucen 1796, fier 2114, tan 2937, mis 5174, conper 7616, but in C

Orthography and Principal Linguistic Elements

77

this is sometimes for the sake of (visual) rhyme: von 3286, retorneron 3287, guerpiron 3284; (xi) The use of mi (< mihi, indirect object) is found in aler mi efira 378 and aler mi istovra 380. (xii) The form si (= reflexive pronoun, se) is found certainly in 2092, and probably in 1507 and 3305 (where si, adverb, «so» or s’i may be intended). In 700, C has s’est in an hypometric first hemistich; V7 reads si est, correcting the metre.

C Features Peculiar to Franco-Italian Phonology (i) Retention of a unchanged is found in C in cals (= cels) 3942. (ii) No diphthongisation of tonic open e in a free syllable: ben is found 72 times, bien 209 times (both very evenly distributed throughout the text); ren occurs 5 times: 362, 4260, 4329, 7070, 7389 but rien 4 times, 1189, 2020, 3258, 7284; tenent is found once, 7279. (iii) Replacement of tonic e with o: possibly found in l’on for l’en, 5028; more frequently non for n’en. Nevertheless, although non is found in 267, 380, 467, 885, 2077, 2110, 3212, 3348, 4445, 4446, 7539 and 8058, it may not always stand for n’en; and 3212 reads: nen plor ou non plange, probably indicating that these are standard alternative graphies at the time. (iv) Whereas there is frequently no diphthongisation of tonic open o in a free syllable in Franco-Italian, in C duel is found 91 times, but dol in 3927, 7369 only. (v) Absence of diphthongisation of closed tonic o in relation to -or is so widespread in literary Old French that its occurrence in C has been treated as normal; only the monophthong in solement 3315 and soe 2116, 4181 has been noted as a specific feature (but soule is found in 1296). (vi) Closed tonic o written u in mult: in C, there are 58 instances of mult, one instance of molt, but 409 of mot, 27 of mout (all in the first half of the text) and one of mut (1852); similarly, pruz is found once, prou 4 times, but proz 30 times, pros 3 times, preuz once, preu 8 times. On the other hand, as an example of closed tonic o becoming u before a nasal, sunt is used consistently throughout C (206 instances), sont once (6419) and son once (2552). In addition, volunté is found in 586, 4796, 8047, 8052; nun for non («name») in 7270; uncle once (6753) but oncle 24 times; unqes (1289, 5396, 7120, 7335), but onqes 48 times and onques

78

(vii)

(viii)

(ix)

(x)

(xi) (xii) (xiii)

(xiv)

Introduction

once (3231); but funt for font occurs only once (2975). The future 6, ouserunt (1949), is rather strange, since one would expect the conditional, ousereient or ouseroent, or the imperfect subjunctive, ousassent. The only examples of oi replaced by e are aver 1924 – but avoir, infinitive, 10 times and substantive, «wealth», 12 times; and poer 943 for pöois (cf. pöoir, 5339, 5341, 5342). The triphthong eau is reduced to au or eu only in the following cases in C: haume 5380, heume 3 times, eume 62 times (+ elme 5 times); bauté is found 7 times, jovençax (= jovençaus) in 5896 only; castax (= castaus) was written in line 25, then i added superscript as a correction. There are 14 instances of intertonic e becoming a in C: one past participle, achaté 7536, and the adverbs: isnelament 3955, 7790, isnelamant 8088, sauvament 4682 and belament 7469, plus the substantives: parlament 4852, demorament 4860, 7414, venjament 5051, confortament 5053, devisament 7123, demandament 7145 and serament 7848. No examples are found in the first half of the text (Section A) and none within the closely assonance-based passages, not even in laisse CCL. The very common practice in Franco-Italian of replacing a final e with an a is found in only 12 instances in C. These are the adjectives: tuta 433, bona 1609, bella 3517, plus quaranta 841; the substantives: batailla 1146, arma 1634; the preposition, da for de 1491 – but possible contamination: da la lance; the verb forms: era, imperfect 3, 1296; perda, present subjunctive 3, 2008; josta, imperative sing., 7558 (as in V4); demora, present subjunctive 3 (?) 7215; and possibly scenbla 1423, either Italian present indicative 3 or French past definite 3. (Most of these are found in Section A, but only da, era and scenbla within assonance-based laisses.) The only instances of final e > o are: no for ne, 235, 1540, 3318, 3559, 8133, and Charllo Meine 4145. Secondary stressed o > e is found in estages (= ostages) line 51 only. As regards consonants, the common Franco-Italian ch for c before a velar vowel is found in: chaschuns 62; auchor (= altor) 1198; choche 5463; eschorcer 8125; chapler 1958, chaplier 2952; chaples 2318, 2523, 3164, 5499, 5701, 7566, 7934; char (= car, conjunction) 2429, 3378; caucher 908 alongside chaucer 6459 and enchauchent 2469, enchaucent 4246, 4284, 6266; enchauchez 4809, enchaucer 6402. The vocalized velar l, frequently spelt l in Franco-Italian, is found in this form in: elme 5181, 7770; elmes 3118, 5739, 6426 – but eume/ eumes/ heume 65 times; albespin 6080, albespines 8089 – but aubespin 3899, aubepine 7086; calt (< caleir) 2011, 2577, 3126, 3187, 5412 and chalt 3114

Orthography and Principal Linguistic Elements

79

but chaut 4 times; alquant… auquant… 392; alcun 7157; alcior 1391, 8029; maldist 3889, maldira 5102 – but 11 instances of maudi-; malmetra 4079; malmis 2594, 5438, 7391, 7897, 7901 but maumis 4 times; maltalant 350, 365, 682, 2786, 4899, 4907, 6670, 6926, 7542, 7582, 7748; maltalent 437, 7910; maltalentis 697, 3456; maltalantos 6402 but mautal- 9 times; malvaise 2844 but mauvais- 8 times; salvee 4226, but sauv- 17 times; salt 5408, salz 1828 but sauz twice. (xv) The representation of l mouillée (normally -ill- in Old French) by the graphy -ll- is found in assallons 3557; batalle(s) 16 times; entallee 4550; fallie 5872, fallir 6213; vallan 3410, vallanz 3433, vallant 4738, 5587, 6278, 7478; ventalles 3118, 5392; vitalle 4333; apareller 6870; asoleller 8126; conseller 73; envermellez 2521, envermellee 5482; esvellez 4864, esvellai 6774; mervelles 139, 7227; mervellos 939, 1667, 2410, 7882; mervellose 1225; mervelle 2330, 4202, 6162, 6471, 7082; merveller 6728, 6890; mervelleose 4290; mellors 3223, 5341, 6834; mellor 5470, 6369, 6465, 6549, 7543, 7786; orelles 4109; orguellos 4695, 5465, 6390; somellant 5952; travellee 6683; velle (= vieille) 3150, vermelles 1328, 2781; the graphy -lliis found in aparellié 7663, ballier 7691, balliant 1525, balliez 6293; and single l is found in mervel 6894 and merveleouse 2394. (xvi) The assimilation of l to r is found in corpe for colpe 1572, but no other examples have been noted. (xvii) Effacement of præconsonantal s is found in: trepasez 664; trepaser 1953, 6862, 6863 and trepasser 4276; trepassa 3823, 3989; trepassement 4129; trepassent 4555; trepasserent 4685; trepasee 6564 – but trespasera 66, trespasé 559. (xviii) The haphazard use of single and double consonants is found in the following words in their various cases and tenses: angoiser/angosser/ angoisser, aquison/aqisson, avaler/avaller, bailier/baillier/bailler/ballier, bailir/baillir, defendant/deffendant, eschele/eschelle/eschile/eschille, froisser/froser/froisier, isi/issi, laissier/laisser/laiser/laisier/laïer, taisir/ taissir, trepaser/trepasser, tresuer/tressuer, vassaus (6342) but vasal, vasaus everywhere else. This list is not exhaustive, but it is representative in showing the preponderance of letters l and s affected. Most words cited above are found relatively infrequently. A double for a single consonant is found in repossant 5977, reposser (= reposer) 6263; also possibly possez = posé 455. The words mesage and mesager are consistently written with a single s (30 instances); and ausi (com) is found 5208, aussi not at all. (xix) Aphæresis of the initial syllable is found in stez 549; this is present 5, ester (cf. Italian stare) but it is spoken by a Saracen in the text, so may

80

(xx)

Introduction

be deliberately «foreign». Cf. also § B, Phonology (ix) above for examples of s impura without prosthetic e. In addition, it should be noted that there is in C great hesitation over the spelling of words containing -n-: on the one hand insertion of n as cited in § B (Phonology (viii): Intrusive inorganic n) above, on the other hand, omission of this letter in graphies such as puig (but see comment on g under Orthography, page 72) and haste. With regard to this last word, whilst hanste («shaft» of lance) is found 2183, 5401, 7565 and anstes 2796, the graphy hauste (due to n mis-read as u?) occurs in 1143, 2057, 2233, 2243, 3424, 3843, 4407 and auste in 1847, 5193. Furthermore, astes is found in 1415, aste in 2267 and haste in 1716, 1847, 2201, 2211. Whilst the latter graphies may be due to omission of the abbreviation ~ (= n), such prevalence of graphies other than hanste (matched on occasion by V7’s reading) may well be a Franco-Italian feature, i.e. etymologically closer to the Latin source, asta.

Morphology In regard to verbs, the following features are found very rarely in C: (xxi) the -om termination (4th pers., Western French and Franco-Italian) is found solely in orom 653, again in Saracen speech, plus avrome 1301, also Saracen speech, and dirome 3394; in all 3 cases, this may be poetic licence to meet the requirements of rhyme; (xxii) eüsem is found 1474, but this is imperfect subjunctive 4, not present indicative 4, although the ending may well be Italian; (xxiii) luisont 1405 may represent luisoent, imperfect 6 but could equally well be the Eastern French past definite 6; (xxiv) past definite in -ui, common in Eastern French and in Franco-Italian, is found only in conuit 1593, 2780; connujit 5715; conuirent 4582, 4587; (xxv) the future forms with e preserved from intertonic a are found in some 8 instances (e.g. tornera, menera, sonerai, ensegnerai), but these are surely normal thirteenth-century forms as is finera 599, finerai 1407; (xxvi) the present participle is found with the ending -ent only in lucent 1033; (xxvii) poissent = puissent is found 5205. (xxviii) However, era, the Franco-Italian form of imperfect 3, estre, is found line 1296. With regard to the definite article, the following features are found: (xxix) le is used for li, subject masc., sing., over 40 times 776, 987, 1024, 1071, etc., on 24 of those occasions perhaps because the verb-subject inversion

Orthography and Principal Linguistic Elements

81

has confused the copyist and in several cases where the noun is part of a familiar title or in apposition to a Proper Name: e.g. Richer le Normant 4973 and Gerart chevauche, le hardi conbatant, 6623; but li sans et le chals 7948 implies that the two graphies were seen as equally valid alternatives. In 3907, le replaces li (sing.) and then li (subj., plur.); le instead of li (subj., plur.) is also found with verb-subject inversion 7348. (xxx) li is found some 20 times in place of le, object masc.sing., but inversion and apposition clearly pose problems for the copyist and the first examples (783, 918 and 932) are found in «Saracen speech»; lo as obj., masc.sing. is extremely common, occurring over 200 times; lo is also found as subj., sing. in 4015 (inversion), 6653 (apposition), 7255 (inversion), 7417, 7870, 7902 and 8120 (inversion);82 (xxxi) le is used as obj.masc.plur. (instead of les) in 188, 270 (+ au for as before sages, i.e. word beginning with s), 596 (les ses…), 1944, 1989, 2352, 2610, 2801, 6357, 7394, in all cases before an initial consonant. In C, li is also used instead of les, obj. masc.plur., in 134, 821 (+ verb-subject inversion 822), 1788, 1828, 2524, 3446, 6768, 7764. Conversely, les is used in error for li, subj.sing., 6276 and on this occasion, exceptionally, has been emended as a scribal error (li sosduiant > le sosduiant because of inversion, then > les sosduiant through repetition of s.). In relation to personal pronouns: a. there is occasional use of Italian ne for (French) en: 65, 3643, 6799. b. On possibly 4 occasions in C, si replaces se as reflexive pronoun, 3rd person (cf. B, Morphology (xii). c. Occasionally (lines 48, 1610, 2236, 2537, 2862, 3204, 4589, 6665), le is used for li, confusing direct and indirect object pronouns. d. There are 87 examples of lo used as the direct object, masculine. In respect of the relative pronoun, there is confusion of qi and qe from time to time, but this seems a general Old French rather than a specific Franco-Italian feature. Thus qe is found as subj.sing. in 361, 746, 1057, 1205, 1364, 1907, 2699, 3514, 3776, 3892, 3915, 3995, 4061, 4381, 4456, 4914, 5376, 6253, 6597, 6679, 6940, 7128, 7224, 7368, 7419, 7804 and as subj.plur. in lines 566, 2168, 3284, 3293, 4187, 4219, 5231, 6299, 7464. Conversely, qi is used as obj.sing. in 345, 499, 1112, 1634, 3086, 3911 and as obj.plur. in 152, 899 (scribal error, metre

 It is characteristic of the C copyist to misread verb-subject inversion, as shown by the use of the wrong person of verbs (e.g. in 546 and 2491) or change of syntax (in 3080), as well as in the cases of apposition and use of a familiar title cited above.

82

Introduction

requires elision), 1860, 1999, 3031, 4937, 5225, 5944, 6339, 7161, 7329, although it is better read as a scribal error for q’il in 156, 1347, 2091. On several other occasions (72, 1934, 2433, etc.) reading q’i allows this graphy to be retained, but the justification is sometimes questionable. Also qi est is on 15 occasions abridged to q’est, 203, 825, 834, etc. Furthermore, qi is used in the MS instead of qe, adverb, «than», in 1721, 6643 and for qe, conjunction, in 2339, 3088, 3418; in the last 3 cases it has been emended for the sake of clarity. Additionally, C has the following occasional features: (xxxii) pouc is found 1423, poc 2147, the Old French pou or poi being contaminated by Italian poco. (xxxiii) Similarly, the forms suigent (present 6) 5794 and siguant 1529, sigant 4589 (present participle) show an Italian influence on the Old French verb sivre (modern Italian seguire). (xxxiv) In C 913, the Old French temple is contaminated by Italian tempio to produce templo or tenplo (C normally using n before p) (xxxv) The graphy jejuner or jeiüner in 4334 and 8097 may show the influence of Latin jejunare or Italian digiunare – or may be used simply to stress that this word is trisyllabic. (xxxvi) The word hardos 4119 (ardos in V7) poses problems as to meaning. Whether it derives from a source reading hisdos («hideous», «terrifying») or from ardans, «burning», or even (with metathesis of vowels) from hordas, «dirty» or «contaminated», it appears to have been modified by the copyist on the basis of Latin arduus and Italian erto so as to produce a word meaning «steep-sided» or «precipitous». (xxxvii) There are a few examples of metathesis, a characteristic of FrancoItalian texts: entres for enters (adjective, «whole»), 8102 (but cf. 5519, ent7re = enterre/entrere/entrire(?)); and possibly hardos for hordas, 4119 (see above). (xxxviii) The graphy apostoles 7152 rather than its equally valid alternative apostles may show Italian influence or simply greater familiarity with the Latin apostolum (via religious texts). Similarly, C’s over-riding preference for the graphies arcivesqe, archivesqe may be due to religious (Latin) rather than purely Italian (arcivescovo) influence. (xxxix) In 7232, Fre may possibly show the influence of Italian Fra, but is more likely to result from accidental omission of the symbol for re, Frere Oliver constituting a regular first hemistich. (xl) Typical Franco-Italian confusion of vowels occurs from time to time in C: asteler for osteier, 195 and 828; par coi 281 for por coi; proz for prez, «ready», 758; avrir for ovrir, 864; poiz for puiz («hills»), 1172;

Orthography and Principal Linguistic Elements

(xli)

(xlii)

(xliii)

(xliv)

(xlv) (xlvi) (xlvii)

83

fuit for fait, 1275; paëstez for poëstez, 2508; Tresq’es for tresq’as, 2789; Divers 2885 for Devers; gitez 2926 for getez («thrown down»); ou for au, 3255; car for c’or, 3317; por for par, 281 and 3579; aparolt for aparelt, 8116. The graphy vesre 7542 for vesrai, future 1, is typical of an early development in South-Eastern as well as South-Central and South-Western French. There is also typical Franco-Italian confusion of voiced and unvoiced consonants: sab for sap, 1369; confenoner for gonfanoner, 123; cache for gage, 892; agreanter for acreanter, 965; vergie for verdie, 1357; grosir for croissir, 2155; eclise (< Latin ecclesia, or influenced by Italian chiesa), 3703; la chité cerchent for la cité cherchent, 5854 (or simply a scribal error?). In 2447, çase for chase is an Italian graphy. The letter z and the combination tz are a Franco-Italian feature occasionally found in C: e.g. zo for ço, 92; douteretz for douterez, 1137; sazez for sachiez, 2066. In 377, el is found as pron.masc.subj. instead of il. This occurs several times in V4 (cf. Beretta 1995, 514–517). However, in C, il is used on 5 occasions when the subject is clearly feminine (Braimimonde or Aude): 4498, 4499, 4838, 4839 and 7121. There is no indication in the Beretta (1995) edition that this is found in V4 and no record of its being a Franco-Italian feature. It may simply show the copyist’s poor comprehension of what he was writing. Very occasionally, the letter e is found as present 3, ester («to be»), e.g. in 916, 943, 1479, 5587, 5729, 8135, but this may represent either Italian è or omission of the slightly domed line regularly used over e to give est in C. In 3254, the epenthetic e in devereien may show Italian influence. The Italian adverb li («there») is used in 379 instead of Old French la. Although the substitution of -is for the -ir infinitive on 7 occasions may be due to poetic licence, it is also commonly found in V4, which may imply that this is a Franco-Italian feature.

Features such as the use of lo are typical of Eastern and particularly SouthEastern French. In South Champagne and Burgundy there was a falling together of the ãi and nasalised ei diphthongs and the development of a palatal glide, producing the rhyming of -aigne with -eigne which one finds in C (cf. Appendix B). These regions also feature the palatalised forms of the verbs tenir and venir, teigne and veigne, as found in the following examples: teigne (C 270), sosteigne

84

Introduction

(C 2340), veigne (C 3341), meitegne (C 3934), ateignantz (C 4285) and conveigne (C 7830). The South and South-East and East were the regions most affected by the development of a palatal glide; this is shown particularly clearly in C in the graphy -ngn-, found in dangner (C 56), foingne (C 3258), s’eslongna (C 6948), vengneson (C 7403), the emended form destrengnant (C 3873) and in pongnant (C 2698, 2736, 4857, 4905, 6695) and pongnons (C 3625). Also found in C are the following features, sometimes classed as Western Region, but which should more accurately be termed North Western, since this covers Anjou, Maine, Touraine, Brittany and Normandy: a. imperfect 6 termination -oent in poent 4312, 6089 (alongside estoient 4312); b. imperfect 3 termination -ot in estot 1159, 2208 (but estoit elsewhere), poignot 4128, avot 7574 (+ levot 6741 where past def.3 leva/levat is needed). Normandy was originally more closely linked to its North-Eastern neighbours than to those in the west and therefore shared some northern linguistic traits. It is surprising to note just how many Franco-Italian features have been found in a text which is still couched in a standard form of Old French. Although over 40 characteristics have been listed, in most cases the number of examples is extremely small, sometimes limited to only one or two in a text of 8,201 lines, and the graphies do not constitute a consistent usage. As accidental lapses into a regional pattern, they reveal the transcriber’s nationality, not the region of origin of his source. Certain distinctive features, such as the use of final -a for -e in 12 instances (cf. § C (x)) leave no doubt that this transcriber was Italian; but, like V7, it is definitely written in Old French and not the strongly Italianized language of V4, La Geste Francor, L’Entrée d’Espagna or many of the other post-1350 mss. The closest match to C and V7 in language and to V7 in illumination is La Bataille d’Aliscans, Holtus (1985, XLVIII–LXIX) assessing it as written in Old French with Italian overtones and D’Arcais (1984, 592) dating it to the later fourteenth century and comparing the stylised foliage around the first page to that in V7. On the other hand, there is also significant evidence (particularly in the past definite 6 verbal forms listed in § A, Morphology (viii)) of regional characteristics of Northern and North-Eastern French; and these are supplemented by the 12 features common to both Northern France and Italy listed in § B, Phonology and Morphology. At the same time, the «local colour» found in some of the Proper Names (personal, historical and geographical) relates very pertinently to the literary activity centred on the confrérie in Arras and spread-

Orthography and Principal Linguistic Elements

85

ing north and east as well as southwards along the Arras-Troyes-Dijon axis83 on the trade route from Boulogne-sur-Mer (and, beyond that, England) to Provence and Northern Italy.

The «North-eastern France» Element in the Redactions of the Roland In ten instances between C 47 and C 4426, Aix is mentioned as Charlemagne’s main seat of government and/or the site where the judgement of Ganelon will take place; this is confirmed by V7, O (3696, 3706, 3734, 3744) and V4, plus PTL in due course. In contrast, Monleon is found only twice before C 7326: C 109 lists it alongside Chartres and Blois; and C 3498 alludes to written testimony to be found there. However, in the final section of the poem, where O retains Ais as the site of judgement, all the other versions in the «new» late twelfth-century redaction agree in locating this in Laon. In C 7326–7595 (270 lines of text) there are 7 references to Monleon: 2 to Pépin le Bref’s great palace there and 5 to Ganelon’s being taken there for judgement. This insistence on Monleon in such a short passage in the final 900 lines of Section C arouses suspicions of a certain regional prejudice by the author in its favour. It is, however, historically logical, for Laon was a favourite palace of the Carolingian kings and the events related took place before Charlemagne was crowned Emperor in 800, i.e. when Laon was the capital of the Western Franks. Local tradition also links Laon to the trial and execution of Ganelon, identifying the «champ de la bataille» and the quartel Roland 84 – but of course these local legends may have arisen later, from the popularity of the Roland. If the author was led astray by pride in his region’s history, he may have based his stress on Laon on anachronism, for the city was also the capital of France 895–988, after Charlemagne’s Empire was divided into three parts. The stress is nonetheless there – and it is intriguing to find that, at some time pre-1500, the P manuscript was the property of the diocese of Laon. One wonders whether it was rescued from the library of the Abbaye de St.Vincent during the conflagration of 1359 or reposed in the cathedral library before that. When the close correspondence of most of P’s first 86 laisses to CV7 (C 101– 180) indicates a shared model, one questions why this lapses in the next 18  Jean Bodel’s Chanson des Saisnes (1197–2000, cf. Jacob-Hugon 1998, 23), Bertrand de Barsur-Aube’s Girart de Vienne (beginning of the thirteenth century) and Adenet’s known works testify to the continuing interest in up-dating the old chansons de geste between 1180 and 1280.  Cf. Martinet 1972 and 1989; Rejhon 1993; Roland Corpus, vol. IV, Paris ms., ed. Rejhon, 2005.

86

Introduction

laisses and in subsequent sections until the beginning of Section C. It may be significant that P 86 (C 180) is not found in O or L but exists in abridged, assonanced form in V4 laisse 134. There was a highly reputed school (under Alcuin) in Laon, with students from far and wide and a considerable copying of manuscripts, both religious and profane, in the scriptorium. The presence of large numbers of Italian canons in Laon between 1250 and 1350 (and particularly 1272–1305)85 may indicate links with Northern Italy which facilitated the transfer of manuscripts such as the CV7 model. Also in Section C, one finds St. Firmin named alongside St. Paul in C 6074. There were two saints of this name, both closely linked to Amiens. The first was martyred late in the third century: one finds his life-story depicted on the marble screens enclosing the choir of Amiens cathedral and his life-size statue on the trumeau of the left-hand portal of the West porch. The second, St Firmin the Confessor, died peacefully in Amiens in 390 AD. He was responsible for the transfer of the Martyr’s remains to a new church, on the site of today’s cathedral. Later, in 1186, relics of Firmin the Martyr were solemnly carried across France to his birth-place, Pamplona, for interment in the cathedral there. The route taken was the ancient Route d’Espagne, via Chartres, the abbey of Boneval in the Loir valley, Touraine and Anjou – exactly as listed in C 7353–7354. Just north of Vendôme and just off this road, one finds the little village of St Firmin-des-Prés with its ancient church containing old frescoes depicting a procession on horse-back and the faded picture of two saints standing side by side, one in red (for martyrdom) and one in blue (as a confessor?). Other regional place-names are found throughout the earlier part of the text: Hurepois with the wider meaning of «Neustria» in C 110, 5488, 5696, as found in Bodel’s Chanson des Saisnes (where the area is specified, «from MontSt.Michel to Château Landon»,86 not the smaller area in the Beauce); Argoigne, a small region east of Paris, in C 5687: Henoier in C 2904 for the inhabitants of Hainault; and a reference to the Jarris or Larris west of Châtillon-sur-Seine, C 5694. The unknown roi Dotaire in C 5759 is not found in other chansons de geste or the reference books; and there is no evidence to link King Lothaire (reigned 954–986) with exceptional wealth. Reading the name as a scribal error for Clotaire, it becomes an allusion to the great treasure Clotaire I is traditionally said to have assembled as tribute in Soissons in the sixth century, l’avoir qi fu au roi Dotaire representing a folk memory.

 Cf. Millet (1982, 60–64) for a correlation of dates and places of origin in Italy.  Cf. Brasseur (1990, 214s.) for the Episode of the Hurepois, ll. 298–392.

Editorial Policy

87

Another factor to be taken into account is the historical involvement of powerful noble families of the North-Eastern region and Burgundy in the Spanish Crusades,87 from the eleventh century. The marriage of Félicie de Roucy to the future king of Aragon just before 1063 was followed by the strong involvement of her great-nephew, Bertrand de Laon, in later Crusades. The alliances and interaction of North-eastern French nobility with the Christian kingdoms in Spain and its active involvement in the fighting there would have been a constant stimulus to interest and a spur to embellishment of the poem.

I. Editorial Policy Principles of Emendation In the «Roland Corpus», Duggan’s stated aim is to re-create the CV7 source version. This assumes that model to have been the (lost) Roland Rimé and V7 to be the earlier and more authentic copy because, according to Segre, C later turned to another, assonanced version in order to make good damage in that model. The evidence set out in I.1 above gives strong grounds for believing that C is the earlier manuscript, containing a text copied uncritically from the CV7 exemplar, and that it is V7 which turns to a different rhymed source in order to make good faults, omissions and inaccuracies of metre visible in C today, but deriving from the CV7 model’s readings. The V7 manuscript contains 8333 lines of text, but this increases to 8397 lines in the Duggan Critical Edition. In contrast, the C manuscript contains 8201 lines. The present edition identifies 38 lines omitted by C but essential to the sense: of these, 27 lines are found in V7; the remaining 11 lines are lacking in V7 (and therefore presumably lacking through scribal error in the shared CV7 model): of these, 4 are inserted from P, 3 from V4, one each from O, T and L and one from C’s repetition. In the present edition, inserted lines are written in italics; they bear the number of the preceding line, plus a (and b on occasion), in the margin, with the source given in brackets and a detailed footnote. Where the C reading lacks a V7 line but makes sense without it, no insertion is made, since this may be V7’s own addition or copied from another source (shared with a collateral version), not a CV7 model reading. Where a C

 Cf. Boissonnade (1923, 13–16) for the Influence des alliances matrimoniales sur le développement des croisades françaises de l’Espagne du Nord, and Martinet (1972, 27–34) for the Laon connection.

88

Introduction

reading makes sense, even though the combined evidence of the other versions makes a misreading by C extremely probable, that reading is retained as indicating the degree of misapprehension shown by the C copyist – but glossed in the footnote. The aim of the present edition has been to produce a fully annotated Châteauroux text with the minimal degree of emendation consonant with comprehensibility and thus to retain those inaccuracies characteristic of the C copyist. The criteria have been whether the text as it stands makes sense and, to a lesser degree, whether there are indications of corruption, such as faulty metre, strange division of sense-units or repetition of identical words or phrases. The evidence of the other versions has been taken into consideration, insofar as each shows any degree of relationship with C in the episode in question, but not given over-riding importance. Where a reading in C derives from a scribal error in the CV7 model but makes sense, it is retained. For example, C 47 and V7 47 both read Aut s’en en France, which clearly derives from a misreading of En France ad Ais (O 36) or Ad Asia en France (V4 40) in the CV7 source. Since this makes sense as present subjunctive 3, s’en aller, it has been retained but glossed. Again, where the copyist has mis-read his original but changed it to something which makes sense in the context, his reading has been retained, even though V7 has apparently kept the source version, as supported by the collateral texts. Thus there are several readings arising from C’s difficulty in recognizing Q as the initial letter of a line: G for Q, in C 149 and most particularly in the alexandrines C 263 and 281, which read San ce por coi ne dient, San ce par coi vos mande, where Quant ço vos mandet (O 222) and Quant il vos m. (O 239) support V7 268 and 286 as the source readings. C’s weaker reading makes sense, so it is retained – in part because it reveals the limitations of the C copyist, which will be a factor in assessing other scribal errors. In contrast, when the same problem of recognition of Q occurs in C 793: Sance ferir (V7 811 Quant de ferir, O 514 Quant por ferir…, as required by the sense), and C 1608, where Saut… gives a superfluous verb and makes the line incomprehensible, C has been emended. De (C 657) has also been emended to Qe; and the manuscript reading Den (according to Bourdillon and Michel) or Ben (Foerster, Mortier and Duggan) in C 32 may be a scribal error for Q’en (see below and also footnote to text) – although De in V7 33 lends support to Den or D’en… in C. Accidental errors in transcription have been emended in accordance with (usually) V7’s reading, supported where possible by reference to the appropriate collateral version. Where the error is the revisor’s (e.g. C 1571, where Sena is emended to Clina by a later hand but V7 1517 reads Seigne and P 20 Saingna), C’s original reading has been retained with detailed footnote.

Editorial Policy

89

C folio 1 throws considerable light on the differing characteristics and attitudes of the two scribes, C and V7. V7 laisse 1 maintains a rhyme in -aigne without variant graphies, whereas C laisse 1 has the same rhyme with variants -egne, -aagne and even -igne; but the text is almost identical in both. The main difference is V7’s insertion of V7 7, lacking in C but not absolutely essential. In laisse 2, rhymed in -er, V7 15–30 and C 14–29 are identical apart from variant graphies; but V7 lacks C 30, which Duggan validly inserts as V7 31 (an extra line at the foot of V7 folio 1 recto); after V7 32 = C 31, V7 has a single, clearly comprehensible line (V7 33: De Val Perdue, si com oï conter) where C 32–33 are extremely corrupt. V7 33 slightly resembles O in giving Blanzardin a place of origin, but there is no real match; and V7’s second hemistich may be an adaptation of that in C 32 (se com orez nomer). C 32–33 have otherwise no match with OV4 or V7; they are peculiar to C. V7 33 reads like a simplification of a difficult reading: it rounds off the laisse but does not fit smoothly or comfortably after V7 32. C’s is the more difficult reading. It poses problems in each hemistich. Thus Ben tot le mõt (or Den…) in C 32 may represent Ben tost le monstre / En tot le mont / D’entor le mostre / Q’en tot le mon/ / D’entoilement; se com orez nomer is elliptical and may read nomer in error for noncier, while the symbol 9 could be read as quan (not com), «however many» to balance tant in the next line; Nen verez hõ could be transcribed as N’enverez hon/hom or Nen verez honc; and tant sages mesajer may need reading as sag’es («clever with messages») or otherwise indicate that the copyist read tant sages not as «such a wise…», but as «so many wise messengers/wise messages», justifying the writing of sages as the plural form. These confused lines do, however, provide the direct speech needed in relation to Quant Blançardin oït in C 34. Both O 24 and V7 34 have a purely descriptive first line in laisse 3: Blancandrins fut des plus saives paiens / E Blanzardin fu molt bon chevalier respectively. In contrast, C 34 immediately carries the action forward. Lines 32–33 demonstrate C’s uncritical approach to the model before him and his readiness to copy what he does not fully understand – and V7’s constant aim of producing a comprehensible text and readiness to excise what cannot be corrected. Since both V7 and C differ from O and V4, the only alternative versions available here, as well as from one another, one suspects that scribal error and/or damage occurred at the foot of folio 1 recto of the CV7 model, rendering the reading incomprehensible. Faced with this situation V7, the more knowledgeable, experienced and confident scribe, reduced the two lines to one, V7 33: De val perdue si com oï conter. This uses but modifies the second hemistich of the CV7 model’s reading (line 32 in C): se com orez nomer. In contrast, C copied the model as faithfully as possible, perhaps not realizing

90

Introduction

how confused the reading was. V7’s is the more thoughtful and more expert transcription of the shared model, but not necessarily the earlier one. Editorial policy here has been to examine all possible mis-readings of the model and, as an exceptional measure, to modify the reading. Thus, the possibility that Ben in C 32 is a scribal error for Q’en must be considered. Since other mis-readings of Q have been identified, this is a tempting hypothesis; whether read as a conjunction or a relative pronoun, it makes better sense than B or D in relation to en tot le mont. It does not, however, clarify the sense of the rest of C 32–33 or provide a clear introduction to the direct speech needed in C 33. If, however, after C 30 the line-order were also changed to read: Q’en tot le mont, si com orez nomer / «Fors Blanzardin – cil ne se vol celer –/ N’enverez honc tant sage mesager», this would make excellent sense. Sadly, there is no other evidence to support this hypothesis, so a different solution has been somewhat hesitantly chosen. Such an emendment has been made in only one other instance in the text: C 7945, where C, V7, P, T and L all have the same reading which quotes a nonexistent proverb irrelevant to the context, due to a scribal error in the common source. This has been emended on the grounds of sense, context, palæography and external evidence from other literature as indicated in the footnote. In all other cases, C’s readings are retained wherever possible. The correction of glaring errors is supported by collateral evidence cited in the footnote. Occasional inconsistencies in the text, such as the remarks in C 3861–3864 which are obviously spoken by Roland (in C) but make sense only in Oliver’s mouth (as in V7) although the latter’s death is recorded in C 3394, are fully dealt with in footnotes but left unchanged, since neither alternative solves the problem. Where C has syntactically faulty forms of words, such as past participles for infinitives, or deformation of endings, such as -or for -on in traïsor (C 1226 and 1398) and ciglator (C 1228), or any other unusual graphies for the sake of the rhyme, the poetic licence has been noted but the readings left unchanged. Out of 78 such instances, 61 occur prior to C 4180 (ie in Section A), 10 in Section B and 7 in Section C. Several of the last 17 are noticeably of a minor nature, modifications of graphies to give a visual rhyme rather than syntactically faulty. One specific feature of C, found 10 times in Section A and (possibly) once in Section B, is the use of -is to represent the infinitive -ir for the sake of the rhyme: this seems to result from a graphy -irs (found in ferirs, C 509) with subsequent assimilation of r.88

 Cf. Pope (1934, 156s., § 396) regarding effacement of præ-consonantal r, especially in the group rs, in Later Old French.

Editorial Policy

91

The copyist uses standard Old French abbreviations, which have been transcribed in accordance with the scribe’s normal orthography. In particular, this results in the following transcriptions: 9 for com or con: in C, used before -e on 18 occasions to signify come (adv. or conj.), written in full 14 times. Otherwise, it always represents con- within a word (even before b or p) or co- before an m (since -mm- is never found in C). As the monosyllabic adverb, conjunction or preposition, it has been transcribed as com to match the graphy in C 513, 1625, 4189, 4470, 4546, (Cõ 5981), 6576 and 7091 (although con is found in full in C 4504, meaning «what a…!»). This also serves to distinguish it from con, preposition meaning ‘with’ (a most unusual form and a pure Italianism, standing for cum), found in full in C1860 and 7301. Similarly, the titulus used for the nasal abbreviation has been transcribed n, even before a b or p, in accordance with C’s practice: the graphy -mb- is found, written in full, in 23 instances, compared with 99 instances of -nb-; -mp- is found in desrompue C 2478, amples 5015, emperere 6878 and empere 7100 compared with well over 50 examples of -np-. 7 for e or et, «and»: Both graphies are used in C, infrequently but indiscriminately, but et in full more often than e, so this symbol is consistently transcribed as et. Unless the spelling is varied to meet the particular case (which the copyist himself does not do), either transcription poses problems of metrical accuracy on occasion. Both 7 and 9 are used as capital letters: 7 in C 942, 9 in C 173, 2972, 4704, 5086, 5402, 6594, 7214, 7241, 7442, 7483, 7721. p with bar through the stem is normally transcribed as per, following the scribe’s usual graphy; but in certain words it stands for par-: e.g. parla C 4941 after parlance (in full) C 4940. P + o (o in superscript over p) is consistently used for por. qe has been transcribed as qe, qi as qi, since qu is only rarely found before e or i in C (68 times, 31 of which involve conquis-, conquir- or conquer-, but the single word que once only, C 6734, qu’ in C 149 and 6482, and qui in C 1728, 2951, 3036); qua- is regularly found, e.g. quãt for quant (less frequently qant), quarte (which may indicate Latin/Italian influence). With regard to Proper Names, the following abbreviations are found: .R. for Rollant is constantly used from C 120 on. The graphy Rollant is not found in full until C 232; it is regularly used for both subject and object. The graphy Rollanz occurs 11 times as subject, once (C 1076, at end of line to provide the correct rhyme) as object. Rollanz for .R. (object) has therefore been adopted in C 4298.

92

Introduction

The abbreviations .K., .K’, .Kll’., .Kll’e and .Kll’n. are all found from C1020 onwards, although .K. is the most common. In the first 1019 lines, Charlles is written in full 14 times, Charlle (subj.) 6 times, (obj.) 5 times, Charllon (obj.) 3 times, Charlon (obj.) once, Karles (obj.) once, Karllon (obj.) twice, but never abbreviated. From C 1020 on, .K. and .K’. are in frequent use; and K has been retained as the initial letter when transcribing the name in full. The predominant graphy of the full name is Charlles (34 times), with Charlle, subject 11 times, Karlles 3 times, Carlle twice and other graphies only once each. For the object, the full graphy Charllon is found 22 times within the whole text, compared with Charlle (4 times as object), Karllon (3 times object, but subject, C 7463), Carllon (twice) and Charlon (once). The abbreviation .Kll’n (.K.’on in C 4435) is normally used for the object. To meet the requirements of metre, .K. and .Kll’n. have been transcribed as Karlle Mene wiith explanatory footnotes in C 4209, 4279, 4371, 4864, 5568, 7854, 7962, 7981, 8000, 8019, 8035, 8132. C 7962, 7981, 8000, 8019 and 8132 all introduce a new speech by Charlemagne, so use of 4: 6 metre may be an intentional ploy to signal this to the audience; on the other hand, Brook (1990, 17–19) agrees with the transcription of .K. as Karlle Mene so as to sustain the alexandrine in this passage in Section C. The only other abbreviations for Proper Names are .B., .G. and .M. The abbrevation .B. stands for Baligans (subject) or Baligant (object), with final letter occasionally modified to suit the rhyme. .G. is found 8 times for Guenes (subj.), 3 times for Guene (obj.) once for Guenon (obj.), the graphy determined by the requirements of metre, and once for Guenellon (C 6449), obj., to suit case, metre and rhyme. It is also used to signify Gerart (de Vienne) in C 6190, 6577 and 6950. .M. is used solely for Marsille, 15 times as subject, 5 times as object, but only from C 3576 on.

Problems of special linguistic features One major problem in the transcription of the text and production of the glossary arises from the fluctuation in the spelling of words in which the countertonic a stood in hiatus to the tonic vowel, e.g. traïtor, traïner. Problems of metre arise if one attempts consistent use of the tri-syllabic traïtor in C 1319, 5549, 6364, 7134, 7368, 7600, 7879, 8103, 8138, 8168 and 8175. Metrical problems in relation to diphthongs also arise in relation to the verbs trainer and the two radicals of aidier; and there is also some slight evidence of scribal confusion over oi in oit nom (C 1182, 1318, 1742) whereas the standard form ot is used everywhere else, parallelled by poit for pot, soit for sot and late twelfth-

Editorial Policy

93

century monosyllabic oit (present 3) alongside oït, past definite 3. Also, in C line 26, peüst makes the second hemistich hypermetric. All these examples are dealt with in the notes to individual lines. In C 1319, the only example involving traïtor in the early part of the poem, emendation as a scribal error is the obvious solution; but all the other examples are found in Section C. Thus disyllabic traitor may represent late twelfth-/ early thirteenth-century contraction of diphthongs, producing a rather fluid and variable pronunciation. At the same time, V7’s use of the graphy trahitor may be influenced by the Italian traditore. Generally speaking, the Châteauroux text is metrically accurate and irregularities often indicate other errors (of omission, mis-reading or insertion). The scribe does not always indicate elision; but this has been left unchanged. The phonological characteristics of the text have been noted, but not emended unless there is over-riding evidence of a scribal error with supporting evidence from the collateral versions, and the morphological features have been scrupulously respected, so that their regional implicationss may be taken into account.

94

Introduction

Appendix A Alexandrines in the Châteauroux ms. .

Line 

.



.



.



.



.



.



.



.



. .

 

.



.



.



.



.



.



.



. .

 

.



.



.



Ars a nus castiax, nus terres fait gaster E je i tremetrai le fil de ma moillier Trespasera li termes qi nus sera donez Et anscenble o lui ot meint autre princer Atant descendirent tuit li .x. mesager Marsille m’a tramis bien .x. de ses privez Jamais ne vesrai jor de la Pasqe florie Onqes nus n’i ala qi en soit retornez Cheü li est a terre, a poi n’est forsenez Marsilions salue, com ja oïr porez Et Gifeüs ses fiz, qi est de grant pöois Lo prist .i. orages, si cum oï conter De paiens Sarazins issent a recelée Eumes lucent a or e broines endosée Plorant li done l’arc, lo gant e lo baston Se cil fust Cristïens, ben feïst a doter E ses compeing i est qi oit nom Estrament Ne croi pas troi mulet portasent la moitié Rosée n’i puet crostr(a)e, ne pluvie Conquistrai Durendal qe de la moie est menbre .vii.c.mil Saracin encontre als ajosterent Olivers est montez sur un pui alcior Pris en a le loier, sachez por veritez

V reading Ars nos castels… Ge trametrai… Trespasera li termes qi est nomez Ensemble od maint riche princier A pié descent… Marsille m’a tramis de ses privez Je n’en vesrai jor de Pasqe florie Nus n’i vet la qi en soit retornez Chiet li a terre, a poi n’est forsenez Li roi salue… E Gifeüs, ses fiz de grant proiz Lo prist orages… Paien chevauchent tuit choi entrecelée Elmes laciez e b. e. (cf. O : Healmes lacez e ceintes lur espées) – Laisse not in V – Fust Cristïens,… (cf. O : Fust chrestïens) E sis compaing iest qu’ot nom Estramant V different version, different rhyme: Qe dui mulet n’en poroient porter n’i qiet mie – cf note to this line V different, different rhyme: Si conquirai Durendart au capler Cent .m. paiens voldront o els mener Oliver monte… Lines – not in V (cf. O –: Pris en ad or…)

Appendix A

.



.



. .

 

.



. .

 

.



.



.



.



.



.



.



.



.



. .

 

.



.



.



.



.



.



95

Ay! Sire enperere! Coment nos Li emperere… vengerez! Esleger nos estuet as bons brans Mais as espées… acerez R[ollant] porte ses armes, mult li sunt avinant BUT cf note to this line Puis a lacé .i. eume ou li ors relu- Puis lace .i. eume… ist cler Et ot ceinte l’espée qi mout fait a E ceint l’espée… aimer Et adornerent lui a guise de baron Et atornerent a g. de b. Hui se desmostreront li bon com- Hui se desmostrent… bateor E resaili en piez, sa broigne reves- Puis saut en piez… tie Ne reprover i puist a hom de mon Ne reprover puist hon mon lignage lignage Mex aiment a morir qe bataille Mex aiment morir… esciver C’aprocherent paien lo trait a un C’aprosment entr’els… archier Tante espée brandir, tant espi Tante spée brandir,… (alexandrine) flamboier Puis escria «Monjoie!», la baniere Puis escria la baniere Charlon Charllon A grant péché somes e traï e Par grant peché somes trahy e vendu vendu Mot fu grant li estor, nus nel pot – V the same – (alexandrine) endurer Ou est donc Auteclere? Por qoi ne Ou est Autecler,…. la traiez? La bataille i ot pleiniers e adurée Bataille i ot,… Qi le veïst venir, mot le deüst – V the same – (alexandrine) pener La teste li toli sanz negun reco– V the same – (alexandrine) vrer La bataille par fu merveilleuse e – V the same – (alexandrine) estreigne A ses armes vermelles e al chival As riches armes… corant Et a son fier visage e au fer conte- A son fer corage… nant Et cil le recoillent, se metent con- – cf. note to this line peignon Puis en oncist .xxiiii. Sarazins de V different version, different rhyme Et Medée .xxiiii. de toz li plus prisez’ (cf. O  and P ): E .xxiiii. de tuz les melz preisez)

96

Introduction

.



.



.



.



.



.



.



.



.



.



.



.



.



. .

 

.



.



.



. .

 

.



.



.



Chevalier tel valor avoir De tel valor doit estre encoragez /Bons dever[e]ien[t] chevalier qui est enparagez Et ensanble o els Girart de Rosion Aprés ocist Girart de Roseillon (cf. O : Ensembl’od els…) Si chalongent lor cors de la gent Si chanchelons et le chors e la vie saracine Li fist outre paser, grant duel en V, different version, different rhyme ot le roi Mais tu nel nonceras el reigne ou V different version, different rhyme tu fuz N’en vanteras n’en laies tolu / A ces paiens qui ci ai tant veü Ne me fesis un mal, ne a toi ne Ne me feist mal, ne a toi…(cf. O : fis pis Ne:m fesis mal…) Toz dis avez esté vasaus e comMult fust jadis hardi e conbatanz batanz Qi de la Virgine nasqui en BalliQui de la Virge nasqui en B. ant Charllemene apele Rabel e Guine- Karles apelle… manz Por combatre sunt bien richement Mult par sunt bien de combatre ordenatornées ées Puis eïs[t] a enfer por sauver Preist en feme por saluer bone gent bone gent Quant Baligans le vit, si devient e Quant .Ball’. le vit, si fu plus noir plus noirs Granz ert la nivole, la poudrere Grant est nuïsse, la poldrine… est levée E de la Virgine nasquis en Baliant Et de la Virge nasquis en B. L’emperere chevauche tot le chief Li emperere chevauce tot primier primier Charlle Mene vit Aude, si la va .K’. vit Aude,… (ie Karles vit Aude) enbracer Qe puisse mais partir mes amor Qu’i partir puisse mes amors de Rolde Rollant lant Ne devez pas avoir ver moi nulle Ne devez avoir vers moi nulle haïne haïne Poi aproiseriez la terïene vie Poi preisierez… Cil sire le destruie qi confonde Line not found in any other version Mahon! Pinabax fu armez tost e isnelaPinabels s’arme… ment Des plus saintes reliqes de la V different version, different rhyme Crestïenté line closest to C is: Karles i fist reliques aporter Cil vus beneïe q’en la crois fu – not found in V – penez

Appendix B

97

PLUS C 7841–8133, with the exception of C 7843, 7851, 7854, 7861, 7891, 7899, 7934, 7981, 7995, 8000 and lines 8123–8125, as discussed in the relevant footnotes. C 8200 poses a problem in that, strictly, it has a 5-syllable first hemistich, since the final e should not be counted at the end of the hemistich. Perhaps «Cil» is a scribal error for «Icil», or perhaps metre did not seem very important in this envoi. The metre in one or two other lines poses some problems, for example «Enterreront» in C 3032 and 3040 appears to be treated as having 3 syllables rather than 4, but these are discussed in the notes, in the context of the individual line. In general, as the above list demonstrates, V7 has the normal 4:6 decasyllabic metre where C has an alexandrine. This is certainly the case in the long passage of alexandrines, C 7841–8133, which is unique to C. However, V7 does have five alexandrines to match C’s, as may be seen in the above table. It would appear that it should in fact have 6 examples of alexandrines, since, in the line equivalent to C 7089, V7’s «Ne devez avoir» contains 5 syllables; this must indicate a scribal error in omitting the «pas» found in C. However, V7’s one glaring error, unusual in so careful a copyist, is the mis-reading evident in its version of C 5317: «Preïst en feme…»!

Appendix B Assonanced Laisses in the Châteauroux ms. Laisse No.

C’s assonance

 Laisses with an «a» assonance  -ance, -ayme, -ande, -ange, -angle, -aigne  -asme, -able, -aigne, -eigne, -eine, -ange, -einte, -ance, -ame  -age, -aille, -arlle, -aste, Aspre, -arde, autre, -ace, -aple  -aine, -enche, -anche, -ange, -aigne, -eigne, -ante, -ande  -age, -aile, -alle, -ache, -ace

O’s laisse No. and assonance

V

: a assonance simi- Smooth -ance rhyme lar to C : almost identical NOT IN V to C : almost identical to C

NOT IN V

: a assonance similar to C

NOT IN V

: almost word for word identical with C

Smooth -ée rhyme

98

Introduction

 Laisses with an «o» assonance  -ose, -oule, -ore, -ome, -onge, -ontre, -onent, -onte, one  -ors, -ois, -ost, -oit, -ols, -ops, and  ( lines): -or, -os  -osse, -oble, -ote, -oche, -oupe, -oste, -one, -oce, -ome, -onche, -ose  Laisses with an «e» assonance  -erne, -erre, -esse, -ere, -este, -aele, -erde, -erte  -igre, -ie, -einte, -endre, -embre (some attempt at -ie rhyme)  -ente, -emple, -ene, -eine, -eindre, -eigne, plus -ande  -ée, -en, -ent, -ien 

: u asson., very similar to C

Smooth -ez rhyme

 + first  lines of : almost identical

NOT IN V

to C : u assonance + -oisset, almost word for word identical to C

: identical to C

Smooth -is rhyme

: ei, i, ie, e assonance, fairly similar to C : almost identical to C

Smooth -er rhyme

: e assonance, rough match -ers, -er, -ez, -iez, -iet : e assonance, close match with C

 Laisses with an «i» assonance  -ille, -ie, -ite, -ine, -igne, -ire, -ime, -ins, -isme

Smooth -ant rhyme

: very similar to C but slight variants

NOT IN V

Smooth -ez rhyme Smooth rhyme in -iez

Smooth -ie rhyme

* Laisse 1 is a borderline case; it has not been classed as assonanced, despite the inclusion of one -igne in a laisse with a pattern of -aigne, -eigne, -agne, -egne endings, because this could simply be a scribal error: omission of a or e. Laisse 165 shows in its brief repetition of lines that barguegne and bargaigne were equally valid graphies. Similarly, laisse 250, which has 11 rhymes in -eigne, 2 in -ene, one in -egne and one in -aigne, has not been classed as assonanced because these graphies all had the same pronunciation by the thirteenth century: cf. Pope (1934, 161, § 407, 179, § 467, 197, § 529). In the case of laisse 1, the copyist may have been unsure whether he was copying an assonanced or a rhymed text, since it would be easy to obtain a better visual rhyme, as V7 does; alternatively, he may have been careless or ignorant. Dilapidation or damage on the first folio of the source text seems unlikely, in view of V7’s better readings.

Appendix B

99

In addition to the above, there are a number of laisses with less-thanperfect rhymes, «pseudo-rhymed laisses»: laisse 85 has 19 lines with a rhyme in -ie, one ending in -ise; laisse 151 has 2 lines ending -anc, one in -ans, in a laisse rhyming in -ant; laisse 175 mixes -ors and -os endings; laisse 192 has -ant twice in a laisse rhyming in -anz; and laisse 196 opens with 4 lines rhyming in -ant, then 4 in -anz, then -ant for one line, finally 13 lines in -anz. The classification of these laisses depends partly on the pronunciation of a final consonant in rhyming termination, partly on the degree of importance one accords (and contemporary readers accorded) to visual rhyme. Many of these apparent «faults» could be glossed over in oral performance. Problems arise over the pronunciation of the intervocalic r in the -ere ending; in laisse 94, this termination is found in 3 consecutive lines in a laisse rhyming in -ee, and in laisse 203 there are 2 similar examples. The letter r is also involved at other points in C’s text, where a number of laisses monorimes only achieve a smooth rhyme if the penultimate r in the vowel + rs ending is not sounded or through an inaccurate (sometimes grammatically inaccurate) spelling. For example, laisse 98 rhymes in -is but has martyrs C 1505; and its -is ending in C 1504 and -irs in C 1509 represent the -ir infinitive. Similarly, laisses 202 and 217 contain 2 instances and 3 instances respectively of the «-is infinitive». Likewise, laisse 297 rhymes in -és but has -ers 5 times; and the ending -oirs is found where the rhyme is in -ois, once in laisses 279 and 279, twice in laisse 320; laisse 300 with rhyme in -oir has one -ois; and manoirs is spelt manois in laisses 270 and 344 (cf. Pope (1934, 156, § 396) on the assimilation of r in the rs group). Although the copyist usually shows a great concern with visual rhyme, these minor variations may well indicate local pronunciation with the consonant slurred or swallowed. Analysis of the pattern of occurrence of assonanced laisses, as shown above, in relation to the divisions of the Châteauroux text (cf. Introduction, 51–58) shows that: i. if one excepts laisse 1 with its rather dubious assonance, there are no assonanced laisses until laisse 75 (beginning C 1211), although laisse 74 comprises only 3 lines in order to rhyme (being equivalent to the first 3 lines of O laisse 67 which has an a assonance). In the initial 74 laisses (1210 lines) there are 6 examples of «incorrect endings», modified for the sake of rhyme (C 105, 246, 631, 780, 850, 1048); ii. there is a cluster of assonanced laisses at the end of Part 1 of Section A (laisses 75–96 inclusive; or rather 73–100, to allow for the pseudo-rhymed laisses and the general cohesion of this passage); iii. there are no assonanced laisses in Part 2 of Section A (laisses 101–196 inclusive) unless one counts as assonanced laisse 175 (7 rhymes in -ors, 3

100

Introduction

rhymes in -os, possibly indicating that the r is silent or almost silent in the pronunciation of -ors, cf. Pope (1934, 156s., § 396) re assimilation of r, especially in -rs group); iv. there is another cluster of assonanced laisses at the beginning of Part 3 of Section A (laisses 197–218 inclusive); but thereafter smoothly rhymed laisses to the end of that Section (laisses 219–247); v. there are no assonanced laisses in Section B (C 4180–5886 inclusive), unless one counts laisse 250 (referred to above, p. 98), or in Section C (5887–end); and vi. within each «cluster» there are also a number of pseudo-rhymed laisses: in the first cluster, 12 (73, 76, 79, 84, 85, 88, 89, 91–94 and 98) and in the second cluster, all 13 non-assonanced laisses contain very minor faults in their rhymes which, strictly speaking, make them pseudo-rhymed. The «clusters» identified above contain the 2 smaller groups of laisses involved in C’s 2 «repetitions» (analysed in detail in the Introduction), although the clusters cover a somewhat wider range. It has been suggested (cf. Segre 1974, 148–165) that C turned to an assonanced source in these two clusters because pages were missing from its rhymed source or too worn or damaged to be legible, but these passages may equally well represent early remaniements in which some laisses proved particularly recalcitrant to conversion to rhyme. (The examples of poetic licence and the number of pseudo-rhymed laisses in these passages are indicative of the problems involved.)

Appendix C Rhymes in Descending Order of Frequency of Occurrence Masculine rhymes -er/-ier -ant -ez -on -é -or -a -ent

-is -anz

 laisses, scattered through text, but  between  and , and  between  and  (one  lines, another  lines long)  laisses  laisses  laisses  laisses  laisses  laisses, scattered through text  laisses, (this is pronounced, as in nouns and adverbs, not the silent verb ending) although -enz occurs in C , , , ;  lines in laisse   laisses  laisses

Appendix C

101

-u  laisses -al  laisses (none in the first  laisses) -ir  laisses -uz  laisses -ois/oirs  laisses -i  laisses -in  laisses -iz  laisses (none after ) -ié  laisses -el  laisses -oi  laisses -ons  laisses -ier/-ngner  laisses (none after ) -ors  laisses (of which  laisses with -ors/-os), plus  laisse in -orz/-oz PLUS  laisse each of the following rhymes: -ai, -alt, -art, -ax (=aus), -eins, -erent, -ers, -és, -ies, -oir, -ont (but with sunt l. in the MS) and -os.

Feminine rhymes -ie  laisses -ee  laisses -age/-aje  laisses -ance/-anze  laisses -ue  laisses -eigne/-egne/-ene/-aigne  laisses -aigne  laisses -aire  laisses -ire  laisses PLUS  laisse each of the following rhymes: -ees, -ele/-elle/-eille, -ente (with -ante l.), -ere, -ine, -ise, -ogne/-oigne/-oine (laisse ), -ure

A total of 348 laisses with masculine rhymes and 89 with feminine rhymes, the remaining 15 laisses being assonanced (cf. Appendix B). Because of the effacement of præ-consonantal r in Later Old French (cf. Pope 1934, 156s., §§ 395s.), found in C particularly in the group -rs > -s, the rhyming of -ors and -os, -orz and -oz has been accepted as standard. Similarly, the varied graphies -eigne/-egne, -ele/-elle/-eille and -ogne/-oigne/-oine have been accepted as phonetically equivalent in the thirteenth and fourteenth centuries.

II La Chanson de Roland (Critical Text and Notes) 

5

10

Charlle li rois a la barbe grifaigne Sit anz toz plenz a esté en Espaigne; Conquis la terre jusqe la mer altegne. En meint estor fu veüe s’enseigne. Ne trove borc ne castel q’il nen plagne, Ne mur tant aut q’a la terre nen fragne; Fors Saragoze au chef d’une montaigne. La est Marsille qi la loi Deo nen dagne; Mahomet sert, mot fait folle gaagne. Ne poit durer qe Charlles ne le taigne, Car il n’a home de lui servir se feigne,

Folio 1a.

4. The C ms. reads: venue, with n inserted superscript in nearly full-size character between the first e and u (i.e. venue rather than veüe); but V7 reads veü. A reviser’s error in C? In either case, fu ve ue ses enseigne is hypermetric – although C treats ronpue C 3169, tenue 3967 and Josue 5055 as disyllabic before a consonant, so venue is probably read as 2 syllables; but vestue 6609 is read as tri-syllabic. V7 reads fu veü sa ensaigne but feminine agreement is often omitted; the alternative would be elision of sa enseigne to obtain the correct metre. C’s ses enseigne seems due to a simple scribal error producing duplication of se: veüe s’enseigne > veüe se s enseigne. 5. V7 reads praigne, present subjunctive 3, prandre, implying that C has confused l and r, but planer/plainer was regularly used to mean ‘tear down the walls’, ‘raze’ or ‘destroy’ in relation to towns; the -agne graphy provides the visual rhyme. 6. After this line, V7 inserts Cité n’i a qi contre lui se taigne, which gives a better antecedent for «Except for Saragossa…» and improves the balance of the text. The assonanced versions are more succinct: O line 5 reads Mur ne citét n’i est remes a fraindre; V4 11 approximates to O. Nonetheless, such an omission so early in C’s text seems to indicate an incredibly careless copyist; the extra line might well be V7’s own invention, improving on his source document. 10. Either ‘He cannot hold out that Charles does not seize him (i.e. Marsile)’ or preferably ‘It cannot last that Charles does not seize it (i.e. Saragossa)’. V7 here reads la taigne, which either indicates the latter meaning or, possibly, ‘…that Charles does not attack it (or him)’, reading l’ataigne (present subjunctive 3, ataindre): l’etaigne would then be either scribal error or variant graphy. 11. C: seigne OR feigne, the initial letter (long s) apparently lacking the bar to make an f, but neither seigner, reflex. ‘to cross oneself’, nor soigner, reflex. ‘to take care’, is apt in this context. V7’s faigne confirms that the reading should be feigne (< feindre, ‘to hesitate’, ‘to be slow’), which also makes it clear that the il in this line is Charlemagne, not Marsile.

104

La Chanson de Roland (Critical Text and Notes)

Fors Gainelon qi l treït por engigne. Jamais n’est jor qe li rois ne s’en plegne.

 15

20

25

En Saregoze ert Marsille li ber, Soz une olive se sist por deporter; Environ lui si demeine e si per. Sor un peron q’il fist a toz lister Monte li rois, si comence a parler: «Oiez, seignor, qe je vos vei mostrer; Consiliez moi coment porai esrer; Desfendez moi de honte e d’afoler. Ben a set anz ne sunt mie a paser, Li enperere c’on puet tant redoter En cest païs entra por conquister. Ars a nus castiaus, nos terres fait gaster; Cité n’avons qi vers lui peüst durer.

12. C: qil ti7t por engigne, with 7 rather than ~ over the i; the reading is therefore treït, a graphy of traït or trahit, past definite 3, trahir – which also gives the correct metre. This is confirmed by V7’s (hypermetric) qi le trahi par engaigne (‘who betrayed him by deceit’), which again indicates Charlemagne as the person betrayed. C’s qil thus represents a reduced form of qi le. 17. C: Sor un peron q’il sist toz lister, with the second hemistich incomprehensible and hypometric. V7 17: q’il fist a or lister indicates C’s s as a scribal error for f (copied from a line just above) and C’s omission of a (with lister: ‘border’, thus ‘gather round’?). V7 relates this line to C 15 (V7 16) and in 18–19 transposes C 16 and 17; but C 17–18 make a better integrated sentence. Instead of C 18 = V7 19, O 14 has Il en apelet e ses dux e ses cuntes (echoed by V4 20) to introduce the speech. 21. C: Desrender, V7 Desfender, but the imperative plural -ez ending seems to be needed, as in Consiliez in C 20 (cf. O 21: guarisez and V4 24: Ganrenta, the Italian imperative). The first conjugation infinitive ending may be due to contamination by esrer (C 20) and Italian difendermi if desfinder moi is seen as in explicatory apposition to esrer, dependent upon porai. 22. ne sunt mie a paser, ‘(which) are not still to be passed’, i.e. ‘which are already past’. 25. C: Ars a nus castiax…, the i being inserted superscript to make castiax tri-syllabic (although it is disyllabic, C 278 and 1025, and -iax is elsewhere a monosyllable in C). This is the first alexandrine to interrupt the normal 4:6 metre. V7 reads Ars nos castels, although ars is strictly the past participle (< ardeir). The source reading may well have been Ars nus castiax (read as 4 syllables), with ars for arst, past definite 3, echoing conquis, line 3, or a scribal error for art (present indicative 3, cf. fait, C 25). It is only in this line that C writes nus as a graphy of the possessive nos. 26. The C reading, peüst, imperfect subjunctive 3, pooir (‘would be able’) appears to make the second hemistich hypermetric; however, in qi Dex peüse garir, C 1860, the -eü- also reads as a single syllable. In contrast, where this word occurs later in the text, peüst (C 3347 and 3862)

La Chanson de Roland (Critical Text and Notes)

105

Mais a vos toz consel vel demander, Par qel enzing porai vers lui aler.» Mal soit de cel qi onsast mot soner,

is disyllabic, and Or repeüstes (C 6442) constitutes a 4-syllable hemistich. (See also Introduction, 92–93 regarding the varying value given to diphthongs.) Here, V7’s reading is the (equally correct but monosyllabic) present subjunctive 3, puist, modifying the meaning slightly. 29–33. In C 30, the cæsura is marked in C for the first time (using a punctus); se is inserted superscript. C 29 and 31 are identical to V7 30, 32 in the Duggan edition, but V7 lacks C 30 and 33 and greatly modifies C 32. Duggan recognizes C 30 as a CV7 source reading and inserts it as V7 31, correcting C’s scribal error (C: levasent where singular levast is needed, cf. osast, C 29; here the idea is ‘stood up to advise’). Foerster indicates in error that this line is already in V7. Thus V7 30, 32 confirm C’s reading (C 29, 31): ‘Woe betide anyone…’ followed by ‘Except Blanzardin’ – lines which are either the author’s comment on the situation or the Saracen nobles’ muttered reaction to Marsile’s request for counsel. Any idea that the writer is here anathematising those Saracens who put forward Blanzardin’s name (and thus set the tragedy in motion) is rendered untenable by the ‘except Blanzardin’ in C 31 – and the two imperfect subjunctives, 29, 30, plus vol, past definite 3, confirmed by V7, make it difficult to read this as a direct response by the Saracen barons. This must also be read in the context of the Saracen nominaion of Blanzardin, implied in C 34: Quant B. oït paiens conseillier’. V7 34 has ‘E B. fu mult bon chevalier’, V4 27 ‘B. est plus saçes civaler’, O 24‘B. fut des plus saives paiens’ in a purely descriptive line where V7 has similarities with V4. Problems arise in C 32, 33: ‘Ben tot le mont, si com orez nomer/Nen verez hõ tant sages mesajer’. After Marsile’s speech (ending C 28, V7 29, O 21, V4 24)) O and V4 have clear and succinct statements: N’i ad paien ki un sul mot respundet/Fors Blancandrins de Castel del Valfunde (O 22–23); No li ert païn che niente li responde/Ma tut lor teste verso la tere inbroçe (V4 25–26). After matching C 29 and 31, V7 33 ends the laisse in disjointed fashion, with a non sequitor: De val perdue si com oï conter. V7 33 apparently uses an assonanced source different from O or V4 in the first hemistich, but modifies the CV7 source in the second hemistich. Thus in this passage the V7 copyist rejects 2 illegible and/or incomprehensible lines (C 30 and 33) and makes what sense he can of C 32. As a result, V7 needs to replace C 34 (which fits well with C 29–33) with a different line. In C 32, Ben tot le mont is incomprehensible as it stands. It could possibly arise from a misreading of Ben tost le mostre: ‘He very soon shows it (i.e. his readiness to give good advice)’, but this involves three distinct errors. Alternatively, Ben might be a scribal error for Q’en, for the C copyist sometimes has difficulty in recognizing Q (cf. C 263 and 281, 267 and 1608, and Introduction, 88); but this does not fit easily into the context, after celer and before direct speech in C 33. However, whilst Foerster and Mortier both read: Ben tot le mont, the earlier transcription of C (quoted in Francisque Michel’s introduction to his 1869 edition of O) was Den – and this accords with the first word of V7 33. Emending B to D (two very similar capitals in C) would offer the following possibilities: (i) D’entor le mot, with mõt written in error in C, could mean ‘From all sides the declaration’, followed by direct speech in C 33 (although this involves considerable ellipsis); (ii) more persuasively, since both O and V7 specify a placename (albeit a different one) in respect of Blanzardin, the reading D’Entor le Mont or, better, D’Entoilemont/D’Entorlement would be possible, with the second t of tot a mis-reading of r or i and an allusion to entouillement (‘entanglement’, even ‘guile’ or ‘deceit’) in the place-name

106 30

La Chanson de Roland (Critical Text and Notes)

Ne qi levast son seignor conseilier Fors Blanzardin – cil ne se vol celer D’entoilement, si com orez nomer: «Nen verez hon tant sag’es mesajer!».

 35

40

folio 1b.

Quant Blançardins oït paiens conseillier, De vasalage fist asez a proisier. Prodome i est por son seignor aidier; Dist a Marsille, «Ne vos qier esmaier. Mandez Charllon, l’orgoillos e lo fier, Foi et salu por vostre mesajer. Tremetez li meint auferant destrier, Faucons müez por aler rivoier; Meutes de chiens li donez por chachier, Ours et lions por lui estragnïer;

(a typical pejorative implication in respect of Saracen place-names); (iii) finally, cil ne se vol celer d’entoilement (‘he did not wish to avoid entanglement/involvement’) makes good sense – and is the simplest and most satisfying reading. In the second hemistich, wear on the page makes nomer nearly illegible, but this was the consistent reading of all earlier editors. Perhaps ‘as you will hear (them) name (him)’ fits better with the Proper Name, but ‘as you will hear (people) nominate (him)’ is equally valid – and leads into speech. In C 33, N’enverez is an alternative transcription, but N’envoierez is the normal thirteenthcentury form – cf. C 220 and Pope 1934, § 973, so Nen verez is preferable; hõ may represent either hom/hon or honc (adverb, ‘ever); the plural object sages may be a scribal error (perhaps arising from tant) but is better read as elision: sage es > sag’ es m’. ‘You will not see anyone so wise in respect of messages (negotiations?)’ stands as direct speech by the Saracens, as they endorse Blanzardin’s nomination, and provides a direct antecedent justifying oït in C 34. The considerable corruption in C here may derive from errors in the CV7 source or from scribal errors in C – probably arising from wear on the first page of a source manuscript. In contrast, the V7 copyist, faced with the same or greater illegibility in the CV7 source, has the expertise and confidence to emend and excise so as to produce a meaningful reading. These lines are also discussed under I.4 Editorial Policy, Introduction, 89–90. 37. C’s qier, present 1, querir, may well be a scribal error for qielt/chielt/chaut (present 3, caleir, impers.). Cf. V7 37: chaut and O 27: ne vus esmaiez (imperative). 42. C reads: Meintes, a scribal error, as Mortier points out, for Meutes, ‘packs’; V7 42 reads moutes. C: Estranier, possibly a graphy of estrangié, past participle used as adjective, ‘foreign’, or an unusual variant of either estraigne/estrange or estraignier/estragnier – even possibly estraïer with inorganic Franco-Italian n. V7 43: esbanoier, similar enough to estranier to be a potential source of scribal error and perhaps more appropriate in a list of fauna to be sent to Charle-

La Chanson de Roland (Critical Text and Notes)

45

50

55

107

Cinqente chars li faites caroier Qi conblé soient de fins besanz d’or mer, Don il pora loer ment soldoier. Aut s’en en France; ben se doit repairer. Vos le jurez, a feste seint Micher Ses hom serez s’il le velt otroier, Trestote Espaigne en tenrez a bailier. S’il velt estajes, faites li envoier Ou .xv. o vint per lui mieus afaitier; E je i tremetrai lo fil de ma moillier, Por nen d’oucire sanz autre recovrer. Meus vel li rois le face detrenchier, Qe nos sofrons d’Espagne cel dangner.» Paien escrient, «Bien fait a otrier.»



60

Dist Blançardins li proz e li senez, «Per me poing destre qe vos ici veez, E per ma barbe dont li pels est meslez, L’ost des François lors desfaire verez. Chaschuns ira el reigne dont il fu nez; Charlles a Ais e ses riches barnez, Ou a Estanpes ou a Paris delez.

magne with their use specified in an infinitive: aler rivoier (C 41),chachier (C 42); but this may be V7’s embellishment. 47. Both C and V7 read Aut s’en en F…, with aut a variant of alt, Jussive subjunctive 3, aler, ‘Let him go away…’. This clumsy construction, with awkward repetition of en, involves treating the second hemistich, most unusually, as a separate entity. Contrast O 36: En France ad Ais s’en deit ben repairer and V4 40: Ad Asia en F… In thirteenth-century Italy, Ais en France might signify Aix-en-Provence, not Aix-la-Chapelle (modern Aachen) and thus imply rejection of Charles d’Anjou, the ruler of Provence controversially made King of the Two Sicilies by the Pope – a highly sensitive political issue. 50. C: terez, with missing nasal; V7 50: tenrez. This form of the future is a typical dialectal trait of the Northern region of France (cf. Pope 1934, 489, § 1320, xiii). 53. The alexandrine is perhaps deliberately used to give weight to the line; V7 53:Ge trametrai… 54. Both Foerster and Mortier read Sor nen d’oucire, but Por is the C reading, supported by V7 54: Par non doucire. Cf. O l.43: Par nun d’ocire; the line is not found in V4. This is the only instance of the spelling oucire in C, which elsewhere has oncire, occire and ocire (in all tenses). Cf. C 177: Por mon deu croire… and footnote.

108 65

70

La Chanson de Roland (Critical Text and Notes)

A seint Michel nos soit le jor donez; Trespasera li termes qi nus sera donez: N’osra de nos noveles ne vertez. Li enpereres est mot de grant fiertez, Qe nos ostages avroit lors degolez. Asez est meus qe vos les i perdez, Qe nos perdons d’Espagne les reignez, Ne q’i sofrons les dous ne les lastez.» Dient paien, «Bon conseller avez.»

2.a

 75

80

85

Li rois Marsille son conseil fait finer. Il en apele Clarin de Balaguer, Et Priamus, Gualane et Babuer, Estomarin et Orebe son per E Loënel et Marprinant de Mer, Et Blançardin por sa raison mostrer. Ce dist Marsille, «Or, baron, del erer! Al seje a Cordes porez Karllon trover. Branches d’olives devez o vos porter; Pais senefient, se:l volent creanter. Se m’i poez par enzing acorder, Terres et fiez vos ferai mot doner, Argent et or qenqe porrez mener.» Paien respondent, «Bien s’en doit hom pener.»

65. C: ne soit…, but the negative makes no sense; ne may be a scribal error for n9 = nos or an example of Franco-Italian ne = French en. 66. Once again, an alexandrine gives added weight to the line. In V7 66 the second hemistich becomes: li termes qi est nomez with elision of qi est to give a decasyllabic line. 69. C: avroit lore degolez, a hypermetric second hemistich, but lore (an unusual graphy of lors) is found nowhere else in C; it has been emended to lors, found 60 times in C. V7 69: avra tost decolez. 76. Two people in C and V7 (Galan e Babuer), but only one in O 65 (e Guarlan le barbét) and V4 66 (e Giraldo li barbé); in C, b is barred before Priamus. 87. C: Naion, perhaps intended as a graphy of Nairon/Noiron (infidels) with -ir- reduced to -i-; but V7 87 and O 77 agree on: paien, indicating a scribal error of a different kind. Elsewhere C uses Noiron (< Neron) only as the name of a Saracen god (cf. Proper Names).

109

La Chanson de Roland (Critical Text and Notes)



90

94a (V7) 95

Li rois Marsille a ses conseaus finez; Dist a ses homes, «Baron, or atendez. Al sege a Cordes sera li rois trovez. Branches d’olives en vos meins porterez; Pais senefie, zo est la veritez. De moië part l’enperaör direz, Per le suen deo, q’il ait de moi pitez; Ja nen sera cist moi passez, Q’a lui irai ou mels de mes casez. Crestïens serai batizez e levez; Jontes mes meins, sera siens comandez, Servirai lo tant qe serai finez.» Dist Blanzardins, «Bons mesages avez.»

 100

105

110

Dis blanches mules fist amener li rois Qe li tramist un amiral cortois. Freins ont a or, les resnes sunt d’orfrois, Seles d’argent, li estrier d’or grezois. Cil monterent qi sages sunt des lois, Branches d’olives porterent ce fu voirs; Pais senefie entre paiene lois. Par ce fu Charlles coroceus et destrois. Deus! qe dolors en France crut le mois, A Monleon, a Chartres et a Blois, Et an Anjou et per tot Hurepois.

2.b

92. senefie, pres.3, here and 106 and 245, because it is the carrying of olive-branches which is seen as the subject; but the normal plural is found in an identical clause, C 83. 94a. O 83 reads: Ja einz ne verrat passer cest premer meis, V4 83: Ançi che passi lo primer mes d’estez; the line, omitted in C through scribal error, improves the sense, but is not absolutely essential. 104. The -ent verb ending must here count as a syllable, which is unusual at the end of a hemistich. This may be past definite 6, but future 6 (monterent = monteront) better justifies the pronunciation. In C 105, porterent is trisyllabic, but in mid-hemistich; contrast C 245, 769 (erent), 1006 (convoierent), etc. 105. C: Blanches, but cf. C 91; voirs as -ois rhyme probably indicates assimilation of r. 106. Cf. C 92 and footnote. 110. C: Anjon, with n for u through scribal error, or possibly intrusive Franco-Italian n.

110

La Chanson de Roland (Critical Text and Notes)



115

120

125

130

Li enperere qi Franc doit justisier Lez fu et bauz, et tot si chevalier. Cordes ont prise, s:en fait les murs briser, A ses pereres abatre et depicer. Tel gaeng ont fait qe nus n’i puit priser, Or et argent et meint garniment cher. Les Saracins a fait toz detranchier, S’il ne vost croire et faire batiser. Charlles li Maines estoit en un verger, Enscanble o lui Rollant et Olivier, Sanses li dus et Anseïs li fier, Gui de Guascogne et Anceume et Guarner, Jofroi d’Anjou qi ert confenoner; Et anscenble o lui ot meint autre princer; De cels de France plus de .xvii. miller, Sus per les rens qi sistrent ou gravier. A tables joent por aus esbanoier; Et auquant d’eus joent a l’escachier; Et escremiscent cil baceler legier, Lancent et gitent por lor cors essaier. Desoz .i. pin, dejoste .i. oliver, S’asist li rois qi France a a bailier.

111. C: Frant; V7: Franç (graphy of Franz). Since Franz is found only once in C, but Franc regularly used as the plural form (and c and t confused on occasion), this graphy has been adopted. 114. C: A seo pereres; seo is either a scribal error for ses or a Northern/North-Eastern variant; V7 117 and O 98: ses. Pereres were projectile-launching machines. 117. In lines 113–117 the mixture of singular and plural verb forms appears to depend on whether the writer thinks of Charlemagne alone or all the French knights as the subject. 124. Once again an alexandrine. 125. C: .vii. miller, making the hemistich hypometric. O 109 reads quinze milliers, V4 107 .xv.miler, so C may have intended .xv., .xii. or even .xvii. (which all give the correct number of syllables), although V7 128 has troi c. milier. Finally, .xvii. hs been chosen as closest to the C reading and close in Roman figures to the O-V4 versions, which C resembles here. 126. ‘Up by ranks’, the French being seated in ascending order of rank or importance, makes sense here; but V7’s sor pailles d’or (‘on carpets of gold’, 129) and O 110 Sur palies blancs (‘on pieces of white brocade’), echoed by V4 108, may well indicate a scribal error in C. 127. C: auses esbanoier, with superfluous es barred. 130. C: cor for cors, ‘body’ (indeclinable); the graphy arises from contamination by lor.

111

La Chanson de Roland (Critical Text and Notes)

135

Cler ot lo vis, le cors grant et plenier; Blanc ot li poil come flor de lorer. Ses fiers sanblans fait mot a resoigner Sol qi:l regarde, ne l’istuet enseigner. Atant descendirent tuit li .x. mesager. Li rois salue Blançardin tot primier.

3.a

 140

145

150

155

Blançardin fist mervelles a loër. Devant le roi s’en est venuz ester; Mot doucement le prist a saluer, «Bau sire rois, cil deus vos puis garder, Qi fist le ciel, la terre et la mer; En seinte croiz laissa son cors pener Et el sepoucre cocher e repouser, Et au tierç jor de mort resusciter, Por cels q’il volst enscenble o lui mener! Per moi vos mande Marsilion li ber Qu’enqis avon la loi por nos sauvier; As cristïens se voudra ascenbler. De son avoir vos voudra mot doner: Chevaus de pris qi mot porez amer, .vii.c. muez ostors por rivoier, Ours et lions et veutres por vener, .L. cars qe vos ferez mener, De fins besans qi vos fera raser; Bien en porez vos soldoier löer. En cest païs ne clamez mais ester, En France a Ais poez bien reposer.

133. In C: grãt 7 b plenier, with b written in error and immediately barred. 136. C: Sol qil regarde lie li stuet enseigner. V7 139 reads Soit qil demant ne l’estuet e… and O 119 S’est kil demandet ne l’estoet e… Thus lie is scribal error in C, due to anticipation of li. 149. C: Gu’enqis, with an initial G – or a most unusual variant Q. 156. C: sis besans qi vos ferai…, but sis should read fins (cf. C 45, Blançardin’s speech to Marsile, with fins besanz in an identical context); scribal error due to omission of cross-bar of f, followed by omission of ~. There is often confusion of qi and qe in Old French, but this is probably q’i with i a thirteenth-century reduction of il. The implicit subject is therefore Marsile, not Blancandrin, and there is no good reason for the first person ferai; it is probably due to a misreading of fera/ferat, future 3.

112 160

165

La Chanson de Roland (Critical Text and Notes)

Marsilions voudra aprés vos aler, Cristïens ert, si s’avra fait lever. Jonte ses meins se velt a vus livrer; De vos tenra Espagne a governer; Servira vos tant cum pora durer. S’il ce ne fait, el neli velt rover.» Out le li rois, soi prist a merveillier; Clina son chief, si comence a penser.

3.b



170

175

Li enpereres un petit se pensa – Sa costume est qe par loisir parla. Dist as mesages, «Bien le savez peça Marsilions ainc gaires ne m’ama; De ceste couse qe il mandee m’a, Confaitement m’en aseürera?» Dist li paiens, «Sire, bien lo fera Por bons ostages qe il vos livrera; O .xv. o vint ou tant com vus plaira. Por mon deu croire, un mien fil i ara; Ja plus gentis de lui un soul n’en i a.

160. To give the correct metre, the two a of voudra aprés need to be elided into a single syllable. 161. ‘so he will have had himself baptised’, ‘so he will have been baptised’. 165. C’s spacing is el ne li velt rover, apparently introducing indirect object pronoun, 3rd person, li into a speech addressed directly to Charlemagne – perhaps through Italian influence. There is no evidence to warrant emending li to vi; but el neli velt r. may be read as ‘he wishes to ask no-one for anything’, i.e. ‘he does not seek to ask anyone for mercy’ or ‘these are his only terms’. If velt is a scribal error for valt (impers.) then ‘it is not worth while asking him for anything else’ (again ‘these are his only terms’) would make good sense. V7, adding an extra line, makes 165 part of a speech by Charlemagne: E respont .K., «Deus en voil mercier!/S’il ce me fait, ne li voil plus rover». This is supported to some extent by V4 127, Dist l’enperer, «Regracia sia Dé!». O 137 reads: Li empereres tent ses mains vers Deu. V7’s second line is not found in V4 or O. 172. C’s reading: mandeemande ma, with the second mande struck through; V7 75: mandee m’a. 177. Blançardin’s affirmation to Marsile, C 54: Por nen d’oucire, matched by O 43: Par nun d’ocire, ‘as one named to die, as one destined for death’ (F. Whitehead’s edition), is carefully modified here, as the envoy speaks to Charlemagne: ‘to give credit to my god’ or ‘as proof of my (good) faith’.

La Chanson de Roland (Critical Text and Notes)

178a*(V7) 180

185

113

En cest païs estes mult grant tens a, Bien a .vii. anz qe vostre jent i entra; En France a Aiz aler vos convendra. Marsilions aprés vos en ira; As boins a Aiz la loi Deu recevra; Crestïens ert, si se batisera; Jonte ses meins, a vos se livrera; Servira vos tant com il vivera.»



190

195

Beus fu li jor, si prist a decliner, Et li solauz se prist a esconser. Li rois comande le mules establer; Doçe serjants les livra por garder; Et les mesajes comande a osteler. El grant vergier a fait son trez lever; Vers Saragoze en fist le chief torner – Ce senefie ne s’en voudra aler; Tressc’une nuit se poise repoiser, Iloc au jor se voudra asteier. Li enperere ne velt mie oblier, Mise et maitines s’en ala escouter;

178a. There is no antecedent to explain i (‘there’) in C 179, but this is remedied by V7’s extra line. The final a of V7’s line may be a scribal error for ja: ‘You have been in this country a very long time already’; however, V7’s a estez may well be an Italian-style formal 3rd person addressed to Charlemagne (the meaning being the same). 182. C: la loi recevra, a hypometric hemistich; but clearly this is the Christian religion; cf. la loi Deu C 3826, 6102; there is no comparable line in V7, O or V4. 194. poise is a variant graphy of puisse, repoiser a graphy of reposer; cf. V7 200: se poisse reposser. Tressc’une nuit implies ‘…until (the next) night’, so that au jor in C 195 means ‘at daybreak’. 195. C: se vroudra asteler, with extra r inserted through scribal error; asteler could be a FrancoItalian graphy of osteler – cf. V7 201: se voldra hosteler: ‘If he can rest for a night, he will wish to lodge there (i.e. in Saragossa)’ – but this does not make good sense in the context. It could be a scribal error for estaler, with a and e transposed: ‘he would like to take up position to fight there’, since Charlemagne shares with Roland the view that the war should continue until Marsile is defeated. In C 828, asteler is an error for osteier (‘to campaign). But here se… asteler is reflexive and the most obvious scribal error is i read as l: se asteier, ‘to hasten’, a variant of se hasteier; this may imply ‘hasten to attack’ (cf. hastie, C 5546). 196. C: ne velt mi, with the final e of mie omitted, possibly because of the elision.

114

La Chanson de Roland (Critical Text and Notes)

Ses barons mande por consel demander; Por cels de France voloit tot jor esrer.

 200

205

210

212a(V7)

215

Beaus fu li jors, li soleuz esclariz. Karles li Maines, qi tant par fu hardiz, En est asis desoz .i. pin floriz. Un faudestue q’est de fin or masiz, Li rois de France demanois iest asiz; Puis fait mander de ses barons esliz. Ogiers i vint, li pros e li gentiz, E l’arcivesqe qi mot estoit noriz, Sanses li dus et ses freres Terriz, Jofroiz d’Anjou et li cuens Enmauriz, Acars li Mors et ses freres Almiz, Gui de Gascogne et Miles li Joïz, Li cons Rollant qi mot fu ses norriz, E Oliver sis conpainz e sis amiz; E si fu Guenes qi toz les a traïz. Dexor comence tes parole et teus diz, Dont douce France torna en grant esilz.

4.a

 Ce dist li rois, «Baron, or entendez! Conciliez moi au meus qe vos savez: Marsille m’a tramis bien .x. de ses privez, Per aus me mande – ne sai s’est veritez –

200. C: li soul est, but li soul is not found elsewhere meaning ‘sun’, est is the wrong tense after fu in the first hemistich, and V7 206 reads li soleuz esbaudiz; C’s reading is emended as a scribal error. 204. C: demanois i est asiz, which is hypermetric. V7 210: iest desus a.. The original reading may have been s’est demanois a… 212a. As Foerster and Mortier both note, the omission of Oliver’s name from this list is rectified in V7 219 although with an hypermetric second hemistich in which sis is perhaps repeated in error. 215. Both C and V7 222 read esliz (‘elect, important’), which makes no sense in the context but provides a good visual rhyme. Metathesis may be due to scribal error; clearly esilz (‘distress’) is needed.

115

La Chanson de Roland (Critical Text and Notes)

220

225

230

De son avoir m’envoiera asez, Ors et lions, veutres enchaïnez, Chevaus de pris, coranz et abrivez, .vii.c. chamels et mil ostors muez, .L. cars d’or et d’arzent rasez. Quant je serai en douce France alez, Il me sivra o mil de ses casez, Crestïens ert batizez et levez, Jontes ses meins sera mes comandez, De mi tenra Espaigne en quietez, Servira moi tant cum ert en santez. Mais je ne sai qels est li suens pensez.» Respont Rollant, «Certes, mar lo cresrez.»



235

240

245

Li enperere a sa raison fenie. Li cons Rollant o la chiere hardie En piez se drice, bien dist qe no l’otrie. «Droiz enperere, por Deo le fil Marie, Ja mar cresrez qe Marsille vos die. Bien a set anz, vostre grant ost banie En ceste terre entra par aaitie; Pris avons Nobles et Merinde saisie, Tote Vauterne et Prince la garnie. Li rois Marsille i fist mot grant bosdie, Qi ses mesages par dire felonie Vos envoia a masnie escherie. Branches porterent qi la pais senefie;

4.b

235. C: nol otrie. Pope (1934, 189, § 502) records nou as a regional development of nel, but no is also a graphy of non, the emphatic negative. V7 241: ne volt mie, O 194: ne l’otriet mie and V4: no llo otrio miga, using mie for negative stress; no in C and V4 may indicate a FrancoItalian influence. 241. C: Nauterne, but V7 247 and O 199 read Valterne. A scribal error in C. 244. Maisnie/masnie originally meant ‘the household or company of barons who constantly attend their lord’, but it is found in the twelfth-century Couronnement Louis and thirteenthcentury Girart de Viene in the more general sense of ‘company, suite, cortege’. Here, Roland seems to imply that this is an insultingly low number of envoys – only 10, despite Charlemagne’s status – but standard phrases such as this are regularly used to meet the requirements of rhyme, so it may not be highly significant.

116

250

255

La Chanson de Roland (Critical Text and Notes)

Par vos barons an fu raisons cellie. Deus de vos contes de plus grant segnorie Li tramesistes, ce fu mot grant folie; L’uns fu Basins et li autre Basie; Li rois Marsille fist mot grant desverie, Les chief en prist el pui soz Autevile. Sonez vos grailles, ne l’entrobliez mie, A Saragoze menez vostre ost banie; Tenez lo sege a tote vostre vie. Si vengiez cels cui joie il a fenie.»



260

Li enperere en tint lo chief enbron, Si duist sa barbe, son chief et son grenon; Toz coiz se tint, ne dist ni o ni non. Trestot se taisent, ne mais qe Gainellon; Il sail en piez, si vint devant Karllon. «Droiz enperere, jamay cresrez bricon, Moy ne autrui, se de vostre preu non. San ce por coi ne dient la gent Marsilion,

246. The form cellie (past participle, fem.to agree with raisons) is unexpected since the verb should be celer/celler. Pope (1934, 164, § 418) draws attention to a change in stem, ceile -ois, cf. Pope 1934, 162, §§ 408 (3), and 495, 1322, xxii) rather than scribal error. 1377. C: li solauz est esclaris, hypermetric, due to scribal error in repeating the syllable es/ est; thus: esclaris (past part. esclarir, used as adj., ‘shining’) with est understood from 1st hemistich.

172

1380

1385

1390

La Chanson de Roland (Critical Text and Notes)

Sonent mil graille por els faire ardis; Granz fu la noisse, si l’oïrent Francis. Dist Olivers, «Conpeing Rollant, oïs! Bataille arons, ce croi, de Saracins.» Respont Rollant, «Deus nos en soit amis! Por son seignor doit hon estre penis, Et endurer et grant calt et grant fris; S’en doit mult bien son cors avoir depris. Or gart chascuns qe grans cols a desplis! Male canzons n’en soit dites ne oïs! Paien ont tort, si com il m’est avis; Mavais exenple n’en doit estre enpris.»

22.a



1395

1400

Olivers est montez sor un pui alcior; Garde sor destre par mi un val erbor. Il veit venir cele jent paienor, Si:n apela Rollant le pugneor: «Devers Espegne voi venir tel brunor, Desoz paiens tant destrier coreor, Icil feront as Cristïens dolor. Guenes li fel en a fait traïsor, Q’il nos juja devant l’enperaör.» «Tais, Olivers!» respont Rollant le jor. «Mes perastre est, ne vel c’ait desenor.»

1385. C: grant calt et grant fris, which is hypometric; T 351 has: les chaux et les froys, V4 947: caldo, with an extra syllable. Reading et grant calt et … gives greater stress and a better rhythm; it also matches the pattern of V7 1448: Et endurer et les fains et les seis. 1388. C: n’en dites ne oïs, with soit inserted superscript. V7 1450 reads: n’en soit dite de neis and O 1014: de nus chantét ne seit. C is probably associating dire and oïr here, with the past part. oïs treated as a monosyllable (cf. Introduction, 92–93) but C’s reading may arise from scribal error, either by omitting the de found in V7 or by misreading neïs, adverb, ‘not at all’. (V7’s laisse rhymes in -eis.) 1395. C: bondor, but one cannot ‘see’ a noise or commotion. V4 956 and T 360 both read brunor, ‘glint of burnished metal’, with part of the verb veïr; O 1021: vei venir tel brunur; V7 1458: brunor in a modified line. 1398. C uses traïsor as a graphy for traïson, solely for the rhyme. V7 1460: Gueines l’a fait, je:l tieng a traïtor, ‘…I hold him to be a traitor’, modifies the line to avoid the mis-spelling. Cf. C 1226. 1401. C: Mes perstastre est, ne vel cain desenor, with extra st inserted in perastre; V7 1464: paraste, O 1027: parrastre, T 368 parrastre; cain might be a graphy of caënz (‘herein’), i.e.

La Chanson de Roland (Critical Text and Notes)

173



1405

1410

Olivers est desor un pui montez, Bien esgarda d’Espaigne les reignez, Vit Saracins qi tant sunt ascenblez. Luisont cil eume qi a or sunt gemez, Et cil escu et cil auberc safrez; Celes eschiles ne puet nus acontez; Tant en i a, toz en fut abosmez. Plus tost q’il poit, dou pui est avalez, Vint as François, devant toz l’a contez.

 Dist Olivers, «Or ai paiens veüz: Onqes enscenble Cristïens n’en vit plus; Cil devant sunt .c.m. o escuz; Eumes ont clers et blans osberc vestuz,

Roland does not want (to be tainted with) dishonour in this connection, but V7 reads chait or ch’ait, indicating c’ait (present subjunctive 3, avoir). O 1027: …que mot en suns (‘say a word about him’), V4 962: ne voi’ che mot ne son and T 368: ne veil ta deshonnour all confirm c’ait as the correct reading here. 1403. C: nuez, ‘clouds’, which makes no sense unless they are ‘clouds of dust’ or figuratively ‘storm-clouds’; V7 1466, V4 964 and T 372: regné, O 1029: regnét support the reading reignez. 1405. Luisont is probably a simple scribal error for luisent (present indic.6, luire), as found in O 1030, V7 1468 and T 374 (lusent in V4 966), but it could be a Franco-Italian graphy of luisoent (imperfect 6); otherwise -ont may possibly represent the Eastern French past definite 6 ending, originally used in conjugation I verbs, but extended to other types. Cf. Pope (1934, 496, § 1322, xxx). 1407. Again a past participle, acontez, where the infinitive is grammatically needed; poetic licence for the sake of the rhyme. O 1034 reads: ne poet il acunter, but V4 970: non poit hom esmeré is close to C. V7 1471 reads: nen seront pas nombré, as does T 376 (both having a laisse rhyming in -é). 1408. C’s reading: toz en sui abosmez, implies either the author’s comment or direct speech by Oliver, but O 1036: en est mult esguarét confirms that this is descriptive, with verb in 3rd person. The correct reading could be est (as in O), but fut is a more likely basis for error. V7, V4 and T modify the line. 1409. Just as oit = ot in C 1182, 1318, 1742, poit is a Franco-Italian graphy of pot, past def.3, pooir. 1411. C 1411–1533 are repeated in a rhymed version, 1534–2070. Cf. Introduction, 20, 28–32.

174 1415

1420

La Chanson de Roland (Critical Text and Notes)

Droites les anstes, ou bons espiez moluz; Bataile avrons, onqes mas tel n’ot nuz. Seignor François, de Deo aiez vertuz! El canp estez, qe non soiez vancuz!» Respont Francois, «Daait ait qe:l refuz! Ja por morir ne vos i faudra nuz.» Ce dist Rollant, «Deus li granz, nus aïuz!»

22.b



1425

Dist Oliver, «Paien ont grant esfor; De nos François me scenbla (mult) pouc avor. Conpeing Rollant, car sonez vostre cor; Quant l’ora Charlles, retornera alor.» Respont Rollant, «Trop fous seroie lor; En douce France en perdroie m’onor. Ainz i ferai grans cous de Durendor; Sanglanz en ert li brans deci a l’or.

1415. C: astes, with nasal symbol ~ omitted; O 1043: hanstes (‘lances’), V7 1482 and T 386: lances. 1416. C: onqes mas tel ne vit nuz, which is hypermetric; O 1044 reads: unches mais tel ne fut and V4 979: ma’unches tel non fu. C’s repetition (cf. laisse 101) omits the line, but V7 1482 includes it: onqes mes tel n’ot nuz. T 388 has: onc plus grant ne fu veüz. The V7OV4 agreement on onqes mas tel leaves the verb as the error in C, with the avrons in the first hemistich suggesting a matching n’ot in the second (i.e. adopting the V7 reading). 1419. C: Respont .R., «De da ait qe:l refuz!», but Oliver addresses the Seignor François in C 1417–1418 and Roland makes a separate response, C 1421. C’s later version (C 1541) reads Et cil respondent, «Ne soiez esperduz», as does V7 1485. O 1047 reads: Dient Franceis, «Dehét ait ki s’en fuit», which implies another scribal error in C’s hypometric second hemistich: the omission of ait from Daait ait (‘Accursed be…’). V4 982 has: Respont François, «Mal abia chi s’en fu!», either treating François as a collective noun or implying ‘each individual Frenchman’. This reading has therefore been adopted. 1421. ‘Great God, help us!’, aïuz being imperative singular, aïuer, variant of aidier. Cf. vencuz, C 1245. 1423. C: me scenbla mult pouc avor, a hypermetric hemistich showing Italian influence with pouc apparently the Franco-Italian form poco for pou, poi and scenbla perhaps = Italian present 3, not past definite 3. O 1050: m’i semblet aveir mult poi also has 7 syllables; but V4 985: me resembla aver poche and V7 (un-numbered line in laisse 92A in Duggan’s Appendix A, with rhyme in -ors): me scembla poi avores (with rhyme in -ors in this laisse) omit mult. The correspondence between C’s 2 versions of this passage is rather sketchy and offers no help here.

La Chanson de Roland (Critical Text and Notes)

1430

175

Felon paien mal monterent as por. Ge vos plevis, toz sunt jugé a mort.»



1435

1440

«Conpeing Rollant, car sonez l’olifant! Si l’orra Charlle, qi est as porz pasant; Je vos plevis q’il tornera esrant.» «Ne place Deu,» ce li respont Rollant, «Qe ja un jor por home m’espavant; Ne por paiens ne vel estre cornant; Ja n’en aront reprocie mi parant. Quant je serai en la bataille grant, Je i ferai mil cols et plus de zant, De Durendart l’acer verai sanglant. François feront, se Deu plait, ensemant; Ja cil d’Espeigne de mort n’aront garant.»

 1445

1450

«Conpeing Rollant, car sonez la menée! Si l’orra Karles de France l’alosée.» Respont Rollant, «Ne plaze Deu lo pere, Ne Marïen, la söe douce mere, Qe por paiens perde mon lois deriere! Ainz i ferai de Durendal ma spée, Jusq’a mes poinz en ert ensanglentée. Felon paien tant mar font ascenblée; Meus vuel morir qe France en soit blasmée.»

1430. C: montera, but felon paien is nominative plural and followed by sunt in C 1431; V7 (laisse 92A) has mar monterent…, O 1057: felun paien… vindrent., with both verbs in future 6.; V4 995 modifies the hemistich. 1432. C laisses 93 and 94 match the order of V4 laisses 82 and 83, reversing that of the other versions. 1433. C: as poz; O 1057: as porz; V4 998: al port. 1445. C: laloee with an s inserted above; the s is now rather faint and may well mark a later emendation of an original la loee, ‘the renowned’, to l’alosee. 1448. Again a Franco-Italian graphy, lois for los, ‘reputation’, ‘fame’; cf. oit = ot, poit = pot. 1449. C: ma bone spee.

176

La Chanson de Roland (Critical Text and Notes)



1455

1460

Dist Oliver, «De ce ne sa je blasme, Car veü ai les Sarazins d’Erable; Vestu en sunt le val et la montaigne Et li lariz et trestote la pleigne; Grant sunt les oz de cele gent estreigne, Et nos avons mult petite conpeigne. Sonez li cor, si l’osra Karlemeine.» Respont Rollant, «Mes corage ne cange. Ne place Deu, ne a seint ne a seinte, Qe ja por moi perde sa valor France! Enz i ferai de mon espié demeine. Por ben ferir l’enperere nus aine.»

23.a

 1465

1470

1475

Rollant est proz et Oliver est sage; Ambedui ont mult tres grant vasalage. Puis qe il sunt as armes sance faille, Ja por morir n’esciverunt bataille; Preu sunt li conte et de mult grant bernage. Paiene gent a grant tort se travaille. Dist Oliver, «Ci n’a mester tenzaille. Cist nos sunt prés, trop nos est a lonz Çarlle. Vostre olifant corner ne le degnaste. S’i fust li rois, n’i eüsem daumage. Garder amont, ça devers les porz d’Aspre:

1453. Present indicative 1, sa je is found regularly in Old French, cf. Fouché (1967, 151 § 74b): sa ge may result from the absorption of y of sai/say into the following prepalatal, or possibly be based on fa(s)-ge). Reduction of ai to a was also a feature of Northern, North-Eastern and Eastern French as well as of Franco-Italian: cf. Pope (1934, 488, § 1320, vii). Sa is found only 4 times in C (C 1453, 3316, 3357, 8061), always with verb-subject inversion, compared with 22 instances of sai. 1458. C’s Et nos avons petite conpeigne is hypometric, but O 1087 reads: …mult petite cumpaigne. 1475. Garder, infinitive used as imperative singular; O 1103: Gardez. At first sight, the Col d’Aspe seems to be wrongly identifed, lying too far to the east when Charlemagne’s army is heading north from Roncevaux towards the Port de Cize. In C 1950, C simply reads: les porz in this context, but les porz d’Aspre in C 1250. However, the worrying sight is that of the Saracen army (and particularly that section under Amauriz which is heading into the mountains to

177

La Chanson de Roland (Critical Text and Notes)

1480

Veoir pöez dolance a reregarde. Qi cest nos fist, jamais n’en fac’il autre!» Respont Rollant, «Ne dire tel outrage! Mal soit dou cuer, qi è ou ventre coarde. Nos remanrons a estal en la place; Por nos seront et li coup et li caple.»



1485

Quant voit Rollant la bataille sera, Plus fierement c’uns lions s’acresta; Franzois escrie, Oliver apela, «Sire conpeing, amis, traiez vos za! Li enperere qi ici nos lassa, Men esciant, ce croi, nos secora. Por son seignor qi se travalera, Je vus créant, bon loier en avra.

23.b

fight Gautier’s men) sweeping round into the mountains to the east so as to sweep down to the attack . 1476. C: Veoir pöez de lance areregarde. In the repetition, this becomes C 1951: La reregarde fait mot a redoter (V7 1899: redopter) – a rationalisation which could imply dotance (‘fear’ or ‘doubt’) here. O 1104: dolente est la rereguarde and V4 1050: dolent la reegarde confirm the readings dolent and reregarde (V4 attempting to rectify O’s hypermetric hemistich). Whitehead suggests reading O’s dolente as ‘apprehensive’, but this still implies that the French rearguard is troubled at the sight of the Saracen army and lacks confidence even before the battle. Since this can hardly have been the original intention of author or redactor, it is tempting to see this as a rationalisation of an early copyist’s error: arier/arer regard mis-read as a rereguarde. In the circumstances, ‘looking back’ (towards the approaching Saracen army) gives Oliver (and Roland) valid cause for concern and even apprehension. In C, dolance a reregarde has been retained for lack of supporting evidence for any alternative, but ‘cause for alarm on looking back’, would be a more satisfying reading. 1478. The infinitive dire again used as imperative singular (cf. C 1475 above), but this time in a negative command, where the usage is more usual (cf. C 951: Por noient n’en prier). O 1106 and V4 1052 both read: dites. In C’s corresponding (repeated) laisse 123, C 1950 reads: devrez a regarder, 1955: Tais, Oliver, ne dois desesperer, with considerable modification and expansion. 1479. The e is found as a single letter in the MS, although the present indicative 3, estre, ‘to be’ is clearly needed. This may be due simply to the omission of the abbreviation sign--above the e (= est) or may be a Franco-Italian form è. In the repetition, C 1956 reads qi velt acoarder, which is close to O 1107: ki el piz se cuardet but of no help in this line. Since another Italianism is found C 1491, the reading è has been adopted. 1481. C: Por nos serons.., the future 4 ending in error because of the preceding nos, whereas seront, future 6, is needed, because ‘both the blows and the sword-thrusts’ are the subject of the verb.

178 1490

La Chanson de Roland (Critical Text and Notes)

Se:l doit on faire, de zo ne dotez ja. Fer da la lance et je de Durenda, Ma bone espée, qe li rois me dona. Se je i mur, meillor vasal l’avra; D’un vasal noble ceste espée fu ja.»

 1495

1500

D’autre part est l’arcivesqe Turpis; Son cheval broche, monte sor .i. laris; François apele, un soul mot lor a dis: «Seignor baron, Charlles nos lasse cis. Por nostre roi devons ben morir uis. As Cristïens devons estre aidis. Bataille avrons, de ce sui je ben fis, Car a vos els veez les Sarazins. Clamez vos copes, si clamez Deu mercis; Asolvirai vos por vos armes garis;

1490. The spelling seems to be zo rather than ço, neuter pronoun, perhaps reflecting the pronunciation indicated in Pope (1934, 327, § 849). There is also the graphy za for ça, ‘here’, C 1485. 1491. Mortier and Foerster agree in reading da la lance; da is either a scribal error for de, due to the influence of la: de la > da la, or the Italian preposition da, here meaning ‘with’ rather than ‘by’. 1493. It is out of character for Roland to say of his sword ‘a better vassal will have it’. When this laisse is repeated in C, C 1973–1974 read: Se je i muert, dire poet qi l’avra, «Boen vasal ert icil qi l’avera» (with avera pluperfect 3) – a reading supported by O 1123–1124: Se jo i moerc, dire poet ki l’avrat… que ele fut a noble vassal, V4 2035–2036: Se nos muron, poront dir in tute part/Ch’in tot li mund no fu tante noble vassal and P 431: Iceste espee, vassaus hom la porta; but V7 1922–1924 read: Se je i moir, dire poez i ja,/Meillor vassal de moi ne la ara./D’un vasal noble ceste espée fu ja is closer to C 1493–1494 than to C 1973–1974 – and suggests that C 1493 reduces the common model of V7 1922–1923 to a single line in error since, if C’s avra is read as an abridged pluperfect 3, the two hemistichs become a non sequitur. The only other alternative is to insert ne before the verb, making the hemistich hypermetric. 1497. C: lor a gris, but in the repetition line C 1977 reads: lor a diz and O 1126: lur ad dit. In view of the copyist’s penchant for poetic licence, the reading dis (for past participle, dit) has been inserted. 1504. Use of the past participle, garis, where syntax requires the infinitive, garir, could be another instance of poetic licence, but Pope (1934, 156, § 396) attests to the effacement of prae-consonantal r and its frequent assimilation in the group rs. (It also seems a feature of V4, so possibly Franco-Italian.) The acceptability of -irs as rhyming with -is is confirmed by martyrs (C 1505), ferirs (C 1509); -is for -ir is also found in C 3218, 3226, 3457, 3463, 3464, 4950, 5691 (cf. Introduction, 83 (xlvii), final paragraph of 90 and Appendix B, 99).

La Chanson de Roland (Critical Text and Notes)

1505

179

Se vos morez, si i serez martyrs, Se en irez a joie en Paradis.» François descendent, si sunt a terre mis; Dist l’arcivesqe, «De Dé vos beneïs.» Por penitence lor comant a ferirs.

 1510

1515 1515a

François se levent, si se misent en piez; Bien sunt asols, qite de lor pechiez; Li arcivesqe Deu les a asseürez. Puis remonterent es chevaus abrivez, Atornent soi com chevalier menbrez, Et de bataille furent bien atornez. Rollant escrie, «Oliver, frere, oiez! «Segnor baron, un petit escoutez De Guenellon, qi ci nos a livrez. Pris en a le loier, sachez por veritez.

1510. C: misent, past definite 6, metre; misent is also found C 1990, but the normal OF form was mistrent or mirent. Exactly comparable with this are fisent (for fëistrent or firent, past def.6 faire) in C 1527 and prisent (< prendre) C 1943. These forms are a feature of Northern French (Wallonia, Artois, Picardy and N. Normandy) – cf. Pope (1934, 490, § 1320, xxx) and Fouché (1967, 288, § 146(d)). Linguistic features of this version are discussed in the Introduction, 75: Principal linguistic elements: Regional characteristics, A: N. and NE French, Morphology (viii). 1512. Before deu, the letter d on its side with comma above has been inserted superscript in darker ink to give the reading de Deu – but this makes the hemistich hypermetric and Deu can be read as indirect object of possession. Next is the cæsura, then C has the hypometric les a seürez, apparently reducing a asseürez in error. In the repetition, C 1991 reads: de Deu les a seignez as the second hemistich, identical to V7 1942, O 1141, which suggests a scribal error in C 1512: seignez mis-read as seürez. 1513. C: monterent, but the final -ent of past definite 6 (normally silent at the end of an hemistich) makes the hemistich hypometric. O’s laisse 90 closely matches C here, but O 1142 reads: Puis sunt muntez, with passé composé (as does V4 1072). In the repetition, C 1992 reads: Puis remontent… It has therefore been assumed that, in C 1513, the initial re-, possibly indicated by the symbol 7, was omitted in error. At the end of the line, abruez is written: that is, abrivez with omission of i. 1515a. In C, there is nothing to indicate who is speaking, although O 1145 has: L:i quens Rollant apelet Oliver. Since this passage is repeated later in C, the corresponding C 1994 has been inserted. 1518, 1519, 1521. These are a cluster of regular alexandrines, which C’s subsequent (repeated) version changes to regular decasyllabic metre in C 1997, 1998 and 2002. See Introduction 20– 21, 26–31 for further details of the first repetition.

180

1520

La Chanson de Roland (Critical Text and Notes)

Ay! Sire enperere! Coment nos vengerez Dou roi Marsille, qi nos a conparez! Esleger nos estuet as bons brans acerez.»

24.a



1525

1530

Es porz d’Espegne s’en est entrez Rollant Sor Velantif, son bon cheval corant. Porte ses armes, mult li sunt avinant, Et son espié va li ber balliant, Contre lo ciel vait mout bien torniant; Fermé i a un confenon luisant; Cors a gaillart, lo vis cler et riant; Si conpeignon aprés li vont siguant. Li arcivesqe de Deu les va seignant; Puis sunt monté en lor destrier corant; Adobent soi com chevalier vaillant, Et de bataille fisent mout bon scenblant.

 1535

Dist Olivers, «Je ai paiens veüz; Nus hon de car onqes mais n’en vit pluz.

1523. After this line, the MS continues with lines 1530–1533 inclusive, although this interrupts the description of Roland and leaves les in 1530 without antecedent, so that it is only by implication, and never specifically stated, that C 1531–1533 relate to the French warriors. However, in the corresponding laisse 129 in the repetition, C 2050–2051 are identical to C 1522– 1523 but are followed immediately by Porte ses armes, qi mult fu ben seanz and an expanded version of C 1525–1529 in C 2053–2061; moreover, when C’s additional 4 lines (Li arcivesqe… mout bon scenblant) are moved to the end of the laisse, the first 8 lines of this laisse exactly match O laisse 91 (C 1152–1160, omitting 1158). Repositioning the lines provides les (1530) with a clear antecedent, Si conpeignon in 1529, words which are also the subject of lines 1531–1533. It eliminates the need for .R. in 1524, inserted in C solely to identify the subject after the tooearly insertion of C 1530–1533, and makes this line decasyllabic. 1529. siguant is an irregular present participle (possibly an Italianism) of sivre/suïr, ‘to follow’; the normal form is sivant or suiant. 1533. See note to C 1510 for comment on the form fisent. 1534. The ‘first repetition’ in C begins here and continues up to and including C 2070 (see Introduction, 20–21, 26–31 for a detailled examination of this passage). 1535. C: onqes n’en vit pluz, a hypometric second hemistich; the matching earlier line (1412) reads: Onqes enscenble Cristïens n’en vit plus, but O 1040 has: unc mais nuls hom en tere n’en vit plus and V7 1478: Nus hon de char onqes mes n’en vit pluz. On the basis of O and V7, mais has been inserted.

181

La Chanson de Roland (Critical Text and Notes)

1540

El primier chief a ben .c. mille escuz, Eumes laciez et blans aubers vestuz, Les lances roides, les forz espiez moluz. Seignor François, de Deu aiez vertuz! El canp estez, qe no soiez vencuz.» Et cil respondent, «Ne soiez esperduz. Ja por morir ne vos i faudra nuz.»



1545

1550

Dist Olivers li gentius e li ber, «Sire conpeing, plaïst vos escouter. Se tuit les autre le voloient otrier, Sus en l’angarde vus convent a monter, Por les grant oz Marsilion mirer. Por cel Seignor qi tot a a sauvier, Lo bruel des lances ne puet nus hon esmer; E je ferai vos François ajoster, Et cez batailles rengier et ordener.» Respont Rollant, «Ne:l qier a refuser. Guenes li cons nus a fait mout pener, Mais por Celui qi tot a a juger,

24.b

1538. After this line, in the earlier version of this passage, C 1416 has: Bataille avrons, onqes mas tel ne vit nuz, which is found both in O 1044 and in V7 1482. In fact, V7 appears to combine C’s two versions of this laisse into one – and even inserts C 1421, which is not found elsewhere (cf. Introduction, 26–31). Once V7 has written (C) 1421 (the last line of laisse 91 and first line of folio 22 verso), it continues with the first 11 lines of (C) laisse 102 (C 1543–1553 inclusive), ending with the first line of folio 24 verso; it then has lines 2–10 of folio 22 verso (C 1422–1430) – and after that line 2 of folio 24 verso (C 1554). The implications of this important passage are discussed in the Introduction, 26–31. 1540. C: qe no soiez, but C laisse 101 duplicates laisse 91 in which C 1418 reads: qe non soiez. The reading no may be scribal error (omission of ~) or an instance of Franco-Italian: Old French ne > no. O 1046 reads que ne seium. 1542. After this, several laisses (102–117, C 1543–1866) are inserted before Oliver urges Roland to sound the horn. They have no counterpart in O and expand the text using standard phrases – but they chiefly describe the arming of the French peers, reminding the audience of their names after an interval. 1545. C: les voloient, with repetition of les found earlier in the line, through scribal error; V7 1490: le. 1554. V7 1500 adds: Se Dex en France me laisse retorner – as does P line 3, indicating a common source. Apart from this, P 1–7 closely match C 1553–1558, although the first words of P 1–3 are missing.

182

La Chanson de Roland (Critical Text and Notes)

1555

Mout durement li qier guesredoner, A trois roncins je lo ferai trainier.» Aprés cest mot, font la messe canter. Li cons Rollant i vait por escouter.

 1560

1565

1570

1575

Quant ot Karles sa grant ost devisee, Vers douce France a sa gent retornee. La reregarde a Rollant comandee; Et cil la fist, ne fu pas refusee, A vint mil homes de France la loee. Va s’en la nuiz, si est l’aube crevee, Beaus est li jor, clere est la maitinee; Li solauz lieve, qi abat la rosee; Cil ousel cantent parmi cele ramee. L’arcivesq ber a la mese cantee; Li cons Rollant l’a de cuer escutee; D’une once d’or l’a li cons honoree. Sena son chief, s’a l’imaige aoree, Ist dou mostier, s’a sa corpe clamee. Vint au peron, si demande s’espee; Cil li aporte cui il l’ot comandee, Et cil la ceinst q’en dona grant colee; A maint paien sera anqi privee.

1556. A line foreshadowing Ganelon’s fate, familiar of course to the audience. The final infinitive is usually found in the form traïner, development to trainer beginning in the thirteenth century; cf. Introduction, 92–93 regarding problems of metre involving -aï- diphthong. Cf note to C 2400. 1558. C: Li cons .R.lai vait…, with la added very faintly superscript, presumably by a revisor. It is required neither by the sense nor by the metre; and V7 1504 reads simply: i vait. 1568. C: arcivesq for arcivesqe, but the unabridged graphy would produce a hypermetric hemistich. V7 1514 reads simply Li archevesque, and P 17: Li arcevesques, both omitting ber. 1571. C: Scliena , with dot beneath s and e to indicate they are expuncted: giving clina (‘bowed’). V7 1517 reads: Seigne, P 20: Saingna – and ‘crossed himself on the forehead’ is a perfectly good reading. It seems sena was misunderstood, probably by a later revisor. The alteration appears unnecessary. 1572. C: corpe, a graphy of colpe (sf. < culpa, ‘confession’); P 21 has corpe, but V7 1518: coupe. It is not clear what mostier (‘monastery’) Roland is supposed to leave, other than the religious house later established at Roncevaux; clearly an anachronism, probably due to the scribe’s automatic association of the Mass with a church or religious house. 1576. ‘Will today be familiar to many a pagan’ – i.e. ‘many an infidel will have felt its force’.

183

La Chanson de Roland (Critical Text and Notes)

1580

1585

Sor Velantif a la crope trisee Salli li cons sanz nulle demoree, Pase avant outre a l’angarde montee. Soz son vert eume a la teste inclinee; A son piz sere sa grant targe roee. Vit des paiens mout grande aünée, Sisante mille a la primere hee; Por qatre liues ont prise la valee. Il les maudist de la Virge honoree: «Deus,» dist Rollant, «qi feïs mer salee, Men esient, ma mort i est juree.»

25.a



1590

1595

Li cons Rollant fu en l’angarde en son, Vit en cel ost meint Saracin felon; Grant .iiii. lieues qe de fi lo set hon Orent porpris entor et environ; El primer chief le roi Marsilion; Bien le conuit Rollant, li niés Charllon, As garniment q’il porte et au dragon. A une lieue erent ja li gloton, Ja porprenoient les terres environ,

1577. C: Sor Velantif atr a la trape trifee, the meaning of the last 2 words being unclear. V7 1523 reads: a la crope truisee, P 26: a la crope truilee, which points to scribal error: trape for crope (an abridgement of cropiere, ‘crupper’) and trifee for trisee, ‘plaited’ (or truilee, ‘woven’). As C 4993–4994 refer to a cropiere d’acer treslice, i.e. made of plaited (or 3-strand) steel fabric, ‘plaited’ may be what the scribe had in mind – but Duggan translates V7’s truisee as ‘rounded’. 1579. In the C ms, e is lightly entered and barred after l’angarde: Roland goes farther on through his forces, to the ‘observation post’ (ie the forward post, nearest to the Saracen army) which is already montée, ‘set up’ or ‘in place’, or possibly, ‘He goes up through his army, [he] has set up the forward observation post’. V7 (1525), conscious that Roland is in charge of the rearguard, changes the second hemistich to: a la garde montée (‘has mounted the guard’), but P 28 has C’s reading. 1584. ‘For four leagues they have occupied the valley’. V7 1530 reads: Por quatre sens… and P 33: Par .iiii. senz…, meaning ‘from four directions’(?), or otherwise sens/senz a mis-reading of leus (found in C). 1589. C: sa5r with the tironian symbol over the a; V7 1545: Saracin, P 38: Sarrazin. 1590. C: de si lo set…, but both V7 1536 and P 39 read: de fi (‘for certain, assuredly’). 1596. C: Ja porprenent, which is hypometric; V7 1542 reads: porprenoient, P 45: porprennoient..

184

1600

1605

La Chanson de Roland (Critical Text and Notes)

Quant li niés Charlle comença s’orison. «Deus,» dist li cons, «por ton santisme non, Qi en la Virge presis anoncion, Seint Daniel garisis dou lion, E Seint Jonas dou ventre dou peson, De mort a vie suscitas Lazeron, Et tu Seint Piere pousas en pré Noiron, Et convertis Seint Pol son conpeignon, Por pecheors venis a garison, E Susanë garis de faus tesmon, La Maudelene feïstes lo pardon, Quant a tes piez se coucha a bandon, Merci reqist par bona entencion,

1597. C: so orison; the cæsura is correctly marked after Charlle, but the repeated o could indicate lack of elision. V7 1543 reading s’oriscon and P 46 s’orison show that the correct metre is maintained. 1599. The nasal consonant or its abbreviation is omitted from anocion in the ms. The reference is to the Incarnation of Christ, as recounted in Matthew i.18–25, Luke i.26–38 & ii.4–11. 1600. A reference to the Book of Daniel in the Old Testament: Daniel vi.16–23. 1601. The familiar Old Testament story, recounted in Jonah i 11 – ii 10. 1602. The New Testament miracle recounted in John xi.1–46. 1603. C: Pere; the normal spelling Piere is found in C 1300 and 4084, and monosyllabic in both cases. V7 1549 changes this hypometric hemistich to: Pierres l’apostre but P 52 has: Et tu saint Pierre. In the second hemistich, pousas en pré Noiron – ‘placed St Peter in Nero’s fields’ – is a reference to the Vatican, Pré Noiron being a transcription of Prata Neronis, the site that became the Vatican: thus ‘made St Peter (the first) Pope’. The whole of this prayer, C 1598–1615, has a strong flavour of popular religion, being a very jumbled list of Old Testament, New Testament and Apocryphal incidents familiar to the audience. 1604. A New Testament reference: Acts ix.1–22. 1606. C: Et Susane, with final e sounded to obtain a 4-syllable hemistich; possibly a scribal error for Susana. The reference is to the ‘Susannah and the Elders’ story in the Apocrypha, in which false witness is given against the virtuous Susannah. C reads reston, a scribal error; V7 1553 tesmon, P 56 tesmoing. 1607. A reference to Mary Magdalene, mentioned by name in Luke viii.2, but also identified with the woman who poured ointment over Jesus’ head (Mark xiv.3) or his feet (Luke vii.37) and also with the ‘woman taken in adultery’ (John viii.3–11). 1608. C: Saut a ses piez, but the line has too many verbs and ‘jumps to her (his?) feet’ contradicts se coucha; C mis-reads a capital Q on occasion (cf. Introduction, 88). V7 1555 and P 59 read: Quant a vos piez, but Christ is usually addressed in the 2nd person singular and tes is closer to ses. 1609. C: bona, another example of Italian influence (cf. tuta C 433, quaranta 841, batailla 1146, etc.), but the final -a serves to indicate that there is no elision. Cf. Introduction, 78 re Italianisms.

185

La Chanson de Roland (Critical Text and Notes)

1610

1615

Tu le feïs gente remission; Totes ses pechiez pardonas as lairon, Quant te pendirent li traïtor felon, Si voirement com nos ben lo creon – Veingier me laisse dou conte Guenellon.» A cez paroles, descendi li frans hon.



1620

1625

1630

Li cons Rollant a s’orison finee. L’aigue dou cuer li est es els montee; Aval la face li est clere colee. Vers douce France a la reisne tornee, Droit a son tref a sa voie astee, Et vit François qi s’arment par la pree: Vint mil furent a enseigne doree. «Franc,» dist Rollant, «bone gent honoree, Sor totes autres cremue et redotee, Com vos voï de seignor esgaree! Vendu i somes par male destinee. La traïson ne puet estre celee; Com qierement sera guesredonee! Bataille en ert plene et aduree; Jamais por home n’ert si pesanz pensee.

25.b

1611. A reference to the thieves crucified alongside Christ, one of whom (according to Luke xxiii.39–43) was promised by Christ that he should that day be with Him in Paradise. 1612. C: quant tedependit li traïtor felon, a valid correction but the sense requires the deletion of prefix de (thus keeping the metre correct). This line has been mis-read and misunderstood by C, since the verb should be plural (cf. V7 1559: Quant te pendirent). Cf. also C 3921: Quant te pendirent li traïtor felon. C is often uncomfortable with verb-subject inversion, particularly with the verb number. 1614. After this, V7 inserts an extra line, 1562: Qi de nos a fait si mortel trahison, supported by P 66: Venduz noz a par male traïson – although this is a variant version in P. 1616. C: so orison, as in 1597; a scribal error, but son orison would make this hypermetric. 1619, 1620. C: a la reisne astee… a la voie tornee, but the normal phrase is aster la voie (‘move forward quickly’); V7 1567 reads la resne tornee, P 70: sa resne tiree – but V7 1568: voie menee (emended by Duggan to tornee). In C the two past participles have apparently been transposed through scribal error. 1622. C: tornee, identical to the original ending of 1620 (cf. note above) and therefore suspect, makes little sense; V7 1570 and P 73 both read: doree.

186

1635

La Chanson de Roland (Critical Text and Notes)

Dieus! C’or ne:l seit l’ost qi s’en est alee, Mar i entrassent ceste gent desfaee. Se plaist Yhesu qi la m’a comandee, L’arma dou cors qi Deus m’i a donee, Anz q’ele soit de mon cors desevree, I ferai tant de Durandal ma spee, Deci c’au poing sera ensanglentee. Aprés ma mort, en ert France doutee.»

 1640

Li cons Rollant descent dou pui aval, Enmi la presse des François ou est engal. Donqes lor conte dou jornal angosal: «Baron,» dist il, «n’en partiron par al.

1631. C: c’or ne:l sent…, but seit (present subjunctive 3, saveir) as in V7 1579 is clearly required: ‘Alas! If only the army which has departed knew of this, then this cursed race would find they entered here in an evil hour’. 1633. C: qi la m’a comencée, clearly a faulty reading. V7 1581 reads: qui ame m’a donée, but combines into one line C 1633 and 1634; in P, 84–85 are almost identical to C and P 84 reads: qui la m’a commandee, with la referring forward to l’arma (another Italianism, ‘the soul’) in the next line. 1640. The second hemistich is hypermetric unless one elides ou est into a single syllable; it is syntactically impossible to transform the line into an alexandrine. The meaning is presumably: ‘in the midst of the French where it is level (ground)’ (cf. igal, ‘flat area’, C 5912 and 8153) or ‘… where he is on equal terms’ (cf. engal, ‘equal’, C 5736, en igal, C 7877). V7 1588 reads: Enmi la presse des François done stal (‘…he holds his own’) and P 191: …des Fransois rent estal (‘…he gives battle’ or ‘.he returns to the charge’), both of which read as rationalisations of a difficult reading and fit poorly in the context of Roland’s actions here. It is possible that e+ ~ (= est) has been misplaced in C, or that two e, each with a symbol above, have been metathesized, en estal becoming est engal – the g being inserted to make sense of the new word; ou would then have been added to link the new clause to the rest of the sentence. Excising ou and reading en estal, ‘standing upright’ or ‘firmly in position’ to correspond in some degree to V7 and P would offer an alternative with the correct metre. But the suspicion remains that all three readings may rather derive from an earlier mis-reading of voisinal/ voissignal, adjective, ‘neighbouring’ or ‘near-by’, with metathesis of uo, iss > est and ingnal/ ignal > ingal; cf. the misreading of voisinage, C 7945. 1641. C: dou jonal angosal, with r omitted from jornal and angosal not known in this form: an unsatisfactory reading, probably due to scribal transposition of angosos jornal with the ending of angosos modified to suit the rhyme (and de le dolors jornal in C 1643 a possible source of corruption). V7 1589 has Donques lor conte del doloros jornal, followed by .de Deu i rois roial 2 lines later; P 92: Adont lor conte d’un angoissoz jornal, omitting C 1642–1643. The meaning in C is: ‘of the grievous day/day’s work? (that lay ahead)’.

187

La Chanson de Roland (Critical Text and Notes)

1645

1650

Or vos semon del doloros jornal Dou roi Marsille, nostre anemi mortal. Bien i ferez a la guise francal; As cols doner soiez tot paringal. Enprés nos mort, n’en soit dit negun mal! En la grant presse m’orez crier ‘Roial!’, L’enseigne Karle, mon seignor natural.» Li cons Rollant ot le cuer mot loial; Desoz son eume a regar de vasal; A voiz escrie, «Amenez mon cheval!»

26.a



1655

1660

Li quens Rollant n’a soing de donoier; Marsilions, cui Deus doinst engonbrier, Cevache a force por les Frans corochier. «Dieus» dist Rollant, «qi tot as a baillier, Ceste grant chace ne puet nus apaier.» Adonc escrie, «Armez vos, chevalier!» Al pié d’un mont, desoz un val plenier, La est armez li cortois Oliver. Il se vestet un blanc auberc dopler,

1643. C: de le dolors jornal, with each syllable needed for the metre; V7 1591 reads de Deu i rois roial, but Duggan substitutes the C reading. 1652. C: A une voiz, but the normal phrase is A voiz. V7 1600, which simply omits une, has therefore been adopted; but P 101 reads A haute voix and this is the last line of C’s laisse, where an alexandrine might well be deliberately inserted for effect. 1654. C: doist, for doinst, present subjunctive 3, doner. 1657. nus is of course a graphy of nuls, nominative singular. 1659. C: Al pui d’un mont, desoz un val plenier, but this is contradictory, with corruption due to an inattentive copyist’s use of standard phrases. V7 1607 and P 109 read: al pié, which has been inserted. C and V7 agree on the second hemistich, although P reads desoz un lorier and ‘under a tree of some kind’ is a standard formula (cf. desoz un pin C 131, 202, 571; desoz le pin 4018, 4139, 4148; desoz un lorer 6763; desoz un arbresel 7919). There may therefore have been corruption in the CV7 model at this point. 1661. C: Il se vestent, the last word written vestet with ~ over the second e; but the3rd person singular is required, to match the verbs in C 1660, 1664 and l665; and vestent at the end of the hemistich makes it hypometric. V7 1608 reads El dos li vestent…, but it has the plural, La ont armé…, in the previous line; P 111 has: En son dos vest…, as found in C 1679 and 1707. The CV7 model’s reading was probably se vestet, with ~ inserted in C in error. C 1662–1665, 1669– 1670 continue with verbs in the singular, as do both V7 and P, which here may offer variant emendations to a hypometric source.

188

1665

1670

1675

La Chanson de Roland (Critical Text and Notes)

Fort et espés et serré et entier; Quarals ne lance n’en puet maille fauser. En son cief lace un vert eume d’acer; Et ceint l’espée dan Girart au vis fier: Desoz Viene, en un vergier plenier, El grant estor mervellos adurer, Rollant le conte en fist agenoier. Prist un escu fort et grant et plenier; El front desus ot .iii. bocles d’or mer. On li amene son corante destrier Et il i monte, n’ot soing de donoier; Trestoz le fait desoz lui arçoier. Turpin encline, son front se fait seigner. Point le cival, outre le fait paser.

1665–1668. The Girart is Girart de Viene, Oliver’s uncle; C 1666–1668 refer to a major incident in the thirteenth-century chanson de geste of this name: the single combat between Roland and Oliver, as champions of their respective uncles – an exemplary fight in which both heroes showed great valour but which neither could win. Roland and Oliver then swore eternal friendship and Roland was betrothed to Aude. If the date of composition of Girart de Viene is correct, this provides a firm date before which these lines in C could not have been written – unless there were earlier oral traditions. There is no reference to Girart de Viene in O, so this is an example of the expansion of one cycle (that of Doune de Mayence and Aimeri de Narbonne) in order to supply additional background to another (the origin of the Roland-Oliver relationship in the Charlemagne cycle). The MS reading adurer is presumably poetic licence for aduré, the standard adjective applied to battles (cf. C 1146, 1629, 2317, 2359, 5395), with mervellos used adverbially, rather than mervellos a durer. 1668. C: fist a enoier would mean ‘…caused …to become exhausted’ but the V7 1616, P 119 reading agenoillier is probably the source reading, mis-read by C: a enoier < aienoier/ajenoier/ agenoier. 1670. On the evidence of V7 1619: La guinche fu d’un vermeil palig chier and P 122: La guiche en fu d’une vermeil paile chier, ‘the strap was of expensive red brocade’, C may have omitted a line here. 1671. corante, inaccurately feminine, supplies an extra syllable, needed unless destrier (normally 2 syllables) is read as tri-syllabic. It may show the influence of Italian corrente; C 1396 reads destrier coreor when 3-syllables are needed. This line is not found in V7 or P. 1672. C: qe n’ot soing…, the superfluous qe making the hemistich hypermetric. V7 and P lack this line. 1673. In the C ms., this line immediately follows C 1670, implying that Oliver bends his shield, but this is a standard formula used of a horse when a knight mounts (cf. C 2093); also C 1674 follows C 1675, again an illogical order. Instead of C 1671–1675, V7 1620–1623 read: El ferant monte, Rollant li tient l’estrier;/Trestot l’a feit desoz lui archier./Torpin encline, son cors li fait seignier,/Ultre s’en passe li tret un archier. As P 124–127 match V7 almost word for word, C’s line-order has been emended.

189

La Chanson de Roland (Critical Text and Notes)



1680

1685

Grant bruit demenent paien et Sarazin. Lez .i. costé, soz l’onbre d’un sapin, Ont adobé l’arcivesqe Turpin. En son dos vest .i. oberc doplentin, Et ceint l’espée dont li poinz fu d’or fin; A son col pent un escu bauvesin. On li amene .i. destrier peitevin. Turpins i monte a loi de palacin; Desoz son eume porte son chief enclin. Lés Oliver s’acoste le meschin; En cez deuus orent paien felon voisin.

26.b



1690

1695

Estouz de Lengres comence a parler: «Siri Rollant, pleist vos a escouter! Vers moi se traient trestoz li .xii. per.» Armes demande por son cors aprester. Un blanc osberc li ont fait endoser, Qarals ne lance n’en puet maille fauser; Puis a lacé .i. eume ou li ors reluist cler – Une tapace ot faite en son fermer – Et ot ceinte l’espee qi mout fait a amer. Devant lui fait son destrier amener; Estouz i monte a loi de chevalier, A aute voiz comença a hucier: «Siri Rollant, faites vus gent haster!

1679. V7 1628 adds: En son chief lace un elme pitein, supported by P 132: En chief li lacent .i. elme poitevin. A possible omission by C, but the line is not essential to the sense – and C 1682 ends with destrier peitevin, where V7 1631 has d. bon e fin, P 135 d. morentin. 1688. It is tempting to correct Siri to the normal Sire, but it occurs again C 1699 and thus may be intended to show a regional variant. 1690. C: demandet + ~ over the second e, but with a vertical bar through both ~ and t to correct this to the singular: demande. 1693, 1695 Both lines are alexandrines because of the compound verb tense; V7 1642, 1644 and P 146, 148 reduce both to decasyllabic lines simply by using the present indicative: lace, ceint. 1699. C: vus gent, with vus written in full, but presumably a graphy of vos; V7 1648: voz genz.

190 1700

La Chanson de Roland (Critical Text and Notes)

Paien chevauchent, ne vos qier plus celer; Pormi ces terres, les voi esperoner; A cez destroiz nus velent enserer.» L’escu au col, lait lo cheval aler; Tresc’a Turpin ne se voust arester.

 1705

1710

1715

1720

Aprés celui adoberent Oton Et adornerent lui a guise de baron; En son dos vest un osberc fremelon; Por maïstrise sunt ovré li giron, Batu a or entor et environ; Nus cols de dart, de lance ne de penon Ne:l forfeïst vaillisant .i. boton. Faiz fu les eumes por grant division: Une tapace i sist ou chief en son; Il l’i lacerent par grant affliccion. A son col pendent un escu a lion; Hanste ot de frasne et vermeil confenon. Un li amene un auferant gascon; De plene terre est sailliz en l’arzon. Atant s’en torne a cuide d’esperon; Le destrier broze; il cort por tel randon, Il vait plus tost qe ne fait un bouzon. Acostez s’est dalez le fiz Odon, Estolz de Lengres a la chiere fazon.

27.a

1701. C: cest terres, with anticipatory t inserted at the end of ces through scribal error. 1706. A deliberate alexandrine, where V7 and P obtain a regular decasyllabic metre by omitting lui. The use of disjunctive pronoun lui is strange, since one would expect a simple Et l’adornerent (4 syllables). It seems that C, having pronounced the final syllable of adoberent, C 1705, now pronounces the final syllable of adornerent as a result of (ungrammatically) inserting lui. 1716. C: Haste, with the nasal n or abbreviation ~ omitted; but cf. note on hauste, C 1143. 1718. At this point, in the first line of folio 27a, begins a new style of punctuation in respect of the cæsura: a punctus elevatus (the inverted semi-colon sometimes marked quite heavily) instead of the simple punctus used from folio 5a onwards. Also, from here on, every i is marked with a comma inverted over it (a kind of light, slanting stroke). There also appears to be a distinct change in the writing, especially in the size of the characters; and new abbreviations are used, eg. cõ for com, C 1733. Does this indicate a change of copyist? 1721. C: qi, but this must stand for qe, adverb, ‘than’; V7 1670 and P 175 read: que.

La Chanson de Roland (Critical Text and Notes)

1725

1730

191

Dist l’arcivesque, «Entendez mon sermon! Qi coart est, n’i vaudra un boton; Mais chascuns face sa vie garison: Tendez vos meins par grant devocion Envers Celui qui sofri pasion, Qi nos gasra de mort et de prison.» Turpins de Reins par bone entencion Lor fist de Deu gente asolucion. Aprés s’estregnent, irié come lion.



1735

1740

1745

Gerins s’adobe com hon de grant puisanze. Il vest l’osberc por la reconoisance, E lace l’eume ou ot mult grant fiance – A Sarazins voudra faire pesance – E ceint l’espée a la guise de France; El destrier monte a force sanz dotanze, Puis prist l’escu et a saissi la lance. Puis prie Deu, qi tot a en balance, Q’il li otroit sanz nulle demorance, De cele gent qi n’oit nulle creance Ferir i poist por söe vigorance; De lor amis i fera grant irance. Atant s’en torne, si a brandi sa lance; Acostez s’est dalez les bers de France.

1729. gasra, graphy of garra/garira, future 3, garir. 1732. C: s’astregent, but g often seems to represent ng or gn in mss. of this period, cf. poig for poign/poing, C 715, and Introduction, 72. 1734. A capital I in black ink, but the l of Il is formed by the (decorative) red line. C and V7 1683 both read: sor la r…, but this makes no sense. The hauberk would enable recognition of the wearer and P 188 reads por la r…, which is adopted by Duggan. 1736. C: voudre; V7 1685: voldra; P 190: fera; future 3. 1738, 1739. C’s line-order is (illogically) reversed in V7 1687, 1688 and P 192, 193. Gerins seems unnecessarily hampered when mounting his horse. C 1739–1740 are inset about 2 cm. from the margin because a fault in the parchment has produced a hole. 1741. C: Qi li or Q’il i, with omission of a second l: thus Q’il li otroit, ‘That he would grant him.’. 1742. oit is here a variant graphy of ot, past definite 3, aver. V7 transposes C 1742, 1743 but Duggan here prefers C’s line-order.

192

La Chanson de Roland (Critical Text and Notes)



1750

1755

1760

1765

1770

«Segnor baron, ceste n’ert pas celee: Li ber de France feront lor aünee.» Gerins s’en torne, ce fu verté provee, L’escu au col, la ventaille fermee. A Oliver a dit raison menbree: «Sire,» dist il, «dites vostre pensee. Pormi ces terres voi celle gent desvee; Ceste grant perde ne poet estre celee; Bataille en ert estote, renomee.» A ces paroles ont lor raison finee; Chascuns soz l’eume a la color muee. Puis prient Deu qi fist la mer salee Qe droit lor face de la gent desfaee. Rollant l’entent, s’a la teste levee, E dit parole qi fu bien ascoutee: «Baron, as armes! N’i ait meis recelee: La gent Marsille ai mult bien avisee; Ja nos sunt prés, a demie loee.» Gerers l’entent, s’a la color muee, De mautalant s’a la car tresuee. Por grant fierté a la broine endosee, Laza son heume, si a ceinte l’espee; A son col pent sa grant targe listee. El destrier saut, s’a la lance cobree; Puis laisse corre tot une speronee, Lez un brolet menuëment ramee. De cleres armes resplendit la contree. Gerers s’acoste lez la gent honoree.

27.b

1747. Strictly, ceste (also in V7 696) should be the neuter, cest, since it relates to the whole clause in C 1748, but it has been made feminine to agree with aünee in order to obtain the feminine rhyme celee. 1749. C: vertu; V7 1698: verté; P 203: veritez. The meaning seems to be: ‘this (speech) was proven truth’ rather than ‘his valour was proven’. 1761. This line, introducing Roland’s speech, is lacking in V7 or P, but inserted from C by Duggan. 1771. C: une speronee: the s impura without prosthetic e is a common feature of both NorthEastern French. Cf. Pope (1934, 492, § 1321, x) and Franco-Italian features, B (ix), 80.

193

La Chanson de Roland (Critical Text and Notes)

 1775

1780

1785

1790

1795

Baus fu li jors, s’est li solauz levez. Li .xii. per furent ja aprestez. Sor toz les autres est Rollant escriez; A vois escrie, «Baron, car vus hastez! Marsilions nus est prés, ce sachiez.» Dist Berenzer, «Mes armes m’aportez!» Et hon si fist, n’est hon pas demorez. Mult li fu tost ses hauberc aportez, E sus el chief ses vert eume fermez, E ceint l’espee a senestre costez. Bone est la lance, li fers fu acerez; Li confenons por maïstrise ovrez. Afichez fu sor les estriers dorez; Lo destrers broche par andos li costez. Croisa la mein, s’est un poi aclinez, Reclama Deu et les soies bontez: «Beau Sire Pere, qi en crois fus penez, Merci de m’arme qe li cors est finez! Se je i moert, de l’arme aiez pitez!» Atant s’en est l’escu al col tornez; Lez Oliver est li cons ajostez.

28.a

1785. Between li and fers, either t or c is barred. 1786. C: Li confenons fu, with the cæsura marked after fu, which makes the first hemistich hypermetric but the second regular. The scribal error could lie in inserting fu in this line, since it would be implicit from C 785. However, it may be an error in positioning the cæsura, since V7 1735 reads: Li confanons fu par maistrie ovrez. Once again the syllabic value of the -ai-, -aï- diphthong is in question, as discussed in the Introduction, 92–93. 1791. Rather a strange, even heretical, variant of Christian doctrine, if taken literally, but the Middle Ages liked to play with verbal paradoxes on the Trinity. In this verbal play on «I and my Father are one», the thought is simply that, since the First Person of the Trinity and the Second are of the same substance, then the First Person (God the Father) may be said to have suffered upon the Cross; but in its eagerness to show that the substance is indivisible, it falls into the error of ‘confusing the persons’. 1793. C: moert represents moerc, present indicative 1, morir, with palæographical confusion of t and c. V7 1742: muer. Cf. O 1122 and Pope (1934, 341, and 489, §§ 900, and 1320, xxviii), also C 1973: muert.

194

La Chanson de Roland (Critical Text and Notes)



1800

1805

1810

1815

Païen chevaucent, ne font pas lonc sejor. L’archivesche escrie par molt grant vigor: «Franc chevalier, soiez bon pognëor! Hui se desmostrerent li bon conbateor; Bataille aront, nus hon ne vit gregnor. Qi i mosra, il ert a joi tot jor En paradis, devant lo Creator.» Ez vus Gerart, un noble pugnëor; De traïson a gardé son honor. Il s’escria per mult ruste vigor: «Or tost as armes, por Deo nostre seignor! Marsilions chivauche por vigor; Granz est li bruiz de la gent paienor.» A lui armer en corent li plusor: Vest un osberc, nus hom n’en vit meillor; Un heume agu li lacent por amor; Une tapaze i sist el son auzor; Puis ceinst l’espee a la guise francor; A son col pent un escu pent a flor. On li a trait un destrier corëor; Sailli i est par force et par vigor; Treis foiz le point, si salt por tel valor,

28.b

1796. C: chevaucen, with the final t omitted in error. 1797. In C, a most unusual metre, for the cæsura is marked after escrie, giving either two equal 5-syllable hemistichs (with archivesche escrie elided) or 6:5 metre (without elision). V7 1746 has the same reading as C, but P 251 has: Li arcevesques s’escria par amor. Elsewhere in the MS, C consistently has Li archivesqe as the first hemistich (lines 1991, 2004, 2020, 2420 and 3469), but adopting that reading here (as Duggan does in his Critical Text) entails excising molt grant to obtain a 6-syllable second hemistich. 1799. C: desmostrerent (future 6, giving a 6-syllable first hemistich and therefore a typical C alexandrine. This could be a scribal error for desmostrent (present subjunctive 6) – cf. V7 1748: demostrent; P 253: demonstrent – but V7 and PTL regularly ‘correct’ C’s alexandrines to decasyllabic metre. 1817–1818. C: Treis fois se point, si salt por tel valor/…qi se tenist cel jor; V7 1766–1767 read: Trois fois le point et cil cort par valor./Soz cel n’a beste qi se tenist plen tor; P 271–272 echo V7 with …le hurte. and …qi se tenist plain tor. In the first hemistich of C 1817, le is far preferable to se: it must surely be Gerart who three times spurs ‘it’, the horse upon which he has just leapt, so that it bounds forward. However, C’s tel is vital later, since C 1818 provides the sequel. The same use of se tenir a, ‘to follow’ or ‘to keep up with’ is found in C 5161: Ne s’i tenist en corant un levrer and se in 1818 has been emended to s’i on this basis. There is no reason to emend cel jor, which makes good sense.

La Chanson de Roland (Critical Text and Notes)

195

Soz ciel n’a beste qi s’i tenist cel jor. Lez Oliver s’acoste par vigor.

 1820

1825

1830

Aprés Gerart s’arme Sanses ses druz; Forment se haste, qe mult est irascuz, E mult li poise qi cols n’i a feruz. Il laza l’eume don li ors est batuz, E por le maistre ses bon branz ot ceignuz; Bons fu l’espié, vermeil fu ses escuz. Tost li fu traiz li auferanz cremuz; De plene terre est el destrier sailuz; Trois fois li hurte, si fist li salz menuz; Plus tost il cort qe qarals destenduz. Aprés François est a l’eslais venuz; Rollant apele, qi bien fu entenduz. Gerart chevauche, a lor jenz est venuz; Sor toz les autres fu bien reconeüz.

 1835

Grant fu li bruiz de la gent paganie. Dist Anseïs, «Dame Seinte Marie, Vertu me done de cele jent aïe!

1824. ceignuz is poetic licence, making this a pseudo-rhymed laisse; everywhere else, C uses ceint, the regular past participle of ceindre. 1829. C: Plus tost li hurte… but this makes no sense and li hurte is clearly copied from the preceding line in error. V7 1778 reads: il cort. 1830. C: a lelais, a scribal error for a l’eslais, ‘with all his force’, ‘at full speed’. V7 1779 ends the laisse at this point, omitting C 1831–1833. 1831–1833. P 285–287 offer a different, more intelligible version: Rollant escrie si que bien fu oüz:/«Or, as chevaus! Lor homes ai veüz!»/Sor touz autres fu Marsiles connus. In C, there is no satisfactory sequel to C 1831, and the mention of Gerart in C 1832 is unexpected, as Sanses is the subject of this laisse. A confused reading in the source would explain V7’s omission of the lines. 1835. C: A deus aïe!, V7 1784: E deus aïe! but P 289: Dist Anseïs. Here each laisse relates to one of the 12 peers, so non-identification of the speaker indicates a scribal error in the CV7 source. 1836. Here, aïe = ‘hated’, i.e. haïe, past participle, haïr, feminine to agree with gent. The graphy may be deliberate, to echo aïe! (‘help!’) in the previous line: from time to time the remanieur indulges in a little play on words.

196

1840

1845

1850

La Chanson de Roland (Critical Text and Notes)

Guene li cons, cui Damedeus maudie, Nus a venduz par sa grant felonie. Cil nos aïut qi tot a en bailie.» Coche s’adenz, durement s’umilie; E resaili en piez, sa broigne revestie, E lace l’eume ou li ors reflambie, Au flanc senestre ceint la spee forbie; A son col pent une targe florie. Un li amene un destrier de Ongrie; Il laisse corre tote une praërie. De son espié a la hanste saisie. Puis prie Deu qi tot a en bailie Qe droit li tiegne de cele gent haïe. Dist Oliver, «Ne:l lasarai je mie, La moie enseigne trestut a resbaudie.» Li renc aprocent par mut grant estoutie.

29.a



1855

Rollant s’arme com hon de grant haïr. Un blanc osberc li font mult tost vestir – Ne l’i a home ne:s pen de lui servir – Et en son chief li heume enbastir. Son cheval fait et rengier et courir; N’i a cropiere de cendal ni de linir;

1841. Another regular alexandrine, reduced by V7 1790 and P 295 to decasyllabic metre as usual. 1843. la spee: once again, an example of the s impura without prosthetic e, cf. speronee, C 1771. 1847. C: haste, with missing n or ~; V7 1796: l’auste; P 302: la hanste. 1850. C and V7 agree here: the l of ne:l refers forward to La moie enseigne in C 1851. 1853. Roland was armed in laisse 103; P 308 reads Hues (ie Hiues/Iues/Ives?) as the twelfth peer. The CV7 source may have mis-read the initial H as .R. 1855. The sense is: Il n’i a hom, but both C and V7 (1804) agree on Nel i a hom, or probably Ne l’i a hom, ‘there is not a man belonging to him’, indirect object showing possession. The verb pen is present subjunctive 3, pener (reflex.), ‘to make a great effort’. 1856. C: li li heume. 1858. C’s hypermetric second hemistich reads de cendal ni de linir, with linir an unknown word. V7 1807 reads: ni de lir, with lir also unknown, but apparently (like linir) a corruption of lin (‘linen’) during conversion of -i- assonance to -ir rhyme. P 313 reads: ne samit, ‘samite’ being a richer silk material than cendal; samis is a possible variant graphy, but there is no

La Chanson de Roland (Critical Text and Notes)

1860

1865

197

En arçon monte, onqes n’i ot leissir. Con lui li .xi. qi Deus puisse garir! Mais Deus n’i plot, qi tot a a bailir: Per grant dolor les estuet morir. Ses conpeinons fait Rollant departir; A vois escrie, «Barons, alons ferir! Pongons vers aus, ses alons asailir!» Por mautalent fait son cheval fremir.



1870

1875

1880

Li cons Rollant ne fu mie esfréez. Devant lui fu Velantis amenez; Li cons i monte com vasal adobez. Dist Olivers, li proz et li senez, «Sire conpeing, envers moi entendez! Meintes foiz fui asaiez et provez, De cohardie ne fui onqes nomez; Vostre olifant, se il estoit sonez, Charlles le oïst, le roi encoronez; Je vos plevis, ja sera il tornez, Secora nos par bones voluntez.» Rollant respont, «De folie parlez! Ja Deu ne place, qi en crois fu penez Et el sepoucre cochez et repoussez! – Fors ad enfern fu nostre Sire alez Por ses amis traire de dolentez. –

record of the graphy samir. For the sake of the metre and because of the close C–V7 connection, V7’s reading has been inserted. 1860. Con (found also in V7) is an Italianism for cum, ‘with’. V7 1809 and P 316 specify .xii. peers. V7 continues: qi Deus poisse garir, with typical Franco-Italian confusion of qi and qe (as in C) but correct metre (P: Deus puisse g.). C’s peüse makes the second hemistich hypermetric. The form puist is found 10 times in C, but puisse (found 5 times) is closer to the V7 and P readings and to C’s peüse. 1862. C: estuet partir morir, perhaps partir being first copied from the end of C 1863 in error. 1867. C and V7 1816: esfrez, in error for esfréez, past participle, esfréer; the error makes the hemistich hypometric in both texts. P 323: esfraez. C repeats this error in 1894 (see below). 1875. C: Charlles le oïst, with oïst imperfect subjunctive 3, ‘would have heard’; the elision is not indicated but is clearly needed for metrical accuracy. V7 1824: le oit; P 331: l’orroit. 1882. A reference to ‘The Harrowing of Hell’, a popular mediæval story, loosely based on scripture. The syntax in C 1879–1884 is confused: 1879–1880 apparently constitute one sentence, followed by 1881–1882 as an aside – but 1884 is the logical completion of 1879. In P, C

198

1885

La Chanson de Roland (Critical Text and Notes)

Felon paien seront a mort tornez, Qe ja li cors soit de par moi sonez! Ainz en ferai de Durendal asez, Ma bone spée qi me pent a mon lez, Toz en sera mes braz ensanglentez. Meus vel morir qe face tel viltez.»

29.b

 1890

1895

1900

Dist Olivers a la chiere menbree, «Sire conpeing, car sonez la menee! Je le vos ai autre foiée rovee. Si l’osra Karles de France la loee; Secorra nos en estraigne contree; La gent d’Espaigne ne vint pas esfréee. Chascuns soz l’eume a la teste enclinee; Bien scenblent gent de bataille aprestee.» Out le Rollant, si redist s’alenee: «Ne place Deu qi fist la mer salee, Ne Marïen la poucelle honoree, Qe por païen i face ja cornee! Ainz i ferai de Durendal m’espee, Deci q’au poing sera ensanglentee. Felon païen mar virent l’ajornee; Meus vel morir qe France en soit blasmee.»

1883 becomes the penultimate line of the laisse; but, though formulaïc and suspect, it makes sense when read with 1884, with initial Qe meaning ‘rather than’. In this position it provides a good antecedent for Ainz, 1885. V7 1828–1832 has equally confused line-order, placing Duggan’s 1832 after 1835. The confusion results from introduction of additional religious allusions into the original text. 1888. At the end of this laisse, V7 inserts an extra laisse (CIX in Foerster), equivalent and very close to C laisse 93. In the Duggan edition, it is transferred to Appendix A as laisse 108a. 1891. C: foée roée, with foée for foiée and roée for rovée, past participle, rover. Presumably the latter is feminine to agree with la menée in the previous line, although the obvious antecedent is le in the first hemistich; on the other hand, roée may result from contamination by foée. V7 1840 reads: Ja le vos ai deus autres fois rovée and P 349: Que je vos ai hui autre fois rouvée. C’s foée roée seems a deliberate echo of the sound of a horn enjoyed by a rather casual, idiosyncratic jongleur. 1894. Both C and V7 here read: esfree, but the verb is esfréer and the past participle is here feminine to agree with gent, so an extra e is needed: esfréee. P 352 reads: esfraee. 1904. C: viel, but with a dot beneath the i to show excision. Also: Fance, without correction.

199

La Chanson de Roland (Critical Text and Notes)

 1905

1910

1915

1920

«Sire conpeing, encor vos vuel rover, Vostre olifant, qe lo facez soner. Se Karles l’out, qe France a a garder, Ge vus plevis, l’ost fera retorner.» «Ne place Deu,» ce dist Rollant li ber, «Qe por païen comence a corner, Ne de ma boche en doi estor geter, Ne douce France crot hon a reprover! Quant je devrai en bataille entrer, Adonc orez ‹Monjoie› reclamer, Per bon corage hautement escrier. Plus de .m. cols ferai a l’ascenbler, De Durendal ferai qi tant est fier, De cele espée qi mult fait a loër, Enprés ma mort les orra l’on conter; Tot en vesrez lo brant ensanglenter. Franc, se Deu plait, voudront aqi joster. Ja cil d’Espaigne ne s’en poront löer; Permi la mort les estovra aler.»

30.a

 1925

Dist Oliver, «N’en doi aver ontage. Je ai veü d’Espeigne le bernage: Covert en sunt li mont et li herbage, E li lariz, environ li boscage; Grant sunt li oz de cele gent sauvage.

1911. C: en doit estor geter, but present 1, doi, is needed to match de ma boche. V7 omits this line, which is intensely condensed; P 369 reads: en doie estureter (= estur jeter). Duggan inserts this line, translating geter as ‘reject’. Here ‘refuse battle’ or ‘throw away the chance of battle’ is preferable: ‘Nor should I, by sounding the horn, throw away the chance of battle’. Summoning reinforcements would imply that Roland refused the challenge of pitching the small French rearguard against the massive Saracen army – and thus undermined his reputation. 1912. crot is a graphy of croit/croist, present subjunctive 3, creire. Reduction of oi to o is frequently found in Northern, North-Eastern and Eastern French as well as Franco-Italian. Cf. Pope (1934, 488, § 1320, vii). 1917–1918. V7 1866 and P 375 combine C 1917’s first hemistich and C 1918’s second hemistich into a single line.

200

1930

1935

La Chanson de Roland (Critical Text and Notes)

Si m’est avis, segont lo mien pensage, Fust i li rois, nos i feron daumage. El cor soner n’est mie grant outrage.» Respont Rollant, «Ne me vint en corage. Ja Deus ne place, qi fist chascuns bernage, Q’i par moi France ait negun reprovage, Ne reprover i puist a hon de mon lignage! Asez vel meus defendre mon aage Qe cil païen aient de moi qevage, Ne nos par als perdon nostre heritage.»

 1940

1945

Rollant fu prouz et Oliver fu ber; Perigal furent et conpeignon et per. Puis qe lor vient a lor armes porter, Meus aiment a morir qe bataille esciver; Proz sunt li conte, haut prisent a parler; Païen chevauchent, si font le rens serer. Dist Oliver, «Or le pöez mirer. Vostre olifant ne degnastes soner; Lons nos est Karles, tart est del retorner; ‹S’il fust o nos,› ben le poöns jurer, ‹Ja cil d’Espaigne ne l’ouserunt penser›.

30.b

1930. Like C, V7 1878 reads ferons, future 4, where ferions, conditional 4, is needed. 1941. Another fault in the parchment causes the line to be inset 3 cm. C reads: Puis qe lor vivent, but V7 1889 has: Puis q’il vient and P 398: Puis que ce vient. The meaning is: ‘Ever since they were of an age to bear arms…’ or ‘Ever since they reached the age…’, but C’s vivent (an historic present), sits oddly alongside lor. The reading vient, found in both V7 and P, is a plausible basis for C’s error and only modifies the meaning slightly, q’il lor vient being impersonal in structure. In C’s earlier version of this passage, C 1467 reads: Puis qe il sunt as armes sance faille, matching O 1095: Puis que il sunt; and this suggests reading Puis qe il vivent… here in C. 1943. The verb prisent is past definite 6, prendre; cf. P 400: haut prinrent a parler. The form prisent is typical of the Northern region of France; it is found again in C 3735 and 3746, and similar forms, misent (< mettre) and fisent (< faire) are found in C 1510, 1990, 4935, 7736 and C 1533, 2865, 7289, 7358 and 7823 respectively. Cf. Pope (1934, 490, § 1320, xxx). 1949. The verb ouserunt is future 6, ouser. The substitution of u for o before a nasal is probably a Franco-Italian feature; other examples in C are mult, funt, volunté, although the frequency of use varies (cf. Introduction, 77, c (vi). This line is thus a quote in direct speech of what they are able to swear: «They will never dare…!». This is unsatisfactory syntax, use of the future being due to the influence of the present, poons, in C 1948, when the conditional, ousereient,

La Chanson de Roland (Critical Text and Notes)

1950

1955

1960

201

Envers les porz devrez a regarder; La reregarde fait mot a redoter. Cil nos feront les corages tranbler; Ja icel jor ne porons trepaser. Deus, pens des armes, qi tot a a bailier!» «Tais, Oliver, ne dois desesperer. Mal ait li cors qi velt acoarder. Quant je venrai as rustes cop doner, Nos remanrons a estal por chapler. La torneront mi demene et mi per; Quant seront mort, ben feront a plorer.»



1965

Quant veit Rollant qe la bataile arra, Tant par fu fiers qe lion rescenbla; A voiz escrie, Olivers apela: «Sire conpeing, e vos ne:l direz ja: Li enpereres qi François nus laissa, Son esciant cohart home n’i a. Por son seignor, quant hon bien l’amera, Doit hon sofrir ce qe lui avenra

or imperfect subjunctive, ousassent, is needed after the imperfect subjunctive 3, fust, in line 1948; P 407 reads: Ja cil paien ne l’osaissent panser, whilst V7 1897 changes the verb to oisisunt, adding an extra line (not found in C or P) in explanation: Qu’encontre nos venist a l’encontrer. Duggan emends V7 to osassent; but C’s ouserunt is unlikely to be a scribal error for ousassunt or ousissunt, cf. Pope (1934, 384, § 1045), past subjunctive 6 – although the -unt ending is attested in Pope (1934, 385, § 1048). 1952. In both C and V7 1900: Cil nos ferons, with cil referring back to cil d’Espaigne, C 1949; but, from C 1950 on, Oliver is warning Roland of the dangers facing them. Whilst C 1951 could mean that the French rearguard inspires fear, comparison with the earlier version of this laisse (C 96) and specifically C 1476 raises once again the question of L’arer regarde, ‘looking back’, instead of La reregarde; and Oliver goes on, C 1953, to forecast the slaughter of the French. P 411 here reads: Cist nos feront, showing that CV7’s nos is pronoun object, not pronoun subject – and the fact that Oliver is in despair here is confirmed by Roland’s response in C 1955. 1953. Præconsonantal s is effaced in trespaser, used here in the sense of ‘go beyond’, i.e. ‘survive’. 1963. A voiz makes good sense, but in C’s earlier version of this laisse, C 1484 reads Franzois. V7 1911 here reads François (and O 1112 has Franceis) – but the second hemistich reads Olivers apela, in all 4 instances. Only P 420 modifies this to A voiz escrie: «Olivier, que feras?» – and Duggan substitutes A voiz for V7’s François in his Critical Text. 1967. C: quant hon mout bien l’amera, a common phrase reduced to permit the 3-syllablefuture.

202

1970

La Chanson de Roland (Critical Text and Notes)

Et endurer lo mal q’il trovera; Lo cuers, les os et la car i perdra. Fier d’Autaclere et je de Durenda, Ma bone espée, qe li rois me dona. Se je i muert, dire poët qi l’avra, Boen vasal ert icil qi l’avera.

 1975

1980

1985

Li arcivesqe, qi proz fu et ardiz, Lo destrer broze, si monte un lariz. Franzois apele, grant sermon lor a diz: «Seignor baron, Karles nos a nosriz; Por nostre roi devons estre esbaudiz; Crestienté n’ait pas de nos mal diz! Bataille arez, si en soiez toz fiz, Car a vos elz veëz vos enemiz. Garde chascun ses pechiez ait gehiz; Quant vos avrai asous et beneïz, Qi puis mosra, saus est sez esperiz.» François descendent des destriers arabiz; Turpins de Reins, qi preuz fu e hardiz, De Deu les seigne, qi en crois fu laidiz Por nos sauver et le nos esperiz.

31.a

1973. muert, present indicative 1, morir, is probably a graphy of the Northern form muerc with palæographic confusion of c and t; V7 1922 reads: moir. Cf. C 1793 and Pope (1934, 341, § 900, and 490, § 1320, xxviii). 1975. In this laisse, C differs slightly from both V7 and P (which also differ from each other). Where V7 differs from (C) laisse 124, it is close to (C) laisse 98, the earlier version of this laisse. 1977. C: apelet + ~ over the second e, thus apelent; but 3rd pers. sing. is needed, the archbishop, not the French, being the subject. 1983. C: Garde be with ~ over e. After V7 1932, V7 adds: Clames vos coupes, si clamez Deu merciz!/Asolvrai vos por vos armes gariz, excised by Duggan. These are identical to C 1503– 1504 (= O 1132–1133). 1985. C: saus est senz esperiz, with senz (‘holy’) in error for ses through scribal confusion with Senz Esperiz (‘the Holy Spirit’); V7 1934 has: salç ert ses esperiz, but P 445 ends: de Deu soit tres bien fis.

La Chanson de Roland (Critical Text and Notes)

203

 1990

1995

2000

François se levent, si se misent sor piez; Li archivesqe de Deu les a seignez; Puis remonterent, si se sunt afichez Sor les chevaus provez et asaiez. Rollant escrie, «Oliver, frere, oiez! Or sai je bien qe veir me diïez, Qe Gainelon nos a toz engignez; Pris en a l’or qi mar en fu bailliez; Li enperere en ert mult corozez; E les Franzois qi Deus ot tant aidiez Nos vengeront as fers de lor espiez. Li rois Marsille de nos a fait marchiez; Mais nos espiés lor estoet asaiez.»



2005

2010

François monterent, ne se volent targer. Li archivesqe les prist a chastier: «Seignor baron, franc nobile guesrer, Bone parole vus voudrai acointer, Qe au ferir vus avra grant mester: Si a Franzois qi perda son dester, Mete le mein a l’espée d’acer, Si s’en defende a loy de chevaler!» Estouz de Lengres comence a hucier, «Seignor baron, ne vos calt d’esmaier

1990. An exact repetition of C 1510, with the same N-E French verb form, misent. Here C has no equivalent of C 1509 and 1511 (= O 1138, 1140, V4 1068, 1070), found as V7 1939 (not in P) and 1941 (= P 452); but C 1987–1989 modify C 1508–1509. C’s omission of 1511 is probably a scribal error. 1992. C and V7 1943: Puis remontent, a hypometric first hemistich, presumably due to the omission of 7 (= er); but, strangely, C’s first version of this passage has the hypometric Puis monterent (C 1513). In a closely matching laisse, O 1142 has Puis sunt muntez. It seems logical to combine C’s two versions and read Puis remonterent in both cases. 2002. ‘It was necessary for them to put our swords to the test’, with asaiez for asaier (poetic licence). 2008. C: Si a sia Franzois qi perda…, perda beingpresent subjunctive 3 with Italian -a ending. 2011–2020. C begins: Seignor baron, ne vus calt d’esmaier/Estouz de Lengres comence a hucier,/E vers .R. mult fort humilier… in a passage identical to V7 1962–1971 in line-order and text (apart from very minor variants). In both mss. the line Seignor baron…, a standard formula

204

2015

2020

La Chanson de Roland (Critical Text and Notes)

E vers Rollant mult fort humilier, Desoz son eume la teste a incliner, E la color et muer et cangier! Or le vesrez ferir et caploier! Entor son cors vos posrez eslaissier E vos chevaus trestoz en sanc baignier! Maint coup ferons au fer et a l’acer, Qe nos de rien n’i arons reprover.»

31.b



2025

Si com li renc se furent encontré, Es Engeler li Gascon tot armé, Sor un destrier richement acesmé. Cors ot gaillart, eschevi et molé; Soz son vert eume a son chief encliné; A haute voiz a fierement crié:

when opening speech, precedes the Estouz de Lengres line containing the verb of speech. When this line-order is reversed (on the model of P 473–474 and assuming error in the CV7 source) this permits the infinitives in 2013, 2014, 2015 to depend on calt (now in C 2012). After urging the French barons not to be downcast (C 2012–2015), Estouz points to Roland as an inspiration in the forthcoming battle and (C 2019, 2020) includes himself in the fighting. In C 2014 = V7 1965, son eume is singular because each individual French warrior is considered, but the plural le teste, found in both C and V7, is then illogical. It has been emended to la teste in C. Here, Duggan has emended V7 1965–1966 by substituting P’s readings: Je vos voi moult enz elmes embronchier, with the final word matching V7 1965, but elmes transposed from the CV7 first hemistich to replace testes, and E vos colors et müer… P 475–476 show Estouz addressing the rearguard directly (with vos replacing son and sa), then 477 brings in Estouz huimself (me replacing le/les) and 478 switches from vos to nos before continuing with the final lines. The concordance between C and V7 shows that the error lay in the shared source – and that P’s is a revised version. Duggan adopts certain features of the P text: the line-order, the first hemistich of P 475 (> V7 1965), vos colors in V7 1966, me in V7 1967 and mon in 1968 – but otherwise preserves the CV7 version. In the present edition, P’s modifications have been rejected because they so obviously result from the corrected line-order. (This also affects one’s view of the character of Estouz, who in P seems too dominantly self-promoting.) 2014. C: le teste a incliner; V7 1965: le teste enbrocher, with C’s superfluous a caused by the transposition of lines 2012, 2013 and therefore following the comence a hucier pattern in C 2012. 2016. C: ferir et cãploier, ~ adding an inorganic nasal (cf. V7 1967 and P 477: chaploier). 2017. C: En toz son cor…, but V7 1968: Entor son cor… makes better sense, with cor a graphy of cors (‘body’, ‘person’).

205

La Chanson de Roland (Critical Text and Notes)

2030

2035

2040

«Seignor baron, entendez mon pensé: De cels d’Espaigne ai bien l’orgoil miré.» Li archivesqe descent enmi lo pré, Contre orient a son chief acliné, Estroitement a Yhesu reclamé: «Damedeus Pere, por ta seinte bonté, Qe en seinte croiz eüs ton cors pené Et el sepoucre coché et repousé, Et au tierz jor de mort resuscité, Si cum ce est voirs, por ta sainte piété, Si garis hui Rollant, nostre avoé, E garis Karle, le fort roi coroné, E de nos cors n’i aions hui vilté!» Aprés cest mot, a son chival tiré



2045

«Seignor barons,» ce dist Turpins de Reins, «De vasalage vus voi hardiz et pleins. Car tendons or vers Damedeu nos meins, Q’il nos garise a ces cols primereins, E repairer nos laist et saus et seins. Conbatons nos a ces fiz a puteins; Prendre nos cuident come mastins soreins; Bien croi en Deo q’avrons toz les desteins.»

32.a

2037. C: Si garis hui h .R., the scribe nearly repeating hui in error. 2038. Similarly, C: E garis h .K…, almost repeating the line above. 2047. Both C and V7 1998 read: mastins soreins. Syntactically, this can only be accusative plural, so the term must be applied to nos, ie the French. Therefore soreins is probably a graphy of sereins, the Saracens seeing the rearguard as ‘quiet guard-dogs’, not alert to danger or not reacting to it and thus posing no problem. However, since P 509 reads: mastins soutains, it may be a scribal error (due to palæographic confusion of t and r) for soteins/soltains, meaning ‘solitary’ (‘few in number’?), ‘isolated’ (because cut off from the main army?) or ‘hidden’ (implying ‘undisturbed’?), although the meaning ‘inferior’, ‘low quality’ is found in fifteenthcentury texts. Alternatively, this could be an attempt to make sense of a difficult reading in a shared source, common to all three mss., and thus a scribal error for soueins, ie soveins, ‘supine’, so that they think to take the rearguard like guard-dogs who are lying down, illprepared for any attack. 2048. C: q’auriont les desteins, in which auriont seems to be a mis-spelling of either avrions (‘we should have’, conditional 4, aveir) with t in error for s, or avroient (conditional 6), -oie- > -io- through scribal error; and desteins here ‘good fortune’ rather than simply ‘destinies’. V7 1999 reads qu’arons toz les destreins (destreins being a scribal error rather than a graphy of

206

La Chanson de Roland (Critical Text and Notes)

 2050

2055

2060

2065

Grant sunt les oz des cuverz sosduianz. As porz d’Espeigne s’en est entré Rollanz Sor Velantif, qi forz fu et coranz. Porte ses armes, qi mult fu ben seanz, E ses osberc fu safrez jaceranz; Ses eumes fu blans et mult reflamboianz, Et ses escuz fu forz et bien tenanz; Espee ot ceinte, dont bien trenche li branz; Hanste ot grosse, et li fer fu trenchanz. Fermée i a un confenon tot blanz; Les langues d’or en sunt as poinz batanz; Cors ot gaillart, les elz vair et rianz. Tote sa jent est vers lui enclinanz, E cil de France dient c’est lor garanz. Vers Sarazins fu fiers e redotanz, Et vers Franzois humels et sospiranz; Il lor a dit .ii. moz mult avenanz:

destreits, ‘distress’ or ‘suffering’, with inorganic n). P 511 has: nostre en iert li gaains (‘ours will be the victory’), a paraphrase of C–V7 as a logical emendation of a difficult reading. Emendation based on V7’s arons toz has been adopted to obtain the correct metre; the future also has a more positive impact, as seen in P’s reading. 2049. C, V7 and P (laisse 28) are almost word-for-word identical in this laisse, which is the last laisse of the first repetition and also corresponds to C laisse 100 and O laisse 91. L laisse 1 begins here. 2050. C: poz; V7 2001 and P 513: pors. L line 1 corresponds approximately to this line. 2053. C: fu serez et jaceranz, a hypermetric hemistich. V7 2004 reads: fu sarrez jacerant and P 518: fu saffrez jazerans, both omitting et. Clearly safrez, meaning ‘damascened’ or ‘burnished with yellow/blue enamel’ – cf. Whitehead (1942, 160) – is the correct reading, to fit with jaceranz, ‘made of Eastern mail’. In view of V7’s and P’s readings C’s et has been excised , correcting the metre. 2054. eumes, originally omitted, was added at end of line with ∧ to indicate the place of insertion. 2057. C: hauste, for hanste; cf. hauste C 1143 (and footnote), auste 1847 and haste 1716. 2061. C: Totes sa gent est enprés lui enclinanz, with the second hemistich hypermetric. V7 2012: Tote sa gent est vers lui aclinant; P 525: Toz ses barnaiges est aprés lui sivant. Since the collective noun, gent, is here treated as singular, totes has been emended on the basis of V7; enprés could be retained if enclinanz were reduced to clinanz, with no change in meaning, but emending to vers gives a better flow to the line (and the en- of enprés could be due to anticipation of en- in the final word). 2062. C: c’es lor garant; V7 2013 and P 526: c’est… 2065. C: I lor… or Il or; V7 2016: Illor; P 529: Il lor.

207

La Chanson de Roland (Critical Text and Notes)

2070

«Seignor baron, sazez en remenanz Cist Sarazins vont lor mort porcazant, Ancui ferons les guaainz issi granz. Nus rois de Franze n’en fist ainc si vaillanz.» A ces paroles sunt les oz ajostanz.



2075

2080

2085

Dist Oliver, «N’ai cure de parler. Vostre olifant ne dignastes soner; Lons nos est Karles, tart est del retorner. Il ne set mot, n’i a colpe li ber; Cil qi la sunt ne funt mie a blasmer. Seignor baron, pensez del chivauchier. L’enseigne Karle non devez oblier; De la morz Deo devroiz bien remenbrer, As cols ferir, recevre et doner.» A ces paroles, font les oz ajoster. Qi donc oïst «Monjoie» reclamer, Cors et bosines et ces grailles soner, A grant merveille le poist hon escouter. Les destriers brozent, mult les font tost aler; Ses vont ferir sanz trestot demorer. Sor Saraçins se velt a esprover. Franc et paiens anqi vesrez joster.

32.b

 Si cum les oz se durent aprosmer, Li cons Rollant ot le coraje fier;

2066. sazez is a graphy of sachiez, ‘know’, with Franco-Italian z for ch and e for ie; the word remeãz in C results from omission of either n or the first ~ . 2069. C: aint, with confusion of t and c; V7 2020: ainc; P 533: ainz. 2071. The first repetition ends at this point; the laisse closely matches V7 laisse 120 and P laisse 29 – and also corresponds toO laisse 92. 2075. The use of sunt… funt is confirmed by O 1174, Cil ki la sunt ne funt mie a blasmer, V7 2026: Cil qui la sunt ne font… and P 539: Cil qui la sont ne font…; this is the sole instance of funt in C. 2077. The C reading non has been retained as a stressed negative (possibly Italianate?), although V7 2028 reads: nen, P 541: n’i.

208 2090

2095

2100

La Chanson de Roland (Critical Text and Notes)

Onqes cel jor ne volst croire Oliver. Ainz q’i nuit soit, le conpara mult chier. Brandist la lance, si ficha as estrer; Desoz lui fist lo cheval archier. Venu sunt tant Franzois et Henoier, C’aprocherent paien lo trait a un archier. Rollant escrie, «Or, baron chevalier!» La veïst on tante lance enpogner, Tante espée brandir, tant espi flanboier; Grant fu la noisse a lor lances baiser. Li niés Marsille lait cosre lo destrier. Devant les altres lo trait a un archier, Va quesre joste por son cors asaier.



2105

Al joster fu la noisse resbaudie. Li niés Marsile ne s’aseüre mie; Tot primereins, devant sa conpagnie, Va demandant partot chivalarie. Mot fierement a aute voiz s’escrie: «Felon François, Mahomet vus maudie! Mot est fel Guenes e plens de tricherie;

2090. C: croire a Oliver, in a hypermetric second hemistich; but neither V7 2041 nor P 553 has a. 2091. C: Ainz qi; V7 2042: Ainz qe; P 554: Ainz qu’en issist…, modifying the line. Possibly a scribal error, but Ainz q’il with reduction of impersonal il to i also makes good sense; cf. Pope (1934, 155s., § 392, and 324, § 841). 2095. C: C’aprocheret François (with conditional 3) but, after the plural tant Franzois et Henoier in 2094, conditional 6, present 6 or past definite 6 is needed; and François in 2095 makes no sense, but must be an accidental repetition of Franzois in 2094. V7 2046 changes this to the 5-syllable C’aprosment entr’els and P 556–557 have a much better sequel to C 2094: Venu i sont a force et sans dongier./Pres sont paien le trait a .i. archier. The verb has therefore been emended to past definite 6 (assuming omission of n or ~ in the ms.) and François has been replaced by paien (taken from P) to give a more sensible reading. 2096. V7 2047 also reads: O baron chevaler, but it is tempting to read both chevalier and chevaler as mis-readings of chevalcer, the infinitive used as an imperative as Roland calls on his peers to ride forward in attack. Lines 2097–2099 seem to support this reading. 2103. ‘As they joined battle, the clash of weapons became animated again.’ 2106. C: por tot, written in full, without abbreviation; V7 2058: partot.

209

La Chanson de Roland (Critical Text and Notes)

2110

2115

2120

Traï vos ha, n’enporterez la vie.» Rollant l’entent, li cuers l’en retentie; Point lo cheval, des esperon lo guie, Brandist la lance o l’enseigne balie, Fier lo gloton sor la tarze florie; D’un chief en autre li a frait et croisie, La soe broine destroite et desertie; Pormi le cors son fort espi li guie; Mort le trestorne, l’arme s’en est partie. Proz est li cons et de grant segnorie; Ja douce France de lui n’iert mais garie. «Ferez, François, Yhesus vos beneïe! En l’onor Deu, le fil sainte Marie, Cist primer cop sunt nostre, Deus aïe.»

33.a

 2125

2130

2135

Un duc i ot q’apelent Fauseron, Frere Marsille, si fu mot riches on; Cil tint la terre Datan e d’Abiron; En tote Espaigne n’en ot un plus felon. Quant son neveu vit gesir ou sablon, Ist de la presse, mist son cors a bandon; Vers François broze par fiere aatison. Il a escrié hautement a un ton: «Hui perdra Charlles de son los grant pezon.» Olivers ot cele fole raison; Lo destrier broce par fiere contenzon; Por tel vertu va ferir lo gloton,

2110. C: non porterez; V7 2062: n’en porterez or rather n’enporterez la vie makes the better sense: ‘you will not carry your life away’, i.e. ‘you will not escape with your life’. 2111. The verb is retenter (graphy of retinter, ‘to reverberate’ or here ‘to thump’), not retentier, but the scribe needs a rhyme in -ie here. 2114. C: Fier for fiert, present indicative 3, ferir, ‘to strike’; the effacement of the final t, especially after r, is confirmed in Pope (1934, 221, § 615, and 338, § 891). V7 2066: Fer; P 578: Fiert. 2126. A reference to Dathan and Abiram, the sons of Eliab, who rose up against Moses and were killed with their families and followers in an earth-quake and fire (Numbers xvi.1–35), thus personifying those who rebel against God and reject His prophets. 2130. C: Vers François p broze. 2135. C: va a ferir…, but the extra a is superfluous and not found in V7 2088 or P 602.

210

2140

La Chanson de Roland (Critical Text and Notes)

Escu ne brogne ne li fist garison; El cors li met lo pan del confanon; Mort le trebuche senz autre raiezon. Une ranpoigne li dist in ratrazon: «De vos menaces ne m’est pas un boton.» Puis escrie «Monjoie», la baniere Charllon: «Ferez, Franzois, car mult car nos vendron!»



2145

2150

Corsabrins fu un rois de grant haïr, De Barbarie doit la gent mantenir. Paiens apele, com ja posrez oïr: «Ceste bataille poons nos bien sofrir; De cels de France pöez mout poc veïr; Cels qi i sunt pöez mult vils tenir. Hui est li jor qe les convint morir.» Turpins l’entent, del sens cuide essir; Lo destrier broche, si fist l’espié brandir; Soz ciel n’a home qi si vint a grandir. Par tel vertu va cel paien ferir,

33.b

2136. C: ne li fu garison, again fu without the final t of the past definite 3: an early form, cf. Pope (1934, 372, § 998) or a scribal error. Although ‘was any protection to him’ makes sense, the usual construction is faire garison, ‘to offer protection’, ‘to protect’, and both V7 2089 and P 603 have fist, a better reading. 2138. C: raiezon, an expansion of raeson/raison, ‘speech’, so as to obtain three syllables; the meaning is: ‘without further speech (argument)’. V7 2091 reads: sanz nulle raenzon: ‘with no redemption’ or ‘with no ransome’ – but this does not fit with C’s autre. P 605 ends this line with: sans nulle autre ochoison, but 606 with: li dist en sa raison; and T 473 reads: Tout par ramposne ly dist une reson. 2139. C: in ratrãzon, with inorganic n; V7 2092 reads: en retrazon, ‘in reproach’, ‘in censure’. Cf also P’s and T’s readings quoted in note to C 2138. 2142. Again a play on words, this time on the word car: ‘For we shall sell ourselves very dear’. 2148. In view of the plural verb, sunt, Cel has been emended to the normal plural form, cels, as found in all other cases in C. 2150. C: ensir and V7 2103: ensir; this is essir/issir (‘to leave’) with inorganic n. Thus, ‘thinks to leave (ie lose) his senses’. P 616: le sens cuide marrir. 2152. C: qi si vint agradir; V7 2105: qi si soit a gradir. Duggan suggests gradit = ‘to thank’ – which may fit the V7 but not the C line. In C, there is a missing nasal, giving either a grandir, adverbial phrase: ‘more nobly’, ‘more impressively’, or possibly the verb s’agrandir (s’i vint agrandir), ‘to enhance one’s reputation’; cf. O 1244: Suz ciel n’at hume que tant voeillet haïr and P 617: ‘…qui tant les puist haïr’.

La Chanson de Roland (Critical Text and Notes)

2155

2160

2165

211

Escu ne broigne ne li pot garentir; Permi li cors li fait l’espié croissir, Enpeint lo bien, si le fait jus chaïr. Garde a la terre, vit lo gloton gesir; Dist tel parole qi bien fait a oïr: «Outre cuvert, trop savez bien mentir! Prodome est Karlle, bien le devons servir. Nostre François n’ont cure de fuïr. Vos conpeignons ferons les cuer partir; Novele mort lor covint a sentir. Baron François, pri vos de bien ferir. Cest primers cols sunt nostre, Deu plasir.»



2170

2175

Estouz de Lengres fu plens de grant vertu; Mot ot lo cuer dolant et irascu, Quant vit les rens qe si prés sunt venu. «E! Guenes fel, cest plet nos as meü!» Donc lasse corre a plen fren estendu. Brandist la lance del fort espi molu, Fiert l’aumanzor devant sor son escu, D’un chief en autre l’i a frait et fendu, L’osberc dou dos desmaillié et ronpu; Permi le cors li met lo fer tot nu; Enmi la place l’abat tot estendu. Outre dist il, «Paien, mal aies tu! Je ne di mie, Charlles n’i ait perdu.

2155. C: grosir and V7 2108: grossir, but ‘through his body he makes his lance grow large/ extend’ makes sense only if it means ‘he increases the size of the wound’ or, preferably, ‘he presses ever more of his lance through his body’. P 621 reads: fait l’espié croissir (‘makes the lance crunch’ or ‘makes the lance break’) and, since confusion of initial c and g is a feature of Franco-Italian whilst oi is reduced to o in Northern, North-Eastern and Eastern French as well as in Franco-Italian, P’s reading has been adopted. 2156. C: E peint and V7 2109: E peint, but P 622: Empoint and O 1249: Empeint. 2159. C: trop soiez, but both V7 2112 and P 625 read savez, which makes better sense. 2161. C: ot, with nasal consonant or symbol omitted; V7 2114 and P 627 both read: ont. 2173. C: frait e ronpu, but finding ronpu at the end of 2 consecutive lines (2173 and 2174) makes the reading suspect; V7 2126: fendu and P 639: fandu also provide alliteration. 2176. C: E mi, with nasal omitted; V7 2129: Enmi.

212

2180

La Chanson de Roland (Critical Text and Notes)

A grant péché somes e traï et vendu.» Li renc s’astrengent, tot ont fait liche nu.

34.a



2185

2190

Gelers fu proz, mot ot le cuer loial; Il laisse corre tot le pendant d’un val; Brandist la hanste au penon de cendal; Si vait ferir Malpui de Mont Pregal. Ses forz escuz ne valt un don de sal; Trestot li fent la bocle de cristal, L’osberc li fause; aprés li fist tel mal, Li cuer li part; mort l’abat dou ceval; Diable ot l’arme par mot grant trebal. Gelers escrie, «Monjoie la roial! Ferez, François, c’or somes comunal!» Donc se restregnent li conte natural. Dist Oliver, «Mult sunt Franc bon vasal.»

2179. The alexandrine (with cæsura marked after somes) gives added weight, dignity and balance. 2180. C: li chenu, but ‘the hoary-headed have done everything’ is unsatisfactory and tot is found nowhere in C as a graphy of tost. V7 lacks this line, although Duggan inserts C 2180 as V7 2133. P 646 reads: tost ont fait li cheü (cheü, past participle, cheoir); but ‘the fallen’ (subject form, plural because of li and ont fait) makes little sense with ‘have soon done’. (Is this ‘they have soon struck down warriors’?). In C, chenu might be a graphy of cheün (‘each’) with n and u transposed to obtain the rhyme, but ‘they have done everything, each and every one’ is only a marginal improvement. It is simpler to read li chenu as liche nu, with the words wrongly divided by the scribe and nu left in the masculine for the sake of the visual rhyme. There are then several possible meanings for liche (sf): (i) as a graphy of lice, ‘barrier, lists or enclosed field for jousting’ or ‘obstacle, hindrance’, thus ‘they have made the field completely bare’ or ‘…completely cleared all obstacles’, ie ‘cleared the decks for action’, ‘made ready to fight’; or (ii) graphy of lesche/lische, ‘metal plate’ or ‘sword-blade, thus ‘they have all bared their metal’, ‘they have all drawn their swords’. If the first hemistich is taken to mean: ‘The ranks come to grips with one another’ (i.e. ‘become involved in hand-to-hand combat’), this last hypothesis is the most apt and attractive. 2181. The word ot has been added very faintly in superscript. 2189. C: trebal; V7 2142: trabal; this Italianate graphy of trepeil, sm, ‘uproar’, ‘agitation’ (substituting voiced for unvoiced consonant) produces a hypometric hemistich. Should this be related to tarabat, ‘noise’, ‘banging’? P 655 reads: batestal and L 118: batistal in a variant line; and Duggan inserts batestal in his Critical Text. 2193. C: Dist lor Oliver, a hypermetric hemistich. V7 lacks this line, but P 653 reads: Dist Olivers…

213

La Chanson de Roland (Critical Text and Notes)

 2195

2200

2205

Grant bruit demenent cist baron chevalier, Et Sarazins, cui Deus doint engonbrer, Seus partiront jusqe a trois .c. meillier; A nostre Franzois livreront estor pleigner. Ce dist Gelins, «Or m’i tieng por lainer, S’or ne m’eslais, ne me pris un dener.» Le destrier broze des esperons d’or mer, Brandist la hanste dont bien trenche l’acer, Fiert l’amoraive sor l’escu de quarter; Teinz ne blazon ne li valt .i. dener; Son confanon li fait el cors baigner; Mort le trestorne par dalez un senter; L’arme de lui enporte l’averser.



2210

Mot fu grant li estor, nus ne:l pot endurer; A grant dolor l’estot a desevrer. Anseïs fist mult fort a loër: Lo destrier broche, mult le fist resaler, Brandist la hanste, va a Torgis joster – De Tortolose se fasoit roi clamer. Sor son escu li va grant cop doner;

34.b

2195. C: engobrer, with missing n or ~; V7 2148: engonbrer. 2196. C: Seus partirent in an apparently hypometric hemistich, but livreront in C 2197 shows that the reading should be future 6: partiront. Seus could be read as a graphy of ceus, ‘these’, ‘the latter’ or the adverb, ‘confidently’. (V7 2149 reads this as S’en partirent in an hypometric hemistich – which Duggan emends to S’en sont parti on the basis of P.) 2197. An unusual division of the alexandrine for C: 5 syllables, then 7, with cæsura clearly marked. The metre could be corrected by emending nostre to nos and reducing livreront to livront through effacement of the intertonic e between r-r: cf. Pope (1934, 366, § 970). V7 2150 modifies the line but retains an alexandrine: A François leviront estor grant et pleigner, with 6:6 division. 2199. The verb ending in m’eslais looks strange, but this is a compound of laissier, for which Fouché (1967, 110 §52) instances lais as present indicative 1. 2201. C: haste once again for hanste; the same graphy is found in C 2211. Cf. note to C 1143 and Introduction, 80 (xx), final item under Phonology. 2204. C: Sor confanon; V7 2157 and P: Son… 2211. See note to C 2201.

214

2215

La Chanson de Roland (Critical Text and Notes)

Teinz ne blasons ne le puet contrester, Ne li osberc garantir ni tenser; Permi lo piz li fait lo fust paser; Mort le trebuce sanz braire ne crier; L’arme s’en va en enfer osteler.

 2220

2225

Ez Engelers li Gascons de Bordele Repoint e broche lo destrier de Chastelle, Brandist la lance et l’enseigne vermeille, Fiert Estormi sor la targe novelle: D’un chief en altre li fent et esqartelle; L’osberc li fause par desus la gonelle; Li cuers li part, mort l’abat de la sele. Dont descent Ote le val d’une conbele – Li cons Rollant le conduist et chaële – De son espié trenche si l’alemele,

2217. Both C and V7 here read brair, which would produce a hypometric hemistich unless read as braïr. The same spelling occurs in C 2746 (sanz brair, sanz crier) with the same effect. The normal spelling is braire or brere and Pope (1934, 406 and 410, § 1063) gives this verb as corresponding in pattern to traire (although the etymology given in Pope (1934, 295, § 750), < rugire + Celtic *brag, may make braïr possible). The traire/braire pattern in these two cases may have been contaminated by trahir. On the other hand, the scribe may have had the verb braier or broier in mind, sanz broier meaning ‘without hesitating, without more ado’; there are in both C and V7 examples of celer treated as an -ir verb (cf. note to line 246). This would again involve omission of e through scribal error. It seems best to emend in each case to braire, assuming the writer’s intention to be one of intensification through use of synonymous verbs, braire and crier, thus ‘before he has chance to make a sound’. 2219. C and V7 2172 have an identical hypometric first hemistich, consisting solely of the Proper Name, but this is due to scribal error in the model. Cf. O 1289: Et Engelers…, P 685: Ez Engelier… 2226. This line is matched by V7 2179: Dont descent ote le val d’une combelle and, less exactly, by P 690: Dont descent Othes le fons d’une vaucelle. Clearly, someone ‘then goes down the slope of a little valley’, but it is surprising to find Otes (the subject of the next laisse, beginning C 2231) rather than Engelers named here. Normally, each French peer has one laisse devoted to him; but the cæsura must be placed after ote and cannot divide tot le val… (P, having mentioned Othes in line 690, ends the laisse after 691 (= C 2227), omitting C 2228–2230 and proceeding immediately to 2231 – presumably so as to continue with Otes. O has 8 lines on Engelers, laisse 100, lines 1289–1296, with 1295 matching C 2225, but O 1296 completely different. V4 and T omit this line.)

215

La Chanson de Roland (Critical Text and Notes)

2230

Ainz q’il retort, ara fait tel novele, Dont Sarazins perdera la boële.



2235

2240

Otes fu proz et vasal mult valant, Lo destrier broche, mot le va semonant. Brandist la hanste del fort espi tranchant, Si vait ferir un paien Estorgant Sor son escu, en la pene devant. Teinz ne blazons ne le valt un besant; Tot li desront son osberc jacerant; El cors li mist l’enseigne verdoiant; Mort le trebuche de son cheval corant. Une parole li dist par avenant: «Otre qivert, ja n’en arez garant.»



2245

2250

Berengers fu corageus et hardiz; Brandist la hanste de l’espié moliz; Fiert un paien qi a non Estormiz. Grant cop li done sor son escu voltiz, Tot li estroe lo teinz et la verniz; Fausé li a lo blanc hauberc tresliz; Son confenon li met par mi le piz; Mort le trestorne entre mil Arabiz. Dist tel parole dont mult fu bien oïz: «Otre qivert, de Deus soiez maudiz!» Des .xii. pers, li dis en sunt ociz; Ne mais qe deus n’en i a remés viz: Ce est Cornubles et li rois Margariz.

35.a

2233. C: hauste. Cf. notes to C1143, 2057, 2201, 3424 and Introduction, 80 (xx), under Phonology. 2236. C: vast; V7 2189: valt; P 697: valut. 2243. C: hauste. Cf. note to 2233 above.

216

La Chanson de Roland (Critical Text and Notes)

 2255

2260

En Margarit ot mot bon chevalier; Bals fu et proz et le cors ot legier. Le destrier broche des esperon d’or mer, Brandit la lance al confenon pleigner, Sor son escu vint ferir Oliver; Lonc le costé li fait paser l’acer; Deus le gari, q’en car ne:l pot tochier; Sa lance froisse, ne le pot desrochier; Outre s’en passe, qe n’i ot enconbrier; Sone son grasle por sa jent ralier.

 2265

2270

2275

La bataille est mult merveillose et dure. Fiert les paiens, mie ne s’aseüre Li cons Rollant, tant com l’anste li dure. Puis trast l’espée, d’or est l’inoldeüre; Fert un paien de mot grant estature: Nom ot Cornuble, nez d’une terre dure; Tot lo porfent tresq’en la forcheüre, E lo cheval de ci qe la jonture; Trestot l’abat el pré a la froidure. Une ranpoine li dist a desmesure: «Deus, qi tot fist, te doinst male venture! Ja de bataille n’averas jameis cure.»

2266–2267. C: aste (also in V7 2219), cf. C 2201, 2211 and note to 2201. C’s line-order is also found in V7 2218–2219 and, whilst it is quite exceptional to have verb-subject inversion stretching over 2 lines, involving two verbs and a Proper Name (and without any grammatical need), this certainly gives added impact to the verbs. However, O 1321 reads: Li quens Rollant mie ne s’asoüret,/Fiert de l’espiét, tant cume hanste li duret, and the same order is found in P 729– 730, T 566–567 and V4 1243–1244. It is therefore highly likely that the two first hemistichs were transposed in error in the C–V7 source document. 2274 & 2275. In both lines, a missing ~ for n: rapoine for ranpoine, C 2274, and doist when doinst, present (jussive) subjunctive 3, doner, is needed.

217

La Chanson de Roland (Critical Text and Notes)



2280

2285

2290

Rollanz fu proz et de mult fier coraje; Tint Durendart par mot ruste bataille; De Saraçins a fait mot grant doumage. Cel jor mostra si ben son vasalage. Qi l’atendit ne fist mië qe saje. La teste i prent, ne demande autre gaje; Sanc et cervelle fait voler en l’erbaje; Tot a son cors sanglant et son visage. Et Oliver de ferir ne se targe; Li .xii. per, qi sunt de haut parage, Ferent et caplent desor la gent sauvage; Murent paien a duel et a hontage. Dist l’arcivesqe, «Nostre gent est mot sage; Bien se defendent a cest estrot pasage. Car pleüst Deu, qi fist oisel volage, Ch’i fust li rois cui avons fait homage.»

35.b



2295

Olivers fu cortoiz et afaitiez, E de bataille hardiz et enseignez. Sa lance est fraite, mot en est corocez; Entre ses poinz remest l’une moitez.

2278. bataille reads oddly in terms of the -age rhyme, but Pope (1934, 153s., § 382) draws attention to the eleventh-century pronunciation of -ille, combining j and l. From the late thirteenth century, this was unacceptable, as the variety of the other readings shows: V7 2230: vigorage, P 742: vasselaige, T 578: barnage. 2282. C: La teste i pert, with pert, present 3: ‘He loses his head’, explaining why it is unwise for a Saracen to linger within Roland’s reach; but this occurs in a sequence of actions by Roland (the implicit subject of demande), so pert reads strangely. It may well be an example of scribal ineptitude. Cf. P 746: a teste en prinst. (past definite 3, prendre) and T 582: La teste prent… (V7 omits this line.). Prendre la teste is a standard phrase and the description in line 2283 certainly implies that this is what Roland is doing to every Saracen he encounters. Since the abbreviations 7 and ~ could give rise to scribal confusion of prent and pert, C’s reading has been emended on the basis of T. 2288. Murent is a graphy of muirent/muerent, pres.indic. 6, morir. 2292. C: cui avons fait domage, but this makes no sense. V7 2244 reads homage and P 757: homaige. 2296. C: Entor ses p…; V7 2248 and P 761: Entre ses p…, which makes much better sense.

218

2300

2305

2310

2315

La Chanson de Roland (Critical Text and Notes)

Fiert Fauseron en l’eume q’est vergez; Fors de la teste li a les elz sachez, E la cervele li espant a ses piez; Quant il l’ot mort, s’en fu joios et liez. Aprés oncit deus autres renoiez: Ce fu Torgins, qi mal fu vengiez, E Liganors, uns valez sorqidez; Ses tronçons brise, se li est esclicez. Voit la Rollant, mot s’en est merveilliez: «Sire conpeing, estes vos enragez, Qi de baston en estor vos aidiez? Acer et fers i est plus resoignez. Ou est donc Auteclere? Por qoi ne la traiez?» Dist Oliver, «N’en sui pas aisiez, Car de ferir sui trop enbesognez.» «Deus!» dist Rollant, «c’est torz et grant pechez. A! Guene fel, com nus as engegniez! Por traïson nos as a mort tornez.» Atant s’en torne, corant et eslaissiez. Oliver s’est ver paiens repairez.

36.a

 La bataille i ot pleinere et aduree. Grant fu li chaples de la gent desfaee. D’ambedeus parz fu forment redotee 2297. Oliver has already killed one Fauseron (brother to Marsile), C 2138. V7 2249 also has Fauseron and so do V4 1270 (Falsiron) and T 597 (Fausseron), but O 1355 reads: Malun and P 762: Nabigant. 2302. Similarly, Torgins or Torgis de Tortolose has already been killed by Anseïs in laisse 138, C 2217. 2303. C: Liganors, and P 768: Lucanor, but V7 2255: li Ganois. 2308. C: i est plus a resoignez, a hypermetric second hemistich, but V7 2260 and P 773 both omit a. 2309. Once again, a regular alexandrine to give balance to the two questions. 2312. C: toiz with dot below i to indication excision and comma for r superscript inserted over o. 2316. Oliver was first written in abridged form: Oliv7. Later, a small u was added after the O, and a small o or e near the top of and just preceding the l – to change this to ou el or ou Oliver. This (erroneous) emendation is not found in V7 2268 or P, which retain C’s original reading.

La Chanson de Roland (Critical Text and Notes)

2320

2325

2330

2335

219

La grant bataille qi la fu devisee; Li plus seürs a la color muee. Et Olivers a la teste enclinee; Por mautalant mist la mein a la spee. Si vait ferir Justin de Valfondee. Trenche li l’eume, la ventaille est fausee; Tot lo porfent deci q’en la coree; Li brans cola sus la sele doree; Au bon destrier a la schine coupee; Mort l’abati devant lui en la pree. Vit le Rollant, mervelle li agree; Li enperere de France la loee Por si fait cols lor a s’amor donee. De totes parz fu «Monjoie!» escriee. Paien fremissent, l’ost est espoëntee; Arier se trait demie arbarestree. Marpris s’en torne tot une randonee; Marsilions la novele ot contee. Li rois l’entent, s’a la teste levee; Mot tost comande qe sa gent soit armee.

 2340

Marsilions, cui tote honor sofreigne, Prist .xx. mil homes laiz et de puite ovreigne; Ne croient Deu ne la suë conpeigne; Per un destor merveillos et estreigne

2325. C: vetaille, with missing n or ~ abbreviation through scribal error. 2328. la schine for l’eschine, the s impura without prosthetic e, very common in North-Eastern French and in Franco-Italian. 2332. C: lors, ‘then’, but the indirect object pronoun, lor, ‘to them’ is clearly required here. 2339. C: comande qi sa gent, but the conjunction, qe, is needed. 2340. The (very worn) ms. reading is sosteigne, present subjunc.3, sosteindre, ‘to support’, ‘uphold’ and it makes no sense to wish honour to Marsile. V7 2292 reads: sofregne and P 810: souffraingne, ‘to whom may all honour be lacking’, which corrects the scribal error. 2341. homes is added, very faintly, in superscript over the cæsura to correct the metre of the first hemistich; the second is also hypometric if one assumes elision in the final phrase. However, coupled with the adjectival de puite ovreigne, laiz reads better as an adjective: the masculine plural of either lai (‘unschooled, ignorant’) or, more aptly and preferably, since C uses z as a graphy of ts, lait (‘deadly, horrible’). V7 2293 (laiz e de puite o.) and P 811 (lais e de pute gaingne) support the insertion of et.

220

2345

2350

La Chanson de Roland (Critical Text and Notes)

Vont a Gauter, qi garde la montaigne. Rois Amauri porte cel jor l’enseigne Per le costé delez une montaigne; Les destroiz garde, devers les porz d’Espaigne. «Deus!» dist Gauter, «Seint Mallo de Bretaigne! De nos a fait dolorose barguegne Guene, li fel, cui Demedeus confreigne!» Ses genz escrie qe chascuns d’als s’estreigne.

36.b



2355

2360

Rois Amauris est sus le monz venuz, O lui .xx. mile des paiens mescreüz. Franzois asaillent a force et a vertuz. Per grant dolor les ont le jor feruz; Toz les ont morz, ocis et confonduz. Sor toz les autres fu Gauter irascuz; L’escu enbrace, si fu traiz li branz nuz; Vers le renc maistre s’en vient, li sauz menuz; A als se joint, ne lor rent pas saluz.



2365

Si cum Gauters est a els ajostez, Paien l’asaillent environ de toz lez. Ses forz escuz li fu fraiz et quassez, Ses blanc oberz ronpuz et desafrez; Trois espiez ot par mi le cors colez, Tot ot percez les flans et les costez;

2346. C: Per le costé d’eus, delez…; V7 2298: Por le costé, devers…; P 816: Il les ajouste par devers… C’s hypermetric first hemistich may arise from a scribal mis-reading of devers (in abridged form: dev7s) as d’eus – and devers occurs immediately below, C 2347, so there may be duplication. In C, delez (‘beside’) avoids having devers twice, in consecutive lines. 2347. C: poz; V7 2299: porz; P 817: pors. 2349. C: de nos as fait, the scribe reading this line as addressed to God or St Malo (in 2348), but the subject of the verb is Guene, C 2350, who has indeed made a ‘grievous bargain’ in respect of them. V7 2302 and P 820 read the present 3, a. V7 2301–2302 transpose C 2349–2350 to give a more logical order. 2357. C: irasus; V7: irascuz. 2358. C: bronz, written brõz; V7 2310: branz; P 828: brans.

221

La Chanson de Roland (Critical Text and Notes)

2370

Li cuers li faut, qe tres foiz est pasmez. Ne pot sofrir Gauters li alosez; O veille o non, s’en est del canp tornez; Mais mult s’en vait corozois et irez. Grant aleüre est del mont avalez. «Rollant,» escrie, «ou estes vos alez? Fiz a baron, et car me secorez!»

 2375

2380

2385

Gauters costie dalez un tertresel; Desor l’arzon li gisent li buël; E Franzois font de Sarazins masel. Li cons Gerins sist el cheval morel, E ses conpaing Gerers fu mot isnel; Ambedui poignent le pendant d’un valcel, Si vont ferir un paien Tinodel; L’uns en l’escu, par desoz en cantel, E l’autre en l’osberc, don d’or sunt li clavel. El cors li begnent les espiez a noël; Mort lo trestornent tres par mi un praël. Aprimereins i fu, li fiz Abel: Celui oncist Enzeler de Bordel. Turpins de Reins reoncist Singlorel L’enchanteör, qi par son grant revel Fu en Enfer por faire son avel –

37.a

2367. C: Li cuers li faut faut… 2371. C: Grant est de aleüre, the scribe inserting est de too early by mistake. 2379. C: valcel, the first l being added superscript. 2384. Both C and V7 2336 here have a singular verb in error, but P 857 reads: trestornent. 2385. The MS reading is A primereins, meaning ‘first of all’, and le fiz Abelcould be taken as the subject of fu, rather than in apposition (cf. Tot primereins, C 2105); but Saracen names often have meanings. V7 2337 reads Aspre mereins (Foerster) or As premereins (Duggan), which Duggan emends to Asprimereins; and P 858 (Esprevaris), T 632 (Espervariins), V4 1303 (Esperçiaris) and O 1388 (Esprieres) all treat this as the name of Abel’s son (perhaps here meaning ‘First-born’). 2386. In C, the pronoun celui is always in the object form singular. Thus in both this line and 2678 there is subject-verb inversion, with celui the direct object: ‘Enzeler killed that one…’. 2387. C: reoncist ci Singlorel.

222 2390

La Chanson de Roland (Critical Text and Notes)

Per artimaisse l’i conduist Jupitel. Dist l’arcivesqe, «Ci a riche cenbel.» Rollant respont, «Ben ferent li doncel. Oliver, frere, itel cop me sunt bel.»

 2395

2400

La bataille est merveleouse et grant; Mesléëment ferent paiens et Franc; Se li uns lasse, l’autre est defendant. La veïst on tant vert eume luisant, E tant escu a or reflanboiant, Tant bon osberc safré et jacerant, Tant destrer brun lor resnes traïnant, Dont li vasal gisent mort par les chans. Deus! tant prodome perdent le jor lor sanc, Qi pois n’en virent ni moillers ni enfant, Ne lor amis, qi sunt as porz pasant.

2390. C: Per artimaisse and V7 2342: Per artimaise, with this graphy perhaps resulting from confusion of the name of the goddess Artemis with the usual graphies: artimaille/artimaire (sf.) or artimai/artimage (sm), meaning ‘magic’ or ‘black arts’. O 1392 reads artimal; P 863 changes this completely to: par droite voie. At the end of this line, both C and V7 read Pinabel, otherwise mentioned only as Ganelon’s nephew and eventual champion (cf. C 527, 2413, 7613, etc.). P 863 and T 637 read Jupitel, V4 1307 Jupiters and O 1392 Jupiter; so the text has been emended. 2394. In C, merveleouse et grant is metrically correct because, although the final e of merveleouse is elided with et, the -eou- diphthong is dissyllabic. In contrast, V7 2346 (merveillouse et grant) and L 151 (merveillouse et granz) correct the spelling but need to avoid elision. P 867 inserts an extra syllable (miravillouse…), T 641 an extra word (merveillouse et moult grant), V4 1310 is an alexandrine with different second hemistich, and O has e merveillose e grant in a matching line (1653) at a later point. However, C later (line 4290) has mervelleose treated as 3 syllables only (mervelleose e coranz , where V7 4403 has merveillose et corrant) – with C’s extra e perhaps inserted through scribal error; it does, however, raise the question whether there is elision here. 2400. C: ∧ lor, with a faint single letter o (‘with’) or 9 (= con, ‘with’), in superscript. This may well be the work of a later revisor, unfamiliar with the earlier 3-syllable form traïnant. The verb is found twice more in C, as trainier 1556 and trainer 2543, and the infinitives are definitely dissyllabic on both occasions. Here, V7 reads …lor resnes detrainanz, deliberately inserting an extra syllable. Nevertheless, the noun traïn is definitely dissyllabic in C lines 2833, 3310 and 3891. Pope (1934, 110, § 242) states that ‘contraction (of aï to ai) began in the thirteenth century…’, so variability of spelling is to be expected in texts of that date. The same inconsistency is visible in C’s use of traitre/traïtre and traitor/traïtor (cf. Introduction, 92–93). 2401. mort is added at the end of the line and its place of insertion indicated by means of ∧.

223

La Chanson de Roland (Critical Text and Notes)

2405

2410

2415

Karle li meine en sera mot dolant. Cui chaut de zo, quan ne lor ert aidant? Li mals traïtre, Guene le soduiant, Mauvais servise lor fist a icel tans, Quant soi meïsme vendi as mescréant; Puis en fu mort par mervellos ahant; Enscenble o lui de ses apartenant. Ce trovons nos es estoires lisant: Qe Pinabel fu par lui recreant. Detraiz fu Guene a forz chevals corant; Si dut il estre, car il fu mescréant, Quant por argent fu itel mercaant.

37.b



2420

En la bataille sor la paiene jent Feri Rollant par mot fer mautalænt; Et Oliver prove son hardiment; Li archivesqe plus de mil cols i rent; Li .xii. per ne se tarzent nïent; Franzois i ferent trestuit comunalment. Morent paien come chaitif dolent. Qi ne s’en fuit, enqi pert son jovent.

2405. C: .K.meine, followed by a punctus for the cæsura, but this is hypometric without the li found in V7 2357. 2407. C: traire; V7 2359: trahitor, which Duggan emends to traïtre; P 880: traitres. 2410. C: par mervellos talant and V7 2362: par m. tallanz. In C 6118, this phrase means ‘with terrible longing’, which makes no sense here. Even if talant is read as meaning ‘appetite for revenge’, there is no mention of the persons to whom it applies. P 883 reads: par merveillouz ahans and T 657: par m. ahant, and ‘by means of terrible torment’ is much more apt. An earlier copyist appears to have mis-read h as l and inserted an initial t, producing a standard phrase – but inappropriate to the context. 2415. C: Si dut estre with ∧ inserted between dut and estre and il superscript. 2418. The æ diphthong in mautalæent is possibly a Franco-Italian graphy of e, but it is more likely to be a scribal correction of -ant to -ent, since a 23-line laisse with -ant rhyme ended only 2 lines earlier. 2420. C: prent, but it is illogical for Turpin to ‘take’ so many blows; V7 2372: rent; P 889: rant. 2422. After Franzois, there is ∧ with i inserted superscript. 2424. C: per, written as the letter p with bar through the tail, a common abbreviation; but both V7 2376 and P 893 read: pert (present indicative 3, perdre) and a finite verb is needed here.

224 2425

2430

2435

2440

La Chanson de Roland (Critical Text and Notes)

Franzois i perdent meint riche garniment, (* * * * * *) Tant bon espi noëlé a argent; De lor espées sunt tot li brant sanglent. Et si i perdent meint chevalier vaillent. Cui caut de ce, char ne lor vaut nïent; Mais ne veront ni amis ni parent, Ne Carllemeine qi as porz les atent. En France crut si doloros torment, Q’i aparut de tonoire et de vent, Pluives et grailles desmesiurement, E chient foudre et menu et sovent; Et terremote i est comunalment, De Seint Michel desc’as porz d’Egricent; N’i a cité dont li murs n’i cravent. En droit midi, tenebres i sunt cent; N’i a clarté se li cels ne l’i rent. Il n’i a home qi mot n’i s’espouënt;

2425. Here C leaves a line blank; V7, P, L and T go straight on with line 2426, but C 2426–2427 > T 671. O omits C 2426–2429; in V4, 1331–1332 replace C 2426, 1333–1334 match C 2427–2428, but C 2429 is omitted. 2428. C: Et si ∧ perdent… with another i superscript, to give the correct number of syllables. 2434. C: Pluiues, probably pluives, but C 1358 has pluvie; V7 reads pluie in both 2386 and 1421. See Introduction, 71–72 under Orthography regarding problems of reading C’s -ui-/-iucombination. 2436. C: Et terre i est mote comunalment, with cæsura after est – which does not make sense. V7 2388 reads: Et teremote i est…, P 905: Et terremeute i est… and O 1427: E Terremoete ço i ad…, indicating how the scribe has displaced mote in C. 2437. C: s barred before desc’as and poz again omits r. In place of this one line, O 1428–1429 has: De Seint Michel josqu’as Seins,/Des Besençun tresqu’as porz de Guitsand and P 906–907: De Bezanson jusqu’as pors de Wissant,/De S. Michiel jusqu’a Rains ausiment, T 679–680 matching P. V4 makes no mention of place-names. Only V7 2389 matches C, apparently omitting two half-lines. 2440. This line was omitted by V7 and P, perhaps due to incomprehensibility; but Duggan inserts it in his Critical Text as V7 2392. T 683 has: se lui ciel ne lui rent. O 1432 reads: N’i ad clartét, si li ciels nen i fent (‘There is no light, unless the skies break/split’, ‘.unless there is lightning’) and V4 1341: Ne a clarité se le cel no se fant. In C, rent (present 3, rendre) may be read as ‘restore’/’allow to escape’, but could be a scribal error for fent, with the preceding pronoun modified to fit. In C, ne appears to be partly erased and ce in very fine script inserted, but this makes no sense and is probably a revisor’s error. 2441. s’espouënt, present subjunctive 3, s’espouënter (espoënter).

225

La Chanson de Roland (Critical Text and Notes)

2445

Dient pluisor qe c’est defeniment: «La fins de siegle ci nos est en present.» Il n’i savoient ni dient voir nïent; Ainz est dolor por la mort de Rollent.

38.a



2450

2455

Fort sunt li signe et li orage grant, En douce France tante çase versant, Tres miédi jusc’al solel cochant; Noiz est oscure; ne vit hon tant ne quant. Jors ni solauz n’i fait clarté luisant; Home n’i a, ne qit morir esrant; Bien puent estre en cel reigne dolant, Car li bon muerent, dont il sunt desirant. A Seint Denise, cui Deus parame tant, La troverez en istoire lisant Ceste dolor por la mort de Rollant; Meudre vasal n’en ceindra jamés brant; Por chevalier ne qiet mes d’auferant.

2442. C: Dient pluisor qi…, but qe is clearly needed; cf. V7 2394: que. Since this is the conjunction, not the relative or interrogative pronoun, where qi and qe are often used indiscriminately, it has been corrected as a scribal error arising from the occurrence of qi immediately above in 2441. 2445. C: por amor de R. in error for por la mort, as is confirmed by the reading of V4 1346, O 1437, P 913. V7 2397 has: lamor (l’amor). Similarly, C 2456 and V7 2408 read: amor, whereas the other versions have: la mort. .R. has been transcribed as Rollent because the C scribe makes a very nice distinction between -ant and -ent rhymes in this sequence of laisses (151, 152 and 153). 2447. C: vont tante çase…, with çase an Italianate graphy of chase (the V7 2399 reading), ‘house, cot’; thus ‘blowing down so many humble dwellings’ – but Duggan emends this to chose. 2449. The word hon (hõ) was originally omitted, then added in superscript with ∧ after vit. 2452–2453. ‘They may well be sad in that kingdom, for the good whom they long to see again are dying!’ – a reference to the death of so many French knights in the rearguard at Roncevaux. 2456. See footnote to C 2445, where the same error occurs. 2458. ‘He never falls from his horse for (any) knight’, i.e. ‘No knight can ever unseat him from his horse’; qiet is here a variant graphy of chiet, present indicative 3, cheïr; cf. V7 2410: Por chevalier nen chesra d’auferanz’ and P 925: Por chevalier ne chaï d’aufferrant.

226

La Chanson de Roland (Critical Text and Notes)

 2460

2465

La bataille est pleignere, durement aduree; D’ambedeus parz est fortment redotee. Ferent François de lor trenchant espee, N’i a celui ne l’ait ensanglentee. «Monjoie!» escrient, l’enseigne renomee. La veïst l’om tante broine safree, Tant espiez escroisir, tante teste coupee, Tant destrier mort a la resne trancee, Dont li vasal gisent mort par la pree. Paien s’en fuient parmi une valee; Franc les enchauchent par tote la contree.

 2470

2475

2480

Paiene gent, dolante et irascue, Par mi Espaigne ont lor voie acolue; Franc les encauchent de la Terre Absolue; Meinte chiere arme i ot le jor perdue. Paiene jent est morte et confondue; Li cons Rollant durement les argüe. La veïst on les plenes si vestue, Tant eume frait, tante tarze ronpue, Tant destrier ont les renes desrompue, Dont li vasal gisent en l’erbe drue. Ceste bataille ont nos Franzois vancue. C’est la primiere qi si lor est venue. Deus! Puis lor est si grant pene creüe, En grant dolor en est France cheüe.

38.b

2464. C: tate, with missing nasal (n or ~); V7 2416: tante. 2465. C: Taint, here surely a scribal error rather than diphthongisation. 2466. The reading al la resne trancee – ‘with the reins cut’, presumably in the fighting – is supported by V7 2418 (Tant destrier mort li resne traincee) and by C 2478 (… les renes desrompue) and is therefore retained; but P 933 has: Tant destrier vont lor resne traïnnee, showing the riderless horses galloping around, trailing their reins, after their riders have been killed. 2471. acolue (identical in V7 2423) is the past participle of acoillir (‘to take’), a slightly unusual form, but quoted by Fouché (1967, 370, § 190). 2482. Dexpuis in the C ms., written as a single word, may well be a graphy of Despuis (‘afterwards’), but V7 2434 reads: Deus puis and P 949: Deus puis…, as separate words, and the dramatic invocation of God is found in T 717: Dieu plus and V4 1372: Deo, si grant…

La Chanson de Roland (Critical Text and Notes)

227

 2485

2490

2495

2500

Nostre Franzois ont feru a baudor; Paien ont mort por lor ruste vigor; De cent meillier n’en sunt dui en retor. Dist l’arcivesqe, «Nostre gent a grant valor; Nus rois en terre n’en ot onqes meillor.» Il est escrit en la geste francor, Droiz est a estre en la Terre Major, Qe proz vasal i ot l’enperaör. Vont par lo canp por rejoster alor. Qi donc oïst lo duel et la clamor, Plorent des els par mot grande dolor. Es vos Marsille et d’Espeigne la flor! Li augalie chivauche par iror; Voit le Oliver, si mue la color; Rollant apele, si li dist par amor, «Sire conpeing, por Deu le criator, Nostre conpaigne partira a dolor. Ja uns de nos n’estordra hui cest jor. Guenes li fel, cui Deus doint desenor, Nus a venduz a la gent païenor.»

2489. C: es en la geste… 2491. C: Qe proz vasal ont l’enperaör with hypometric second hemistich; V7 2443 has the metrically correct: Que proz vasal i hont l’enperaor. Reading Qe (rel.pron.) as referring back toTerre Major (C 2490), vasal (subj.plur.) and enperaör (indirect object, sing., for possession?) gives the meaning: ‘which valiant vassals possess (in vassalage to) the emperor’, with a (al) or de(del) perhaps omitted in error. P 958 and L 345 have: Que vassal soient avec l’emperéor, and T 726: Que vassaux soient avec l’empereur, keeping vassal as the subject, but V4 1381 reads: Qe bon vasal oit nostre inperaor (oit, past definite 3, aveir) and O 1444 either Que vassals ad li n. e. or Que vassals est li n. e. (taking vassals either as substantive, object plur., or as adjective, subject sing., meaning ‘valiant’). This implies a mis-reading of subject-verb inversion in an early exemplar. C 2484–2495 closely correspond to O 1438–1448, laisse 112, and emendation to vasals ot nostre emperaör (cf. emperaör, subject sing., C 5562) permits insertion of V7’s i so as to obtain the correct metre. In this confused ‘Margariz episode’, C lacks O 113 and 116, but O 114–115 = C 176–177 and O 117–118 = C 163–164, though O lacks C 165. 2500. C: partira hui a dolor, hypermetric through hui being copied in error from line 2501. P 966 reads: partira a dolor. V7 2452 has: partirons a tristor (emended by Duggan to the CP reading). 2501. C: hui de jor, an unusual structure, apparently a scribal error for hui cest jor, ‘today’ (or possibly jor de hui with inversion?). V7 2453: en cest jor (Duggan emends to hui cest j.), P 967: d’ui cest jor.

228

La Chanson de Roland (Critical Text and Notes)

 2505

2510

2515

Li cons Rollanpt est fortment dementez; Maint rustes cols a feru et donez; Et Olivers li proz et li senez; Li .xii. per resunt de grant bontez, Ronpent et fendent par dure paestez; Saracins ont a martire livrez. De cent millers, n’en est uns escapez, Fors Margari, le traïtor provez. Se il s’en fuit, n’en doit estre blasmez; Mot grant besoing est par lui demorez: De qatre parz est entr’el cors navrez, Ses brans d’acer i est toz desertez, E ses osberc runpuz et desmailliez. Envers Espaigne est ses chevals tornez; Al roi Marsille a toz les faiz contez.

39.a

 2520

Rois Margariz est toz sous retornez Et ses osberz ronpuz et depecez;

2508. paestez is the Franco-Italian version of the normal Old French pöestez, ‘force’ (V7 2460). 2510. C: escãpez, past participle of escape with ~ added in error, perhaps influenced by camp. Inorganic n is common in Northern and North-Eastern French and Franco-Italian. 2513. This line apparently means: ‘Very great need is put off by him’, i.e. ‘He postpones attention to his own very great need’, ignoring his wounds (demorer = ‘to delay’). The line is not found in V7, P or L; O and V4 omit the entire incident; but T 739 reads: Mot grant essoigne a de son corps naffré (‘He is greatly delayed by his wounded body’), combining C 2513–2514. Duggan inserts this line into the Critical Text as V7 2465 and as part of the original (CV7 model) reading. 2514. De qatre parz, ‘from four sides’, ‘from four directions’ makes reasonable sense, but may well be a scribal error; V7 2466 and P 979 both read espiez (‘lances’) instead of parz – as does C 2522. 2515. To avoid C’s hypometric second hemistich, V7 2467 reads: i est t…; P 980 has: ensanglentez. 2519–2524. These lines are simply a modified repetition of lines 2511–2518 and V7 omits them, continuing the identical -ez rhyme without a break and thus making a single laisse of C’s laisses 157 and 158. V4 1387–1400 has a version very similar to C’s, with the same repetition. P 986–993 omits C 2513 (as V7 does) but writes C 2512 twice (P 978 and 984) and makes a new laisse as C does, the words varying only in minor detail. T 742–751 omits C 2514–2516 inclusive in laisse 157 but includes these details in laisse 158 with added lines but no change in the general sense. In L, lines 368–375 roughly correspond to C from C 2512 on, but in abridged

La Chanson de Roland (Critical Text and Notes)

2525

2530

229

Ses bons brans est de sans envermellez, Et il meïsme de qatr’espiez navrez; Il vient del canp ou li chaples fu fez. Li rois Marsille a toz li faiz contez, Mot fierement li est cheüz as piez: «Bons rois d’Espeigne, esrament chevauchez. Les Franc de France troverez enuiez De cols ferir, de vos cors martiriez. Toz les plusors troverez ensegnez; Perdu i ont meint chevalier prisez Et de lor gent plus de l’une moitez; Li remananz est mot afebliez: Il n’en ot armes, de verté lo sacez.»

 2535

Marsilions fist forment a loër; Dist as paiens, «Je vos puis mot amer. Li cons Rollant fait mot a redoter; Qi le veïst venir, mot le deüst peser. Per deus batailles se nel pöez mater,

form with omissions and paraphrase. O omits Margariz’ return to report to Marsile. After C 2519, P adds (line 987): Sa lance est frainte et ses escuz perciez, which fits extremely well before C 2520 and is supported by V4 1394. 2528. This is apparently the past participle, martiriez, ‘martyred by your attacks’ (cors = cours, ‘expeditions, raids’) – cf. T 754: de nos martiriez – but martirier, infinitive dependent on enuiez (C 2527) ‘hard pressed to martyr your bodies’, in parenthesis to ferir in the first hemistich, is a better reading. V7 2480 emends this example of poetic licence, with syntax subordinated to rhyme (cf. P 996: de nos cors martirier), by substituting sunt mult entalentez , a standard formula – but Duggan inserts the C reading. 2529. ensegnez, a graphy of ensaigniez. 2533. Il stands for Li remananz, C 2532, so ot, present 3 is grammatically correct, although ont feels more appropriate. P here inserts an extra 125 lines (1003–1127) corresponding to V4 1410– 1452. 2536. A dot below a in the ms. indicates excision, but a redoter is needed for both sense and metre; perhaps a revisor’s error. 2537. This alexandrine has very well-balanced hemistiches (veïst: deüst), matched exactly in V7 2489; but Duggan reduces this to normal 4:6 metre in his Critical Text on the basis of P and T. (P 1183: Qui le veult vaincre, il s’en doit mout pener; T 806: Qui le veult vaincre, moult se doit pener; L 441: Qui le mal voit, moult se covient pener.) 2538. C: se les pöez…, but a negative is needed, as in V7 2490: se nes pöez…, or more accurately se nel p… so as to refer specifically to Roland.

230

2540

2545

La Chanson de Roland (Critical Text and Notes)

Trois en ferai, se:l volez creanter. Les .x. eschelles remanront por garder; Les autres .x. iront as Frans joster. Hui perdra Charlles l’orguel q’il selt mener; A mot grant honte vesrez Franzois trainer.» Un confenon q’il ot fait adober Dona Grandoine por les autres garder; Et cil le pris, qi mot s’en vet pener.

39.b



2550

2555

2560

Desor un mont laisse Marsilion; Va s’en Grandoine, il et si conpeignon, Permi un val o fiere aatison. A trois clos d’or lacie son confenon. A vois escrie, «Car chivauchiez, baron!» Mil graille sonent, mult en sont cler li son; Dient Franzois, «Deus pere! Qe feron? Tant mar veïmes la bonté Gaynelon: Vendu nos a par mortel traïson.» Li arcivesqes a parlé par raison: «Li home Deu, hui recevez le don! En paradis avrez benecion. Mais li cohart n’i aront ja pardon.» Respont Franzois, «Comunaument l’avron, Qar por la loi esauchier conbaton.

2542. C: se∧t with l in superscript, to give selt. 2547. Marsilion is also found as subject form in C 148, 612, 2799, although normally object (17 instances); thus, with laisse, present 3, laschier, this gives ‘Marsile lets go the reins’, ‘halts’. Cf. V7 2499: remest; V4 1465: roma se; and T 816: si vint. However, P’s laissa (line 1144) may imply that Grandoine 2548 is the subject. Or might C’s laisse be imperative, addressed to the reader? 2548. C: Vai s’en, an obvious scribal error; Va s’en in V7 2500 and P 1145. 2549. In the C ms., the mi of Permi is inserted in superscript with ∧ beneath. 2552. C: en son cler li son, the final substantive influencing the spelling of the verb, son in error for sont. V7 2504 takes li son as subject form singular, reading en fu cler, but P 1149 has en sont cler. 2554. Apparently a bitter comment from the French, la bonté Ganelon being ‘Ganelon’s courage’, ‘the service Ganelon has done us’ or, ironically, ‘Ganelon’s bounty/benevolence (towards us)’. However, V7 2505, P 1151 and T 822 read le conte G and V4 1472: li cont G.: i.e. simply ‘Count Ganelon’.

231

La Chanson de Roland (Critical Text and Notes)

2565

Ja Deu ne place qe chel pechié fazon, Dont nos son reigne perdre i puison.» Turpins de Reins, par bone entencion, De Deu lor fist gente ausolucion. Lor ont estreint meint auferant gascon; Puis remonterent, irié come lion. Ver les paiens par fiere entencion A vois escrie, «Monjoie la Carllon!»

 2570

2575

2580

2585

Li rois Marsille en deus les fist partir: Les .x. eschelles en fist o soi tenir, Les autres dis chevauchent por ferir. Mil graile sonent, si se firent oïr; Dient Franzois, «Deus! O poron garir? Li .xii. per, che porons devenir?» Li arcivesqe ne se velt plus taisir: «Li home Deu, ne nos calt de morir! Huy vos ferai coroner et florir, En paradis richement aseïr. Mais li cohart n’i porront mais venir.» Respondent Franc, «N’i devons pas faillir, Car chascun jor penons de lui servir. Poi avons gent, mais mot ont grant ahir, Coment qe soient esmaié au venir.» Jhesus de gloire, qi tot a a baillir,

40.a

2580. C: …n’i porront mais issir, but this makes no sense, since cowards will be unable to go to Heaven, not leave it. The line is not found in O or V4, but V7 2532 reads venir, supported by P 1176 and L 484, and by parvenir in T 845. Since aseïr occurs at the end of C 2579, it may be the source of the error, the scribe accidentally repeating it in a variant graphy: aseïr> essir > issir, essir being a graphy of both verbs. (Thus mais s’essir would mean: ‘… never be able to rest there’.) 2581. C: Respondent François, a 5-syllable hemistich; but C 285 and 4325 have the normal 4syllable Respondent Franc. Elsewhere in C, the word respondent occurs only at the end of an hemistich, where the -ent ending is not counted as a syllable. V7 2533 reads Respont François, treating this as a collective singular, but P 1177 has Respondent Franc and T 846 Responnent Frans. 2583. Both C and V7 2535 may read ot because gent, though a collective noun, is singular; but the present tense, not past definite, is needed: ont. So ot must arise from scribal omission of ~.

232

La Chanson de Roland (Critical Text and Notes)

Lor fist a toz tel ardiment collir, Jamais por home ne les vesrez fuir.



2590

2595

2600

2605

[P] 2608a

Nostre Franzois voient lor enemis. Per la Deu grace qi en la crois fu mis, Fu chascuns proz, corageus et hardis; Les destriers brozent, sunt les espiez brandis, Hardiement ont paiens envaïs; Grant cops lor donent sor lor escuz voltis; Desor les bocles les ont fraiz et malmis, Les blans osberc runpuz et desartis, Permi les cropes des destriers arabis. Mais d’une cose soiez vos ben certis: C’al joster ot tel noisse et tel cris, Desoz lor piez est li monz retentis; Mot fu vasaus qi n’i fu esbaïs. Li estors fu durement esbaudis; Comenciez est li deus et li estris: Si grant domage en sera a Paris, Bien en morunt mil chevalier eslis, Tante pocele caste de lor amis E tante dame veve de lor maris, Tant bon enfant de lor pere orfenis; Quant iert seü el reigne seint Denis, Mainz en ara grans dolors et grans cris.

40.b

2589. C: Per la Deu gre with˜ over final e, but reading grent/grant makes no sense and gré (‘goodwill’, ‘favour’) is masculine; the V7 2541 reading, grace, has therefore been adopted. 2593. Initially omitted, cops was added at the end of the line with a sign for its insertion before lor. 2595. C’s reading, desarsis, is a variant of deserti, ‘shattered’ or ‘cut to pieces’; V7 2547 reads dessafris, a form of dessafrez modified for the rhyme and meaning ‘stripped of its damascened finish’; but the best reading here is dessartis, ‘unravelled’, ‘torn to pieces’. 2602. est was added in the very common form e + ~, in superscript, above the ending of comenciez. 2608. C and V7 2560 read: Mainz en ara el reigne seint Denis, but this is not completely satisfactory as a summing up of the tants of the preceding 3 lines. P’s 2 lines (1206–1207): Quant iert seu ou regne saint Denis,/Moult i aura grans dolors e grans cris make much better sense and are supported by T 878–879 and L 515–516. The CV7 shared model seems to have telescoped two lines into one.

La Chanson de Roland (Critical Text and Notes)

2610

233

Li Franc de France, del segnoril païs, Ben se defendent o le brans coloris, Qe chascuns d’aus voïst estre garis; Mais en cel canp les covint estre ocis.



2615

2617a (V7)

Un Sarazin de Saragoze i a, De l’ost conduire fist chil qe comanda. Cel q’il comande mot tost on li fera. En Saragoze uns lor paiens esta; A Guenelon icil s’aconpagna, Quant a Marsille le message porta E li plevi qe Rollant traïra.

2611. C: voïst, a variant graphy of volsist/vosist, imperfect subjunctive 3, voleir, used to express the conditional. Cf. Pope (1934, 375, § 1011** (North-Eastern French) and Fouché (1967, 34, § 171 (b)). 2613. This may be Grandoine, to whom Marsile gave the gonfanon, C 2544–2545, although he led off his men in line 2548; alternatively, it may be the same Saracen referred to in C 2616, who can be identified as Cliboïs. Certainly, O 1526 is identical to C 2613 and O 1528 states: ço est Climborins (variant graphy: Cliboïs). C 2616 is not found in O. 2614–2615. C’s text appears to mean: ‘He who commanded (= Marsile) appointed him to lead the army; People will very quickly do what he orders there’. In this case, the scribe has misunderstood the syntax of C 2614, taking De l’ost conduire as dependent on comanda instead of on fist (de being superfluous with faire). The other rhymed versions show considerable variation here, partly coinciding with C, partly differing: cf. V7 2566: Del ost conduire poestis il se fa/Ce q’il comande mout tost hon li fera; P 1212–1213: Sa volenté font cil qu’il commanda/ Quant il commande, la seingnorie en a; and T 885–887: De la moitié fist ce qui commanda/De Sarragoce les pers i ajousta/Quant bon ly est, la seignorie en a. There seems to be considerable confusion; certain identical phrases occur in each, but in a different order, as if each were trying to make sense of a faulty source. If this is the case, then C’s second hemistich in C 2614 may be explained as (i) the omission of 2 half-lines by C, so that this is matched to the wrong first hemistich (T’s 3 lines and V7’s quite different text give some credibility to this); as (ii) C’s transposition of the 2 second hemistiches, although this does not solve the problem of faire de + infinitive; or (iii) a mis-reading of a Proper Name as fist chil qe. If one takes hypothesis (ii), mot tost could well be a scribal error for moitie ost (cf. T’s moitié and O 1527: meitet), relating to Marsile’s division of his army and his appointment of Grandoine to lead one half (C 2570– 2572). In hypothesis (iii), the name in question could be Cliboïs (cf. notes to C 2613 and 2616) or, more credibly, fis Capoé, i.e. Grandoine (cf. C 2750), with a slightly different graphy of Capoé. 2616. The information in C 2620 enables this Saracen to be identified as Clibois; cf. C 925, laisse 56. 2617. The extra line, V7 2570, is needed to make sense of C 2618.

234

2620

2625

2630

La Chanson de Roland (Critical Text and Notes)

Per amisté sa boche l’en baisa; En guesredon un eume li dona; Aprés li dist e mot bien li jura, Terre Major a onte livrera, L’enpereör fortment avilera, E la corone del chief li ostera. Sist el destrier, Barbemor apela; Mot bien lo broche, la resne li lassa; Fiert Enzeler la o il l’encontra, Cel de Gascoigne qe Karles tant ama. L’escu li fent et l’osberc li qassa, Permi le cors son espié li guia, Plene la lance del destrer lo geta; Morz est li quns, des siens un soul n’i a. Dient Franzois, «Deus! Qel dolor ci a!»

 2635

2640

Li cons Rollant apele Oliver: «Sire conpeing, ci a grant destorber! Perdu avons lo Gascon Enzeler, Cel de Gascoigne qe li rois a si chier; Nus n’i avons nus meillor chevalier.» Respont li cons, «Deus le me last vengier!» Lo destrer broche des esperon d’or mer, Tint Auteclere dont li brant fu d’acer, Desor son eume q’il fist a or verger Fiert lo gloton, qe ne:l volt espargnier.

41.a

2624. There is an insertion mark after del in the ms. and chief has been added at the end of the line. 2626. The words la resne are added at the end of the line, with a mark for their insertion after broche. 2632. des siens un soul n’i a appears to mean ‘there is not a single one of his men left’ (perhaps implying ‘to protect him’), but siens is also found as a graphy of sens (‘senses’), C 2712, and this must be the meaning here, too: ‘not a single one of his senses remain’. V7 2585 reads de siens, but P 1231 has de son tans plus n’i a (‘he has no time – ie life – left’). The diphthongisation of open, blocked e is characteristic of North-Eastern France (cf. Pope (1934, 491, § 1321, iii) and not unknown in Franco-Italian. 2638. A play on words, the first nus being the first person plural subject pronoun, ‘we’, the second, written n9, a graphy of nuls, adjective, ‘no’; grammatically the object form, nul (as in V7), is required. The phrase n’i is written n∧ with the i in superscript.

La Chanson de Roland (Critical Text and Notes)

2645

2650

2655

235

Tot lo porfent deci qe el tenpler, Trenche lo cors, si oncist lo destrier; Si l’abat mort prés d’un gasté moster. Den autre ocist qe Marsilles ot chier: C’est Rapadans, qi mot fist a priser. La teste li toli sanz negun recovrer; L’arme de lui enportent averser. Set Arabis fist les arçons voider: Ne sunt mais preu por estor comencer. Ce dist Rollant, «Or vos voi mot irer. Vus ne devez altrui aconpagner; Encontre moi devez apareilier. Li enperere qi France a a bailier Por ites cols nos aime et nos a chier.» «Monjoie!» escrie por sa gent esforcer: «Ferez, François, Jhesus vos puist aidier!»

 2660

La bataille par fu merveilleuse et estreigne. François sunt las, n’i a nul ne se pleigne.

2647. C: Deus en reocist qe merveilles ot cher, with a 5-syllable first hemistich in which ‘two’ is illogical because only one Saracen is subsequently named (Rapadans, C 2648) and an equally illogical second hemistich, since Oliver would not be wonderfully fond of the Saracen. V7 2600: Deus en ocist qe Marsilles ot cher reduces the first hemistich to 4 syllables and shows merveilles to be a scribal error, easily emended; but V7 too lists only one Saracen. Although O (1554–1555) and V4 (1529–1530) name 2 Saracens who are killed, P 1246, T 916 and L 555 read .i. where C and V7 have Deus, a logical emendment – but P subsequently names a second Saracen in 2 lines not found in the other versions, P 1248–1249: .i. autre encontre c’on apelle Turfier,/De Hauteclere isnellement le fiert. P also changes the order of the second hemistiches (P 1246–1247): .i. en r’ocist qui mult fist a proisier/… qu’il n’avoit gaires chier, although T 916– 917 keeps C’s order: …que il n’avoit pas chier/… qui moult fait a prisier. There are two possibilities: (i) that C’s source omitted a line or lines similar to P 1248–1249, or (ii) that Deus en reocist is a scribal error for Den autre ocist, ‘afterwards he killed another’. The latter hypothesis makes better sense, for Oliver has already killed Cliboïs, C 2642–2646, and Den autre ocist could easily be mis-read, if worn, as Deus en reocist (with scribal confusion of u and n). O 1554 begins: Puis ad ocis. 2660–2682. A very confused laisse in C, with lines 2669–2673 repeating word for word C 2663– 2668 (but not 2667), and with C 2679–2682 (cf. note below) inserted completely out of context (as they are in V7 and P). The variant graphies (Oliver, Olivers; dolerose, dolerouse; barguegne, bargaigne; toz, tuit; mosrons, morons; recovreigne, recovregne; genz, gent) simply show acceptable alternatives. The direct cause of the repetition appears to be the 2 lines (2662 and 2668) beginning Marsilions: un saut de même en même. The same laisse is found in V7 and P, with

236

2665

2670

2675

La Chanson de Roland (Critical Text and Notes)

Marsilions o la chiere grifeigne Fu as destroiz, devers les porz d’Espeigne. Dist Olivers, «Ci a male gaaigne, An cui feron dolerose barguegne: Toz i mosrons, sanz nulle recovreigne.» Rollant li cons de bien ferir se peine. Marsilions prist sa genz d’Equiteigne; (Fu as destroiz devers les porz d’Espeigne. Dist Oliver, «Ci a male gaaigne, En cui ferons dolerouse bargaigne: Tuit i morons sanz nulle recovregne.» Marsilions prist sa gent d’Equiteigne,) Frans envaï, n’i a cel ne s’en pleigne; Permi un val lor sort cele conpeigne: Vint mille furent a une vert enseigne; Pinaus les guie et Roudez d’Aqiteigne: Celui feri Turpin parmi l’antregne. Il est escrit en meinz lius en Breteigne,

41.b

no repetition, but V7 changes the line-order considerably. P 1264–1280 match C 2660–2668 + 2674–2677 + 2679–2682 almost exactly, though with 7 variations in detail; V7 omits C 2667 (which Duggan inserts as V7 2620), while P 1271 makes it part of Oliver’s speech: Compains Rollans, dou bien ferir te painne. In C 2660, par (‘very much’, ‘extremely’) gives added stress to the two adjectives, though transposed to the first hemistich for the metre. V7 2613 reads: La bataille fu…, a metrically inaccurate 5-syllable hemistich; P 1264, La bataille est miravillouse et grande does not fit the rhyme, but it matches O 1653 (the first line of O laisse 125a) exactly. Cf. Introduction, 50 and 54, for a discussion of the implications of these variations. 2676. Vint mil furent in both C and V7 2624, which is hypermetric; the best solution is to emend mil to the more usual graphy, mille – and Duggan also does this. 2677. The mention of a baron from Aquitaine – and of Marsile’s gent d’Equiteigne in C 2668 and 2673 – on the Saracen side seems strange. In O, Aquitaine is mentioned only once (O 2325), as one of the lands Roland conquered with the help of his sword, Durendal; so, for O, it formed part of Charlemagne’s empire. The reference here may reflect more recent history, namely the warfare between the French crown and the English (Plantagenets) in the twelfth and early thirteenth centuries. 2678. Celui, ‘the latter’, is the direct object, followed by verb-subject inversion. Cf. C 2386 and note. 2679–2682. These lines are found in an identical position in V7 2627–2630 and P 1277–1280, pointing to an error in an early source: they relate quite clearly to Roland’s attempts to destroy Durendal (recounted in C 4054–4112 but always with a rhyme other than -eigne). Indeed, P adds an extra line (P 1281), Por Durendart que il volt qu’elle fraingne. The second hemistich of C 2681 echoes the second hemistich of O 2312, el perrun de sardonie, the first line of O laisse 172, whose -a- assonance includes several -aigne endings; but O’s laisse is correctly positioned just prior to Roland’s death.

La Chanson de Roland (Critical Text and Notes)

2680

237

Cil qi requirent Seint Jaqe en Espeigne Voient les cols el peron de Sartegne, Si com Rollant parti de sa conpeigne.



2685

2690

2695

2700

Grant bruit demoine la gent Marsilion. «Deus,» dist Rollant, «qi sofris pasion, Mot m’a fait Guene doloros guiredon; Vendu nos a par male traïson. E! Charlle, sire, com grant perdicion Recevras hui del meus de ta maison!» Dist Olivers, «Laissez vostre sermon, Car n’en donroie vaillant un esperon!» Lo destrer broche par tel division Et cil li cort par merveillos randon; En la grant presse mist son cors a bandon. «Deus,» dist Rollant, «se pert mon conpeignon, Ne pris ma vie vaillisant un boton.» A cez paroles esmoevent la tenzon Envers la gent Apolin et Mahon. Devant les autres vint pongnant Valdebron; Sist el destrier qi Marmoris ot non; Mot ben le broche par fiere aatisson; Cil cort plus tost c’ars n’en gete bouzon. Sor son escu ala ferir Sanson, Un duc de France qi mult est riches on.

2695. C: vaillant un boton, a hypometric second hemistich. V7 2643 and P 1292 agree on vaillant un esperon, which corrects the metre but repeats C 2690. The standard construction valeir un boton occurs 4 times in C, 3 times with a finite verb (C 269, 1723 and 2756), but the 4th time (C 1711) using the present participle of valeir in its alternative form, vaillisant. It seems likely that C here intended to avoid repetition of esperon, but omitted -is- from the verb in error. [Duggan emends V7 to une boton.] 2696. C: esmoeve with ~ over the final e to indicate the -ent ending. V7 2644 reads: qui movent, which Duggan emends to esmoevent. 2699. C: Sist il destriers qe… with 3 grammatical errors; but Sist el destrier qi is found in C 2727 and 2847, Sist el destrier C 2752. This is probably scribal error, with transposition of e and i (el… qi > il… qe) due to fatigue, but may show Franco-Italian features. Cf. also lo destrers, object singular, C 1788, and el destriers, object singular, C 6396, and destrier, object plural, C 40, 1396, 1531, 3122. 2701. A rare simile: ‘runs faster than a cross-bow shoots a bolt’.

238

2705

2710

La Chanson de Roland (Critical Text and Notes)

Escu ne brogne ne li fist garison; El cors li mist plein pié del confenon; Mort lo trestorne sanz nulle autre aqisson. A voiz escrie, «Toz i mosrez, gloton! Ferez, paien, car mot bien le vencron!» Dient Franzois, «Deus pere! Qe feron? Or vont li nostre a grant destrucion.»

42.a



2715

2720

Rollanz regarde, si voit Sanson morir; Lors a tel duel, des siens qida issir. Velantif broche per merveillos aïr, Trait Durendart c’onqes ne volt guerpir; Desor son eume q’il fist a or sartir Fiert lo gloton, nel volst pas mescosir: Tote la teste li fist en deus partir; Del front li fist andeus les els issir E durement a la terre cheïr. Dient paien, «Cist fait mot a hahir!» Respont Rollant, «Ne vos puis chier tenir.» «Monjoie!» escrie por sa gent esbaudir: «Ferez, Franzois, Deus vos puist garantir!»

 2725

D’Aufriqe i est uns Aufricans venuz: C’est Malqidanz, li fius al roi Malduz; Sor toz les autres est el estor venuz.

2706. aqisson is a graphy of achoison, ‘reason’, ‘business’. 2708. C: mot bien le vesron, but veïr, future 4 does not fit the context. Cf. O 1578 veintrum, V4 1552 vinceron, V7 2656 veintron, P 1310 vainceron, T 946 vaincron: ‘we shall conquer’. 2712. siens is a graphy of sens, ‘senses’, as in V7 2660, P 1314 and T 949. Cf. C 2632 and note. 2715. The MS reading is definitely Desoz, ‘beneath’, with a final z, but in C the final r of desor closely resembles z, except that the fine tail curls left from the left side of the base of the r. ‘Upon his helmet’ makes far better sense and V7 2663, P 1318 and L 590 all read: Desor. 2725. At this point, V7 (lines 2674–2675) inserts Contre soleuz flamboie sis escuz/N’a garnimant nen soit a or battuz before C 2726; P 1330–1331 is identical but T 964–966 reverse the lineorder, as do O 1552–1553, V4 1569–1570; L 598 has V7’s 2nd line only. This could be C’s error of omission, but it may be deliberate and the lines are not necessary to the sense.

239

La Chanson de Roland (Critical Text and Notes)

2730

2735

Sist el destrier, qi fist les sauz menuz; Broche lo bien des esperons agutz; Fiert Anseïs de l’espié moluz. Contre s’acer ne pot durer li fuz; Li osber ert fausez et desronpuz; Li cuers deu ventre li est en deus fenduz; L’arme s’en vait, li cors qiet estenduz. Dient Franzois, «Bon vasal, mar i fuz! En tant estor as esté coneüz.»



2740

2745

Pongnant i vent li arcivesqe ber; Onqes meudres ne pot mese chanter, Qi de son cors feïst tant a loër. Dist au paien, «Deus te puist vergonder! Tel as oncis qe t’estuet conparer!» Lo destrier broche, mot le fait tost aler; E trait l’espée, un coup li va doner. Fer ne acer ne li pot arester, Ne li hauberc garantir ni tenser; Tresqe l’arzon ne li poet contrester: Mort le trebuche sanz braire, sanz crier; L’arme s’en vait en enfer osteler.

42.b



2750

Grant bruit demenent cil Sarazin felon. Eç vos Grandoine brochant a esperon, Fil Capoé, a un paien felon; De Capadoce tenoit la region. Sist el destrier, c’apeloit Garamon;

2731. In the MS, the symbol for et was added, very finely and lightly in superscript, with insertion mark just after the cæsura, but this is superfluous and makes the second hemistich hypermetric. 2746. Cf. C 2217: Mort le trebuce sanz braire ne crier and the note which discusses whether the ms. reading, brair in both cases, should be read as braïr rather than emended to braire. V7 2696 here reads sanz braire et sanz crier. 2749. C: brochant e a esperon, the e being barred by the scribe.

240

2755

2760

2765

La Chanson de Roland (Critical Text and Notes)

Mot fort lo broze par fiere atison, E fiert Gerin en l’escu a lion, Q’il en ostroe lo tent et lo blason; Lo blanc osberc ne li valt un boton; El cors li mist lo fer et lo penon; Lez une roche l’abat mort el sablon. (* * * * * *) Aprés oncist Gerer son conpeignon E Berenzer de Gascoigne, et Guion. Puis va ferir Antoine d’Avignon, Qi tint Valence et la roche environ; Ses blans osberc ne li fist garison: El cors li mist les pan del confenon; Mort le trestorne sanz nule aquison. Aprés oncist Estouz, li fiz Odon. Dient Franzois, «Deus pere, qe feron? Or vont li nostre a grant destrucion!»

 2770

2775

Li cons Rollant tint Durendart sanglante; Deus, en tant lius icel jor la presente! Dist au paien, «Deus te doint grant tormente! Teus as ocis dont mere en ert dolente; Plorer feras tante belle jovente.» Point lo destrier qi de cosre n’alente; Au Sarazin fera ja male entente. Li qeus qe vence, la bataille est gente.

43.a

2758. After this line, C leaves a line blank, although V7, P, L and T all go straight on with their versions of C 2759. There does not appear to have been any line omitted. 2766. There may possibly be a fine vertical bar in black ink through the o of Estouz, which would change the spelling to Estuz, but elsewhere the graphy is either Estoz, Estouz or Estolz. 2771. C: tremente, a metathesized form of tormente (V7 2721), or possibly tremor, ‘fear-filled quaking’, with ending modified by poetic licence. P 1377: male entente; T 1000: mal entente. 2773. ‘You will make (people) weep for so much fair youth’, i.e. ‘You will cause so many fine young men to be bewailed’. 2774. C: de cosre valente, but a finite verb is needed. V7 2724 and T 1002 read n’alente (‘does not slow down’), indicating scribal confusion of n and u in C. P 1380 reads: s’avance; L 641: a talante. 2776. Li qeus qe vence, with qe the relative pronoun, subject of the verb. This is typical Old French confusion of the forms qe and qi.

La Chanson de Roland (Critical Text and Notes)

241



2780

2785

2790

En Grandoine ot proudome et ben vaillant, Et vertuos et hardi conbatant. Enmi sa voie a encontré Rollant; Ainc meis ne:l vit, se:l conuit au scenblant, A ses armes vermelles et al chival corant, Et a son fier visage et au fer contenant, As els c’ot vairs et al cors avenant. Et de l’espee vit tot lo brant sanglant; Mot volontiers s’en fust torné fuiant. Li cons lo fiert per itel maltalant De Durendart sor son eume luisant, Escu ne broigne ne li furent garant; Tresq’es arçons va s’espée colant; Trenche li cors, si oncist l’auferant; Tot l’abat mort; paien en sunt dolant. Dient Franzois, «Cist cols n’est pas d’enfant!»



2795

Par la bataille ot mot grant envaïes. Deus! Tantes testes i ot par mi parties, Osberc desroz et broines desarcies! Brisent cez anstes sor cez targes flories. François i ferent per lor chivalaries; Felon paien tot i perdent les vies: A vois escrient, «Marsilion, aïes!»

 2800

Felon paiens ferent hastivement, E le Franzois ne se targent nïent.

2778. C: Et untoos…, the scribe probably misreading u7 as u with ~ above. V7 2728 reads: vertuos, P 1384: vertuouz and T 1006: vertueux. 2781–2782. C’s two regular alexandrines give a noble cadence to the description. V7 2731–2732 and P 1387 have the normal decasyllabic metre. 2789. Tresq’es is a graphy of Tresq’as, showing a typical Franco-Italian confusion of vowels. 2795. C: desarcies, past participle of desarcir, variant graphy of desartir/desertir, here used as adjective and feminine plural: ‘cut to pieces’. V7 2745: desarties; P 1400: dessarties; V4 1640

242

2805

2810

La Chanson de Roland (Critical Text and Notes)

La veïsiez si grant caple de gent, Tant chevalier abatu et sanglent, Dont li cheval fuient mot durement Permi les chans travaillé et sanglent, Lor resnes rotes, dont li fren sunt d’argent. Muerent paiens come chaitif dolent; Ariere tornent tost et isnelement; Tresc’a Marsille n’i ot reteniment. Franc les encaucent per lor fer hardiment. Qi atenz fu, de mort n’en ot garent; Ou velle o non, tot i pert son jovent.

43.b



2815

2820

Fiert i Rollant come chevalier fors, Et Olivers qi si a gent le cors. Tote lor gent un poi n’ont de repos. Chevaus ont il bruns et bauzens et sors: Tot ont perdu lor valor et lor cors; Revenu sunt toz as pas et as tros; En sanc vermeil furent jusq’as genos. Les brans d’acer ont debrisez et tors; Cil q’i sunt mort, en ont mot le cuer gros.

and L 659 with singular noun: desartie. (Foerster and Mortier here read desarties in C, but no stroke to form the upper part of t is visible today.) Cf. also C 2595 and footnote. 2810. C: par lor fer m hardiment. 2811. The word atenz is a graphy of ateinz, past participle (subject sing.), ateindre (ataindre), here meaning ‘overtaken’. The same graphy is found in atendantz, present participle, C 3610 and atensist, imperf. subjunctive 3, C 7521. Cf. Pope (1934, 179s., § 467s., 197, § 528, 285 § 717, and 285, § 719) re the spellings ai, ei, e and nasals. 2815. Lor gent apparently refers to the French, who return wearily from the charge, but V4 1661 refers to Qui de Marsilio, similarly described in the subsequent lines. C 2821 must surely refer to the French. 2821. To make sense here, the first hemistich must be taken as an explanatory clause in parenthesis to en in the second hemistich: thus, ‘Those who are dead, they have a very heavy heart over them’, i.e. ‘They are heavy-hearted over those who have died’. V7 2771 reads Cil qi i sunt en ont molt les c.g., ‘Those who are there.’ (i.e. ‘remain alive’), and L 684 has: Cil qui sain sont.’; but P 1421 reads: Cil qui les ont (?Cil qui le sont/Cil qui la sont?) and T omits this line. O has no matching laisse and V4 quite different wording. It is possible that, in C, the scribe first omitted i from cil qi i sunt, then, lacking a 4th syllable to end the first hemistich and confused by ont mot in the second hemistich, copied mot as mort; indeed, his source may

243

La Chanson de Roland (Critical Text and Notes)

Trençent cez piez et meins, testes et cors: Tresc’a Marsille va li traïns des mors.

 2825

2830

2835

2840

Marsille vit sa gent si mal bailie; Sonent les cors, grant en fu la bondie: Puis chevaucha o sa grant ost banie. Mais per devant tote la baronie Vint uns paiens de mot grant segnorie: Abisme ot non, si qiert chevalerie; Ainc n’ama Deo, lo fil Sainte Marie; Mahomet sert, si fait mot grant folie. Ainc per son cuer ne fu jor sanz bosdie; Mot s’est pené de mal et tricherie; Asez plus l’eme qe tot l’or de Ongrie; Plus c’une poiz li est la car norcie. De vasalage a mot grant estoutie; Del roi Marsille a partot druerie: Son dragon porte, a cui sa gent ralie. Li arcivesqe ne:l amera ja mie, Puis qe il a si grant force acolie. Dist tel parole, qe nus on l’a oïe: «M’aïudans Deus, lo Fius Seinte Marie,

44.a

have read mout en ont le cuer gros, making the transfer of mout (as mort) to the first hemistich even more comprehensible. 2822. This is the French reaction to the killing of their comrades: they ‘slice through’ or ‘hack at’ the feet, arms, heads and bodies of the fleeing Saracens. Trençent is a graphy of trenchent. 2826. C: chevauchent in a hypometric hemistich; but the subject is Marsile, so the singular is needed. V7 2776 reads: chevalcha, V4 1674: çivalça; P 1452: chevaucha; this corrects both errors. 2829. C: Abinie, but this makes the hemistich hypermetric; V7 2779 has: Abisme, P 1455: Abismes. The poet regularly gives the Saracens names with a pejorative meaning (cf. Fausiron, Malqidanz, Maltalant, etc.), so Abisme makes good sense of a scribal error. 2833. C: de mal e de tricherie; and the same hypermetric hemistich in V7 2783. It may be corrected by dropping the second de – but de male trecherie would read.better and a source reading de mal e t. would furnish a basis for scribal error. The line is not found in this form in any other version. 2842. V7 2794 supports C’s reading Ha iudas Deus; of the other versions, only P 1464 has a line at all comparable: Deux, dist, peres, qui iestez fiuls Marie. Mortier and Foerster transcribe this as Ha, Judas Deus, but it is theologically impossible to couple the name Judas with Christ as the Incarnation of God, as invoked here, since it was Judas who betrayed Him. Again, judas

244

2845

La Chanson de Roland (Critical Text and Notes)

Cist Sarazins est toz plens d’eresie! Onqes n’amai malvaise cohardie. Se Deu pleüst, ne deüst estre en vie.»



2850

2855

Li arcivesqe recomença l’estor. Sist el destrier, qi fu de grant valor: En Danemarche lo conquist per vigor; Envers Cesaire, lo riche enperaör, Uns clers estoit de Deu nostre Seignor; Tant par est baus q’il n’a soz ciel meillor Et est si bons qe nus n’a sa valor. Li arcivesqes lo broce per baudor; Del frein a or li a fait gran laissor. Fert lo paien, vers qi il ot iror, Desor la tarze qi fu jaune a flor: Pieres i a de diverse color Et escarboncles qi getent grant ardor,

as a graphy of judise or juïse (‘Jews’ or ‘Jewish’) makes little sense: Turpin would hardly invoke ‘God of the Jews’ in this context. If it were a scribal error for judex, ‘judge’ (used in C 8062 in the phrase, Judex de la Loi), then judas could be an allusion to the Last Judgement, a popular subject in mediæval church carvings and theology. Even if one tried to read Judas as an epithet applied to Abisme (C 2829), reinforcing the description in 2833–2834, this would require an hiatus in the first hemistich so as to link Deus to the second hemistich – making any reading aloud or recital impossible. This would cause major problems in an ms. without punctuation, particularly as the cæsura follows Deus. This is probably a scribal error, for Godefroy records a standard formula, commonly used in Old French when swearing an oath: Si m’aist Dieus or M’aist Dieus (‘So help me God!’). The verb aidier (variants: ahider, aiuer, aier) was regularly used in appeals to God; Pope (1934, 92, § 190) lists a conjectural GalloRoman form aj(u)dare and comments (§ 759) on the frequent use of the present subjunctive 3, Deus vos adjutet! (‘May God help you!’) and its resistance to analogical re-creation; she also lists in § 925 the double radical aid- and aiü- and in § 1058 the present indicative, aïde. Fouché (1967, 14) mentions a form aiuder. Assuming a faulty division of the word by the scribe gives the reading aïudas – or, if the omission of ~ is also postulated, aïudans, present participle, subject form, aïuder, ‘helping’. V7’s reading, Ha Iudaus is even closer to aïudans and H very similar to M: i.e. M’aïudans Deus: ‘God helping me’, thus: ‘As God the Son of Saint Mary is my help’, a very appropriate invocation. Cf. C 1835: A deus aïe, V7 1784: E deus aïe, a very similar phrase, though used in error in the earlier context. 2844. C: ama, but this is Turpin speaking and the first person singular is needed; V7 2796 has amai. Reduction of ai to a is a feature of Northern, North-Eastern and Eastern French. Cf. Pope (1934, 488, § 1320, vii).

245

La Chanson de Roland (Critical Text and Notes)

2860

2865

2870

Come solel gete grant resplendor; En Val Mortal, ce dient li plusor, Uns vis diables li dona per amor; Cil le tramist Galafre l’amanzor. Turpins i fiert o sa ruste vigor. Aprés cel cop, n’ot onqes puis onor; Escu ne broine ne li fisent honor; Mort lo trestorne entre dos cent de lor; «Monjoie!» escrie, c’est l’enseigne francor. Dient Franzois de la Terre Major, «Cist arcivesqe val mielz q’enpereör! Car pleüst Deo, le nostre creator, Q’il eüst auqes de teus en cest estor!»

44.b



2875

2880

Li cons Rollant apele Oliver: «Sire conpeing, si:l volez otroier, Cist arcivesqe fait fortment a priser! Por amor Deu, car li alons aidier!» Cil li respont, «Ne fait pas a tarzer.» Cascuns d’als broze son bon corant destrier; Aprés cest mot, le vont a comencer. Grant sunt li cop et li caple sunt fier; La veïst hon un estor si pleinier, Tant escu frait, tant espié briser,

2859. The word gete was added at the end of the line, with its correct place of insertion marked. 2864–2865. That both lines end with the same word (onor; honor) arouses suspicion of scribal error. V7 in 2817 and 2818 has: seignor… honor. T 1084 closely matches C 2865 but ends: firent retour. Neither substitute seems particularly appropriate, so the duplication has been left. 2867. C: Mojoie, with omission of n or ~ through scribal error. 2877. In the C ms., a hypometric first hemistich with the cæsura marked after als, followed by the normal 6 syllables: broze son bon destrier. All other versions divide the line after broche. V7 2830 reads: Chascuns d’els broche son coreor destrer; P 1528: Chascuns brocha son aufferrant destrier; T 1096: Chescun d’eulx broche son aufferrant coursier; and L 750: Chascons brocha son bon corant destrer. Placing the cæsura after broche in C makes the second hemistich hypometric, so L’s 2-syllable corant has been inserted (V7’s coreor having the same meaning but too many syllables). Nevertheless, the first hemistich does not read smoothly and, despite the evidence of V7PTL, the suspicion remains that it should in fact read:Cascuns d’entr’als, with C’s cæsura correctly placed, or Cascuns brocha..

246

2885

La Chanson de Roland (Critical Text and Notes)

Tant blanc osberc deronpre et desmailier, Tant bon ceval fuir tot estraier, Dont li vasal gisent mort el senter. Devers Franzois crurent li enconbrier; Anzois q’il muirent, se vendront il mult chier.



2890

2895

2900

Li Franc de France, de la Terre Absolue, Mainte chiere arme i ont le jor perdue, Mais encor ceint chascuns l’espee nue. Paiene gent est morte et recreüe; Envers Marsille ont lor voie tenue. Respont li rois, quant cele ot entendue: «Terre de France, tu soies confondue! La toie jent a la moie vencue! Li enperere a la barbe chenue Conquis a Rome, Calabre a retenue, Costentinoble et Saissoigne sa drue. Meus vel morir sor cest herbe menue, Sa grant fierté ne soit hui abatue! Se Rollant muert, mot chier li ert vendue; En grant dolor en ert France cheüe; Ja por Charllon n’ert mais jor retenue.»



2905

Felon paien por lor rustes vertuz Fierent de lances et d’espiez moluz; Franc les destreignent a lor brans esmoluz. Adonc fu bien li caples mantenuz: Brisent lor lances, si fendent lor escuz,

45.a

2885. C: Divers, but C elsewhere always uses the spelling devers, found here in P 1536, T 1105. 2886. C: se vendront mult chier, which is hypometric; V7 2844 reads: se vendront il mult chier and T 1110: ils se vendront… 2891. C: vie (‘life’); V7 2849 reads fuie and ‘taken their flight’ makes better sense, but P 1544 and T 1118 agree on voie – and ‘have made their way’ is the obvious reading in the context. 2892. cele refers back to the paiene gent in C 2890, therefore a feminine singular although the translation would be ‘… when he had heard them’; the feminine is later essential for the rhyme. 2907. In the C ms., Brisen with the final t added faintly.

La Chanson de Roland (Critical Text and Notes)

2910

2915

247

Fausent ces broignes ou li ors est batuz. Deus! Tantes testes i sevrerent des buz! Tant pié, tant puing i ot le jor perduz, Et tant vasaus oncis et confonduz, Dont li cheval fuient per les paluz! Li cons Rollant mot en est irascuz; Quant voit morir ses amis et ses druz, De pité plore li vasaus coneüz. Et de son oncle li est il asez pluz, Par cui il est amez et chier tenuz. En la grant presse est as paiens venuz; Mot durement s’i est li ber venduz.

 2920

2925

Li cons Rollant est en la presse entrez; Desor son chief est ses escuz gitez; Tint Durendart, dont li pons fu dorez; De bien ferir fu mot entalentez. Deus! Tant bon eume en a esquartelez, E tant escu et fendu et troëz; Tant Sarracin i a hui mort gitez, Teus .c. paiens i a le chief copez; Li pire d’als est forment alosez, De vasalage essauciez et prisez.

 2930

Cons Olivers d’autre part s’en torna. Por bien ferir iloqes s’aresta;

2909. C: sevrerent des buz, a hypometric second hemistich. V7 2867 reads: i ot sevré del bu and P 1567: i ot sevré dou bus, the verb (il) i ot being used impersonally. The construction i sevrerent is a little clumsy, compared with the smooth flow of i ot sevré (or possibly i ont sevré, to match C’s verb in number), but there seems no good reason to modify C further. 2917. C: Por cu il…, ‘Because he .’, but V7 2876 reads Par cui and ‘By whom he .’ makes much better sense. Mis-reading Par as Por causes the copyist to reduce the repeated i of cui il to a single i. 2926. C: sar with abbreviation symbol above to indicate this stands for Sarracin, as written in V7 2885. The final word, gitez, is a graphy of getez (‘thrown to the ground’). 2930. Olivers begins the line in C, leaving the hemistich hypometric; V7 2889 has: Quns Olivers, V4 1762: Dux Oliver, T 1151: Quant Oliver; the graphy cons is more common in C.

248

2935

2940

2945

2950

La Chanson de Roland (Critical Text and Notes)

Tint Hauteclere, qe il forment ama; D’or est li pons, bons fevre la forja; Cent tanz valt d’or qe ele ne pesa; Sanz Durendart, soz ciel meillor n’en a. Li cons la tint qi mot bien l’asaia. Deus! Tant bon eume le jor esquartela, E tant escu fendi e peçoia, Tant bon osberc ronpi et desmailla, Tant pié, tant puig, tante teste copa; Li sans vermeus jusqe es piez li cola. Dist li Rollant, «Conpeng, traiez vos za! Nostre amistié hui cest jor partira. Li enperere, quant nos reconosra, En douce France jamais tel duel n’ara; Bien ait prodon qi por nos proiera, En sainte eglise orison en fera.» Lo destrier broçe, la resne li lassa; Il a escrié, «Conpeing, traiez vos za! Li uns sanz l’autre, se Deus plet, n’i mosra!»

45.b



2955

Qui donc veïst Rollant et Oliver De lor espées ferir et chaplier; Li arcivesqes ne se volt atarzier. Cels qi sunt mort bien puent repouser, Il est escrit, a Seint Denis mostier. Ce dist la geste, ainc n’i ot recovrier. Li .xii. per en ont mort .x. meller;

2933. C: fuere, due to metathesis of u and e. V7 2892 reads: fevres and T 1154: fevre, meaning ‘blacksmith’. Thus: ‘a good smith forged it’. Reading fuerre (‘sheath’) would ruin the sense. 2937. C: tan, with omission of the final t, but tant is consistently found in the next few lines. 2942. C: Conpeg, with missing n; V7 2901 reads: Compeing. 2944. The word quãt has been squeezed in, on the line, by the scribe, as if (partly?) omitted at first. 2951. Qui donc veïst…: ‘Then one could have seen…’. 2955. C: a Seint Denis el moster with either i or 7 (= ri) inserted after t. Mortier reads mostrier, Foerster mostier; moster (occasionally mostier) is the graphy found elsewhere in C. This is a hypermetric hemistich in the C ms., but V7 2915, P 1629 and L 793 agree in omitting el (‘in the’).

249

La Chanson de Roland (Critical Text and Notes)

2960

A qatre estors se puent bien aidier; Li quinz aprés fist mot a redoter. Nostre Franzois orent grant engonbrer; Tot sunt ocis li baron chevalier, Ne mais qe .c. qe Deus vost espargner; Ja de cel jor n’istront sanz engonbrier.

 2965

2970

2975

Li cons Rollant voit des suens tel daumage, Tel duel en ot, per un petit n’enrage. A Oliver en a dit son corage: «Sire conpeing, je vos ting per mot sage. Tant bon vasal voi mort sor cest herbage, Dont l’enperere a perdu son omage; Pleindre devons douce France la large E Charllemeine, le roi de haut parage. Confaitement porrons avoir mesage, Por cui saüst le nostre grant daumage, Qe nos sofrons en cest desert sauvage?» Dist Oliver, «G’en dirai mon corage: Meus viel morir, qe g’en die hontage.

46.a



2980

2982a(V7)

«Ge sonerai le cor,» ce dist Rollant, «Se l’osra Karll’ qi est as porz passant. Je vos plevis, senpres ert retornant; Secorra nos per lo mien esciant.» Dist Olivers, «Blasme en avrez grant, E reprover aront vostre parant! Quant jel rovai, nen feistes niant Ni ja per moi ne serez mais cornant.»

2972. The abbreviation 9 = con is here used, unusually, as a capital letter to begin the line. 2973. Por cui, ‘by means of which’. This may be contrasted with Por cu = Par cui, ‘by whom’, C 2917. 2978. C: .Kll’., a new abbreviation for Karlle or Karlles, possibly used to stress that Karlles is monosyllabic at the end of the hemistich. 2982a. V7 2943 paraphrases O 1708, V4 1806, matches P 1649 and approximates to L 810. It justifies Ni (beginning of C 2983), so its omission is probably C’s scribal error.

250

La Chanson de Roland (Critical Text and Notes)

 2985

2990

2995

Ce dist Rollant li proz et l’alosez: «Sire conpeing, merveillos tort avez. Por amor Deu, por qoi me ranpoinez? Nostre bataille est fort bien, lo veez; Por devers nos en est li noals alez. Ge cornerai, se vos bien le volez.» Dist Olivers, «Vos en serez blasmez; Vostre lignage en sera meins amez. Quant je:l vus dis, se li cors fust sonez, Charlles fust ci et ses riches barnez; Marsilions ne fust ja si ousez, Q’encontre lui eüst paiens mandez. Mais per ma barbe qe vos ici veez, Se puis venir el reigne o fui nez, Ma seror Aude, jamais ne la verrez; Entre ses bras nule foiz ne gisrez!»

 3000

3005

3010

3015

Li cons Rollant o la chiere hardie Oït Oliver, qi si le contralie. Mot doucement li dist et se li prie: «Sire conpeing, por Deu lo fil Marie, Vos me portez iror et felonie.» Dist Oliver, «Bien l’avez deservie! Meus valt mesure qe ne valt estoutie! Mort sunt François per vostre legerie. Se de ton corn fust faite la bondie, Ci fust li rois qi France a en bailie, Pris fust Marsille, ou il perdist la vie; Nostre bataille fust pieza defenie; Vostre proeze ert anqi chier merie. Vos i mosrez, France en ert afeblie; Li enperere de nos n’arra aïe.» Li uns vers l’autre plore per conpaignie.

46.b

2994. The word ia is written superscript with insertion mark beneath. 2999. gisrez is a variant graphy of gerrez, future 4, gesir, ‘to lie’. 3014. C: de nos arra, but the negative required here is supported by V7 2978: n’avra ahie.

La Chanson de Roland (Critical Text and Notes)

251



3020

3025

3030

Li arcivesqe les ot contrarier; Point le cheval des esperons d’or mer, Mot doucement les prist a chastier: «Sire Rollant, et vus, sire Oliver, Por amor Deu qi tot a a baillier, Li uns vers l’autre ne se doit corochier, Car hui cest jor morons sanz recovrer. Corocement ne nos ara mester. Loing nos est Karll’, tart ert son repairer, Mais neporquant, si porra il aidier, Vendra li rois, si nos voldra vengier; Ja cil d’Espeigne n’escanperont entier. Franc troveront nostre grant enconbrier, Ploreront nos q’il nes voudront laissier, Prendront nos cors, si nos feront cochier Desor les bieres qi porteront somer, Enterreront nos en arche de moster; Sauvages bestes ne nos porront mangier.» Rollant respont, «Ci a bon civaler!»

3023. ‘Anger will be of no use to us’. 3029. C: qi nes voudroit laissier, but all the other verbs in this sentence are future 6 with Franc (C 3028) the antecedent of qi. Conditional 3 makes sense only if the implicit subject is either Charlemagne (rather distant in C 3026) or ‘each one (of the French, 3028) who would…’; here, nes is a graphy of neis, ‘not at all’. V7 2993 reads: qu’il nen volront l…, ‘for they will not wish to leave any’. There was clearly confusion in a source text: in both P 1693 (que nel porront l…) and T 1257 (qu’il ne pourront l…) there is uncertainty over the final letter, either t or s in P and possibly a deleted t or s in T (with similar forms verron, prendron, ploreron, etc.). So it is doubtful whether nos was seen as the antecedent of qe or qi; and PT’s use of porront suggests that this clause may have originally referred to the weeping. C may here reproduce an error in the source text, corrected in one way in V7, in another in P. C’s use of qi may be due to common Old French confusion of qi and qe but could also be read as q’i; however, given the identical readings in V7PT, q’il nes voudront seems most appropriate. 3031. C: qi porteront; V7 2995 and T 1256 read: que p… 3032. C: Ent7reront nos, with the cæsura marked after nos both here and in C 3040, expansion of the abbreviation producing a 5-syllable hemistich. Since this occurs twice, one must assume that the repeated er of Enterreront was elided into a single syllable: Enterr’ront, to give a regular metre. Here, V7 2996 reads Et feront nos sepellir en enterrer, which Duggan emends to Et metront nos en arches de mostier (on the basis of C). Then V7 3004 has En feront nos en artres por repos, again emended by Duggan to: Enterront nos en aitres por repos. Cf. O 1750: Enfüerunt en aitres de musters and footnote to C 2040 below.

252

La Chanson de Roland (Critical Text and Notes)

 3035

3040

«Sire Rollant, se sonez est li cors, Karles l’osra, qui a passé les porz; Secorra nos, et il et ses esforz. E sachiez bien, meinz poinz en ert detorz, Quant nos vesront ensanglentez et morz. Enterreront nos en arches por repos; Ne nos menjucent cez bestes et cez porz.» Rollant respont, «Ci a mot bon conforz.»

47.a



3045

3050

3055

Li cons Rollant o la chiere menbree Oït la raison, merveilles li agree, Qe l’arcivesqe li a dite et contee. De l’olifant la lumiere a tornee, Mist l’a sa boche, si sona la menee; Enpenst lo bien per mout fiere alenee; Tentist li porz, si respont la valee; Granz .xv. liues en est la voiz alee. Karles l’entent de France la loee, O sa grant ost qe il ot retornee; Dist a ses homes, «Nostre gent est meslee; As Sarazins ont bataille jostee.» Guenes a bien la parole escoutee E dist au roi, «Ceste avez vos trovee! S’autre el deïst, menzoigne fust provee.»

3038. Having begun the line Sachiez… in error with capital S, the copyist added E in the margin. 3040. See footnote to C 3032 with regard to problems of metre here. Where C reads arches (implying a church or cloister), V7 3004 has artres, O 1750 aitres, ‘cloisters’ but also ‘cemetery around a church’. C probably misreads this through ignorance. In both C and V7, por repos is an unsatisfactory rhyme, but it may reflect pronunciation: i.e. effacement of r in the -ors/-orz endings. Cf. Pope (1934, 156, § 396). 3041. menjucent is present subjunctive 6 of mangier/manjier. 3045. C: conitee; V7 3009 and P 1709: contee. 3048. C: Eupenst, written in full, with confusion of u and n; O 1754: empeint; V7 3012: Speint. 3056. Use of the feminine, ceste, may be justified by relating it back to parole, C 3055, or in anticipation of menzoigne, C 3057; it might also relate implicitly to pensee, ‘thought’, ‘idea’. It is needed so as to justify the feminine singular of the past participle, required for the rhyme.

253

La Chanson de Roland (Critical Text and Notes)



3060

3065

3070

Li cons Rollant a la chiere hardie A grant dolor, car forment afeblie; La sois l’argüe et li chals le coivrie. De l’olifant a faite la bondie; La maistre vene a ronpue et partie. Paiene gent en est mot esbaïe; Li plus seürs a sa voie acolie, Quant Blanzardins a l’enarme saissie A un destroit mist sa grant conpaignie. Dist Oliver, «Por Deu lo fil Marie, Or voi grant bruit de la gent paienie. Or del ben faire, qe Monjoie est fenie!» Ensus se traient une archie e demie.

47.b



3075

Blanzardins fu et forz et conquiranz; O lui .m. homes, Sarazin et Persanz. Per mot grant pienes, per merveillos samblanz, E per dolor sona son cor Rollanz; De son cervel est la tenple ronpanz; Pormi la boche li ist fors li cler sanz. Fort fu l’alene, la voiz en fu mot granz; Karles l’entent, qi est as porz pasanz. Nayme li dus en est mot sospiranz

3059. The feminine singular, afeblie, shows that car is the substantive, ‘flesh’, not the conjunction. Laisse 191 is found only in C and V7 (laisse 181), not in OV4PTL. 3060. C: e li chals le coniurie, a hypermetric hemistich in which conjurie (‘exhort’, ‘invoke’) makes no sense and convirie is not recognisable. V7 3024 reads: coivrie, present 3, coivrier, ‘to torment’. C’s error probably arises from a (typical) mis-reading of the -iu- combination in the source. 3068. C and V7 both read: Or voi grant bruit…; although ‘noise’ must be heard, not seen, Baligant’s words in C 5228 show that this was a common colloquial expression: Je voi la noisse, si com vont hon disant. Since veïr has a second meaning, ‘to be present at’, ‘to witness’, this may mean ‘I am aware of’. 3071. Both et are written as symbols, but the first is inserted superscript, in the same ink as the text. C 3071–3072 are not found in the other versions. 3075. C: teple, with ~ omitted; V7 3039: temple; but C normally uses n before a p.

254 3080

3085

3090

3095

3100

La Chanson de Roland (Critical Text and Notes)

Et dist al roi, «Cist cors est conbatanz! Rollant lo sone, ce est ses olifanz. Ainc ne:l sona, se ne fu besoing granz.» Guenes respont, li leres sozduianz, «Ja estes vos veus, chanuz et feranz! Iteus paroles rescenblent ben d’enfant! Bien conoissez l’orguel q’i a Rollanz; Il est mout bauz, si est fous ses talanz; Grant merveille est qe Deus l’en est sofranz. Ja prist il Nobles tot sanz les vos comanz; Fors s’en issirent li Sarazin as chans; Tuit s’entrocissent a lor espiez trenchant. Rollant li fers, li hardi conbatanz, Se fist lever enz es pré verdoianz, Saisi les cors a toz les conbatanz; Q’il volst li sans ne fust aparisanz. Sor toz les pers est il ore gabanz; Aprés une lievre est tote jor cornanz! Chevauchez, rois, ne soiez atarjanz! Terre Major, qi tant per est vaillanz, Loinz est encore, ne soiez detrianz. A mot grant piece n’i serez sejornanz.»

3080. V7 3044, P1725, L 849 match C’s first hemistich, apparently due to non-recognition of verb-subject inversion (Ce dist li roi) in O 1768, V4 1874, T 1284, but C 3084–3085 read better as addressed to Nayme, not Charlemagne. Cf. also Introduction, 80–81 (xxix), Franco-Italian: Morphology, definite article. 3082. C: besoig, with n or ~ omitted, unless the g should be transcribed as ng; V7 3016: besoing. 3084. C: ver, a graphy of vair, ‘grey-haired’, which may imply ‘showing signs of age’, ‘becoming old’; V7 3048 reads veus, graphy of viels (adj.subject form sing.), and O 1771: veilz. Since use of 3 synonyms (ver, chanuz, feranz) seems excessive, ver has been taken as a scribal error for veus. 3088. C: qi, V7 3052: qui, but T 1291: que. 3089. C: tot sanz le vostr comanz, a rather strange phrase in which the final e is dropped from vostre for metrical accuracy (cf. V4 1882: seça vost comant), while le is oblique form, singular and comanz, oblique plural. Neither V7 nor P has such confusion, V7 3053 reading: toz sanz les voç comanz and P 1736 omitting tot/toz and reading: sans le vostre commant. Since the plural ending -anz is needed in C for the rhyme, emendment has been based on V7. 3095. Both C and V7 3059 read en fust aparisanz, indicating that Roland wanted the blood to be visible; but O 1775–1779 relates how Roland took Noples without Charlemagne’s authority – and washed away the blood from the meadows afterwards to hide evidence of the battle (lever C 3093 and V7 3057 = laver). Here, he removes the bodies for the same reason, en being a scribal error for ne (due to metathesis).

255

La Chanson de Roland (Critical Text and Notes)



3105

3110

Li cons Rollant sa boche ensanglenta; De son cervel la tenple ronpue a; La maistre vene de son cuer desevra; Permi la boche li clers sans li cola; E prist son cor, terce foiz li sona. François l’oïrent et Karles l’escouta. E dist li rois, «Cist cors grande alene a!» Respont dus Neyme, «Car frans hon le sona. Cist fel l’a mort, qi feindre nos rova.» Franzois desendent et Karles s’adoba; Si secorra ses homes qi sunt la E son neveu q’en tel dolor laissa. Cui chalt de ce, quant meis ne lo vesra!

48.a

 3115

3120

Li enperere fait ses grasles soner. François desendent por lor cors conréer; La veïssiez tant osberc endoser, Elmes lacer et ventalles fermer; Ceignent espées dont li pon sunt d’or cler. Qi les veïst sor cez chevaus monter, Tantes enseignes en cez lances fermer! Les destrier brochent, mot les font tost aler; Hastivement volront as porz passer. S’il venent la ou Rollant est, li ber,

3102. Laisses 193–196 are found in identical form in V7 laisses 183–186, but C alone repeats the same incidents in laisses 197–201, in an assonanced form closely corresponding to O’s version. Cf. Introduction, 23–25, 38–48 for more detailed examination of this repetition. 3106. C: tres foiz li sona, ‘three times he blew it’, a hypometric hemistich, but V7 3070 reads: tiercha foiz…, P 1750 and T 1302: autre fois. Each time Roland sounds the horn, there is a reaction from Charlemagne: the first time (laisse 190), Ganelon dismisses this as an invention (C 3056–3057); the second time (laisses 191, 192), he argues against taking it seriously (C 3086– 3101). This is the third time, when Naymes accuses Ganelon of treachery and the French army turns back. C’s tres is probably a scribal error for terce (as in C 4976, 5272, 5289, etc.) rather than the PT reading autre. 3117. C: osberec, with an extra e inserted in error, which would make the hemistich hypermetric. 3120. ‘Then one could have seen…’; cf. C 2951 and footnote.

256

La Chanson de Roland (Critical Text and Notes)

3125

Chascuns de cels se volra mot pener. Qi calt de ce, trop sunt tart a aler!



3130

3135

3140

Por dan Karllon fist Deus droite raison; Li jors s’eclaire, car de fi lo set hon. Duel a li rois de Rollant le baron; Karles chevauche per fiere contençon Et li Franzois en sunt en grant friçon; Chascuns en plore desoz son eume enbron. Li rois fait prendre li conte Gainellon; Si lor comande per tel division, S’il lor eschape, ja n’arront raiençon. Idonc le prist uns quus de sa maison: Maistre cous fu, si l’apelent Bovon. De totes parz corrent si conpeignon; Guenellon pristrent, qi q’en poïst ne qi non, Et poinz e piez li lient de soion. Aprés li pelent la barbe et le grenon; Sor un somer leverent le gloton, Se li lacerent el col un chaënon;

48.b

3125. C: Chascuns d’els, a hypometric first hemistich, as in C 2877; but V7 3089: Chascuns de cels. 3127. The word dan was inserted superscript, in the same ink, as a correction. 3135. C: Si lor…, exactly as in the line above, but there Si = ‘So’. S’il lor… is the better reading here and one l might easily be dropped in error in this context. Duggan corrects V7 3099: Si lor to S’il lor. 3136. C: qus + ~ over u; V7 3100: quns; both C and V7 regularly use these standard graphies of cons, ‘count’. V4 1921 reads: quanti; but O 1817: cous, ‘cook’. The scribal error (probably in the CV7 model) arises from a mis-reading of quus through confusion of n and u. 3137. C: cons (written 9s, the normal abbreviation) is again a scribal error for cous, as in C 3136. The same word occurs in C 3197, when this incident is repeated in C. The ‘master cook’ in charge of victualling Charlemagne’s army would be an official of considerable importance. 3139. The word poïst is added at the end of the line, with its point of insertion indicated after en. It is apparently treated as monosyllabic; cf. Introduction, 92–93 regarding -aï- and -oïdiphthongs. 3143. C: chainon, but 3 syllables are needed for the metre and the normal graphy is chaënon (as in V7 3107). C 3206 reads chaënon and confusion of e and i seems more likely than omission of additional a, with ai a graphy of e (chäainon). The same confusion occurs in the plural chaines, C 4427 and 6109.

La Chanson de Roland (Critical Text and Notes)

3145

257

Son cors ont mis a grant destrucion; Tant l’ont gardé q’il rendirent Charllon.



3150

3155

3160

3165

Karlles chevauze, tristes de mautallant, Sa blance barbe sor sa broigne gisant. «Dieus,» dist li rois, «beaus Pere roiemant, Per Guenellon me crost pene mot grant! En velle geste est escrit de lonz anz Qe traïson firent forz et pessanz El capitoile de Rome a escianz: Jule Cesar, qi tant per fu vaillanz, Mordrirent il o lor grefes trenchant; Puis en morurent per merveillos ahanz, En feu grezois en fu chascuns ardanz. Cist fel retrait a lor apartenanz, Qi ce loa Rollant fust remananz, La reregarde de mes homes faisanz. Ja se je:l pert, de ce sui bien creanz, Jamais n’ert jors n’en soie meins dotanz Ne sor mon chief n’ert corone seanz.» Ainz qe venist Karles li rois puissanz En Rencivals, o li chaples fu granz, Avoit vencues .vii. batailles Rollanz; Marsille fu vencuz et recreanz; De l’un des braz ne fu il pas portanz.

3145. C: redirent, omitting n or ~; cf. V7 3109: qel le rendrent, O 1829: que:l rendrent, V4 1933: chi lo renda and T 1348: l’ont rendu. Cf. Fouché (1967, 264s., §§ 132s.) re Latin reddere, found also in early Italian. In C’s repetition of this incident, C 3208 reads: Tant lo garderent qe:l redirent Charllon, ‘until they handed him back to Charles’. In C 3145 qil may be a contraction of q’il le, which would have made the hemistich hypermetric, or le may be implicit from the first hemistich. C here conforms to the earlier version of the story, found in O; this line is contradicted by Section C’s later remaniement. 3155. C: emorurent, for en morurent, ‘they died for this’, since V7 3119, P 1805, T 1370 and L 911 all insert en. Otherwise, this could be a graphy of esmorurent, ‘they died’. 3157. ‘This villain belongs to their company (is of their kin)…’. 3161–3162. ‘Never a day will pass without my being unsure whether my crown rests securely on my head’: Charlemagne recognizes that, without Roland, his authority is weakened. Meins is used to emphasize the negative.

258

La Chanson de Roland (Critical Text and Notes)



3170

3175

Li cons Rollant a sa boche sanglente, De son cervel li est rompue la tenple; L’olifant sone a traval et a pene. Karlles l’oï, qi l’intent a grant peine; Ce dist li rois, «Cest cor a grant alene». Respont dux Nayme, «Cil vasal est en peine. Bataille i a, per lo mien escïentre. Cil soit ocis, qi se vos rouva feindre! Adobez vos, s’escriez vostre enseigne! Si chevauchiez, qe Rollant nos demande.»

49.a



3180

3185

Li enperere a fait soner ses cors; François descendent, si adobent lor cors; Escuz ont genz et espiez granz et fors, E confenons blans et vermoil et blois. Es destriers montent, qi les porterent tost; Brozent et pongent tant com dura li pors. N’i a celui qi d’autre non parloit: «Se dan Rollant veions ainz qe fust mors, Ensanble o lui i ferirons grant cols.»

3168. At this point, C repeats in assonance the incidents contained in laisses 193–196. Cf. Introduction 23–25, 38–48 for analysis of this repetition and its relationship to the assonanced versions. 3172. C: elenae, possibly elence, since c is often closely attached to the next letter. This is a scribal error for alene, ‘breath’ or ‘blast’ of a horn (cf. C 3077 and 3108). O 1789: Cel corn ad lunge aleine. 3174. C: esciente (sf), but lo mien indicates that this is masculine. This (assonanced) laisse corresponds very closely to O laisse 135 in which 1791 reads: Bataille i ad par le men escïentre. C’s reading has therefore been modified to escïentre (sm). 3175. C: qi se vovra feindre or qi se voura feindre, a hypometric hemistich. O 1792 reads: ki vos en roevet feindre, ‘who asks you to act in a craven or half-hearted fashion’ (roevet, present indic.3, rover) and C 3110 (in the earlier version of this laisse) has qi feindre nos rova. Thus, vovra may arise from a telescoping of vos rovra; alternatively, voura could be an error for rouva, with transposition of r and v – and the initial v of the error arising from the missing vos. This gives: ‘Let him be killed, who thus asked you to act in a craven… fashion’, a direct reference to Ganelon. 3185. C: ainz qe il fust mors, with il barred. 3186. C: ferirons, with the r in superscript in a different style of script, so a later insertion.

259

La Chanson de Roland (Critical Text and Notes)

3190

Cui calt de ce, car il demorent trops! Esclariz est li vespres et li jors; Contre solel reluisent cil aume a ors E cil escu qi ben sunt peinz a flors E cil espi qi d’or ont l’esplendors.



3195

Li enperere chivauche par iror, Et les François dolenz et corozos; N’i a celui qi durement ne plor, Car de Rollant si ont toz grant paor.



3200

Li rois fist prendre li conte Gainellon, Se le comande as cous de sa maison; Tot le plus mestre en apela Sanson. «Bien lo me garde com encrisme felon; De ma masnee a faite traïson.» Cil le reçoit, s’i met .c. conpeignon

49.b

3189. The second hemistich is hypermetric unless aume a ors is read as 2 syllables; but a or (‘with gold’) or aor (‘now’) is always disyllabic elsewhere in C. The rhymed version of this laisse (194) offers no help. O reads, lines 1808–1809: Cuntre le soleil reluisent cil adub, Osbercs e helmes i getent grant flambur, implying that C’s line derives from a much modified remaniement of the earlier text. 3191. C: espli, a scribal error for espi, ‘spears’. 3192. C’s 4-line laisse 199 should form part of laisse 198 with the same assonance; but iror… plor… paor (all lacking final s) may have caused C to read these lines as rhymed (despite corozos, C 3193). 3197. C: Se la comande, but the feminine pronoun is clearly wrong and emendation to Se l’a comandé would make the hemistich hypermetric. O 1817 reads: Si:l cumandat, but C 3134 has: Si lor comande. In the context of this line, le is the only possible reading. As in 3137, the ms. reading: 9s = cons (‘counts’) is emended to cous (‘cooks’). Se, here a graphy of si, ‘thus’, has not been modified. 3198. Sanson here, but Bovon, C 3137, Besgun in O 1818 and Begon in V4 1922. Sanson, one of the French peers, has already been killed (C 2711). This ‘master cook’ (possibly ‘quartermastergeneral’?) may correspond to the ‘rich baron’ Sanses in C 6326 and to ‘count Sanson’ in 6453. 3201–3202. C: Et cil le recoillent, si metent conpeignon/Cil de la coisine,… This gives a 5-syllable first hemistich in each line, unless the oi diphthong is read as 2 syllables: recoïllent. But 3202’s first hemistich remains obstinately 5-syllable. At this point, O reads (O 1821–1822): Cil le receit, s’i met .c. cumpaignons/De la quisine… and V4 1925–1926: Celu lo guarda a si cento compagnon/ De la cuxina… C’s earlier version of the laisse has, in 3138, a considerably modified line: De

260

3205

La Chanson de Roland (Critical Text and Notes)

De la coisine del meus de sa maison; Icil li pelent la barbe et li grenon; Cascuns le fiert quatre cols de son poing; Forment le hurtent o fust et ou baston. Puis li ont mis el col un chaënon; Sor un somier le trosserent en son; Tant lo garderent qe:l rendirent Charllon.

 3210

3215

Par grant iror chevauche Charle Meine; Desor sa broigne li luist sa barbe blenche. Pongent a force tot li baron de Franche; Il n’i a cel qi n’en plor ou non plange, Quant il ne sunt ou Rollant en conpaigne. Il se conbat as Sarazins d’Espeigne; Mort sunt li sien, ne sunt qe soul .l.; Asez est fous qi ja meillor demande.

 Rollanz regarde les monz et les laris; De cels de France i vit des morz jesis totes parz corrent si conpeignon. T 1340 has Si le receit o lui ceu compagnon. Certain features are common to O, V4 and T: the verbs are 3rd person singular, as reference to C 3198 leads one to expect; .c./cent (100) companions are mentioned (taking T’s ceu as a scribal error); the first of these lines, but not the second, begins with Cil (Celu in V4, an aurally equivalent Si le in T). C’s first hemistich has therefore been emended on the basis of O; in the second hemistich, metent is seen as a scribal error for met .c. or met cent, arising from the (faulty) use of the plural verb earlier; finally, in C 3202, Cil is taken as arising from the earlier errors or copied in error from the previous line and therefore deleted. C thus closely matches O (as is the normal pattern in this passage) and retains standard 4:6 metre. 3208. C: redirent. Cf. note to C 3145 above. 3210. C: li luist sa barbe; O 1843 reads: li gist; T 1362: Sa… barbe… jesans. It appears that li luist (< luire, ‘to shine’) arises from a scribal misreading of li iist (= li jist, li gist < jesir/gesir, ‘to lie’) with repetition of li + extra i producing lu-; but barbe blanche makes this an apt and effective modification. 3212. C: ou nõ b plange, with b written in error and barred. 3217. From this point, C once again corresponds to V7, P, L and T, after the short repetition. 3218. C: vit des morz jesis, a hypometric second hemistich, with jesis (for the rhyme) as poetic licence for jesir (‘to lie’). V7 3133 solves the problem with vit tant gesir ocis; P 1819 has: eni vit moult jesir; T 1385: voit gesir tant d’ocis; L 920: i vit maint mort gesir; and O 1852: i veit tans morz gesir. The addition of i is thus common to 3 versions and the infinitive, gesir, common to

La Chanson de Roland (Critical Text and Notes)

3220

3225

3230

261

Et il les pleint come vasaus gentis: «Seignor baron, de vos ait Deus mercis, Totes vos armes o lui en paradis O les apostres et ses autres amis! Mellors vasals ne vit vostre ancedis. Si longement m’avez toz jor servis; Per vus a Karlles ses grant reignes conquis; Li enperere si mal vos vit moris. Terre de France, mot fustes dou païs; Huy es deserte de tant home de pris! Baron François, por moi vos voi ocis; Je ne vus puis garentir n’estre aidis; Aït vos Deus, qi onques ne mentis!

all. The grounds for possibly emending jesis to jesirs are set out in the notes to C 1504 (cf. also notes to C 3457, 3463 and 3464, plus Introduction, foot of 41 (i), 43, 83 (xlvii) 90 and Appendix B, 99). 3221. Strictly, this line needs a finite verb, as is found in the other versions: V7 3136: mete en son paradis, with T and L the same; P 1822: metra en paradis and O 1855: otreit li paradis. In C, a fold or fault in the parchment produces a gap in the word: oli ii with the usual slanting stroke over the final i and the preceding letter (or half letter) no more than a short vertical stroke with slightly curved top and base. It bears no resemblance to mete; and taking it as a scribal error for otreit (present subjunctive 3, otrïer, ‘to grant’), as in O, requires emendation of en paradis to li paradis. The other versions offer no support for reading oli ii as a graphy of aliet (< alever, ‘to raise’); but o lui (‘with him’) reads quite well as an elliptical construction, with either soient or ait (C 3220) understood. 3226. Here, vos vit moris is the same construction as in C 3218, with the infinitive (dependent on vit) spelt with -is, not -ir, to meet the requirements of rhyme; i.e. ‘saw you die’, with poetic licence. O 1860 reads: tant mare vos nurrit (‘in such an evil hour brought you up in his service’) and P 1828 tant mar nos a norris, L 926 con mar vos a noriz, T 1392 si mal vous a nourris; but emending moris to noris, ‘brought up in his service’, means vit needs emending. V7 3141 has vit… mortis (perhaps for mortris/mordris, past participle, murdrir, thus ‘killed’?) pointing to a variant (or an error: vit for ait?) in the CV7 source. This is found at the very beginning of the second ‘cluster of assonance’, in a laisse where C and V7 are almost identical, so V7 could well still be using the same shared model at this point. (cf. Introduction, 73, Orthography, and 90, Editorial Policy). 3227. C: dor païs (or possibly doi païs), but V7 3142 reads douz païs, P 1828 dou païs, O 1861 dulz païs, L 927 dou païz and T 1395 des païs. 3231. C: Bau sire Dex, a standard invocation, but here a finite verb is needed; indeed, one expects to find another line to complete the sentence, after the relative clause. V7, P, L and T all omit this line, but O 1865 reads: Aït vos Dex and V4 1973: Deus vos aït – and the wish ‘May God help you!’ fits the context extremely well. There is no close match between the two phrases as grounds for scribal error; it seems to arise simply from the substitution of one standard phrase for another.

262

La Chanson de Roland (Critical Text and Notes)

Oliver, frere, vos ne doi estre eschis; De duel mosrai, quant ne puis estre ocis. Sire conpeing, ferir alomes is!»

50.a

 3235

3240

3245

Li cons Rollant a Oliver mostree Une raison li a dite et enseignee: «Ensanble o Frans devon morir, beau frere, Por lor amor en estrainge contree. O, douce France, com es deseritee De bons vasaus, confondue et gastee! Mot grant sofraite en ara l’enperere.» Por quatre foiz a ‘Monjoie!’ escriee. Prent l’olifant, si sone la menee; Lo cheval broche tote une randonee; Puis va ferir de sa trenchant espee.

 Li cons Rollant va ferir de s’espee, De Durendart q’il a mult bien provee. 3232–3296. The vellum of this folio is very thin, particularly towards the foot of the page, where the writing on the reverse side shows through and makes deciphering difficult; there is a mended tear in the edge and erasure on C 3260 has produced a hole along the line of script, so that some words are illegible or practically illegible. 3234 . The word is in the text is a graphy of i (‘there’), the inorganic s being added to give the rhyme. 3236. A rare example of apo koinou in C, with une raison the object of both mostree (C 3235) and dite e enseignee (C 3236). The second hemistich is hypermetric unless li a is elided into one syllable. V7 3150–3151 read Li quns Rollant a Oliver contee/Une raison li a dit e monstree, which may be the source reading (the feminine contee referring forward to raison as does C’s mostree), but C’s reading makes sense. V4 1977–1978 has: Li cont Rollant Oliver en appelle/Une raixon li a dit e membré. P 1836, T 1401 and L 934 read: A Oliver a dit raison membrée. 3237. C: Ensanble Frans, but the normal phrase is ensanble o (‘together with’) and this is used (with a variety of spellings) throughout C. V7 3153 reads: Ensemble ot F (corrected to od by Duggan), P 1829 has Ensamble as…, T 1404 Ensemble es… Insertion of o does not affect the metre in this line. 3238. C: estraige, with omission of n or ~. V7, P, T and L all read: estrange. 3246–3261. Laisse 204 should in fact be two laisses. The first, 5 lines ending in -ee, has the same rhyme as the preceding laisse and its first line is an echo of the last line of that laisse; but it deals with a new, single, separate incident. It is followed by 8 lines rhyming in -ien/en. This laisse falls within a ‘cluster of assonance’ – cf. Introduction, 23–25, 38–48 and

La Chanson de Roland (Critical Text and Notes)

3250

3255

263

Fabrin d’Espaigne i a permi trenchee; Puis en oncist .xxiiii. Sarazins de Medee; Jamais n’iert hon qe honor si agree. Si com li porz s’en fuit devant les chien, Devant Rollant s’en fuient li paien. Dist l’arcivesqe, «Asez lo faites bien! Chevalier, tel valor avoir deverien! Qi armes porte et ou cheval sit bien, En grant bataille contenir se doit bien,

97–100 (Appendix B) – and corresponds quite closely to O laisse 141 (-e- assonance); indeed, it shows the problems of converting certain assonances to rhyme. In V7, T and L this laisse rhymes in -ez; P has -ee rhyme throughout. 3249. This is a regular alexandrine in C. The number .xxiiii. is confirmed by O 1872 and P 1871 (V7, T and L omitting this line), but in a decasyllabic line: O’s E .xxiiii.de tuz les melz preisez is matched exactly by P. In C, it seems odd to separate the number from its accompanying noun and suggests that C (or its model) may have reduced 2 lines to a single line, omitting 2 separate hemistichs in error. 3250. ‘Never will there be a man to whom honour is so suited’, i.e. ‘… more deserving of honour’. Here, qe is a graphy of the indirect object, cui. 3251. The erasure of the first letter of this line left a hole, so that nothing could be substituted for the initial error. Both O 1874 and V4 1991 read: Si. It seems likely that the scribe (or a revisor) suddenly realized that, with a change of rhyme (cf. note to C 3246), a new laisse was needed – which would of course begin with a larger capital. He therefore erased the standard size S already written. 3254. In C, the cæsura is marked after tel, indicating that the Italian scribe read this as ‘Such knights’ (i.e. ‘knights of such valour’) followed by …deverien read as conditional 6: devereient, minus final t, ‘ought to have’; but deverien/deverient would not produce a rhyme in -ien/-en (though an Italian scribe might make this error). It must therefore be conditional 4: deverien standing for devriens/devrions (with Chevalier the addressee, vocative plural). This is a 4-syllable North-Eastern French form with epethentic e: cf. Pope (1934, 347, § 919 **, 367, § 972 **, and 490, § 1320, xxii). Thus the cæsura is marked in error after tel, dividing adjective from noun. Relocating it after valor gives a regular alexandrine. In the other mss., O 1877 reads: Itel valor deit aveir chevaler, which V4 1998 and P 1877 closely match; T 1417–1418 are almost identical to V7’s smoothly rhymed: De tel valor doit estre encoragez/Bons chevalier qui est enparagez (V7 3177–3178). 3255–3256. C: Qi armes portent e ou cheval si bien/En grant bataille contenir se doit bien; but (i) the mix of singular and plural verb forms does not make sense, (ii) nor does the second hemistich of C 3255 and (iii) both lines end in bien, a word already occurring at the end of C 3251 (and vouched for there by O 1876). O 1878 reads en bon cheval set, suggesting that all verbs should be singular and that si represents present 3, sedeir (seoir): siet reduced to sit with final t dropped. This leaves 2 consecutive lines ending in bien – suggesting that C 3255 should terminate with se tien, though no evidence is available. O 1878–1890, V4 1998–1999, P 1877– 1878, V7 3177–3178 and T 1418 have all verbs in the present indicative 3. C 3255 reads ou for au: either a scribal error or a Franco-Italian graphy.

264

3260

La Chanson de Roland (Critical Text and Notes)

O altrement certes ne:l prise l’en. De soi vengier ne se foingne rien!» Iloc ont pris un itel hardimen, Rollant respont, «Frere m’es et par…!» A icest mot ferent communalment.



3264a[P]

Hom qi si fiert ja n’ara reproçon; En la bataille fait grant defension; Bien s’i contint a guise de baron. Ez vos Marsile brochant a esperon

50.b

3258. The verb foingne is a graphy of the (jussive) present subjunctive 3 of feindre: ‘Let him not show any cowardice in taking revenge’, ‘Let him not be half-hearted in…’. 3259. C: un itel with ∧ below the i, followed by a word, probably hardimen (‘act of courage’, ‘act of daring’), partly lost in a hole in the vellum. To read this as a scribal error for aroiement (haroiment?), meaning ‘battle arrangement’ or here ‘fighting stance’, would fit with ont pris. The line sits oddly here; it can hardly be part of Turpin’s speech, but it may well belong after C 3261. Cf. O 1884–1885: A icest mot l’unt Francs recumencét/Mult grant damage i out de chrestiens. V4 has Turpin’s speech immediately followed by Roland’s, then (V4 2001) A cest mot l’ont François començer as final line of the laisse; T 1422 matches this. Like O, V7 has 2 lines (3182– 3183) ending the laisse (rhymed in -ez): A icest mot fu li estor comenchez/Maint Saracins i fu a mort jugez, matched by L 953–954, paraphrased by P 1881–1882. 3260. The end of the line is illegible because the scribe, in attempting to erase an error, made a hole in the parchment; but the first word of the second hemistich is definitely frere, written in full. According to Foerster, the next word is niés but the normal slanting mark indicating an i is not found, so mes is a better reading. This is followed by the usual abbreviation for et (‘and’) and a word beginning pa- with indecipherable ending. Foerster transcribes this as paciii, but par~ is a possible reading. Thus C’s initial reading (before erasure) may well have been Frere m’es et paren (final t being dropped): «You are my brother (in arms) and kinsman!», Roland recognizing Turpin’s valour. The attempt to erase these words is easily explained by reference to the other versions: O 1883 has: Ferez, ne.s esparignez, V4 2000 Ferez né nul esparmier, V7 3181 Ferez a force, gardez nes espargnez. Thus the copyist or a revisor sought to emend C’s reading to something approximating to Fere, ne.s espargnien!, ‘Strike, let us not spare them!’ – taking espargnien as present subjunctive 4: Pope (1934, 359–364, §§ 951–963) cites -ons and -iens as alternative endings) with final s dropped (cf. deverien, C 3254) – but was defeated by the hole. 3262. The final syllable of this line is illegible; the rhyme requires reproçon, which fits the sense in C. O 1886 reads: Home ki ço set que ja n’avrat prison, matched by V4 2002 and T 1423; V7 3184 and L 956 have qil n’avra garison, P ends line 1884 with raenson. All have a different first hemistich. 3264a. It is clear from the context that C 3265–3266 refer to Marsile; C 3278 confirms this and O 1889 (As vus Marsilie…), V4 2009 (E vos M…), T 1427 (Si vint M…),V7 3189 (Ert vos M. brochant a esperon), P 1890 and L 961 (Ez vos M. brochant a esperon) supply the missing line.

La Chanson de Roland (Critical Text and Notes)

3265

3270

3275

(V7) 3279a(V7&O) 3280

265

Sus el ceval auferant et gascon; Broche lo bien, si va ferir Hugon: Cil estoit dux de Balne et de Digon. L’iscu li fause et l’osberc ageron; Mort l’a abatu sanz nulle arestison. Puis a oncis et Ivoire et Ivon, Ensanble o els Girart de Rosion. Li cons Rollant ne li est gaires lon; Dist au paien, «Damedeus mal te don! A si grant tort me tols mes conpeignon; Coup en aras ainz qe nus departon, E de m’espée el cors savras le non.» Vait lo ferir a guise de baron; Trenchié li a li cons lo destre poing, Que a la terre chaï jus el sablon. Puis prist la teste de Girfales le blon: Cil ere fius au roi Marsilion. Paien escrient, «Aïdez nos, Mahon! Li nostre deus, aidez nos de Karllon,

3269. C’s first hemistich makes good sense but has 5 syllables unless the two a are elided. Cf. O 1894: que mort l’abat.., V4 2012: …l’abat mort…, P 1898: Que mort l’abat, but V7 3195: Mort le trabuche and T 1432: Mort le trebuche, changing the verb to obtain 2-syllable present 3 and avoid elision. 3271. In C, this is written: Et ensanble oels, suggesting that the scribe read this as 4 syllables. However, since Et is written in full, this could well be an alexandrine, with ensanble o (normally trisyllabic) read without elision. V7 3197 reads: Aprés ocist…, O 1896: Ensembl’od els, V4 2014: Insemble ot els and T 1434: Ensemble o eulx, but P 1900 has Et en aprez. and L 966 E avec auz. 3272. C: Li oors .R.; V7 3198: quns; V4 2015: cons; and O 1897: quens. At the end of the line, C reads: bon, which is not apt here. V7 and V4 read: lon (‘far away’) and O: loign. 3279. C: La teste en fait voler en el sablon, followed by the identification: Cil ere fix in C 3280; but 3278 ended with Roland cutting off Marsile’s right hand and Marsile does not lose his head at this point. A line is missing in C’s confused version of events. V7 3208 reads: Puis prist le chief de Girfales le blon, O 1903: Puis prent la teste de Jurfalen le blund, supported by V4 2022, P 1920 and T 1442. V7’s preceding line, 3207, Que a la terre chaï jus el sablon, provides justification for el sablon in C. Presumably the scribe misread la terre as la teste, possibly through anticipation of the next line, and modified the line accordingly. 3281–3282. C: aïdez nos, Mahon! with aïdez here tri-syllabic. This corresponds to O 1906: Aïe nos, Mahum! and V4 2024: Aïa nos, M! The normal 2-syllable aidez is found immediately below in C 3282, but the value of the -ai-/-aï- diphthong is variable in C; cf. Introduction, 92–93. V7 3210 has: Car secor moi, Mahon!, not found in any other version. This could be V7’s own modification of the CV7 model, indicating later rejection of tri-syllabic aïdez. P 1927 reads:

266

3285

La Chanson de Roland (Critical Text and Notes)

Qi in Espaigne nos a mis tel felon, Qe por morir le canp ne guerpiron!» Dist l’uns a l’autre, «Et car nos en fuion!» A icest mot, tel .c. miler s’en von; Qi a c’il plait, ja ne retorneron.



3290

Li rois Marsille le poing destre a perdu; Encontre terre a geté son escu; Lo cheval broche des esperrons agu, Droite la reine vers Espeigne s’en fu. E bien .c. mille s’en alerent o lu, Qe onqes au plait ne torneront mais plu. Dist l’uns a l’autre, «Li niés Charlle a vancu.»

 3295

3300

De ce qe chaut se fuiz est Marsille? Remés i est ses oncles l’augalie. Cil tient Cartage, Oliferne e Caudie, Etiopës, une terre maudite; La noire gent ot en sa conpaignie: Encore sunt plus de .l. mille; Grant ont les els et le cors et l’eschine. Icil chevauchent estroitement per ligne,

51.a

Aidiez, sire Mahon!, echoed by T 1444. C’s repetition of aidez nos in consecutive lines with different pronunciation is unique to C. 3283. C: Qi in E…, the in possibly due to repetition of the i of Qi, otherwise a clerkish error arising from scribal familiarity with Latin in. V7 3212 reads: en Espegne, PTL all have en Espaigne; V4 2026 has: in tot Spagne, whilst O 1908 reads: En ceste tere. 3284. C omits the final letter of guerpiron (future 4 or 6) for the sake of the rhyme. Compare von, C 3286, and retorneron, C 3287. The problem does not, of course, arise in the assonanced versions, but V7 3213 modifies the whole line to avoid the difficulty. P 1931–1932 give these words to the French, with future 4; T 1447 closely matches C; L lacks this line. 3288. This short laisse, matched by V4 laisse 155 and smoothly rhymed, is not found in O, although it lies within a ‘cluster of assonance’. In contrast, laisse 207 matches O laisse 143 almost exactly.

La Chanson de Roland (Critical Text and Notes)

3305

3310

267

Puis escrierent l’enseigne paienie. Ce dist Rollant, «Or recevrons martire, Mais huni soit qi bien ne si vent prime! Ferez i, Franc, de l’espee forbie! Si chalongent lor cors de la gent saracine, Qe douce France n’en ait d’als reprocine! Quant en el canp venra li rois, mon sire, E de paiens i vesra tel traïne, Contre uns des nos i avra plus de .xv., La sara Charlle qe nos bien la faïsme.»



3315

Quant Rollant voit la contredite gent, Qi plus est noire qe n’en est arement – N’en ont de blanc fors les dens solement –

3303. C reads: Puis escrient which is hypometric but matches O 1921 exactly; V7 3242 reads: Tuit escrient, but emendation to escrierent, past def.6, corrects the metre. Duggan also adopts this solution. 3305. C: .si vent. prime, with a full stop after vent and prime then added. 3307–3308. After the direct command, C 3306: Ferez i, Franc…!, C’s alexandrine, Si chalongent lor cors…, present indic. 6, is unsatisfactory as an aside interrupting Roland’s speech, but so too is present (jussive) subjunctive 6: ‘So let them defend their bodies…’. The 3rd person plural is confirmed by lor, C 3307, and d’als, C 3308. All the other versions have a 10-syllable line. O 1926–1927 reads: Si calengez e voz mors e voz vies,/Que dulce France par nus ne seit hunie!. V4 lacks this line. V7 3249 reads: Si chanchelons et le chors et la vie (corrected by Duggan to: si chalonjons et le cors…) but lacks C 3308 (V7 3250–3253 being a paraphrase of C 3309–3312). P laisse 106 has 2 versions: the rhymed P 1967–1968, a paraphrase: Chascuns preudom doit chalongier sa vie,/Que nus mauvais soz son mantel n’en rie and the assonanced P 1980–1981: Si chalongiez et vos cors et vos vies,/Que douce France ne soit par nos honnie much closer to C. L 1016 has: Si sauverons e nos cors et nos vie, T 1472: Si calengon le corps et la vie!, with imperative 4 like V7, then both echo P 1968. C’s initial error, chalongent, arises from misreading of the -ez ending in the source (cf. O 1926, P 1981); the copyist then changes vos to lor, par nos or par nus (de nuls ?) to d’als. The C reading shows the C scribe’s typical ‘line by line’ approach and limited comprehension; it has therefore not been emended. 3309. C: en el cãp, with insertion mark below el – all in much finer script. 3310. Although traïne is needed for the visual rhyme, the context requires traïn (sm. ‘massacre’), not traïne (sf. ‘betrayal’). This may be poetic licence, with final e added to avoid a nasalized vowel in a laisse assonanced in -i-, not -in-, rather than an error due to scribal confusion of traïn/traïne. 3312. faïsme is a graphy of feïsmes (past definite 4, faire), with final s omitted for the rhyme. 3315. The word ont (õt) is added above the line.

268

3320

La Chanson de Roland (Critical Text and Notes)

Ce dit li cons, «Or sa ge veraiement C’or nos morons per lo mien escïent, Mais huni soit qi primes no se vent. Seignez me, Franc, car je le recomenz.» Dist Olivers, «Dahait ait le plus lent!»



3325

3330

Li paien voient qe François i a poi; Entr’aus en ont et orguel et bofoi; Dist l’uns a l’autre, «Li rois a tort, ce croi.» E l’augalie sist el cheval turqoi; Mot bien le broce, ce vos creant per foi. Fiert Oliver el dos derere soi; Lo blanc osberc trés parmi le ronpoi; Parmi lo piz son espi a besloi Li fist outre paser, grant duel en ot le roi. «Charlle de France mar vos laissa rer soi! Mal vus a fait, si com je cuit et croi; Vengié en ai tot ceaus de nostre loi.»

51.b

3316. C: sage, but see note to C 1453 regarding sa as present indicative 1, savoir, when there is verb-subject inversion. O 1935: sai jo, V7 3258 and P 1988: sai. 3317. C: Car nos m…, with car for c’or (‘that now’), possibly due to (Franco-Italian) confusion of vowels, but probably a simple scribal error, since O 1936 reads: Que hoi murrum…, P 1989, T 1482 and L 1027: Que hui, and V4 2059: Que… anchoi. 3318. ‘But may he be dishonoured who does not first sell himself dear’. In C, cher (‘dear’, ‘dearly’) is implicit, but O 1924 reads: Mais tut seit fel cher ne se vende primes. C’s no may be a scribal error for ne, or for non with omission of n or ~, or a Franco-Italian form. V7 3260 reads: …qui primes ne s’en vent; V4 2060: ..que primer non s’en vent. 3321. Laisse 209 is almost identical to O laisse 145, once allowance is made for changes arising from assonance-to- rhyme conversion, also to V4 laisse 158. P’s laisse 109 is assonanced like O. 3329. C: Li fist autre…, where outre is needed; the tail of the a linking it to the u is clearly visible. Later in the line: grant duel en ot le roi, with insertion mark ∧ beneath en. 3330. C: Charlle meine de France; O 1949: Carles li magnes; V4 2072: Carlo de France; P 2030: Karles de France. 3331. C’s cuit is a graphy of cuid, < cuider, present indicative 1 being flexionless: cf. Pope (1934, 340, § 898).

La Chanson de Roland (Critical Text and Notes)

269



3335

3340

3345

3350

Voit Olivers qe a mort est feruz; Tint Hauteclere don l’acer est moluz, Fiert l’augalie sor l’eume irascuz: Pieres e flors l’en a creventez juz; La teste li fent deci as denz menuz. Aprés li dist, «Paien, mal aies tuz! Je ne di mie, Karles n’i ait perduz, Mais tu ne:l nonceras el reigne ou tu fuz.» Puis en apele Rollant qe veigne a luz. Voit Oliver qe a mort est feruz; De lui vengier est fortment aveüz; En la grant presse, se fiert tot esperduz. Qi lor veïst Sarazins desronpuz, Un mort sor l’autre a la terre estenduz, De bon vasal remenbrer li peüst. L’enseigne Charlle oblier non volt pluz: «Monjoie!» escrie, mot est bien conëuz. Rollant apele, son ami et son druz,

3340. The alexandrine gives greater weight and stress to this line. 3341. C: ve igne, the word veigne being split because of a fault in the parchment. 3342. This is an exact repetition of C 3333; at this point a new laisse begins in O (laisse 147) with change to -er/-et/-ez assonance, and in V7 (laisse 197D) and V4 (laisse 160) with change to -er rhyme. C 3342–3352 continue laisse 210, maintaining the -uz rhyme but corresponding to V7 3308–3318 in the first hemistichs. In lines 3342–3343, C has accidently re-copied the first two lines of the model of V7 195D: he then continues with C 3344 = V7 3310 with final words modified for the -uz rhyme – and continues with the same pattern, based on V7 197D. Both 195D and 197D begin Sent Oliver; C confuses the 2 openings because the 2 laisses are consecutive (as they are in OV4PTL). Laisse 196D is unique to V7. It must also be later in date than C’s model because it would otherwise prevent C’s insertion of V7 197D as the second part of C 210. 3343. O 1966 reads: De lui venger ja mais ne li ert sez in a new laisse; V4 2086 has: De si vençer no se vols tarder but also inserts earlier (line 2074, after V7 2073 = C 3333): De sei vençer tarder no se vol plu. All the other versions insert this line between C 3333 and C 3334, with a variety of second hemistiches: V7 3282: mout entalentez fu; P 2034: ne fu mie esperdus; L 1054: ne fu pas esperduz; T 1507: tieus hons ne fu. In C, the reading is aveüz/a veüz, with -üz needed for the rhyme, 3 syllables for the metre. To fit the context, the meaning must be ‘determined’, ‘desirous’ or ‘committed’, ‘vowed’, but no verb meets all the requirements – although veüz, past participle, veïr may have influenced the pronunciation of -üz as a separate syllable. Godefroy lists an adjective, aveu, apparently meaning avisé (Mod.French); ‘committed’ would fit very well. 3350. C: so ami with omission of n or ~: son ami in P, T and L, sun ami in O 1975.

270

La Chanson de Roland (Critical Text and Notes)

Se li a dit, «Enscanble n’irons pluz.» Li uns a l’autre fait duel qi n’i poit pluz.



3355

3360

Rollanz regarde Oliver el visage: Teinz l’a e pers, descoloré et paile; Li sans tot clers fors de son cors avalle, Encontre terre envolent li esclache. «Deus!» dist li cons, «Or ne sa ge qe face! Sire conpeing, mar fu vostre bernage! Tant mar veïstes proëze et vasalage! O France douce, com voi cest jor sauvage; De tes barons ja n’aras estorage. Li enperere en ara grant doumage.»

52.a



3365

3370

3375

Or est Rollant sor son cheval pasmé Et Olivers i est a mort navré; Tant a seinné, li oil li sunt troblé, Ne loinz ne prés ne puet mas esgardé; Ne coneit il negun home charné. Son conpeignon, quant il l’a encontré, Feri l’en l’eume qi est a or gemé, L’une moitié li fent jusc’al nasé, Mais en la teste ne l’a mie adesé. A icest cop l’a li cons regardé, Se li demande doucement et soé: «Sire conpeing, feïstes vos de gré? Je sui Rollant qi tant vus a amé;

3354. Although the scribe wrote epers as one word,Teinz l’a e pers refers back to visage, C 3353: ‘He has it discoloured and livid’. Cf. O 1979: teint fut e pers, ‘he was purple/discoloured and dark blue/livid’. P and V4 paraphrase this: P 2089: Tout le vit taint…; V4 2099: Tut lo vid descolori. 3356. C: escache, with l omitted, a scribal error; O 1981 reads: esclaces, ‘splashes’ (of blood). 3357. C: sage for sa ge. Cf. notes to C 1453 and 3316 regarding this usage. 3366. Another example of poetic licence: esgardé, when the infinitive, esgarder, is needed. 3371. C: laste with te added superscript above the a, apparently by the scribe, to give la teste.

271

La Chanson de Roland (Critical Text and Notes)

3380

Per nul meschie ne vos ai oblié.» Dist Olivers, «Or vos oï al parlé. Feru vos ai, char me le perdoné.» Respont Rollant, «Ne sui point enpiré. Je vus perdoins ici e devant Dé.» A icest mot, l’uns a l’autre encliné; Por tel amor es les vos desevré.



3385

3390

Olivers sent qe la mort mout l’angosse; Andui li oil en la teste li troble; Si pert l’oïr et la lumere tote. Descent a pié, contre orient se coche, De ses pechiez si reclame sa coupe; Contre le ciel andeus ses meins ajoste, Si prië Deu qe paradis li done, Si beneïse Charllon e France doce, Son conpeignon Rollant desor toz home. Faut li li cuers, li eumes li enbronche; Trestoz li cors a la terre li joste; Morz est li cons, plus ne vos en dirome.

52.b

3376. C: nulle meschie, but meschie (a graphy of meschief) is masculine and nulle makes the hemistich hypermetric. 3377. ‘I hear you by your speech’, i.e. ‘I hear who you are.’, ‘I recognise you by your speech’ – Oliver can no longer see clearly. V7 3382: or vos oï parlé and O 2002: Or vos oi je parler; C should have parler, but has modified the ending for the sake of the rhyme: poetic licence or Segre’s pseudo-rhyme. 3378. Char (with inorganic h) is a graphy of car, adverb, used to strengthen the imperative 5, here found in a rather strange form, perdoné, for the sake of the rhyme (pseudo-rhyme). 3380. C: ici devant e dere, but ‘here in front and behind’/’here before and after’ makes little sense. V7 3385 reads: ici e devant Dé; O 2007: ici e devant Deu; V4 2125: ci davant De. 3392. C: Faut li cuers, but cuers is monosyllabic; the meaning is ‘his heart fails him’. An extra li has been omitted through scribal error. Cf. O 2019: Falt li le coer, V7 3401: Faut li li cuers, T 1601: Li cuers li fault. 3393. C: Trestoz li cuers, but cuers is found immediately above in 3392 and does not fit this context. V7 3402, O 2020 and P 2156 all read cors and T 1602: corps. 3394. This is one of the occasions when the writer intrudes into the narrative, communicating personally with his audience: ‘we shall not say anything more to you about him’. The -ome/omes ending for present 4 was in general use in twelfth-century French, but particularly in the North-East both earlier and later: cf. Pope (1934, 339, § 895). V7 3403 has: ni vet plus demorant; O 2021: que plus ne se demuret; V4 2135: qui plus no se demore; P 2159: n’i a plus de son tans.

272 3395

La Chanson de Roland (Critical Text and Notes)

Voit le Rollant, qi:l plore et lo dolose; Ja en nul leu n’estoit plus dolant home.



3400

3405

Li cons Rollant, quant vit mort son amis Gesir adenz, contre oriant lo vis, Ne puet muer ne plor et n’en ait pis. Si doucement a regreter lo pris: «Sire conpeing, de vus ait Deus mercis! Enscenble avons esté meint anz e dis; Ne me fesis un mal, ne a toi ne fis pis. Quant tu es morz, a mot grant tort sui vis.» Al duel q’il ot, li cons pasmez s’est is Sor son cheval, qi ot non Velantis; Afermez est es bons estriers d’or fins; Qel part q’il aut, li cons ainc n’en chaïs.

 3410

Ainz qe Rollant se soit aperceüz, De pasmeson gariz ne revenuz, Mot grant doumage li est apareüz:

3400. C: lo pris, with inverted w over the o to give lor/lors (‘then’), but O 2025 reads: a regreter le prist and the direct object pronoun makes better sense; pris is past definite 3, prendre, cf. Conquis C line 2, mentis C 3231. 3402. A hypometric second hemistich in the ms.: esté meint dis. V7 3413 reads: estés maintes dis (dis, s.masc.or fem.), but this remains hypometric unless it is read as 5:5 metre; P 2170: Ensemble o soi soit la moie toz dis; T 1614: Ensemble ay esté a vous tous dis and L 1162: Compainz avons esté ensamble de toz dis; but O 2028 is closest to CV7: Ensemble avum estét e anz e dis, as is V4 2144 (et ans et dis). C’s reading has been emended to meint anz e dis. 3405. C: pasmez s’est ils, with what appears to be a second subject pronoun in the wrong number added at the end of the line for the rhyme. V7 3416 reads: li quns chaï pasmis and T 1618: le duc c’est pasmis, changing the verb from 1st to 2nd conjugation; O 2031 has: se pasmet li marchis, echoed by P 2173 and L 1165 – but no line corresponds exactly to C’s. Since is was found as a graphy of i, ‘there’, in C 3234 (in the earlier laisse rhyming in -is), presumably ils stands for is = i with pseudo-rhymed ending. 3408. C’s chaïs could show a past definite 3 ending, cf. notes to lines 2 and 3400; but V7 3419 reads: n’i ert chaïs, suggesting that C’s n’en is a scribal mis-reading of n’ert. 3410. C: ne revenuz with re added in finer script and lighter ink, with insertion mark beneath.

273

La Chanson de Roland (Critical Text and Notes)

3415

3420

3425

Mort sunt Franzois, toz les a il perduz Fors l’arcivesqe et dan Gauter de Luz – Repairez est de la montaigne suz, O cels d’Espaigne ou mot s’est conbatuz; Morz sunt li suen, si l’ont paien vancuz; O veille o non, a Rencevals s’enfuz. – E si apele Rollant q’i viegne a luz: «O gentis cons, com vallan on es tuz! Onqes n’en ot poor, la o tu fuz! Je sui Gauters, qi conquis Maleüz, Li meudres hon as vels et as chenuz. Per vasalage sui del estor issuz; Ma hanste est fraite, pecié est mes escuz, Et mes osberc desmailié e ronpuz; Permi le cors o lances m’es consuz. Senpres mosrai, mas chier me sui venduz.» A icest mot Rollant l’a coneüz; Le cheval broce, si va corant a luz.

53.a

3417. C: de Rencevals…, but Gautier should not flee from Ronceval, since that is where the main battle against the rearguard takes place. V7 3428 reads: aval est descenduz, P 2185: est aval d…, T 1630: en val est d… O 2043: desuz cez vals s’en fuit and V4 2159: sor son civals s’enfu offer the clearest source of scribal error, but a difficult choice of reading. C has therefore been emended to a Rencevals. C’senfuz for enfuit shows pseudo-rhyme and also North-eastern reduction of diphthong. 3418. Having begun the line with Si, mistakenly including .R. in the first hemistich, the copyist then had to add E in the margin (cf. C 3038). Perhaps qi should be read qe (or q’il?) rather than q’i. 3419. C’s reading makes perfectly good sense and matches V4 2161 (vailant vasal es tu), but O 2045 reads: vaillanz hon, u ies tu?, whilst V7 3430, P 2187, T 1632 and L 1174 agree on: que es tu devenus? 3423. The second hemistich, peculiar to C, is hypermetric: me sui del estor issuz. After the first hemistich, common to all versions, V7 3434 reads: soloie estre tis druz, very close to O 2049: suleie estre tun drut and echoed by P 2191 and T 1635. C’s metre has been corrected by omitting me. 3424. C: hauste for hanste. Cf. footnotes to C 1143, 2057, etc. and Introduction, 80 (xx), Phonology. 3426. C: o lances mes cossuz, which is very corrupt; cossuz is surely consuz/conseüz, past participle, consivre (‘to strike’, ‘to attack’), not cosuz, past participle, coudre (‘to sew’). The meaning is supported by V7 3437: ai trois espiez feruz, O 2052: ot lances sui ferut, P 2194: en .vii. lieus ferus, plus V4 2169 (…feru) and L 1179 (…feruz), but nothing resembling C’s verb is found. The word mes makes no sense as the adverb, ‘more’; it is probably m’es, ‘here I am’.

274

La Chanson de Roland (Critical Text and Notes)

 3430

3435

3440

3445

3450

«Sire Gauters,» ce dist le cons Rollanz, «Bataille as fait, prové estre creanz; Toz dis avez esté vasaus et conbatanz. Mil chevalier m’amenastes vallanz; Prestez les moi, qe besoing m’est mot granz.» Respont Gauters, «Nes vesrez mais noianz. Les ai laisez en cel doloros chanz: De Sarazins nos i trovames tanz, Orgalifes, Asragons et Persanz, Turs et Hermines, d’Esclavons et Jahanz. Une bataille nus ont faite pesanz De Baldestoz toz lor meillor jahanz. N’i a paien devers al qi s’en vantz: .l. mille en i a mort gisanz. Nus i avons perdu trestoz nos Franz; Bien sunt vendu a lor acerins brans. De mon osberc m’ont ronpu li geranz; Plaies ai tant en costez et en flanz, De totes parz m’en salt fors li cler sanz; Tretot li cors me va afeblianz; Senpres mosrai per lo mien escianz.

3431. C: provees recreanz and V7 3443 reads: provees et recreanz. There is no feminine plural noun to justify provees, so this must be read as prové es – which makes V7 hypermetric. There remains the problem of recreanz, ‘ready to confess defeat’, ‘weary’, a rather discouraging comment. In both C and V7, if this is read as prové esre (or preferably estre) creanz, then ‘proved to be worthy of trust’ makes good sense, with as understood from the first hemistich. 3434. C’s qe besoing m’est granz is hypometric; P 2230: li b. en est granz and T 1645: car b. en ai grant imply that m’est stands for m’en est and it is tempting to adopt this reading; but V7 3448 reads: que b. m’est mout g. in a laisse (V7 204D) where C and V7 have a majority of matching lines. 3435. Nes = ne les: ‘You will never see them any more’. 3439. Jahanz (‘giants’) is repeated in C 3441, where V7 3454 also reads jajans; but here V7’s second hemistich is et les Valibrondanz. P, T and L list a variety of Proper Names at this point, but no giants. 3440–3441. In all the other versions (apart from O, which does not have this laisse), the list of names is completed before the battle is mentioned. The repetition of jahanz (3439, 3442) may indicate error. 3446. C: ganz in an hypometric hemistich, but these are geranz: ‘skirts’ or ‘flaps’ of the hauberk (cf. P 2248: pans, L 1225: panz). C has omitted the symbol for er; cf. giron C 1708, geron C 3268.

275

La Chanson de Roland (Critical Text and Notes)

3455

Je sui vostre on, si vos ting a garanz; Ne m’en blasmez se je m’en vois fuianz, Mais car m’aidez a trestoz mon vivanz.» Ce dist Gauter, «E car m’aidez, Rollanz! Ja, se Deu plast, ne vos faudrai nianz.»



3460

Rollanz a duel, si est maltalentis; En la grant presse comence a feris; De cels d’Espaigne i gete mort teus dis. Gauters i est et l’arcivesqe ausis. Dient paien, «Felon homes a cis! Gardez, seignor, qe il n’en voisent vis! Nos ont tant fait, n’en devons prendre pris. Tot soit honi qi les laira garis!

53.b

3452. V7 3474 reads: se j’ai esté fuianz and the past tense makes better sense; but it seems Gautier is excusing his (involuntary) flight from the other battle, although je m’en vois fuianz, ‘I am fleeing away’, reads rather strangely. Duggan emends V7 on the basis of P 2252: se je m’en sui fuianz. 3457. In order to obtain a perfect -is rhyme, the scribe has substituted feris for the syntactically correct ferir; similarly, he writes garis C 3463 and envaïs C 3464, where the infinitives garir and envaïr are the logical reading (see notes below). Additional examples of pseudo-rhyme in respect of the -is rhyme are found in C 1504, 3218, 4950 and 5791. The effacement of praeconsonantal r and its frequent assimilation in the group rs are attested by Pope (1934, 156, § 396); see also the note to C 1504 and Introduction, 27, 83 (xlvii) and 90 (Editorial Policy). 3458. C: tex dis/teus dis, ‘approximately ten’, a very weak statement of number in a genre prone rather to exaggeration – but V7 3523 reads: lor a jeté morz dis. O 2058, P 2259 and L 1236 all have .xx. and V4 2201 has vint (while T 1664 has .c. occis). Given the OPL unanimity, it seems possible that tex is a graphy of dex (i.e. deus dis, 2 tens), due to scribal confusion of t and d – or simply scribal error. Although teus makes poor sense in the context, it has been retained because deus dis is so clumsy. 3459. C: aisis, which reads better in the context as a scribal error for ausis, ‘also’, rather than a graphy of issi, ‘so’. This very weak, pedestrian line is almost certainly corrupt. V7 3524 has: E Gauter quatre e l’archevesque six, following on from the number killed by Roland in 3458; and numbers feature in all the other versions: P 2260: Et Gautiers .xv. et l’archevesque sis, T 1665: G. .vii. et l’a… dis, L 1237: G. .xv. et l’a… .x., O 2059: E G. .vi. e l’a… .v. and V4 2202: E G. septe e l’a… çins. C’s very different first hemistich makes any such emendation impossible. 3463. C: garis, which might possibly be read as the past participle, masc.plur., of garir, i.e. ‘saved (from death)’, and qi les laira garis as meaning ‘who will leave them alive’. However, it is more logical to read garis as a graphy of the infinitive, garir, particularly since this graphy occurs on a number of occasions in C (cf. note to C 1504 and 3457 above). So the meaning is

276

La Chanson de Roland (Critical Text and Notes)

3465

Recreantz est qi nes va envaïs.» Adonc comence et li huis et li cris; De totes parz si les ont asailis.



3470

3475

3480

Li cons Rollant est aduré et fers; Gauters de Lus est mot bon chevalier, Li archivesque provez et asaiez. Per grant vertu les asaillent Turchez; Mil Saracin les asaillent armez, Et es chevals bien sunt .xl. millez. Lancent lor lances et lor trenchant espiez, Quarals et dars et engeins afaitez. A icest mot, per ont oncis Gauter, Turpins de Reins ses escuz especiez, Son eume frait, si est navrez es chiez, Et ses osberc runpuz et desmailliez, Permi le cors feru de .iiii. spiez. Dedesoz lui git a mort son destriez. E! Deus! qel duel, quant l’arcivesqe chiet!

‘who will allow them to escape’. The graphy garis is then either a mis-spelling of garirs (cf. ferirs, C 1509) or a deliberate modification of garir to meet the requirements of rhyme. 3464. C: qi nes va envaïs, but the infinitive, envaïr, is definitely needed: ‘who does not go to attack them’. O 2062 reads: .ki ne.s vait envaïr, and V4 2206: ne li va invairis, the final word either an odd graphy of envers or invaïr with -is added for the rhyme. V7 omits the line; all the other rhymed versions keep the past participle envaïs but change the syntax to make it fit: P 2273: ont les nos envaïs, T 1669: ont Franceys envaïs, L 1245: ont les .iii. envaïs. For further examples of an -is infinitive in C, cf. notes to C 1504 and 3457 above and Pope (1934, 156, § 396). 3467. C: est aduré e ferers, which is hypermetric since the consecutive e of aduré e are difficult to elide; but ferers appears to be a scribal error for fers, fier (‘fierce’), due to repetition of er. V4 2210 reads: est molt ardiç et fer; P 2274: fu moult hardis et fiers; T 1671 ends with: le corps gent et fier; whilst V7 3531 (with a different rhyme) reads: fu mult de fier corage. 3471. The repetition of les asaillent (identical in C 3470) has a certain effectiveness as a literary conceit, but it may be due to scribal error; indeed, it offers no contrast to Et es chevals in C 3472, whereas O 2071 reads: Mil Sarrazins i descendent a piet and V4 2215: …en desendent a pier, while P 2278 has .xx.m.descendent por lor cors dammaigier. The horsemen are specifically mentioned in C 3472. 3479. C: Pemi; V7 3578: Parmi. The reading .iiii. spiez may show Franco-Italian s impura without prosthetic e; cf . qatr’espiez (qatre spiez?) in C 2522, but qatre espiez in C 3957. 3480. C: De duel soz lui, but duel is also found just below in C 3481 and the horse dies because it is struck by four spears, so for it to die of pain (duel) seems a tautology. The other versions agree on desoz in this context: O 2081: Dedesuz lui…; V7 3579: desoz lui.; P 2291: Et desoz lui.;

277

La Chanson de Roland (Critical Text and Notes)



3485

3490 3491a(V7)

3495

Turpins de Reins, quant se senti cheü, De .iiii. espiez permi le cors feru, Isnelement li ber resaili su, Delivrement en bracie son escu, Per grant vigor sor piez est revenu. Rollant esgarde, li cuers l’en est creü; Puis li a dit, «Ne sui mie vancu! Ja bons vasaus n’ert issi recreü.» Traite a Almice, onqes tel brant ne fu, En la grant presse .c. cols en a feru, Dont maint paien fu a terre abatu, Alquant trenchié, auquant permi fendu; Meint en i a qi teste n’ont sor bu. Ce dit la geste et cil qi el canp fu Et Karlle meine quant il fu revenu, C’onqes tel clerc n’ot oï ne veü. Por Karllon fist Deus tante de vertu, A Mont Leon est escrit cest salu; Qi ce ne croit, ne l’a preu entendu.

54.a

T 1684: Et dessoulz lui.; V4 2224: E desot lui.. Turpin’s horse ‘lay fatally wounded’ beneath him, so: git a mort, though gita mort (‘threw down dead’, giter = jeter) is also possible. 3485. C: en bra, as Mortier agrees, although Foerster reads enbraça; an extra syllable is certainly needed, and also a verb. V7 3587 has enbrace, but the normal form of the verb is enbracier, so the reading has been emended to enbracie. The line is not found elsewhere. 3487. The meaning of creü, past participle, croistre, must be ‘increased in size’, ‘swollen’, but hearts tend to be ‘broken’, crevé, so this looks suspect; but V7 3588 reads cresu with the same meaning, and all other versions modify this hemistich considerably. 3491. C: E la grant presse, with scribal omission of n or ~; O 2090 has En, V4 2234 In; V7 and P: En. Although not absolutely necessary to the sense, V7’s additional line (3593) provides a useful antecedent for alquant, C 3492, much as it is found in O and V4’s rather different version of these lines. P and L have a different paraphrase here. Since V7’s line is the 2nd line of V7’s folio 59 recto, i.e. at the very point where C and V7 coincide again (cf. Introduction, 54–56), its insertion is justifiable.

278

La Chanson de Roland (Critical Text and Notes)

 3500

3505

3510

3515

3520

Li cons Rollant fu mot de fer talent; Devant paiens tenoit son brant sanglent; Ne fuïst mie por or ne por argent, Ne li fuirs ne li valsist nïent. Cent .m. paiens et plus, se je ne ment, Ot devant lui, li cors Deu les crevent! De lui oncire ont merveillos talent; Mais tant redotent son tres fer ardiment, Del aprosmer n’osent faire nïent. Lancent li lances et espiez plus de zent – N’i li font mal, car Deus ne lor consent. Se il vesquist auqes plus longement, Aude eüst prisse, la poucelle au cor gent; Ja ne vesra icel mariëment, Car Deu ne plaist, qe meint el firmament. Per als chaï douce France en torment. Quant le sara li rois qi France apent, Li cons Gerart, bella Aude ensement, Qe li dui cons ont pris defeniment, Ja a nul jor n’avront confortement, Deci q’a l’ore q’aront pris vengement De Guenellon, qes traï malement.

54.b

3500. C: tanlent, with inorganic n, where talent (V7 3602) is the obvious reading. This is probably a simple scribal error, -a- > -an- because the next syllable is -en-, but could be intrusive n, common in North and North-Eastern French, and found in Franco-Italian. 3507. C: redote, but the plural verb is essential; cf. V7 3609: redotent. 3512. To modern taste, au cor gent (‘with a noble heart’) reads very aptly, but V7 3614 reads cors and C 6541 has al gent cors honoré and 6839 al cors honoré, so ‘body’ may well be intended here. 3517. bella Aude may show Franco-Italian influence here, but the final a could be a scribal error due to anticipation of the A of Aude or, more credibly, deliberately used to show there is no elision, since belle Aude ensement with elision would be hypometric. The phrase belle Aude occurs 15 times in C, lines 6541, 6561, 6619, 6631, 6796, 6804, 6821, 6839, 6895, 6906, 6975, 6996, 7259, 7264 and 7265 with elision (bell’Aude) and only once (7125) without elision (to obtain correct metre). Here, V7 3619 reads simply: e belle Aude ensement. 3518. C: ot, with omission of n or ~, since the plural (as in V7 3620) is clearly required. 3520. C: Deq’a l’ore, but the hemistich is hypometric. V7 3622 reads: Desci qu’a l’ore, ‘until the hour when…’, ‘until such time as.’. C’s usual graphy is deci q’a… 3521. qes is an abridged form of qi les.

La Chanson de Roland (Critical Text and Notes)

279



3525

3530

3535

3540

3545

Li cons Rollant l’arcivesqe esgarda; Tresüez fu por le caut qe il a E la colors et li sans li mua. C’est por l’angoisse del cor qe il sona. Por tel vertu li ber s’i apoia, Qe la cervele plen doit li sosleva, E la veüe del chief li mesala. Quant d’Oliver li cons li remenbra, Lors ot tel duel, a poi ne forsena. Quant il se pense qe insement mosra, Pur sol la mort un poi se conforta. Mais savoir velt se Karles vegnera. Prist l’olifant, foiblement le sona. Li enperere estut, si escota: «Deus,» dist li rois, «si foiblement nos va! Rollant mes niés hui cest jor nos faudra! J’oi a son cor qe gaires ne vivra.» A haute voiz a sa gent escria, «Montez,» fait il, «qi onqes jor m’ama! Creim c’ocis soit ainz qe nos soions la.» Qi estre i volst, isnelement monta; Sonent li grasle, qenqe per l’ost en a: .l.x. mile qe nus ne se tarja; Bruient li monz et de ça et de la. Paien l’oïrent et Marsiles parla; Dist a ses homes, «Charllon aron nos ja.»

 Dient paien, «L’enperere repaire! De cels de France oi meint olifant braire, 3527. ‘that his brain was lifted a full inch’ – i.e. by the pressure inside his head due to the effort of sustaining the blast. 3533. vegnera is a very strange form, a combination of present subjunctive and future 3, venir, probably a corruption of vendra/veindra. Cf. O 2103: …se Charles i vendrat, V7 3635: …se Karles y vendra. 3541. C: ainz qe soions la, which is hypometric; V7 3643 inserts the subject: ainz que nos s… 3547. C: Charllon aron ja; V7 3649 has the same hypometric hemistich, but O 2114 reads: Karlun avrum nus ja. T 1715 and L 1303 also insert nos, although P 2346 has Karlemaine avrons ja.

280 3550

3552a(V7)

3555

La Chanson de Roland (Critical Text and Notes)

Mainte busine et soner et retraire! Franzois retornent, cele gent de puit aire! Se Charlle vient, nostre ert la perte maire, Perdu arons tote Espegne e Baudaire E Saragoze, qe donai en doiaire A ma moiller, qi tant est de bon aire! Vez la Rollant, bien voi a son viaire Ja n’ert vencuz teus hon nasi de maire. Car l’assallons, pui arons meins a faire! Se plus i somes, je:l vos di sanz contraire, No fu teus perde puis le jugement d’aire.»

55.a

 3560

Li Sarazin furent mot esfréé, Qe por voir sevent qe Franc sunt retorné. De ce sunt il trestot aseüré: Se Karles vient, a lor fin sunt alé.

3550. C and V7 both read soner et retraire, but the phrase is not found in any other version. Here, retraire must mean either ‘draw out their sound again’, ‘prolong their peal’ OR ‘draw back again’ (with the sound diminishing and fading as the breath fails, producing a wail), rather than meaning ‘reproach’. 3552. C: perce; V7 3652 and O 2117: perte; scribal confusion of c and t occurs occasionally in C. 3553. This line makes sense only if an extra line is inserted before it. O 2119 reads: Perdud avuns Espaigne, nostre tere, V4 2262: Perdu avons clere Spagne la belle. V7 3655 has: Perdu arons tote Espegne e Baudaire, the latter being presumably a graphy of Baudas, ‘Bagdad’ (possibly linking it to the verb esbaudir, ‘to give joy, delight or courage’?). In C 3553, the last word is written dotaire, but with both the t and the i marked as i, with a small slash above– apparently a scribal correction to doiaire, ‘dowry’ (as Foerster agrees); V7 3656 reads doaire. 3559. Mortier reads: daire, with a lower case letter as it is written in both C and V7 3662, but no such monosyllabic word exists and this must clearly refer to a well-known, easily recognizable and calamitous ‘judgement’ or ‘punishment’ some considerable time earlier. Foerster reads: Daire, i.e. Darius, the Persian king whose overthrow (with tremendous losses) by Alexander the Great is prophesied in the Book of Daniel (chap.xi). Alternatively, this may allude to the massive defeat of the army of Darius I at the battle of Marathon, with losses put at 200,000 men (cf. Herodotus, Histories, bks.6 and 7). Duggan (2005, 48) relates this to the trial of Daire le Roux recounted in the Roman de Thèbes, but the disgrace and punishment for treason of this Daire has no relevance to the loss suffered by the French at the death of Roland. However, the lower case d of daire may be significant: if one reads this as d’aire and takes into account the ancient and catastrophic punishment of and judgement upon Adam (familiar to mediæval Christians), then aire (substantive, fem., meaning ‘origin’) or graphy of aive (masc., ‘forefather’) takes one back to Original Sin and the loss of the Garden of Eden.

281

La Chanson de Roland (Critical Text and Notes)

3565

3570

3575

3580

.xl. mille s’en vont fuiant al gué; As porz de Sebre isi fu apelé Li rois Marsille o son meillor barné. .lx. mille sunt el canp aresté. Rollant estoit de l’autre part el pré; Sor Velantif a son chief acliné. E l’arcivesqe estoit a son costé; Tant a seinié, li oil li sunt troblé, Car li paien l’orent a mort navré. E! Deus! Quel duel q’il sunt issi grevé! Cel jor chaï douce France en vilté, En grant poverte et en grant orfenté! Li rois Marsille a sa gent escrié: «Franc Saracin, q’avez vos en pensé? Vez la Rollant, a poi q’il n’est maté; Jamais frans hon n’ert por lui honoré. Hui a perdu Karles son avoé. Espeigne est nostre, or en ait il malgré; Car se vient ci Karles o son barné, Toz li plus cointes a son tens afiné.»

55.b

 3585

Li cons Rollant, quant il les vit venir, Tant se fait fier, Cil le puisse garir Qe le profete fist del poison issir! Li cons fu mot de merveillos aïr; Ainz i mosra q’il lor veille fuir.

3567. After this line, C repeats 3565 in error, but it has been barred by the scribe (or a revisor). 3574. C: vite; V7 3678: vilté. 3576. An inconsistency, since Marsille is supposed already to have fled the battlefield – cf. C 3291. 3579. C’s por lui results from typical Franco-Italian confusion of o and a; V7 3683 reads: par lui, thus ‘honoured by him’/’given favours by him’. 3585–3586. C: garnir, with inorganic n inserted through scribal error. Roland looks extremely fierce (or possibly ‘haughty’, his hubris in danger of nemesis), therefore the writer interjects: «May He who released the prophet from the fish protect him!» – a reference to the Old Testament story of Jonah and the whale (cf. Book of Jonah). Tant is used here for stress: ‘so’ or ‘so much’. 3588. ‘He will die there sooner than wish to flee them’. C’s: qi lor is a scribal error for q’il lor, with reduction of two l to one. V7 3693 reads: que il voille, V4 2270: que il voile, T 1727: que il vuille.

282

3590

3595

3600

3605

La Chanson de Roland (Critical Text and Notes)

Sist el cheval c’on cleme Velantir, Broche lo bien, mot li vint a plaisir; Por grant vigor les va toz envaïr, E l’arcivesqe qi ne li volst falir. Ce dist Rollant, «Mot vos aim et desir. Hui est li termes, qe nos devons morir; Ne nus en puet, fors Damedeus, garir, Ne mes li Rois qi tot a a bailir. Se tant vesqisse qe peüse venir En douce France qi tant fait a joïr, De Guenellon qi tel plait m’a bastir Je m’en venjasse, Deus le puist maleïr! Sire arcivesqe, j’oi les terres bondir; De cels de France pöez les cors oïr; Karles chivauche, qi tant aim et desir. Paien s’esmaient, ja les vesrez fuir. Mais nos qe chaut, nos somes au fenir!»



3610

3615

Li cons Rollant fu mot proz et vaillantz, Et orgoilos et vasaus conquiranz; Onqes n’ama coarz ne mesdisanz, Ne chevalier qi trop s’alast vantantz. Sor Velantir va paiens atendantz; Ainz i mosra, q’il se cleim recreanz; Ja de sa boche n’ert tel parole issanz, Ne ja reproche n’en aront si parantz. Turpin de Reins apelle en oianz: «Sire arcivesqe, por Deu, venez avantz!

56.a

3593. mot vos aim e desir, a standard formula used to express grief, is found in identical form in V7 3698: ‘I am very fond of you and grieve over your death’. (Cf. C 3603.) 3599. This is once again poetic licence or pseudo-rhyme: the -ir ending is needed for the rhyme, but the infinitive, bastir, is written when the past participle, basti, is syntactically required. 3606. The ms. graphy vaillant has a ‘tail’ or cedilla added to the t in a lighter ink, so possibly by another hand; it appears to be either the scribe’s own graphy of z or a revisor’s correction of t to z. The same tz graphy is found in C 3609, 3610, 3613, 3615 and scattered throughout the ms. It may represent a Franco-Italian feature or simply be used as an alternative graphy of z. Cf. Introduction, 72 and 83 (xlii).

La Chanson de Roland (Critical Text and Notes)

3620

3625

3630

3635

283

A pié alez, je sui sor l’auferanz, Sor Velantif qi tant par est coranz. Por vostre amor serai ci en estanz. Or me seignez, si:n serai plus dotanz. Puis icest jor ne serons meis vivanz. Cil gart les armes qi fu aparisanz As .xii. apostres qi l’aloient qeranz! Cil Saracin se vont auqes dotanz. Ne vos lairai tant com soiez vivanz. Pongnons a aus, si ferons maintenanz; De Durendart vel esprover le branz, E vos Almice ou a des letres tanz. Qi ja jogleres male chanzon n’en chanz! Karles mes oncles vient a force poignanz, Et li Franzois qi en sunt desiranz De nos vengier ont merveillos talanz; Secorront nos ainz le solel cochanz. Je conois bien paiens a lor scenblanz; Per un petit q’il ne s’en vont fuianz. Trop sunt ici, ja n’i aront garanz Q’il nel conperent ainz conplie sonanz.



3640

Dient paien, «Si mar i fumes nez, Si pesmes jor nos est hui ajornez! Perdu avons qi ja n’ert recovrez: C’est l’augalie qi tant est redotez; Li rois Marsille est si defigurez, Li destre braz li est del cors sevrez; De nos paiens .xx. mil est afolez.

3621–3622. ‘May He who appeared to the 12 apostles who were seeking Him, protect the souls!’ – a reference to Christ after the Resurrection; cf. St Mark xvi.14, St Luke xxiv.33–36, St John xx.19. 3629. The C ms. reading is apparently poiganz, but the graphy g may signify gn (cf. Introduction, 72, Orthography). V7 3734 reads: pungnant. 3643. C: .xx. mil nest afolez. A negative makes nonsense of this statement, but this may be an Italian ne = French en, ‘of them’, used because of the numeral. V7 3748 reads: .xx. millers afolez, T 1739: en y ont d’afolez. Since en is superfluous after De nos paiens, the n has simply been dropped.

284

3645

3650

3655

3660

3665

3670

La Chanson de Roland (Critical Text and Notes)

E! Mahon, sire, com estes oblïez! Secor nos, sire, qe trop somes grevez! Et Trivigan, quant vos esveillerez? Ja somes nos toz a honte livrez! Tot serons mort, se ci somes trovez! Karles chevauche tot les chemins ferez; Avoc lui a .c.m. homes armez De ceaus de France qi sunt vasaus provez. Se ci nos trovent, mieus nos venist asez Qe nos fuisons noié ou esorbez! Plus de mil grailles de ci oïr pöez. Seignor paien,» font il, «car esgardez! Por devers France nos vient si grant nertez, Nos serons ja trestot acovetez. Bataille arons; Karlon est retornez.» Dist Murgalans, li viels chanuz barbez – Il se dist voir, ja mar lo mescrerez – «Vez en la mil qi ja sunt avalez! Lor armes getent environ tel clartez, Tot cist païs en est enlumenez. Li cons Rollant est de male fertez; Ja n’ert vencuz por home qi soit nez; Lancez a lui, e si vos en alez!» Et il si font, darz aguz et penez, Carrals de fer qi bien sunt acerez. L’escu Rollant fu fraiz et estroëz, Et ses osberc ronpuz et desmailiez; Ses chevaus fu desoz lui decoupez, De dis espiez fu feruz es costez, Qe la boële en est cheüe as piez; Li chevaus chiet et li cons est versez.

56.b

3656. C: nuetez (graphy of nuitee with ending to adapted to rhyme?); but V7 3761: nertez, ‘darkness’. 3658. C: .K’n., therefore transcribed as Karllon, a graphy found in subject form in the abbreviations .Kll’on. (for rhyme, C 5562) and .Kll’n. (C 5697, 5753, 5799); it is otherwise the oblique form. 3660. The word se here represents the adverb si = ‘thus’. 3670. Here V7 3775 adds: Mes il n’en fu ne bleciez ne navré; O 2159: Mais enz el cors ne l’ad mie adeset; V4 2302: Mais in son cors ne l’ont mie dané. This is probably a scribal omission in C, but P has no such line and it is not essential.

285

La Chanson de Roland (Critical Text and Notes)

3675

3680

Desor ses piez est Rollant relevez; Son chief covri, mot fu bien dotrinez, Contre paiens corut toz abrevez. Fiert Rubion, un roi de Balaguez: Permi le trenche, bien est l’osberc fausez, Deci q’en terre en est li brans colez. «Monjoie!» escrie li frans clers onorez, «Sainte Marie, Dame, a mi pensez!» Qi tel cop done, bien doit estre honorez Devant Jhesu en ses grant majestez!

57.a

 3685

3690

3695

3700

Or est Rollant enmi la praërie; A pié descent, mot sofraitos d’aïe. Tint Durendart, s’espee o mot se fie; Fiert un paien de mot grant segnorie: D’une terre ert qi est outre Nubie. Rollant le fiert, q’il nel mescosi mie; L’eume li trenche, la broigne a desarcie, Tot l’a fendu; l’arme s’en est partie. Outre dist il, «Li cors Dé mal te die! Ne:l nonceras el reigne d’Aumarie C’a ceaus de France aies fait vilanie!» Puis reclama Jhesu lo fil Marie; A lui comande e son cors et sa vie. Li arcivesqe dalez lui se ralie; Au quan q’il poit, «Monjoië!» lor escrie, Mais mot li ert la parole falie; Li sans li raie desor l’erbe florie, A poi ne chiet, qe del cor ne sent mie; Jamais eglise n’en ert par lui servie.

3691. C: desacie, but V7 3796 reads desartie and the standard verb for use in descriptions of hauberks and cuirasses being ‘shattered’ in battle is desertir/desartir/desarcir. Cf. also C 2116, 2595, 2795, 5446 and 5604 for various graphies of this verb. 3702. C: del cor ne sent mie and V7 3807: del cuer ne sent mie, but reading cor as a scribal error for cors would improve the meaning: ‘he feels nothing at all of his body’, ‘he has no feeling at all in his body’ rather than ‘he feels nothing at all of (with?) his heart’. This laisse is peculiar to C and V7. 3703. C: eclise, either a Latinism (< ecclesia) or Franco-Italian confusion of c and g. V7 3808: iglese.

286

3705

3710

La Chanson de Roland (Critical Text and Notes)

E Karles vient o tot sa baronie; Sonent lor cor, grant fu la taborie, Grant fu la noisse; si loinz en va l’oïe Qe l’on les ot une lieue e demie. Paien l’entendent, mot fu grant l’estormie; Al plus hardi est la chiere rojie; Dist l’uns a l’autre, «Ne nos atarjons mie!» Atant s’en torne la gent qe Deus maudie, Tot droit al Sebre qe l’on passe a navie.

57.b



3715

3720

3725

Paien s’en fuient mot effrééement. Dist l’uns a l’autre, «Vencu nos a Rollent! Oëz les grailles de la hardie gent: Li enperere repaire voirement; Fiz est de mort qi a cop les atent! Meint gentil roi a bailli malement; Jamais Marsille ne nos sera garent; Perdu avons Espaigne voirement Se l’amirauz por nos ne la defent. Li rois Marsille est bailli malement; Jamais por lui n’arons defendiment. Fil a putein, paien, fuions nos en, Car Charllon vient o son esforcement! Je oi les cor et menu et sovent. Quant il vesra le doloros present De cez François qi sunt mort et sanglent,

3708. C: gnãt, to signify grant. V7 has: gnt + ~, the abbreviation usually found in both mss. 3709. C: roie, but the rhyme calls for the ending -ie, so a second i (= j) is needed, giving the reading: rojie, past participle (feminine singular), rojir. V7 3814: rogie. 3713. C and V7 3818: effreement, making the second hemistich hypometric; but this adverb is based on effréé, past participle of effréer, effraer, so an additional e needs to be added. This provides the extra syllable. Cf. P 2459: effraéement, T 1756: effroiement. 3714. C: .R. as usual, but presumably this scribe would write Rollent here for the sake of the rhyme. 3718 & 3722. The same final words, bailli malement, ‘in a bad state’, makes both lines suspect, but V7 3823 and 3827 have exactly the same reading. P, T and L omit these lines, having a shorter laisse. 3724. ‘Sons of whores’ is a strange epithet to find the Saracens using of themselves. Is this a reference, perhaps, to the number of wives allowed under Muslim law?

287

La Chanson de Roland (Critical Text and Notes)

3730

Ce ert merveille se il de duel ne fent! Ja n’arra joie mais jor a son vivent. Si ara pris de nos son vengement, Don li plus cointes arra le cuer dolent; Deci a Meqe ploreront la nostre gent.»

 3735

3740

3745

3750

3755

Paien s’enfuient – Deus lor doinst enconbrier! Enmi Espaigne prisent a repairer; Mot sunt dolent, n’i ot qe corochier; N’i a celui, tant soit hardi ne fier, E n’ait poor de la teste couper. Li cons Rollant nes pot mais encaucer, Car perdu a Valentir, son destrier; O velle o non, remest ou canp arier. E l’arcivesqe, qe Deus avoit tant chier, Totes ses plaies comencent a seiner; Lors li comence la color a muer E la cervelle li prist ad espessier, E tuit li menbre li prisent a froisser; Enmi le canp s’ala ajenoiller. Rollant le vit, se:l corut a aidier; Son eume a or li prist a deslacer, Puis li a trait le blanc osberc leger Et son bliaut li prist a despecier; Dedenz l’en bote une alne et un quarter, E por desus le prist fort a lïer. Permi les flans le corut enbracer; Sor l’erbe vert le fait söef cochier. «Sire,» fait il, «gisez por refroidier, Se ja vos plaies porroient estanchier. Sire arcivesqe,» dist Rollant au vis fier, «Por amor Deu, ja vos vel je proier.

58.a

3735. C: E mi, with omission of n or ~ (as in 3747): ‘into the middle of Spain’. Starting from Roncevaux, this makes better sense than V7 3840 and O 2165: Envers, V4 2317: Invers, ‘towards’. 3747. C: E mi, with omission of n or ~; V7 3852: En mi. 3757. Se is used with the meaning ‘so’, ‘thus’, ‘in this way’.

288 3760

3765

3770

La Chanson de Roland (Critical Text and Notes)

Ja est ocis le cortois Oliver, Li gentius cons qe j’avoie si chier, E tuit li autre, et Gerin et Gerer, Ive et Ivoire, Oton et Berenger, Gauter de Luz, le vaillant chevalier; Il sunt tuit mort; je nes i viel laisser, Je vel aler porqesre et porchacier, Et aporter devant vos et rengier.» Dist l’arcivesqe, «Bien fait a otrïer. Alez a Deo, pensez de l’esploiter. Li chans est nostre, bien nos devons priser. La morz m’aproce, n’i a nul recovrer; En paradis, o sunt li aumosner, Sunt li lit fait o nos devons cocher.»

 3775

3780

3785

Rollanz s’en torne, tot seus sanz conpeignon, Cerche les vals entor et environ; N’encontre home qe li deïst sermon. Meint Sarazin a trové sanz menton, Sanz oil, sanz teste, sanz braz et sanz talon; .lx.mil en gist per lo sablon. Rollant regarde dejoste un Esclavon, Si a trové et Ivoire et Ivon, Trové i a Anseïs et Sanson; Puis a trové Engeler le Gascon, Ensemble o lui Gerart da Rosion. «Deus,» dist Rollant, «par ta redenpcion, Si voirement com tu Saint Lazaron Resuscitas, ce nos dist la leçon,

58.b

3765. C: lasser with i added above the -as- to correct this to laisser. Another i has been omitted, making C’s: je ne:s viel laisser hypometric. V7 3870 reads: je nes i voil l…, O 2179: ne:s i devuns l…, T 1783: les y devons l… 3766. C: porchater, with scribal confusion of t and c, so a graphy of porchacier, ‘to search’ or ‘to hunt’. V7 3871 reads porcerchier, ‘to examine’ or ‘to survey’, in error. 3767. The symbol for et was inserted superscript with insertion mark, apparently by the copyist. 3773. C: chocer, with transposition of ch and c; V7 3878: colchier. 3786. The raising of Lazarus from the dead is recounted in St John’s Gospel, xi.1–44.

La Chanson de Roland (Critical Text and Notes)

3790

3795

289

La Maudalene feïs mot gent perdon – La te servi, s’en ot bon gueredon – Si com je croi ta resurecion, Verasement, sanz fause entencion: En ton servise sunt mort mi conpeignon; Gardez lor armes de la maliëzon Qe Deus dona a Caïn le felon – Bien dut aler a grant perdicion, Enz en enfer, dedenz cele maison, Quant il ocist son frere en traïson.»



3800

3805

Rollanz ot duel, je ne me merveil mie; Trois fois se pasme desor l’erbe florie. Quant il revint, a haute vois escrie, «Damedeus pere! Dame Seinte Marie! Ou est la mort, quant ne me tol la vie? O douce France, com estes deguerpie, Des bons vasaus voidié et apovrie!» Lors li remenbre d’Aude qi fu s’amie; Si la regrete o parole serie: «Amie douce, com dure departie! Mais ne serez de tel home servie. Qeqe je face, Jhesus vos beneïe,

3789. C: Il te servi, s’en ot mot gent pardon. C’s first word Il is not here used as gender-specific, since it cannot refer back to Lazaron, C3786. It relates to Mary Magdalene (C 3788), named in Luke viii.2 and also identified as the woman who poured ointment over Jesus’ head (Mark xiv.3) or His feet (John viii.3–11) and the ‘woman taken in adultery’ (John viii.3–11) – but the feminine pronoun, ele, would make the hemistich hypermetric. V7 3894 reads: La te s…, la being used with demonstrative value in place of celle. Cf. Pope (1934, 322, § 834). The woman’s reward was forgiveness of her sins (Luke vii.48; John viii.11). C 3788 and 3789 both end with mot gent pardon, which raises suspicions of scribal error. There is no such repetition in V7, which reads: si:n ot bon guererdon (‘so had a good reward for this’) in 3894. C has been emended, using normal graphy gueredon as in C 6061; the other versions omit this passage. 3792. mort, found in V7 3897 but omitted by C, improves the sense and corrects the hypometric metre. 3794. A reference to the story of Cain and Abel, found in Genesis iv.11–12 and very popular in mediæval times, as witnessed by morality plays, church carvings, sculptures and stained glass.

290 3810

3815

La Chanson de Roland (Critical Text and Notes)

Qi en enfer ala por Geremie; Fors en gita le profete Ysaïe.» Lors vint as contes, si ne:s mescosi mie; Toz un et un les porta sanz aïe Devant Turpin, qi mot sot de clergie. Si:s arasna enmi la präerie. Turpins en plore, lors n’a talent q’il rie; De Deu les seigne, en qi il mot se fie, Q’il lor otroit la pardurable vie.

59.a

 3820

3825

3830

Li cons Rollant ne se volst pas targier; Delivrement va le canp recerchier, Qerre les autres qe il n’i volst laiser; Sor toz les autres va qerant Oliver. Il trepassa un val et un rochier; Tuit erent mort, ne se porent aidier. «Deus,» dist li cons, «ci ot bon cherpenter; Cist voloit bien la loi Deu esaucer. Sire conpeing,» dist Rollant au vis fier, «Per ci endroit corut vostre destrier.» Il garde avant, desoz un aiglenter: La le trova sor son escu d’or mer. Rollant li vit la face nercïer;

3810–3811. A reference to the words of the Creed: ‘He descended into Hell’. According to the popular mediæval story of ‘The Harrowing of Hell’, Christ – after His death upon the cross on the Friday and before His Resurrection on the Sunday – went down to release the souls of the prophets (here Jeremy and Isaiah) from Hell. This was the Redemption of the Just who had died prior to Jesus’ coming. 3818. C: Qi lor… for Q’il lor…, with scribal omission of one of two consecutive l. Cf. notes to C 1118, 1741, 3588. The line ends with viæ, with dot beneath the a indicating scribal correction to vie. 3823. C: Il trepasse, a hypometric first hemistich; V7 3928 reads: Il trepassa, past definite 3. 3824. It is clear from the context and Roland’s comments that Tuit erent mort relates to the Saracens killed by Oliver, but there is no specific antecedent to identify them. 3831. C: la face vergier, which makes no sense; nercïer (‘turn black’), as in V7 3936, is more apt and the scribal error would arise from confusion of initial n with u. Presumably, if Oliver’s face is turning black, he is already dead. This fact was stated in C 3394 and confirmed by 3397, in which Roland vit mort son amis. However, there is considerable confusion in the next 35 lines.

291

La Chanson de Roland (Critical Text and Notes)

3835

3840

3845

Delivrement li corut enbracer; Per grant angoisse l’en aporta arier A l’arcivesqe, qi jut soz le lorer. Devant les autres le mist jus sor l’erber; Al gentil clerc le fist .iii. fois seigner. Mais sa parole prist a afebloier, Plore des els q’il ne s’en pot targier. Rollant le voit, cui il avoit mot chier. «Sire conpeing, bien devroie enrajer. Vos fustes fius al bon conte Rainer, Qi tant fu proz por ses armes baillier; Por hanste fraindre, ne por escu pecier, Ne por hauberc desronpre et desmaillier, Et por prodome loiaument conseillier, En nulle terre n’ot meillor chevalier.»

59.b



3850

3855

Li cons Rollant comence a plorer Por Oliver, qi tant pooit amer, E por les autres q’il ne pot oblïer. Son conpeignon prist fort a regreter: «Ahi, conpeing, tant feïs a loër, Mot te devroient toz prodome honorer. A si grant duel nos covent desevrer; Ce m’a fait Guenes, qe Deus puist mal doner, Qe nos vendi as paiens d’outremer. Sire conpeing, ou porai je aler? Ja aprés vos ne doi un jor durer;

3837–3840. It appears to be (and surely should be) the mortally wounded Turpin whose voice weakens as he blesses Oliver’s body and who begins to weep; in that case, in C 3839, one would expect the le to refer to Turpin ‘of whom Roland is very fond’ – but, in C 3840, Roland begins a speech addressed to Oliver (as is made plain by the reference to the bon conte Rainer in 3841, since Renier de Genes, the brother of Girart de Vienne, was Oliver’s father), so le, C 3839, is Oliver, ‘whom he (Roland) held very dear’. As Oliver was killed earlier (cf. C 3394 and 3397 and note to 3831), Roland is addressing his dead body. 3840. C: enraier, but the the standard graphy is enrajier/enragier (enraiier); C does on occasion omit one i where ii should occur and consistently writes enrajer, enrager – cf. note to C 3709. 3843. C: hauste; cf. note to C 3424, where the same graphy occurs. 3848. pooit, imperfect 3, poeir. Cf. C 780: poois.

292

3860

3865

La Chanson de Roland (Critical Text and Notes)

Ançois c’on oie la conplie soner, Por la dolor m’estora devier. Sire Oliver,» dist Rollant au vis cler, «Ma arme ne puet sanz la vostre aler. Se vostre boche peüst a moi parler Et vos beaus els veïr et esgarder, Bien vus cuidoie, beas amis, marier.» O veille o non, l’estuet .ii. foiz pasmer. Dist l’arcivesqe, «Tant mar i fustes, ber!»



3870

3875

Li arcivesqe, quant vit pasmé Rollant, Or a tel duel, onqes mais n’ot si grant; Tendi ses meins, se a pres l’olifant. En Rencivals ot une aive corant; D’aler a l’aive estoit mot desirant; Doner en volst au pogneor vaillant Por la grant soif qi l’aloit destrengnant. Tant s’esforça qe il fu en estant; Son petit pas aloit tot en avant; Tant per fu foibles, n’en pot faire niant; Del sanc qi:l laisse, li va li cuer faillant.

60.a

3860–3864. At this point, V7 3965 reads: Sire Roland, dist Oliver and then gives the subsequent speech to Oliver, despite the fact that the latter is already dead (cf. note to C 3837–3840 above). The cause of the confusion is apparently marier, the final word of C 3864, but the question of ‘marrying’ Oliver (to some-one unknown?) only creates greater problems. (The other versions offer no help: P and O have a very short laisse here, omitting this speech, whilst T and L omit the laisse completely.) If Oliver were speaking to Roland, marier = ‘marry (to Aude’) would make sense. Can this mean ‘ally myself to you by marriage (to Aude)’? Or was there some project for Oliver to marry (in a double wedding with Roland)? Alternatively, one could read m’arier as a graphy of m’arire, ‘to smile at me’: thus ‘If your mouth could speak to me and your eyes see and look at me, I would expect you to smile at me’ or ‘… I would think… that you were smiling at me’. 3866. C: aarcivesqe. 3869. pres is a graphy of pris, past participle, prendre (cf. O 2224); this may be due to confusion of vowels, but cf. Pope (1934, 386, § 1055) regarding modification in the radical of a participle; Fouché (1967, 364) draws attention to the form presu (> pris) which may also have influenced the spelling. 3873. C: destrengant, but g frequently appears to be a graphy of gn, especially where the combination -ngn- should occur. Cf. C 3629. V7 3979 reads: destreignant.

La Chanson de Roland (Critical Text and Notes)

3880

3885

293

Quant ot alé un arpent meintenant, Pasmez chaï sor le pré verdoiant; Ne pot aler ne arier ni avant. Quant il revint, garde vers oriant, Deu reclama, qi fu aparissant As trois Maries qi l’aloient querant. Fist sa proiere, si la dist en morant: «Damedé pere, m’arme et mon cors vos rant. Je sent la mort qi me va angosant; Ne vesrai ja la conplie sonant.»



3890

3895

Li arcivesqe se jut el pré sovin; Forment maldist le lignage Apolin, Qi de Franzois Charllon le fil Pepin En Rencevals firent si grant traïn. «Cil gart Karllon qe de l’eigue fist vin Quant fu as noces de Seint Archedeclin! Sire Rollant» ce dist le clerc Turpin, «L’aive est trop loinz, ne je ne voi chemin; Ne puis aler, trop sui prés de ma fin. A Seint Michel et a Seint Aostin Comant je hui le conte palazin Qi la se pasme desoz cel aubespin!»

3878. C: meitenant, with scribal omission of n or ~; V7 3984: maintenant. 3882–3883. C: Deu reclame, which is hypometric; V7 3988: reclama (not reclame, pace Foerster). The reference is to Gospel accounts of the Resurrection, Matthew xxviii.1, 9, Mark xvi.1– 9, John xx.1, 14. 3885. C: mon arme et mon cors a vos rant, an 8-syllable second hemistich; the cæsura is clearly marked in the ms. after pere (a regular 4-syllable first hemistich). V7 3991 reads: m’alme e mon cors vos rant. C’s mon arme is unusual: nowhere else in the text does C use mon as the fem.sing. possessive adj.; vos is the normal indirect object, so C’s a is superfluous. 3890. Charllon is the indirect object indicating possession, since the ‘followers of Apollo’ (ie the Saracens, since mediæval Christians confused the Muslim and Ancient Roman religions) ‘made such a great massacre of Charlemagne’s Frenchmen’. 3893. The name Archedeclin is a corruption of ‘Archetriclinus’ = ‘master of the feast’. This was made into a Proper Name in error in the Middle Ages and identified with the bridegroom at the marriage-feast at Canaa, who thus became ‘Saint Archedeclin’ (cf. New Testament, John ii.1–11: Christ’s first miracle, turning the water into wine – as alluded to in C 3892).

294

La Chanson de Roland (Critical Text and Notes)

 3900

3905

3910

3915

3920

3925

Li cons Rollant, quant vint de pasmeson, Sor piez se drece, ne sente se mal non. Tel angoisse a, ne pot dire raison; La boche ot plene de sanz et de limon; Il ot de soif si grant destrucion, Tote la boche li creva environ. Si ot enflé lo vis et lo menton; Li cuers li bat, le foie et le poumon; Prés est de mort, n’en arra garison. Contre un arpent, devant son conpeignon, La vit gesir un nobele baron: C’est l’arcivesqe, qi Deus mist en son non. Bati sa colpe por voire entencion; Leva son chief, son vis et sa fazon, Contre lo ciel tent ses meins a bandon; Puis prië Deu, qe par anoncion Vint en la Virge, si sofri pasion Por pecheors venir a raënçon, Q’en paradis le mete en sa meison. Lors s’en part l’arme, qe ne dist plus raison, Cil en soit garde qi dist « Consumaton » Quant le pendirent li traïtor felon. Mort est Turpins el servise Charllon. En grant bataille et en bone orison, Contre paiens fu tot tenps canpion. Deus li otroit sainte beneïçon!

60.b

3901. C: drecæ, with dreça (either past definite 3 or Franco-Italian historic present 3) emended to drece by the addition of an e, in order to correct the metre – probably the scribe’s own emendation since the a has been erased and is hardly visible, and the ink and script are the same as in the main text. 3906. C: lo vis e lo vis. 3919. C: qe ne dist pas raison, but the negative is contradictory here, since Turpin clearly was seen as speaking the truth in religious matters; V7 4025 reads: qui ne dist plus raison. 3920. C: qi dist bone raison, almost duplicating the line above. The scribal error may also be due to careless use of a common phrase, not very apt in the context. V7’s reading (V7 4026), consumaton (‘final completion’), has been adopted, since it corresponds to ‘Consummatum est’, said by Jesus as He died (John xx.30) and C 3920–3921 clearly refer to the Crucifixion. These lines are peculiar to C and V7.

295

La Chanson de Roland (Critical Text and Notes)



3930

3935

Quant Rollant vit l’arcivesqe morir, Fors d’Oliver tel dol ne pot sofrir. Et dist un moz qi li vint a plasir: «Beaus oncles Karles, chevauchez a loisir, Qe vos jamais ne me verrez morir, Ne Oliver qi tant fist a cherir. Per grant dolor nos estuet departir. E! Doce France! Tant vus aim et desir! Cil vos meintegne qi fist a son pleisir Lo bien en terre por son pople garir!»



3940

3945 3945a (V7) 3945b (V7)

3950

Ce dist Rollant, «Beaus oncles, car venez! Un petitet estes trop demorez. Si pesmes jor vos est hui ajornez, Jamais en France a joie n’entrerez, Ne par Rollant un seul secors n’avrez, Ni d’Oliver ne serez esgardez, Ne de toz cals qe vos cargié m’avez; Jamais en vie un seul ne raverrez. Mais neqident ben somes alosez: Li rois Marsille s’en est fuiant tornez; De bons ostages nos a laissé assez, Sesainte mille de ses paiens barbez; Toz li plus sains est si desfigurez, Ne:l conistroit nus hon de mere nez. En Rencivals, el canp de dolantez, Plus de .c. mille en i a de versez, Qi trestot gisent per terre ensanglentez.

61.r

3934. C: meitegne; V7 4041: manteigne. This line and C 3935 allude once again to Christ, this time in his role as healer and Redeemer. 3942. cals is a variant graphy of cels (‘those’), found in Northern, North-Eastern and Eastern France. Cf. Pope (1934, 489, § 1320, xvii, and 494, § 1322, ix) but also common in FrancoItalian . 3945. This laisse is not found in OV4PTL, so there is no supplementary evidence, but V7’s extra lines (4053–4054) make better sense of C 3946–3047. 3949. C: .cc.mille, but this makes the hemistich hypermetric; V7 4058 reads: .c.m.

296

La Chanson de Roland (Critical Text and Notes)

Beaus sire rois, de ce vos confortez: Per uns des noz, bien .c. paiens prenez; Ci a bon cange, Deus en soit aörez!»

 3955

3960

3965

3970

Mot fu Rollant corozous et dolenz; A l’arcivesqe corut isnelament, Qi desor l’erbe gisoit mort et sanglent. De qatre espiez est feruz mortelment Et sa boële jut a terre en present; Permi les els la cervele li pent. Rollant s’aprosme al cors mot doucement; Ses mains li croise sor son piz bonement – Plus söef olent qe encens ne piment. Per bon corage a regreter lo prent: «Sire arcivesqe, mar fu vostre jovent! Je puis bien dire, si m’en sivront .v.c., Meldre profete n’en ot batisement. La loi Jhesu as tenue droitement. Beaus siri Deus, tot issi voirement Con vos la loi donastes Moïsent En Synaï, qi est vers orient, Garisez l’arme de pene et de torment! Saint Michel sire, faites en un present A Yhesu Crist, qi maint el firmament.» Rollant regarde vers France doucement;

3954. The grammatically correct subject singular form dolenz rather than dolent (as in V7 4063) is surprising, since the copyist is usually punctillious over aural and visual rhyme. This may be scribal error, for it occurs at the end of the first line of a laisse, with the previous laisse’s rhyme ending in -z. 3965. C: suiront, with the i clearly marked with slash above. This could be future derived from suïr, the variant graphy of sivre, or due to metathesis, but the copyist regularly misplaces the slash in -iu- or -iv-. Cf. Introduction, 71–72 and C 4283, suianz for sivanz. 3968. C: siri, the same graphy being found earlier in C 1688 and 1699; V7 4077: sire. 3968–3970 This is a reference to God’s giving of the Ten Commandments to Moses, as related in the Old Testament, Exodus xx.1–17. 3969. C reads: Con (in full), meaning ‘when’, ‘as’. V7 4078 has: Com. 3972. sire was added superscript, in fine lettering but the same script, having first been omitted.

297

La Chanson de Roland (Critical Text and Notes)

3975

De l’aler la ot merveillos talent, Mais il ne puet, qe la mors le sosprent.

61.b



3980

3985

3990

3995

Li cons Rollant estoit mot entrepris En Rencivals, entre ses enemis; .lx. mille et plus, ce m’est avis, En ot lés lui detrenciez et ocis. A Damedeus, qi fist perdon Longis, Comande l’arme d’Oliver le marqis Et l’arcivesqe et sis autres amis. Lors se dreça, son olifant a pris Et Durendart, don meint Turc ot oncis. Vers douce France avoit torné son vis; A Deu comande Karllon de Seint Donis. Fors de la presse, com il poït, s’est mis; Le val trepasse, se monte en un laris; Desus cel tertre avoit deus pins floris E deus perrons qi sunt de marbre bis. La vint Rollant, mais il fu si aquis Qe la cervelle li ist per les sorcis; Pasmez chaï, Jhesu li soit aidis, Qe fors d’enfer a jeté ses amis!

 Grant sunt li pin, beaus sunt et ben foillu; Desoz se pasme Rollant, qi tant mar fu; Sor l’erbe vert jut a terre estendu. Prés de lui ot un Sarazin crenu;

3980. This must be read as: lés lui, ‘near him’, ‘in his vicinity’, not les lui, ‘his men’. Longis was the Roman centurion who pierced Christ’s side with his lance. Cf. John xix.34 and note to C 7139. 3988. The 2-syllable past definite 3, pooir: poït is needed here for a 6-syllable hemistich. 3993. C: cuelle, with omission of the abbreviation 7 to indicate er; V7 4102 reads: cervelle. 3995. Another reference to the ‘Harrowing of Hell’. Cf. note to C 3810.

298 4000

4005

4010

4015

La Chanson de Roland (Critical Text and Notes)

Entre les autres s’ert tapi et repu, Qe il n’i fu ne oïz ne veü; Des sanc des autres ensanglentez se fu; Ne voloit pas q’il fust aperceü. Toz estoit sains, mot per i ot geü; Quant il parloit, n’oï noisse ne hu. Son chief dreça, Rollant a coneü; La ou le voit, sore li est coru; Por son orguel a tel plait esmeü, Don li seront andos li oil tolu; Por le nasel lo pris de l’eume agu; Si li escrie, «Rollant, je t’ai vancu! Plus ai conquis qe n’avoie perdu! Ta bone espée rendras per ton treü A Baligan, onqes tes ber ne fu.» Per le grenon l’a pris lo mescreü; Vers lui le sache; Rollant est revenu De pasmison, ou ot tant esteü.

62.a



4020

Rollant se jut desoz le pin pasmé. Li Sarazins, qi tant estoit desvé, Le tint au cercle de son eume gemé. Per tel vertu a le grenon tiré, Li sans li raie jusc’al neu del baldré.

4000. The ms. reading repu is a graphy of repus, the past participle of repondre, ‘to hide’. This verb is a compound of pondre, ‘to place’, for which Pope (1934, 406–411, § 1063) shows a range of past participle forms: post, pos, pus, pons (ponu and pondu), and Fouché (1967, 381) cites repus as the past participle. The final s has been dropped for the sake of the rhyme. (V7 appears not to have recognized this word, for V7 4109 reads: ranpu, which is not known, unless it is a very strange form of rampé, ‘crawled’.) 4007. C: La ou il le voit, which is hypermetric. V7 4116 reads: La ou le voit, following common practice in omitting the subject pronoun, so this has been followed. 4018. se jut, past definite, gesir, ‘to lie’ (as found in C 4025) must certainly be the reading here, but both C and V7 4127 have se uit with the i clearly marked as the second letter. This is probably simply scribal carelessness, since neither se vit (past definite 3, se veoir) nor sevit (past def.3, sivre, makes sense in this context; but the question remains whether seuit might not be an odd graphy of the past definite 3, seoir (normally sesit/seït), i.e. ‘he sat’. 4022. C: basdre. Mortier and Foerster read bardre, but the graphy read as the first r is not found as an r anywhere else in the ms. (Foerster describes it as having the form of an s lunga, the hook level with the top of the d, the tail longer than in the normal s.) V7 4131 reads:

299

La Chanson de Roland (Critical Text and Notes)

4025

4030

4035

4040

4045

4050

A voiz escrie, «Rollant, je t’ai maté! En poi de terme t’avrai le chief copé!» Li cons Rollant se jut adenz el pré; Un poi se fu de son mal tresalé Et auqes ot son cors resvigoré; Ovri les els, s’a celui esgardé Qi si li ot son grenon dessiré. Sus est sailli per mult ruste fierté; Dist au paien, «Trop es desmesuré! Cil cui tu sers t’a mot bien encanté. N’es pas des noz, or t’ai bien avisé. Quant me saisis, mot as fait grant fierté; Ainz qe je muire, t’arai si coréé Qe ja mesage n’en ert por toi porté A Baligan, lo qivert desfaé.» Prist l’olifant, grant cop l’en a doné. L’eume li froise et les os sunt quassé, Andui li oil li sunt del cef volé E la cervelle, Deu en a mercié – Ainc par Mahon ne pot estre tensé. L’arme de lui enportent vif maufé; A cent diables soit le cors comandé. Li cons Rollant i a mot meschevé; Se il vesqist, donc mot fust adolé: Son olifant a frait et eströé, Qe li cristaus a or en est volé. Li quns Rollant a son chief encliné, Al Roi de gloire merci li a crié Por ses pechiez, q’il ne soit enconbré. De Durendart fu forment effréé, Qe Sarazin n’en aient poësté.

62.b

baldré, meaning ‘belt’ or ‘cross-belt’ – and this reading has been adopted, as l and s lunga could easily be confused. 4035. C reads coréé (V7 4145 coreé in Duggan) and the forms coréez (C 513) and coréent (C 1369) are found; the verb coréer is a graphy of conréer, here meaning ‘to deal with’. Where con- was intended, the C scribe normally used the symbol 9, not cõ. The word si was added in superscript. 4039. C: e li oil sunt quassé, but ‘eyes’ are not likely to be ‘broken’ or ‘crushed’; and li oil is found in the next line, in a more appropriate context; a scribal error for V7 4149’s reading: li os, ‘his bones’.

300

La Chanson de Roland (Critical Text and Notes)

 4055

4060

4065

4070

Rollanz senti qe la mort mot l’argue; Sor piez se lieve, qenq’il pot s’evertue; De son visage a la color perdue. Prist Durendart, s’espee tote nue: Devant lui a une piere veüe; Ne la meüsent li buef d’une chesrue; Grans cols i fert, per grant dolor s’argue; Crost li acers qe point ne se remue, De Durendal a la piere fendue. Rollant a dit, «Espee coneüe, Tante bataille en ai faite et vencue! As Saracins vus estes chier vendue; Au roi Marsille tel colée ai rendue, Dont il a hui corné la recreüe. Durendal clere, belle trenchant ague, Deus! tante terre en ai je conbatue, Qe Karles tient a la barbe chenue. Ne vos ait hon qi por autre remue!

4058. C: une pree, but this is clearly a ‘stone’ or ‘rock’, too massive and heavy to be moved (C 4059: meüsent, imperfect subjunctive 6, moveir): une piere (the graphy being confirmed in C 4062, 4077 and 4111). V7 4168 and V4 2456 read: une piere, O 2300: une perre. 4061. C: Crost ni acers, but crost is a verb and ni does not make sense. V7 4171 reads: croist li acers and O 2302: Cruist li acers: ‘the steel grated…’. 4064. C: faite, the i added in lighter ink but the same script, with finer pen-stroke. 4067. C: corne/corné deceüe, a hypometric second hemistich. Reading corne (sf) meaning ‘force’ or ‘power’ would justify the feminine ending of deceüe, taken as a graphy of decheüe, past participle, decheoir, ‘fallen’, ‘brought down’; la corne is then syntactically necessary, supplying the missing syllable. However, V7’s reading: corné la recreüe (‘sounded the retreat’), V7 4177, makes better sense and corrects the metre (although it may be a rationalisation of a corrupt or misunderstood original). This line is found only in C and V7. 4071. C: qi port autre mue, which is hypometric and makes no sense. V7 4181 reads: qi por autre vos mue, correcting the metre but giving an unsatisfactory meaning, since there is no question of exchanging Durendal for another sword. O 2309 has: ki pur altre fuiet and V4 2464: per altres non fue, ‘who runs away because of anyone’: i.e. Roland does not wish Durendal to become the property of a coward. This is much more appropriate. Clearly port is a scribal error for por, but there remains a missing syllable. There is little justification syntactically for autre ne mue (with pleonastic ne based on V4’s negative). It is tempting to emend radically to por autre s’en fue on the basis of V4, but reading autre remue (remuer, ‘to retreat’) as reduced to autre mue through omission of one of the two consecutive re (or of the abbreviation 7) gives a reading matching O and V4 in sense.

301

La Chanson de Roland (Critical Text and Notes)

A mon vivant ne me serez tolue; Si bons vasaus toz tens vos a eüe, Jamais n’ert teus en France l’asolue.»

 4075

4080

4085

4090

Li cons Rollant tint s’espee forbie; De briser la ot merveilose envie; Fiert en la piere, qi ert grant et fornie: Crost li acers, amont est resortie. Quant voit li cons q’il ne:l malmetra mie, Fortment le pleint, je ne m’en merveil mie: «E! Durendart! Bone spee sarcie, Per tantes terres avez esté m’amie; En l’ori’ pont a dé Seinte Sofie, La dent Seint Piere, del sanc Saint Dïonie, Il n’est pas droiz qe paiens vos balt mie; De Cristïens devez estre servie. Mainte bataille en a esté fornie Et meinte terre dont France a segnorie, Qe Karles tient a la barbe florie; Il n’est pas droit, por la fé qe vos die, Hon qi te port en face coardie. Deus, ne sofrez qe France en soit honie!»

63.a

 Quant voit Rollant qe si est deceü, En Rencevaus a paié grief treü: 4078. C: Crost ni acers; V7 4188: Croist li a…; this is a repetition of the scribal error in C 4061. 4079. C: malmet + a form of inverted w superscript over the t, usually the symbol for r or re, but here it must stand for ra to give the future 3 so as to meet syntactical requirements. V7 4189: malmetra. 4084. C: sã Doinie with metathesis of o and i. Dionie, as in V7 4194, is a standard graphy for ‘Denis’, needed here to obtain a 6-syllable hemistich. 4085. C: vos ralt mie, which does not make sense. V7 4195 reads: vos balt mie, balt being present subjunctive 3, baller/bailler/baillier, ‘to touch’, ‘to lay hands on’, which is apt. Cf. also O 2349: te baillisent (< baillir, ‘to control’, possibly ‘to wield’). 4090. C: N’est pas droit; por la fé qe vos die, a line peculiar to C. Il (on the model of C 4085) is needed to correct the hypometric first hemistich; then ‘in faith as I tell you’, i.e. ‘I swear to that’.

302 4095

4100

4105

La Chanson de Roland (Critical Text and Notes)

Li .xii. per i sunt mort e vancu; Li rois de France en ert mot irascu; En orfenté en est son cors cheü. Rollant estoit en son un pui agu, A ses deus mains en ot son piz batu: «Deus, moie cope per la toie vertu Des grant pechez, dont qit estre perdu Cist las pechere, dés l’ore qe nez fu Tresc’a cest jor qe ci est conseü!» Son destre gant a contre mont tendu; Li cels ovri, les angles i sunt venu, Qi metront s’arme en joie et en salu.



4110

4115

4120

Quant voit Rollant qe la morz l’entreprent – Car per les els li cervals li descent, Per les orelles n’ot il mais ne entent – Tint Durendart al poin d’or et d’argent, Fiert en la piere, bote pié et estent; Ne la pot fraindre, qe Deus ne li consent. Quant voit Rollant ne l’i forfait nïent, Sor destre garde contre demi arpent; Si a coisi un fontenil rovent, Plein de venin et plein d’intoschement. Deus ne fist home des lo tens Moïsent, S’il en bevoit, ne fust mort esrament; Mot ert hardos et perfont et pulent. La vint Rollant coroceus e dolent; Entor lui garde, li scoisi nule gent; Durendal prist, par son fier hardiment

63.b

4109. ‘Through his ears he no longer hears or takes in…’. 4112. C: faindre and V7 4221: feindre, both omitting an r vital in the context; cf. O 2314: freindre. 4113. C: forfaint, with inorganic n; this is present indicative 3, forfaire, here ‘to harm’. 4115. The ‘bright red’ hue of the water probably implies a chalybeate spring, impregnated with iron. 4119. C reads: hardos and V7 4228 ardos. Looking at the qualities of the water likely to be specified here, the most likely source reading must be ardans or another adjective based on the verb ardeir; therefore ‘corrosive’, ‘producing a burning sensation in the mouth’. Duggan suggests ‘scalding’.

La Chanson de Roland (Critical Text and Notes)

4125

4130

4135

303

Dedenz la gete, car la mort le sosprent. La gent del reigne en trai vos a garent: Cil nus ont dit, se l’estoire ne ment, Q’encor i est, por voir certanement, Et essera deci au feniment. La morz l’argüe et poignot mot sovent; Mot estoit prés de son trepassement; Sor l’erbe verde s’est cochez plorantment, Son vis torna vers Espegne la grant – Por ce la fait, qe il velt voirement Qe Karles die a trestote sa gent: «Li gentil cons est morz conquiranzment» – Cleime sa cope et menu et sovent: «Damedeo pere, pater omnipotent, Sainte Marie, m’arme et mon cors vos rent; En son cest mont vos en faz un present.»

 4140

Desor le pui se jut li cons Rollant; Son vis torna vers Espeigne la grant.

4127. C: esera; V7 4236: essera. This is the original future form of estre/essere, before the initial e was dropped; cf. Pope (1934, 368, § 977). Alternatively, in C a corruption of i sera. 4128. C: poignot; V7 4237: pungnoit, confirming that this is imperfect 3; -ot is a variant graphy, more commonly found in the twelfth century and applied to both the first and third conjugation, in Central and Northern as well as Western France; cf. Pope (1934, 345s., §§ 914–916). The verb poindre (‘to stab’) may be more apt than poigner (‘to attack’). Cf. notes to C 3629 and 3873: g as a graphy of ng. 4133. O, V4 and V7 all read …e trestote…, stressing that Charlemagne and all his people should say this. 4134. C: conquirazment, lacking one n or ~. 4138. C: En sonc, with no apparent reason for the final c unless it is due to anticipation of the c of cest, or unless -nc is a scribal error for -m, since som is an alternative graphy. V7 4248 reads: En son and en son un pui agu is found in C 4098. 4139. C: Desoz le pui, ‘beneath the hill’; but V7 4249 reads: Desor le pui, ‘upon the hill’, echoing the position given in C 4098 and 4138. The other versions agree on ‘under a pinetree’: desuz un pin in O 2375, soz un arbre foillu in P 2694 and soulz .i. pin fielluz in T 2023. Reference back to C 3990 (Desus cel tertre avoit deus pins.) and to the inaccurate marking of i in the -ui-/-iu-/-in graphies shows how confusion of pin and pui arose. Thus, C 3996–3997, Grant sunt li pin… Desoz se pasme .R., and C 4018, .R.. se jut desoz un pin, but C 4098, .R. estoit en son un pui aigu. The decision here must be arbitrary – either to emend desoz to desor and follow V7, or to emend pui to the very similar pin and follow O, P and T – and the first alternative has been chosen.

304

4145

4150

4155

4160

La Chanson de Roland (Critical Text and Notes)

De meintes coses se vait lors remenbrant; De Durendart dont terres conquis tant, De douce France et d’Aude la vaillant, Niece Girart de Viene la grant, De Charllo Meine qi est as porz passant, Qi le nosri söef por bon talant, E d’Oliver qe il laissa gisant Lés l’arcivesqe, desoz le pin sanglant. Lors se pasma, li cuers li vait faillant; Quant il revint, lors fist un duel pesant, Qe la cervele li est del chief issant. Or set il bien ne puet aler avant Qe il ne muire orendroit maintenant; Bati sa cope, mot fu ben repentant, De ses pechiez fu voir regeïssant. Lors reclama le Glorios puissant, Qi de la Virgine nasqui en Balliant: «Si voirement, come je sui creant, Qe convertis Seint Feron lo tirant, Saint Pol li chauf qi de mal fasoit tant,

64.a

4141. C: …se vont…, the present 6 form perhaps because of the plural meintes coses immediately preceding the verb, but the present indicative 3 is clearly needed. V7 4251: s’en veit. 4145. C: poz, but V7 4255 reads: porz. 4157. C: Qi de la Virgine, the 5-syllable first hemistich probably arising from scribal error, Virgine for Virge (a variant graphy). V7 4267 reads: Qui de la Virge… but C’s error may reflect Italian pronunciation of the final e or possibly an ecclesiastical Latin usage, since the almost identical C 5972 begins: E de la Virgine. C uses the graphy Virge in lines 1585, 1599, 3915 and 5067 without metrical problems. The reference in this and the preceding line is of course to the birth of Christ in Bethlehem. 4159. C: covertis, with ~ omitted, perhaps influenced by covenir/convenir. St Feron (usual spelling St Faron) was a high official at the court of Dagobert I who took religious orders later in life and founded an abbey at Meaux. His existence is attested by his signature on various seventh-century documents, but there is nothing in his life to justify the appellation lo tirant, ‘the obstinate’ or ‘the tyrant’. In fact, the (largely spurious) biography written by Bishop Hildegaire of Meaux in 869 (vol. II of ‘Acta sanctorum ordinis sancti Benedicti’, published 1669) has a high moral tone. The epithet may derive from confusion of Faron with the Egyptian title ‘Pharaon’, familiar to the mediæval world through the story of Joseph (Genesis xxxix seq.) and the account of Moses’ life, the plagues of Egypt and the escape of the Jews, given in Exodus. Cf. Bédier (1921, vol. 4, 290). The story of the Ten Plagues of Egypt witnesses to Pharaoh’s obstinacy in refusing to allow the Jews to leave. 4160. C reads St Policarf, but Polycarp was an early Bishop of Smyrna, martyred there in 155 or 156 AD. He knew St John, whose disciple he was. At his trial, Polycarp himself said that he had been a Christian for over 80 years, so the second hemistich is clearly inapt: there is no

305

La Chanson de Roland (Critical Text and Notes)

4165

4170

4175

De la fornase ou furent li enfant Tuit sain et sauf s’en issirent joiant, Et a Jonas qi aloit preïchant, Qe la balene transgloti en esrant, Al port d’Orcaise, desoz la garillant, Soz Niniven ou erent mescreant, La le geta, une aube aparant; Vostre miracle furent aparissant: Saint Lazaron, qi ere vostre servant, De mort a vie lo feïstes parllant: Damedeu pere, tot issi voiremant, Come ge:l croi et sai a esciant, Garisez m’arme per le vostre comant!» Lors s’aclina sor son escu vaillant; Il joint ses meins, l’arme s’en va cantant; Angle enpené l’enporterent atant, En paradis le poserent riant Devant Yhesu, ou a de joies tant, Nel vos pot dire nus clerc tant fust lisant.

64.b

record of his having done anything evil. V7 4270 reads Seint Pol li chauf, a reference to the Apostle Paul who, as the Pharisee Saul, persecuted the early Christians (cf. Acts of the Apostles, viii–ix, First Corinthians, xv and Galatians, i.). This is another instance of the C copyist transcribing thoughtlessly, without considering meaning. According to tradition, St Paul was bald. 4161. C: la fornas, but fornais/fornaz is masculine. However, Godefroy also lists a feminine noun meaning ‘furnace’: forneise/fornase, and V7 4271 reads: la fornasse. The reference is to the Old Testament story of Shadrach, Meshach and Abed-nego, who emerged unscathed from the fiery furnace into which they had been thrown when they refused to worship Nebuchadnezzar as a god (Daniel iii). The story is again referred to in C 5117 and 5971, and on all three occasions the three Jewish men are referred to as ‘children’. The old translations of the Bible referred to ‘children’, whereas the modern translations of the Book of Daniel make it plain that these were ‘young men’ of good education and of royal or noble rank. 4163–4167. Once again, a reference to the story of Jonah and the whale, as told in the Old Testament Book of Jonah, but with extra details added, including possibly items of local colour acquired via the Crusades. Orcaise may well correspond to the modern Ahwâz, near Basrah (cf. Proper Names). 4166. C: errent, present 6, errer. C seems to have become thoroughly confused between esrer, ‘to wander’, line 4164, and estre, ‘to be’, in this line. V7 4277 has: erent, ‘were’, a much better reading . 4169–4170. Another reference to Lazarus’ being raised from the dead; cf. note to C 3786. 4170. ‘Brought him from death to life, by a word, by speaking’.

306

La Chanson de Roland (Critical Text and Notes)

 4180

4185

4190

4195

4200

Morz est Rollant li franc cons, li proissez; Deus en ait l’arme per les soe pitez! Li enperere ne s’est pas atargiez; En Rencivals est entrez tot irez; O lui .c. mille de chevaliers proisiez. «A! Deus!» dist Karles, «Qel duel et quel pechiez! Ci voi mes homes ocis et detrenchiez, Qe mantenoient mes honors et mes fiez. Ahi! Fel Guene! Traïtre! Renoiez! Com sui remés per vos afebliez! Cest jor est mot li miens pris abassez! Jamais nul jor n’ere joios ne liez, Des qe je soie de vostre cors vengiez Et ocis aie cez qiverz renoiez; Par als sui mot grevez et daumajez; Bien en doi estre desvez et enrajez.» Plore li rois qi mot fu corocez; Sa barbe tire dont li poil est treciez; Tot ses dras a ronpuz et depeciez. Plorent François por le roi q’est irez, N’i a celui qi puise estre sor piez.

 Li enperere demene grant dolor; N’est pas mervelle s’il a al cor iror; 4180. This line begins with an ornate capital M for Mors and there is further ornamentation in the margin (see Introduction, 5–6), marking the beginning of Section B in C’s text. 4185. C reads: qel…quel…, both words written in full, with two different spellings in one line. 4188. C: Ahi fel Guene, e traire renoiez; V7 4299: Ahi fels Guenes, traïtes renoiez, with ï necessary for the metre. Thus both C and V7 imply the word traïtre/traïtres (subject form for the addressee, ‘traitor’) but neither uses an exact graphy of this word. The graphy traïte for traïtre is found in C 8049, but poses problems of metre; here emendation of traire to traïtre makes the hemistich hypermetric unless the preceding (and superfluous) e is omitted. The metrical value of the -aï- diphthong (especially in relation to the oblique form, traïtor) is discussed in the Introduction, 92–93. 4198. C: Tot ses dras . a ronpuz…, with cæsura inserted after 3 syllables, leaving a 7-syllable second hemistich – but this avoids division of the compound verb. 4201. This laisse, like the preceding one, is found in almost identical form in V7 (laisse 239). It is at this point that a marked divergence from the OV4 assonanced version begins: whereas,

307

La Chanson de Roland (Critical Text and Notes)

4205

4210

4215

4220

N’i a baron ne soit en grant tristor: Mort sunt li duc, li prince et li contor Por Gainellon le cuvert traïtor, Cui Deus otroit et onte et desenor, Qi les vendi a la gent paienor. Plorent François, cui Deus otroit onor, Por Karllemene, lor naturel seignor, Qi son nevou regrete por amor: «Beaus niés,» fait il, «mis m’avez en tristor Et tote France est cheüe en dolor. Ja n’avrai mais bien ne joie a nul jor, Quant voi qe mort sunt li bon fereor Qi les paiens metoient en freor. Deus! Qel daumage m’avint en un sol jor Quant j’ai perdu mon ami le meillor, Qi ainc n’ama ne repos ne sejor Ne nus paiens qe m’ont mis en error. Damiselle Aude, vos en arez iror; Icez noveles vos torneront a plor: Mort est Rollant, n’i a mais nul retor, Et li baron don France ert a plor.»

65.a

 4225

«Beaus niés Rollant,» fait li rois Karllemene, «Se un petit vos fust la mort lontene E Deus eüst salvee la nostre conpeigne,

in the latter, Charlemagne passes swiftly over the battle-ground in pursuit of the Saracens and immediately returns before fighting Baligan, in the CV7 version Charles returns to Roncevaux only after his battle against Baligan (which takes place after some delay). As a result, CV7 makes a longer pause on the battle-field at Roncevaux at this juncture and the speeches of grief are divided between the two occasions. 4205. C: Por Gainellon et le…, but the superfluous et (in abridged form), not found in V7 4316, makes the second hemistich hypermetric. 4209. C: .Kll’n.; expansion to Karllemene (the graphy in C 4512) is needed for a 4-syllable first hemistich. Cf. lines 4279, 5568 and, in the passage of alexandrines, C 7854, 7962, 7981, 8000, 8019, 8035, 8132. Where this is necessary to obtain a 6-syllable second hemistich, cf. C 4371, 4864, 6756. Similarly, .Kll’n. C 5356 and .Kll’e. 8146 have to be read as Karllemene. Cf. Introduction, 92. 4213. C reads simply joie a nul jor, a 4-syllable second hemistich; but V7 4324 has: bien ne joie…

308

4230

4235

La Chanson de Roland (Critical Text and Notes)

Ainz quatre mois fuisez vos rois d’Espeigne, Je vos feïsse seignor e cheveteigne. Per tot le mont dotast on vostre enseigne. Morz vos a pris, qi grement me meheigne. He! Tote joie, com vus m’estes estreigne! Je ne pris mais lo siegle une casteigne; Tote proëce est hui veve et breheigne! Seignor baron d’Anjou et de Breteigne, Et cil de France et tuit cil d’Alemeigne, Alez alors fere vostre gaaigne: Morz est Rollant et li autre catheigne; Mes certes Guenes mar vit iceste ovregne!»

 4240

4245

4250

4255

4260

Li enpereres fait ses grailles soner; Per grant iror a fet sa gent armer. Maint bon osberc veïssez endoser, Per grant dolor sor les chevaus monter; Plus de mil grasles i oïst on soner. Paien les oient, qi pensent del esrer: A plen s’en fuient por lor vies sauvier; Franc les enchaucent qi ne:s puent amer. E Karllemene prist fort a speroner, Aprés paiens les fait toz aroter. Lores les trovent, pristrent soi a hastier, Car li solauz ert mot prés d’esconser. Quant Karles vit q’il devoit avesprer, Mot fu dolanz, si comence a plorer; Coche s’a terre, merci prist a crier Au Roi de Gloire, qe il pot tant amer: «Deus,» ce dist Charlle, «un don vos vel rover, Qe lo solel facez por moi ester, La nuit targier e le jor demorer.» Ez li un angle qi selt a lui parler, Qi li a dit, «Rois, ne te dementer; En grant duel faire ne puez ren conquister.

65.b

4230. C: me me meheigne, with accidental repetition of me; V7 4341 reads: me mehegne. 4248. C: le fait toz, but the plural form les is clearly required. V7 4359 reads: les.

309

La Chanson de Roland (Critical Text and Notes)

4265

4268a(V7) 4268b(V7) 4270

4275

Li Rois de Gloire a oï ton penser: Chevauchez fort, n’i avez qe ester. Deus le vos mande, n’avez qe demorer. Clarté et joie vos fera, preu, doner. Vengez vos, sire, des paiens d’outremer Qi ont ocis la flor de vo barner: Ocis i sunt trestot li .xii. per.» Lors monta Karlle come gentil et ber, Jhesui de Gloire comence a mercier. Lors se comencent François a desevrer, Desoz lor piez font la terre tranbler, Per les grant conbes la poudrere lever. Ainz qatre liues, si com j’oï conter, Les ont ateinz a une eve paser, Ou il ne puent chalan ne nef trover. Je ne voi pas com puissent eschaper, S’il ne la boivent ou ne sevent noer; Mais .xv. lieues duroit a trepasser. Cil les confonde qi se laissa pener Enz en la croiz por Cristïens sauvier!

66.a

 4280

4285

Por Karllemene, qi tant fu conquiranz, Fist Deus miracles totes aparissanz En Rencevals, ainz q’il en fust issanz, Car li solauz fu longemenz estanz. Paien s’enfuient et Franc les vont sivanz; Fort les enchaucent, ne se vont pas tarjanz, En Val Tenebre les vindrent ateignantz;

4268. C: .Kll’e., transcribed as Karlle. Before C 4269, an antecedent is needed to identify the ‘they’ referred to in lor piez. V7 reveals the omission by C of 2 lines equivalent to V7 4380– 4381, due to the repetition of Lors in the CV7 model, just as it is the first word of V7 4379 and 4381. 4275. Both C and V7 4388 read bovient, with the i clearly marked, instead of boivent. 4276. duroit, found also in V7 4389 in this form, must be read as impersonal: ‘there remained’ to explain the singular. 4279. C: .Kll’n., for Karllemene, as written in full in C 4512. Cf. note to C 4209. 4283. C: suianz, with the slash to indicate i misplaced, a feature of this copyist’s script. Cf. C 3965.

310

4290

4292a(T)

4295

La Chanson de Roland (Critical Text and Notes)

Vers Saragoze les enmenent feranz. As cols primiers les vont Franc ocianz. Tolent lor voies et lor chemin plus grantz; L’eve de Sebre lor estoit dedevanz: Mot ert perfonde, mervelleose et coranz; Il n’i avoit ni barge ni challanz. Paien reclament Mahon et Triviganz, Puis saillent ens, mez nul n’y ont garant. Li haubergié furent li plus pesanz: Tot droit al fonz alerent li auquanz; Li autre vont en contreval flotanz; Plus de vint mil en chieent meintenanz; Ainz par Mahon n’i furent secoranz. Franzois escrient, «Mar veïstes Rollanz!»

 4300

4305

Quant voit li rois q’il les a fait noier En la grant eve o n’a point de gravier, Plus de mil Turs – Deus lor doinst engonbrier! – Qi mot estoient et orgoillos et fier, D’entrer en l’eve n’orent nul desirer; Contre François se velent redricer. Ja lor feront un estor mot plenier, Ainz q’as poissons se laissent si mangier. Ne il ne puent ne fuir ne mucer, Car enclos erent et devant et derier. La veïst l’on tant escu enbracer,

66.v

4292. Some reference to the Saracens’ entering the water is needed here, to make sense of lines 4293 – 4298, but V7 is identical to C at this point. T 2118 provides the extra line, which is vouched for by O 2469 and V4 2661; an identical first hemistich is also found in P 2854 and L 1676. 4296. Both C and V7 4409 read: chiet, probably due to omission of ~, but chieent, present 6, cheoir is needed to match the other verbs in this sequence and for the correct metre. 4298. Strictly, Rollant, oblique singular, is needed, but the rhyme requires -anz. Perhaps Rollantz? V7 4412 transcribes this as Rollant, since V7 laisse 242D rhymes in -ant (instead of -anz as in C). 4301. Another scribal omission of ~, this time from the common formula, Dex lor doinst…

La Chanson de Roland (Critical Text and Notes)

4310

4315

4320

4325

311

Tant cop ferir desor eume d’acer, Et tant paien morir et detrenchier, Lasé estoient, ne se poent aidier. En Ronchivaus, o fu le destorber, Li rois de France les fist toz detranchier E per les testes en l’eve trebucher. Mot grant eschec en ont si chevalier. Li rois de France descent de son destrier, Le Roi de Gloire comence a mercier; Quant se redrice, solauz prist a cocher. Dist l’enperere, «Tans est de herbergier. En Rencivaus tart ert de repairier; Nostre cheval sunt las, ne:l qier noier; Ostez les seles, mot en ont grant mester: Per toz cez prez les laissez refreschier.» Respondent Franc, «Bien fait a otrier.»



4330

Li enperere illoc se herberga Desor un tertre, sor le Sebre deza. Qi cheval ot, la sele li osta; Franc et Loereng onqes ren n’i laissa. Maint chevalier a terre se cocha; Qi lasez ert, volunters repousa. Maint en i ert qi onqes n’i menja: Qi n’ot vitalle, la nuit se consira; De jejüner grant deree fera,

4310. Mortier reads desoz, but Foerster desor (as required by the sense); the final letter is certainly r, the delicate ‘tail’ curling down from the front of the base of the letter in the usual way. 4334. Pope (1934, 110, § 245) quotes Latin jejunare, so this graphy (found also in C 8097, V7 8292 and V4 5936) may indicate a Franco-Italian feature. Godefroy translates grant deree fera (V7 4448: grant denree fera) as ‘undergo to excess’ but with this line quoted as a hapax legomenon. Since denree normally means ‘a denier’s worth, a small amount’, grant is singularly inapt, so V7’s reading is questionable. If C’s reading is correct, deree may be read as a variant of derei, graphy of desroi, meaning ‘damage’, ‘disorder’. Taken in conjunction with 4335, this makes good sense: ‘He will suffer great damage through fasting, when all he has to eat is what another will give him’. This laisse is much shorter in the other versions: 8 lines in O and V4, 11 in P, 7 in T, 3 in L, with nothing to correspond to these lines.

312 4335

4340

4345

4350

La Chanson de Roland (Critical Text and Notes)

Qi n’a qe prandre a autrui qe donra. Ainc eschargaite la nuit ne s’esveila. Li rois de France pas ne se desarma; La nuit i jut deci q’il ajorna; De garison ainc son cors n’i pansa. Mot ot grant duel, a son neveu pansa Et as barons q’en Rencivals laissa; Ja nulle joie dedenz son cuer n’ara, Tresc’a cel ore qe son neveu tanra: Ou vif ou mort, .c. fois lo baisera. Et Oliver tant per les chans quesra, E l’arcivesqe, qe trovez les ara. En doce France les cors enportera, A grant honor il les enterera. De Guenelon, ce dist, les vengera, De cil traïtres qi si mal les mena. A Haiz en France tel justise en fera, Don toz li siegles aprés lui parlera.

67.a



4355

4360

Li enperere fist forment a loër: Quant il ert liez, bien s’en poïst joier Uns petiz enfes, mais q’il seüst parler; Quant aucuns on l’avoit fait eschaufer, Mot fust hardis qi l’osast esgardier. Desor le Sebre fist sa gent osteler; Onqes n’i fist nul paveillon drecer, Ne cele nuit ne se vout desarmer, Ne son osberc ne volt il pas oster, Ne son grant eume qi si relusoit cler. Ceinte ot Joiose, onqes ne fu sa per: Qi chascun jor la vousist esgarder,

4341. C: lassa with i in superscript at the appropriate point. 4350. It is unusual to find the form traïtres as the oblique singular in C, but the normal traïtor would make the hemistich hypermetric. It may possibly be an early instance of the unusual development of the modern word traître, based on the frequently used vocative (subject form) instead of the oblique: cf. Pope (1934, 110, § 242, 286, § 725). Although the 2-syllable traitor is found some 10 times in C (cf. Introduction, 92–93), all instances (apart from that in C 1319) occur after C 6364; traïtres has therefore not been emended here. V7 4464 reads del traïtor…

313

La Chanson de Roland (Critical Text and Notes)

4365

4370

Mainte color i veïst remuer. Asez savons de la lance parler Don Nostre Sire se sofri a navrer: Karles en ot la pointe fait garder, En l’oré pont la fist bien seëler; Por les reliqes q’el pont fist enfermer, La fist ‘Joiose’ Karllemene apeler. Baron Franzois ne:l devez oblier, Ce fu l’enseigne de «Monjoie» escrier; Por ce ne:s puet nulle gent contrester.

67.b

 4375

4380

4385

Clere est la nuit et la lune luisant; Charlles se gist, mas duel a de Rollant E d’Oliver, lo hardi conbatant, Des .xii. pers qe il perama tant, Q’en Rencevals lassa mort et gisant. Ne puet muer n’en ait le cuer dolant; Deu a proié, qe maint en oriant, Q’il gart les armes et q’il lor soit garant. Las est li rois, penes i sofri grant; Endormi soi, ne puet mas en avant. Per toz cez prez s’endorment meintenant. N’i a cheval qi puisse estre en estant: Qi l’erbe past, si la prent en gisant. Auqes ot Karles vengié son mautalant

4366. The word la was added superscript before lance. This ‘lance’ is the one used to pierce Christ’s side on the Cross, thus an extremely holy relic. Cf. John xix.34 and note to C 7139. 4371. The abbreviation .K. must here be expanded to Karllemene to obtain the full 6 syllables in the second hemistich; V7 4485 has: Karlemeine. Cf. note to C 4209 and Introduction, 92. 4375. C: e li solel luisant, but the copyist confused this line with the common formula, Clers fu li jor e li solel…, whereas it was in fact night. All the other versions agree on la lune here. 4383. C: La, but the word needed here is Las, ‘weary’, as found in V7 4497 and O 2519 (with a variant, lais, in V4 2710). 4385. It is not made clear in C or V7 exactly who is the subject of s’endorment, but O 2521, V4 2712 and P 2924 all refer to the French at this point: eg. O: ..or se dorment li Franc. 4386. C and V7 4500 read: N’i a celui…, but a person is unlikely to eat grass (qi l’erbe past, C 4387). O 2522 reads N’i a cheval, V4 2713: civals, P 2925: Nus de chevax – which is much more apt.

314

La Chanson de Roland (Critical Text and Notes)

4390

Des Sarazins, qi Deu ne sunt creant; Lors s’endormi li rois demeintenant.



4395

4400

4405

4410

4415

Charlles se dort, qi mot fu travaillez. A Damedé en prist mot grant pitez, Saint Gabrials fu toz jor envoiez: Deus li comande qe il soit bien gardez; Li angle jut tote nuit a ses piez. Charlles estoit coroceus et iriez; En Rencevals avoit ses gens laissez, Les .xii. pers ocis et detrenchiez. Li rois se dort; ainz q’il fust esveillez, D’une avison fu forment travailliez: Karles regarde amont vers les hals ciez, Vit les tonoires et l’air qi est negiez, E voit jalées et gresius agoisez Et grant orages; mot s’en est merveillez. Et feus et flambe i est apareilliez; Sor l’ost chaï, mot les a daumagez: Ardent cez hanstes, cez confenons vergiez Et cil escu qi sunt bien vermeilliez; Croissent hauberc et ces tranchanz espiez. Illoc vit Karles ses gens mot mesaisez; Ors et liparz véoit toz desliez, Serpenz et guivres et dragons enrajez; Grifons i a .xxx.m., ce sachiez: As Franzois corent et devant et dalez, E cil escrient, «Karles, car nus aidiez!» Li rois de France en fu mot corocez; Aler i vost li rois, s’il fust laissez – Car uns lions venoit tot eslaissez;

68.a

4401. C: hals siez; V7 4515: hauz siez. Rather than indicating ‘high seats’ (or ‘high places’?), siez is a strange graphy of ciez, cielz, ‘skies’. Cf. O 2532: envers le ciel; V4 2733: envers lo cel; P 2934: vers le ciel; T 2165: devers les cielx. Thus, ‘the lofty skies’. 4407. C: hauste, for hanste (cf. Introduction, 88 (xx), final item under Phonology); also confenos with the symbol ~ omitted from over the second o. 4412. C: gruves, which is unintelligible, but V7 4526 reads guivers and O 2543 guivres, ‘vipers’.

La Chanson de Roland (Critical Text and Notes)

4420

315

Mot estoit fers, orgoillos et prisiez; Li roi asalt, qi ben fu batisez; A braz se prenent, mot fu Karles blicez; Luitient et sachent, mais ne sui aisïez De nomer vus qi remest plus haitiez. Li enperere ne s’est pas esveilliez.

 4425

4430

4435

Aprés iceste li vint autre avison: Q’il ert a Ais sor le mestre peron. En deus chaaines si tenoit un brohon; Devers Ardene, ce truis en la leçon, Coisi .xxx. ors venir tot le treton; Li uns a l’autre disoit bien sa raison; Au roi escrient, «Laissez nos le baron, Car il est droiz qe nos l’en remenon! Nostre amis est, et si le secoron.» Atant e vus un veltre en un landon, E descendoit del grant palais Karllon;

4422. C: Lui tient, as Foerster confirms, but a plural verb is needed to match sachent. C does occasionally divide words wrongly and V7 4536 reads: Luitent, ‘struggle’. 4423. To obtain 6 syllables in the second hemistich, haitiez (‘healthy’, ‘sound’, ‘unharmed’) might be read as trisyllabic, cf. aisïez in the line above; but omission of plus by C is more likely: cf. V7 4537: qui remest plus harciez, ‘who remains more shaken’. Duggan substitutes C’s haitiez for V7’s harciez. 4425. C: L pres at the beginning of the laisse but, as Mortier points out, there is a small a in the margin to indicate which (large) capital was needed. The error is the rubricator’s. 4426. Both C and V7 4540 read: Q’il ert alez…, a common formula, but alez is a scribal error for a Ais. O 2556 reads: Qu’il ert en France ad Ais a un perrun and P 2968: Que il estoit a Ais… 4427. C: chaines, but the normal graphy is chaaines, V7 4541 reads: chaenes and the extra syllable is needed for the metre. 4434. C, V7, P and L all here read: un autre – i.e. ‘another (bear)’, relating back to ors, C 4429; but Charlemagne’s young champion is symbolized by uns veltres, ‘a hound’, in O 2563. It was probably a simple scribal error which transformed veltre into altre > autre. However, brohon (C 4427) is ambiguous: it can mean ‘bear cub’ or ‘young dog’. If the source of the rhymed versions therefore took brohon as ‘young dog’, it would have grounds for changing veltres (‘hound’) C 4434 to autres (‘another young dog’). The objection to this must be that, between brohon in C 4427 and autre in C 4434, the word ors (‘bear’) occurs in C 4429, making the change to autre less justifiable. (Alternatively, V4’s reading, ventre, V4 2753, for veultre may indicate an error in an early source which autre attempts to correct.) That veltre is the correct reading is shown by the typical hound-like attack, C 4436–4437.

316

4440

4445

La Chanson de Roland (Critical Text and Notes)

As ors cort sore per mot fiere tenzon; Tot le plus mastre saisist por lo brohon. La vit li rois un estor mot felon, Mais il ne seit li qes veint ne qi non. Li angle Deu ce demostre au baron; Dormir le fait en mot grant sospeçon. Li rois Marsille s’enfuit a esperon Vers Saragoze, a son mestre donjon; En la cité qide avoir garison. Mais non ara, ne:l puet garir Mahon Q’il non conpert la mortel traïson Qe il a faite de Rollant le baron E d’Oliver et des autres baron.

68.b

 4450

4455

4460

4465

Li rois Marsille a sa voie acolie Ver Saragoze, a sa cité garnie. Tant a esté, devant la tor antie Est descenduz voiant sa baronie. Soz une olive qi ert belle et florie S’espee rent, el fu tost recolie; Lor le desarment la gent q’il ot norie; Puis est cochez sor l’erbe qe verdie. Del destre braz ne de la main n’ot mie; Per som le coute li fu del cors partie; Li sans en chiet sor l’erbe qe freschie; Trois fois se pasme voiant sa baronie. Atant es vos Braimimonde, s’amie: Plore et gaimente, a haute voiz escrie; O lui vint mil de la gent paenie; Carllon maudient, car il ne l’aiment mie, Et les Franzois qe il a en bailie.

4437. Here, brohon = breon (V7 4551), variant graphy of braion, ‘the fleshy part of a limb’, therefore usually either ‘calf’ or ‘thigh’; it must be distinguished from brohon, ‘bear cub’ or ‘young dog’, C 4427. 4453. florie, but being ‘covered in flowers’ is not a distinctive feature of olive trees; V7 4567 foillie, ‘covered in leaves’, accords far better with en l’umbre (O 2571) and en l’ombre (P 2989 and T 2193). 4461. Braimimonde is here Marsile’s wife, as in O. See note to C 961.

317

La Chanson de Roland (Critical Text and Notes)

4470

4475

4480

4485

4490

Puis sunt coru a lor mahomerie; Apolin trovent et Mahon qi brunie; De l’or d’Espeigne fu l’image plastrie. «He! Mauvaiz deus,» fait la gent barbarine, «Com nostre lois est hui abastardie! Dahaiz ait deus qi ne muit ne ne plie! Per vos est hui nostre gent si onie, Ainc per vos n’orent ne force ne aïe. Li Cristïen qi ne vos aiment mie Sunt en Espeigne, per force l’ont saissie.» Marsile prist un grant baston d’olie, Tervagan fiert, qanq’il poet, lez l’oïe: Li ors qassa et la teste est croissie; A tere qiet, mais ne s’en garde mie Uns rois paiens qi est d’Esclavonie: Tervagan chiet sor lui a une hie! Permi les els la cervelle est sallie; L’arme s’en vait et Noirons l’a saissie; El puis d’enfer a un piler la lie. A Mahomet ne la lasserent mie, La grant corone qi de l’or reflanbie. Forment le batent la pute gent haïe. Desoz la tor, en une ostelerie, La l’ont geté per mot grant felonie. Lors fu la ville de grant duel replenie. Li rois se pasme per mot grant estoutie Qe il a fait devant sa baronie.

69.a

4467. The word brunie (found also in V7 4581) is a graphy of the present indicative 3, brunoier (‘to sparkle’, ‘to shine’) with the o dropped for the sake of the rhyme. 4476. C: dalie, possibly a strange graphy of dalez, ‘close by’, or d’alie, ‘of sorb apple’ (the fruit), implying ‘of apple wood’. V7 4590’s d’olie (variant of d’olif), ‘of olive-wood’, fits the context better. 4478. Li ors is either the normal subject form, singular with the final s absorbing the reflexive pronoun: ‘The gold broke’, or an unusual oblique singular, as in C 3189: ‘He smashed the gold’. 4484. C: au p., but the single syllable makes the hemistich hypometric. V7 4598 reads: a un piler, which makes better sense. The copyist has simply omitted ~ over the u. 4485. The word la is explained by the next line: ‘they did not leave…it, the great crown…’.

318

La Chanson de Roland (Critical Text and Notes)



4495

4500

4505

4510

Li rois Marsille revint de pasmissons, Sor piez se drice, mot fu las et enbrons. Dedenz sa chanbre qi ert painte a gerons S’en est entrez li rois Marsilions, E la roïne qi ot les chevols blons. Le duel q’el mene orendroit vos dirons: Ses chevols trait o ses doiz q’el ot lons; «Lasse chaitive,» se cleme agenollons Plus de .c. foiz voiant toz les barons. A sa voiz clere escrie a haut tons: «Hé, Saragoze, bone en totes saissons, Con bon seignor vus a tolu Mahons! Tant est batuz de fuz et de bastons, Froissié li est la teste et li mentons; Jamais par mire n’en arra garisons. Li amirauz de cui nos fez tenons, S’il ne vient ci o tot .m. conpeignons, Por Mahomet, en qel terre fuirons? Ja par ton fil mais nul secors n’avrons! Se nos ici Karlle Mene atendons, Nus serons pris et la cité rendrons.» Ce dist Marsille, «Ne sai qe nos fazons.»

69.b

4495. If painte is read with its normal meaning, ‘painted’, then gerons poses a problem. These could possibly be the heraldic devices which decorate hauberk flaps, rather than the flaps themselves, but there seems no reason why Marsile would have heraldic devices painted on the walls of his room. On the other hand, Godefroy lists girons as meaning ‘the panels of material from which tents were made’. Given the nomadic, Bedouin origin of many of the Moors in Spain, the use of such panels to make a palace room more sumptuous and comfortable makes very good sense – and it also coincides with the European use of tapestries for the same purpose at this period. Whilst there could be a ‘tent-effect’ created in paint, it seems more reasonable to suppose painte to mean ‘decorated’ here – or perhaps a confusion between paindre and pendre. There is a play on two different meanings of the word pent in C 1814: A son col pent (= ‘hangs’) un escu pent (= ‘painted’) a flor. The phrase would then make good sense as ‘hung with panels of fabric (or tapestries)’. V7 lacks this line, which Duggan inserts from C. 4498, 4499. C: q’il mene… doiz q’il ot…, but the subject is Braimimonde; V7 reads elle in each case (V7 4612, 4613). The graphy el avoids making the first hemistich hypermetric. Il = ele/el is also found in C 4838–4839, 7121.

La Chanson de Roland (Critical Text and Notes)

319

 4515

4520

4525

4530

4535

Li enperere par sa grant poësté .vii. anz toz plens a en Espegne esté; Prist i chastiaus, si prist mante cité. Li rois Marsille n’a pas mot demoré, Li jor meïsme a meint brief séëlé: En Babiloine a Baligan mandé – C’est l’amirals, li viels d’antiquité, Qi mot estoit de grant auctorité – «C’a Saragoze, anzois un mois passé, Soit o .c.m. de cels de Val Troblé; Secore nos par sa grant poësté! Et s’il ne:l fait, ce sachiez de verté, Il guerpira sa creanze et son dé, Q’il li failli en bataille canpé, Totes ses ideles et sa solemnité; Et si cresra sainte Crestïenté Et de Karllon tenra son erité.» Baligan l’ot, forment l’en a pesé; Loinz ert d’iloc, si a mot demoré; Mande sa gent de per tot son reigné, De vint roiames et d’une ducheté. Son grant navie ot mot tost apresté,

4520. It is ironic that, historically, Charlemagne attacked Saragossa in 778 in support of the Abbasid Caliph in Damascus, from whom Arabs in Spain were claiming independence. He was persuaded to embark on this expedition by Sulaymánibn Yaqzan al-Arabi, governor of Saragossa, but, when that city resisted his attack, he decided to return to France with hostages and treasure, his rearguard or baggage-train being attacked once he had left the territory covered by the strict conventions of hospitality. 4523–4531. These lines express Marsile’s wish, using a jussive subjunctive for an indirect command and presumably quoting direct from the letter written in formal style: ‘Let him (Baligant) be at Saragossa… Let him succour us…!’, but then continuing in a confused mixture of indirect and direct speech: ‘And if he (Baligant) does not do so, then know this for truth, that he (Marsile) will abandon his faith and his god who failed him…’. Such confusion is typical of the period. 4530. cresra is a graphy of crerra, future 3, creir. V7 4644 reads: creira. 4532. The verb is impersonal: ‘it weighed heavily upon him’, the l’ representing the indirect object, li. V7 4646 reads this personally, but inaccurately: l’en a pené, which Duggan emends.

320

4540

La Chanson de Roland (Critical Text and Notes)

Barges, galies et meint dromon fesré. Soz Alixandre a un havre mot lé, Tot lor navie ont illoc asanblé. Ce fu en mai, el primer jor d’esté, Q’en mer se metent, n’i ont plus demoré.

70.a



4545

4550

4555

4560

Granz sunt les oz de la gent desloiée; En mer se metent quant l’aube est levee. Puis icele ore qe la lois fu jugee, Plus grant navie ne fu apareilee Com Baligans mena cele foiée. Des Turs de Cople, de cels de Valsegnee, Qatre cent nés: la pire est si chargee N’en sofrist plus a icele foiée. N’i avoit mast, neïs verge entallee, O n’eüst cerge ou lanterne enficee. Grant joie mene la pute gent turqee Por la clarté q’il ont apareilee. Perse costoient, la terre ont eslongee, La mer trepassent sanz pene et sanz hasqee. Deus! Tante voile i ot le jor levee! En Saragoze, en la tor batellee, Estoit Marsille o mil de sa masnee. E Brasmimonde s’ert as murs apoiee; La clarté voit, mot fu joiose et liée; Marsille apelle, un poi fu esclairee.

4537. C: Bargies et galies, with the normal graphy of barges contaminated by the ending of galies and an extra et inserted, making the hemistich hypermetric. V7 4651 reads: Barges, galies… 4543. The abbreviation for est, ē, was inserted superscript before levee; but the e of aube must also be pronounced to avoid hypometry. V7 4657: quant l’aube est sclaree, emended by Duggan to esclaree. 4544. This is presumably a reference to the Law of Moses, ie to the giving of the Ten Commandments to Moses by God (Exodus xx seq.), thus ‘from very early times’. 4547. C: Valseignee, with dot beneath the i for excision. 4550–4551. ‘There was not a mast, not even a pointed (or carved) rod, which did not have a candle or lantern fixed on it’, so that the whole fleet was illuminated 4556. C: i ot i ot…

321

La Chanson de Roland (Critical Text and Notes)

4565

«Sire,» fait ele, «joie t’est aprosmee! Baligans vient, s’a sa terre laissee. Or sui seüre, n’ai pas poür qe cee Ceste citez ou je sui herbergiee.»



4570

4575

4580

Li Sarazin esploitent mot d’esrer; Il se navient le travers de la mer, Perse costient, l’Archant et Balaguer, Et Port Palart, Orabloi et Belcler; Deci c’al Sebre ne volsent arester. .c. mille cerges orent fait alumer E mil carboncles en son les maus fermer; De Saragoze les puet hon veoir cler. Laissent Marbrie et Marbroie sa per, E Tortelose dont fu rois Josuer; Amont lo Sebre font les voilles torner. Ne vos chet plus la chanzon demorer: Ver Saragoze, el havre Guinemer, Font lor navies et lor barges entrer. L’aube creva, si prist a ajorner; Lors font grant joie Saracin et Escler, Quant il conuirent li uns l’autre au parler. Cil les confonde qi se laissa pener As faus Jeüs por Cristïens sauvier!

70.b

4564. ‘I am not afraid lest this city fall…; cee is a graphy of chiëe, present subjunctive 3, cheoir; V7 4678 reads chee. 4572. The word max/maus is a graphy of maz, oblique form plural of mast, ‘mast’. V7 4686 reads: nef, which Duggan emends to masts. 4577. The final letter of the verb is not clear, so this may read cher, a graphy of quier, present indicative 1, quesre/qerre, thus ‘I do not seek (wish) to delay the recital of events any further’, but looks more like chet, a graphy of chelt/chalt/caut, present 3 of caleir (impersonal verb): ‘There is no need to delay…’. V7 4691 reads quiert, emended by Duggan to quier. (The graphy qier is found in a similar context, C line 37.) In either case, these are words addressed direct to the audience, pre-empting any impatience caused by the long list of names. 4578. C: el a havre. 4582. C: conuirent; for conurent, past definite 6, conoistre (V7 omits C 4582, but see note to C 4587 below regarding the same verb); for the infinitive, conuistre, cf. Pope (1934, 416, § 1066). Late retention of üi diphthong was a feature of North-Eastern French

322

La Chanson de Roland (Critical Text and Notes)

 4585

4590

4595

4600

4605

4610

Bauz fu li jors et li solauz luisanz; Cil oisel vont per les eves cantant, Quant il conuirent le paien Baligant. Li amirauz issi de son chalant; .x. et set roi le vont aprés sigant, E conte et duc i furent ne sai quant. Soz un lorer ramu et verdoiant, Un faudestue i ont mis d’olifant; Desus asistrent le felon sosduiant, E tuit li autre sunt remés en estant. Li sire d’als en a parlé avant; Sa gent apele, si lor dist en riant: «Franche masnee, je vos ai amé tant; Nori vos ai toz jor per bon talant. Karlles, li rois de France la vaillant, Ne doit mengier se je ne li comant. Per tote Espeigne guesre m’a faite grant; Il a perdu son chier neveu Rollant, Les .xii. pers ou il se fioit tant; E tant des autres don il est mult dolant; Jamais n’ert liez nul jor a son vivant. Or vel aler vers France conquirant. Ne finerai a trestot mon vivant, Tresqe je l’aie o mort o recreant.» Sor son genoil en fert son destre gant. Lors s’esbaudirent li Turc et li Persant;

71.a

4587. C: conuirent, with the same graphy as in C 4582, in what may be a careless repetition of the first hemistich of that line; V7 4700: conurent. This verb reads strangely in the context, implying that the birds sang when they recognized Baligant. If il is intended to refer back to the Saracin e Escler, C 4581, the description in lines 4585–4586 may symbolize the Saracins’ state of mind. Nevertheless, since Baligant is being escorted up the River Ebro (for in line 4588 he disembarks), conuirent may be a scribal error for either conduient, present 6, conduire, or convoient, present 6, convoier – or possibly conjoient or conjuïent, pres.6, conjoïr, ‘to welcome’ (or past definite, conjurent) mis-read by both C and V7. This could arise from -iu- misread as -ui-, frequently found in C and to a lesser degree in V7. No other version has these lines: V4, P and T proceed directly from C 4586 to C 4589. 4609. C: fer; V7 4722 and O 2664: fiert: i.e. present 3, ferir.

La Chanson de Roland (Critical Text and Notes)

323

En haut s’escrient li petit et li grant: «Chevauchez, sire! Alons nos l’i qerant!»



4615

4620 4620a(O)

4625

Baligant fu un petit corocez; De ce qe ot dit est forment afichez; Ne lassera, ce dist li renoiez, Q’il n’aut sor Karle, qi tant est resoignez. Si home en sunt trestoz joianz et liez. Puis apela deus Sarazins presiez, L’un Clariel, l’autre fu Effragiez: «Vus fustes fil au roi Matragiez, Qi mes messages soleit faire volentiez. Je vus comant c’a Saragoze alez, Marsilion de moie part nunciez, Contre François sui venuz toz haitiez. Se ge:s ateng, toz seront detranchiez. Se li donez cest gant qi est ploiez; El destre puing voiant toz li fichez E cest baston qi est d’or entaliez. Puis viegne a moi reconostre ses fiez. En France irai; Paris est aseigiez;

4612. C: il qerant, although the pronoun, oblique form singular, lo, le or lu, is needed. V7 4725 reads: Nos n’alons al querant; but C’s il reading seems to be a scribal error for l’i (‘him there’), due to metathesis. Or this may represent ‘foreign speech’. 4614. ‘On what he has said he is firmly resolved’, with elision of qe ot for the metre. 4616. The word aut is a variant of alt, present subjunctive 3, aler; aut sor = ‘gets the upper hand of’. 4618. presiez < presier, variant graphy of proisier/priser. 4620. In C and V7 4733, roi Matragiez seems to be mentioned to no purpose. P and T have a completely different version of these lines and nothing more is added in L. However, O 2672 adds: Ki mes messages soleit faire volenters, echoed by V4 3027: Que me messages me fari volunter. O’s line has been inserted with slight modification to the spelling to match C’s usual graphies and rhyme. 4629. C: Paris e aseigniez, the e having the slightly arched line above, normally used to signify est in C; and the present may have been deliberately chosen to make a strong affirmation of immediacy. Foerster and Mortier read this as en, which does not fit. The C scribe may have intended ert, future 3, essere, but used the wrong symbol; V7 4742 has the strictly accurate iert, future 3.

324 4630

4635

La Chanson de Roland (Critical Text and Notes)

Se truis Karllon, mort est et meshegnez. S’a ma merci ne se coche a mes piez E ne renoie ce q’il fu batizez, La grant corone dont il ert sorhauchez Li ert tolue, et si ert essiliez.» Paien escrient, «Voz diz soit otrïez! D’aler en France somes apareiliez.»

71.b



4640

4645

4650

Dist Baliganz, «Car en alez, baron; L’uns port le gant et l’autre le baston.» Cil li respondent, «Cher sire, nos iron.» Le congié prenent, n’i ot autre sermon. Tant chevaucherent la voie et lo sablon C’a Saragoze sunt venu le treton. Pasent .x. portes li Sarazin felon, Set ponz passerent, puis passent un donjon; Les rues passent ou sunt li Esclavon; Enmi la ville pervindrent li gloton, Ver li palais qi fu au roi Gibon, Un Saracin del lignage Buiron: Cil ot mordri son pere en traïson. Dedenz la sale oïrent grant tenzon,

4630. C: mort e, with line over e, then the e is barred and est written in full, followed by a symbol for et. 4633. C: g//rãt, with heavy vertical bars between the g and r of grant – for no visible reason. 4638. C: Li uns, V7 3751 Le uns, but this gives a hypermetric hemistich; the normal graphy is: L’uns. 4643–4645. The verb paser/passer is used 4 times in the same sentence in these 3 lines, and to considerable effect, but in a mixture of (historic) present and past definite tenses. Unfortunately, in changing the tense to suit the requirements of metre, the scribe has become confused over the value to assign to the -ent ending (present 6 and past definite 6); thus, in C 4644 (puis passerent un dojon), he repeats passerent from the (metrically correct) first hemistich into the second hemistich, making it hypermetric; V7 has passent on the second occasion (V7 4757). In C 4645, passent is metrically correct, since rues is disyllabic. At the end of C 4644, both C and V7 read: dojon, the ~ being omitted. 4649. C: Cil ot mort, an hypometric hemistich, but with a slanting line over the cæsura as though it were an i. V7 4762 reads mordri, past participle, mordrir (vb.transitive), ‘ to kill’ or ‘to assassinate’.

325

La Chanson de Roland (Critical Text and Notes)

4655

4660

4665

4670

Tant cri grezois et si grant plorison: Brugent et crient per grant aïreson, Pleignent lor peres et lor freres a aut ton, Q’en Rencevaus ocist li niés Karllon Et Oliver et li autre baron; Nul n’en laisserent venir a raënzon; N’orent garant se des ches perdre non. Lor deus maudient, Apolin et Mahon, Et Tervigan et Jupin et Noiron. Dist l’uns a l’autre, «Chaitif, qe devendron? Sor nos cors chiet male confondison. Perdu avons lo roi Marsillion; Li cons Rollant, qi ait maleïçon, De son braz destre li a fait un tronzon; E l’augalie qi tint Cafarnaön, Tote la terre Dathan et d’Abiron. Nus n’avons mie de Virfalle le blon; Tuit sunt ocis, ne sai qe’l celison; Trestot Espeigne ert acline a Karllon.» Endementers q’il dient lor raison, Li dui mesage descendent au peron.

72.a

4651. C reads: grezois, V7 4764 greçois, apparently meaning ‘Greek’, but this is puzzling: are these ‘Greek cries’ because they are in a foreign language, therefore incomprehensible? There is, however, the word gregeos, a variant graphy of greveos/grevois, adjective, ‘grievous’, ‘full of pain’, which fits well here. The scribal error may be in the CV7 model or both copyists may have misread that model. 4652. C’s brugent is present 6, bruire (‘make a din’) with substitution of g for j (i), but may be a scribal error for bragent (V7 4765), a graphy of braient, pres.6, braire. 4666. A reference to Dathan and Abiram, the sons of Eliab, who were condemned to death by God for challenging Moses’ divine appointment and rebelling against his leadership (cf. Book of Numbers, xvi.1–33). This thus indicates part of Palestine, with an overtone of wickedness; but the formula Dathan et Abiron was much used in mediæval contracts. 4668. Cf. footnote to C 246 for evidence of the existence of the verb celeir/celleir. C and V7 (line 4781) agree in reading celison, which here is present subjunctive 4; celisum, present indicative 4, is found in the Lais de Marie de France, ‘Milun’, line 139. The hemistich therefore means: ‘I do not know how we may hide it’ . 4669. C: aclinee, past participle, feminine, meaning ‘made subject (to)’, but this makes the hemistich hypermetric, since the final word of the line must be Karllon for the rhyme. This is

326

La Chanson de Roland (Critical Text and Notes)



4675

4680

4685

4690

4695

Li dui mesage descendent esrament; Lor chevaus tint uns paiens Rualent. Cil montent sus el plus aut mandement; Trovent deus Turs del Castel Boidïent: Cil qatre estoient e cosin et parent. Li mesager lor dient esrament, «Seignor cosin, com vos est covenent?» Et cil respondent, «Seignor, mauvaissement. En Rencivals fumes el canp dolent; Desconfi somes et livré a torment. De .c. millers n’en vint un sauvament. Venez au roi, cui tote Espeigne apent, Si vos dira com lui est convenent.» Lors trepasserent le palais et la gent; En une chanbre qi fu faite a ciment – El primer chief esteit escrit coment Li poinz del siegle fu faiz primeremenz. Desor un lit qi fu d’or et d’argent Gisoit Marsille et ploroit tendrement Por la grant perde qe il ot fait de sa gent. Cil le salue söef per bon talent: «Cil damedeus qi fist et mer et vent, C’est Tervagans, se la letre ne ment: Il salt Marsille a l’orguellos talent E la roïne Brasmimonde ensement! Mesager somes devant vos en present; Tel chose osras qi ert a ton talent.»

thus a scribal error for the adjective acline (feminine), ‘subject (to)’, ‘in subjection (to)’. V7 4782 reads: acline. 4678. C: covenent, graphy of covenant with the ending modified for the rhyme. ‘How are you?’ 4682. C: n’en vint un a sauvament, which is hypermetric; but sauvament is here the adverb, ‘safely’ or ‘safe and sound’, not the substantive, so the a is superfluous. V7 4795: …uns salvement. 4684. C: dirai, but dira, future 3 as in V7 4797, is needed in the context. 4688. Li poinz del siegle: ‘the beginning of life on earth’. This inscription on the wall of Marsile’s chambers, relating how life on earth was first created, may be an allusion to the greater scientific knowledge of the Arab world or simply to the Creation story in Genesis (but this implies that the writer was more aware of links between Islam and Judaism than is indicated elsewhere).

327

La Chanson de Roland (Critical Text and Notes)

 4700

4705

4710

4715

4720

Dist Braimimonde, «Venez avant, Seignor. De vus noveles dites l’enperaör, Car torné somes a duel et a tristor. Tote a perdue Mahomez sa vigor E Terviganz i a grant desenor. Com grant vertuz il firent l’autre jor: En Rencivals faillirent mon seignor. Mal dehait ait tel deu ne lor honor! E je si soie honie sanz retor Se je jamais nul de tel deus aör! Honie soit lor force et lor vigor! Quant cil de France, li noble pogneor, .xx. mil estoient, ce dient li plusor – E li nostre erent.c. mil comenceor – Toz les ocisent li Franzois a dolor. Al port del Sebre vindrent fuiant li jor; Iloc noierent li grant et li menor. Li rois Marsille fu trop entré lez lor: Li cons Rollant o la fiere vigor Le destre braz li trencha sanz retor; Perdu en a sa force et sa valor. Ahi, Espeigne, terre de grant dolçor, Or t’arra Karles qi tient Terre Major.»

72.b

4700. ‘Tell the king your news’; the word enperaor is here used to designate Marsile. 4702. C: Mahomez et sa v…, but et is superfluous to both sense and metre; it is not found in V7 4815. 4704. The symbol for com or con, 9, is here used as a capital letter to begin the line. 4710. C: li nobile pogneor is hypermetric with nobile an Italianism. V7 4823 has: li noble poigneor. 4712. C: E li nostre, a 3-syllable hemistich; V7 4825: Et li nostre erent, although Foerster omits the (clearly legible) erent from his edition. 4716. This makes sense provided one reads: fu trop entré lez lor, ‘had gone in too far alongside them’. V7 4829 fu trop entre les lor, ‘. too far in amongst their forces’, was probably what the scribe intended. 4720. C: deçor, but V7 4833 reads: douçor, ‘pleasantness’, ‘beauty’. C 5998 has the graphy dolçor.

328

La Chanson de Roland (Critical Text and Notes)



4725

4730

4735

4740

Dist Clariaus, «Dame, ne parlez tant! Mesager somes au fort roi Baligant. Marsilions, ce dist, sera garant, Si l’en envoie lo baston et lo gant. El Sebre somes ou .iiii.m. chalant, Batals et barges, meinte galie estant, Dromonz ferez el primer cief devant. Li amiraus est riches et mananz; En France ira Charlle Mene querant.» Dist Braimimonde, «Ja mar en ira tant. Prés trovera le fort roi conbatant; E sunt o lui bien .c.m. conquirant. Espeigne ert lor, de ce se vont vantant. Prou sunt Franzois, bien part a lor sanblant. Nus ne s’i prent qi ja s’en alt gabant. Tel poor ai, li cuers me va faillant. Il asalront ceste ville vallant; De nos feront la justise si grant, N’i garra mie li peres son enfant, Qe ja li rois n’ira del canp fuiant. Meus valt lor Deus qe ne valt Tervigant Ne Apolin, ne Mahon le puant.»

73.a

 4745

«Laissez ester!» fait Marsille li rois. Dist a sez homes, «Mot sui en grant sospois.

4727. C: Batas, for batals, graphy ofbatels, ‘boats’; V7 4840: batels. 4735. This line is not found in any other version (V7 omits C 4735–4737 inclusive). In the second hemistich, the C ms. reading part (written in full) must represent pert, present 3, impersonal form, of paroir, ‘to appear’: ‘it can well be seen from their appearance’ or ‘it is clearly shown by their deeds’. 4740. C: garira, but this makes the hemistich hypermetric. V7 4850 reads: garra, the usual future 3, garir. 4742. C: Tuigãt with the er symbol omitted over the u; V7 4853 inserts the 7 to give Tervigant. 4744. C: Laissez e, with line above the e. Foerster simply records the symbol; Mortier reads en, but this makes the hemistich hypometric; so, too, would the normal transcript in C, est, but an additional symbol (omitted through scribal error) would make this ester, ‘Let be!’ – a common phrase which fits the context. Cf. O 2741: Laissez ço ester!; V7 4855 and P 3169: Laissiez ester; V4 2930: Lasez ester.

329

La Chanson de Roland (Critical Text and Notes)

4750

4755

Arere irez orendroit demanois, A l’amirant direz, car ce est voirs, Desconfiz sui, cheüz est mes bofois, Mort sunt mi home et mi Saracinois, Mes fius ocis qi tant per ert cortois – Aprés ma mort, eüst toz mes manois. Rollant l’ocist, n’a encore pas .ii. mois. Mon seignor dites qe ça viegne manois; Per tote Espeigne, la ou corent mes lois, Tot li doin qite, ne:l faz pas sor mon pois. Dés icest jor le gart contre Franzois! De Saragoze les clés li porterois, E de Karllon li direz sanz grezois, Jamais n’ira a Chartres ni a Blois.»

 4760

4765

Ce dist Marsille, «Oiez raison menbree: Karles de France a mot sa gent menee; Morz a mes homes et ma terre gastee; Meinte contree m’a fraite et violee. Desor le Sebre a sa gent aünee; N’a qe .v. lieues la ou ele est jostee; Trover l’i puet demein ainz la vespree. Per vos li mant qe bataille a trovee. Per roi franzois n’ert jamais reclamee,

73.b

4751. C: manois, for manoirs, the r being dropped possibly simply for the sake of the rhyme, but also because of effacement of præ-consonantal r and its assimilation in the group rs; cf. Pope (1934, 156s., § 396); also note to C 1504, Introduction, 73, 90 and 99. 4755. C: nel faz per…, per being written in abbreviated form as p with bar through its tail; but V7 4869 reads pas (although it omits faz in error): ‘I do not do it against my will’, ‘I do it willingly’. 4767. O 2761 here reads: Par vos li mand, bataille i seit justee, ‘Through you I ask him that battle should be joined there’ (and V4 2950: Se per nos ert la bataille justee). In C, however, qe bataille a trovee seems to indicate that mant (< mander) here has the alternative meaning, ‘declare’: ‘Through you I declare to him that he has found battle’ – ‘…that a battle awaits him’. 4768. The adjective, reclamee, needs a suitable antecedent to justify the feminine singular and this can only be bataille, C 4767, and thus here signify ‘claimed as his (victory)’. Marsile sounds over-optimistic in his forecasts, C 4768–4769, but he cannot assert that battle ‘will never be demanded by the French king’, for he knows that Charlemagne is only too eager to avenge the death of Roland.

330

4770

La Chanson de Roland (Critical Text and Notes)

Ne Saragoze a nul jor regardee.» Aprés cest mot, n’i ont fait demoree; Prenent congié a la gent desfaee; Per tans sera la novelle nonciee A l’amiralt, qi tant l’a desiree. De Saragoze ont fait lor desevree.

 4775

4780

4785

4790

Li dui mesage se sunt d’iloc torné, Delivrement furent andoi monté, A l’amiraut en vont tot esfréé. De Saragoze ont pris la seürté; Les clés li donent; cil les rechoit en gré. Dist Baligans, «Qe avez vos trové? Ou est Marsille qe j’avoie mandé?» Dist Clariel, «Je:l vi a mort navré. Per nos vus mande, ne soiez effréé, Li rois de France est sor Sebre aresté; A Saragoze venra, ce est vanté; Prendre la qide ainz q’iver soit passé. Ja por Marsille ne sera trestorné; Li cons Rollant l’a si mal atorné, Lo destre braz li a del cors sevré En Rencevals, el chanp maleüré. La fu ses fius Girfalli mort geté Et l’augalie a son tens afiné; Mal l’a baili ceste Crestïenté.

4770. C: n’i oit; V7 4884: n’i ont, as required by the sense. 4772. C: nonovelle nonciee, with an extra no accidentally repeated in the alliteration. 4774. C: De Saragoze ont pris la seürté, which has the wrong rhyme and is out of place. It is repeated at its proper place, as C 4778, where it again follows a line beginning A l’amiralt, or rather A l’amiraut. Instead of C’s reading, line 4774, V7 4888 has the same first hemistich but ends with ont fait lor desevree. This V7 reading has been adopted. 4785. C: vaté, with missing ~, where V7 4899 reads vanté. 4786. C: ainz qe ver…, but ver, meaning ‘Spring’, is supposed to have died out in Gallo-Roman, so (unless this is a late clerical survival) qe ver must be a scribal error for q’iver (yver in V7 4900), changing the season to Winter. According to C 4540, Baligant’s fleet set out in May, on the first day of Summer. 4789. C: Braz, with capital B written in error, probably because the Proper Names Baligant and Braimimonde occur frequently at this point in the text.

331

La Chanson de Roland (Critical Text and Notes)

4795

Tote sa terre nos laïet en qité; Jamais per lui n’en serez destorbé, Qe n’en facez la vostre volunté. Secorez lo, a poi q’il n’est oltré!» Balligans l’ot, si a un poi pensé.

 4800

4805

4810

«Sire amirant,» ci li dist Clarïez, «Saluz vos mande li rois, et amistez. En Rencevals sunt paien detranchez, E lor Franzois lor sunt si daumajez: Morz est Rollant qi tant estoit prissez Et Oliver i fu morz, ce sachez, Li .xii. pers dont Karles est irez – De lor Franzois bien .xx.m. pezoiez. Li rois Marsille i est mot enpirez, Del destre braz a esté mesheignez, Et de Karllon fu asez enchauchez; Si home furent a grant dolor neiez. Per nos vus mande, n’en eschapa un piez. Desor le Sebre sunt Franzois herbergiez; En cest païs nos ont trop aprochez; Si vos volez, li repaire ert a grez.»

74.a

4794. C: nos lait en q…, with ~ over the i to give laient; but a singular verb is needed. Pope (1934, 338, § 887) lists present 3, lei-ier (‘to leave’), as leiet (final t optional) and Fouché (1967, 110, § 52) quotes laïet as present 3, laïer, and this last reading has been adopted. V7 4908 reads: lais, emended to lait. 4795. C: sere for serez, fut.5; V7 4909 reads: serez. 4797. C: otrié, ‘granted’, a word frequently used in this narrative, but inappropriate here. V7 4911 reads: oltré, a ready basis for scribal error, which gives the meaning ‘for he is almost dead’. T 2344 echoes this sense with: car moult est meschevé, ‘for he is badly wounded’. 4810. C: menez, ‘led’ or ‘brought’; although adequate, this lacks the impact of neiez, ‘drowned’, found in V7 4924 and O 2798, or P’s noiez (line 3220); T 2357 reads: noier. This is C’s scribal error. 4814. C: li repaire ert a grez, but V7 4928 reads: li repaire i ert guez (emended by Duggan to guiez); O 2801 reads: li repaires ert grefs (= ‘grievous to them’) and P 3223: lor repaires iert griez. In view of the general consensus here, it seems best to retain the meaning that the French would suffer a damaging retreat if Baligan decided to give battle. If C’s grez is read as a graphy of the substantive, grief, then a grez means ‘with suffering’, ‘with damage’, ‘with grievous losses’; if it is a reduced form of agriez, then it means ‘tormented’. In either case, the meaning matches that of the adjective in the other versions.

332 4815

4820

La Chanson de Roland (Critical Text and Notes)

Et Baligans s’i est mot afichez, En son corage en fu joios et liez; Del faudestue se dreza sor ses piez; Puis si escrie, «Baron, ne vus targiez! Issiez des nés, montez, si chevauchiez! Se ne s’enfuit Karles li sorqidez, De duel mosrai, se il n’est esilliez.»



4825

4830

4835

4840

Paien d’Erabe sunt fors des nés issu, Puis sunt monté, n’i ont plus atendu. Li amiraus apele un son dru Qi avoit non Fergalem de Mont Nu: «Condui ma gent a force et a vertu!» Cil li respont, «Voluntiers, a salu.» Baligan monte sor un destrer crenu; Enscanble o lui .xxx. roi mescreü; Tant chevaucha c’a Saragoze fu. Sor un peron de marbre est descendu; Catre baron li ont l’estré tenu. Per un degré qi de vert marbre fu Montent amont li paien mescreü. Encontre vint Braimimonde a vertu: «Gentius hon, sire, mon seignor ai perdu! Karles de France l’a mort et confondu!» Chiet li as piez, car el pert la vertu; Quant el revint, n’i ont plus atendu, Enz en la chanbre sunt ensanble venu.

74.b

4823. C: ni ot…, V7 4937: n’i ot. Duggan accepts this as an impersonal construction: ‘there was no more delay’, but after two verbs in passé composé 6, ont seems the better reading. Cf. also C 4839. 4838, 4839. These lines, peculiar to C, read very strangely with .il pert… il revint.; it is Bramimonde, not Baligan, who faints and comes to her senses again. Emendation to el rather than ele keeps the line metrically accurate. In the glossary to his edition of V4, Beretta (1995, 514– 517) lists as many examples of el, pronoun, 3rd person, masc. subject.sing., as of il, masc, subject sing., but the use of il here may be due to poor comprehension rather than to confusion of il, el, ele. But cf. C 4498, 4499 and also C 7121, where il refers to Aude. As in C 4823, ot… atendu has been emended to ont…atendu.

333

La Chanson de Roland (Critical Text and Notes)



4845

4850

4855

4860

Baligans fu pleins de mult grant mautalent; Dedenz la chanbre est entrez esrament. Li rois Marsille fu navrez durement; Il en apele un roi de Bonevent: «Prenez mon braz, dreciez moi bonement.» Au puing senestre prist un baston d’argent; A Baligant le rent mot bonement. «Amiraut, sire, tote Espeigne vos rent E Saragoze et l’onor q’i apent. Moi ai perdu et trestote ma gent.» Et cil respont, «Le cuer en ai dolent. Ne puis o vos tenir lonc parlament, Car je sai bien Karles s’en va fuient.» Le gant reçut et la terre ensement; Plorant s’en torne d’iloc isnelement; Par les degrez jus del palais descent; Monte el destrier, vint pongnant a sa gent. Il lor escrie a sa voiz clerement: «Montez, seignor, tost et astivement! E chevauchiez, sanz nul demorament, Aprés Franzois tot arotéëment! Li rois s’enfuit o tote France apent.»



4865

Al maitinet, quant il dut esclarier, Esvellez est Karllemene au vis fier. Il en apelle et Neymon et Ogier: «Or tost as armes, nobile chevalier! En Rencivals vel anuit herbergier;

75.a

4841. Most unusually, this is a 5-syllable first hemistich, but Baligans is written in full and the cæsura clearly marked after pleins and followed by a 6-syllable hemistich. V7 4955 reads: Balligans fu plens de mout fier talent, with cæsura marked after fu. The easiest explanation is the omission by C of Et as the first word of the line: i.e. an alexandrine in the source (corrected to 4:6 metre by V7). Alternatively, C’s cæsura may have been displaced: if it occurs after fu (as in V7), then the 3-syllable mautalent must be accommodated in the second hemistich, by dropping either mult or pleins (always a monosyllable in C). Although both these words are found here in V7, they are standard formulæ and not absolutely essential.

334

4870

4875

4880

4885

4890

4895

La Chanson de Roland (Critical Text and Notes)

La troverai lo mortel enconbrier, Le duc Rollant et le cons Oliver, Et l’arcivesqe qi tant fist a priser, Les .xii. pers qi m’avoient mot chier. Ahi, fel Guene,» dist Karllon au vis fier, «Jesus de gloire te doinst mal enconbrier! Se Deus ce done, qe puisse repairier En douce France, je m’en qit bien vengier!» Lors s’arme Karles, le fort roi droiturer, Et tuit li autre, qi n’ont soing de targer. A cez paroles, este vos un paumer: Nez fu de Chartres, si avoit non Reiner; Espi au col et baston de pomier; .xvii. anz l’ot en prison Noitïer, Uns rois paiens mot orgueillos et fier; Eschapez fu per Brehas sa moiller. Devant lo roi se vait agenoiller: «Sire,» fait il, «novelles voi nuncier: Rois Baligan vos suit au dos derier; Ensanble o lui bien .c.m. chevalier. Celi oï conter et fiancher Q’il vos venroit durement corocer. Marsillion volent de vos vengier.» Li rois de France se prist ad enbronchier, Puis se dreza, si s’afiche a l’ester. «Hé, Deus,» dist il, «ne l’en descorager! Se je sor aus puis mon duel esclairer, Donc puis je dire qe Deus m’a forment chier.»

4873. C: doist; V7 4987: dont; but doinst, present subjunctive 3, doner, ‘let Him give!’, is needed. 4877. C: soig, with ~ omitted, or with g as a graphy of ng; V7 4991: soing. 4878. At this point O, V4, P and T make Charlemagne return to Roncevaux before the battle with Baligant. C and V7 reverse this order, so that Charlemagne’s long mourning becomes a continuous theme from Roncevaux through the journey across France (cf. J S Crompton, Manchester MA thesis on the Châteauroux MS, April 1951). With the new emphasis on the «Belle Aude» episode, it is poetically more effective to sustain the grief-laden atmosphere, rather than let the battle episode intrude. 4885. voi reads best as present 1, voleir: voil, as in V7 4999.

335

La Chanson de Roland (Critical Text and Notes)



4900

4905

4910

4915

4920

4925

Venir s’en velt l’enperere puissant En douce France, dont mot est desirant. Mais ne qit mie q’il s’en alt a itant, Ainz vengera auqes son maltalant, Qe li ont fait li gloton sosduiant, Car li paumiers li va toz jor contant Qe aprés lui chevauche Baligant. Li Saracin erent ja venu tant, Devant Karllon vont l’avangarde prenant. Atant es vos un Saracin pongnant, Devant lo roi en vint esperonant; Ne:l salua, ainz dist per maltalant, «Diva, fels rois, vois t’en tu donc fuiant? Jamais en France n’osras sermon ne chant! Tu as perdu Oliver et Rollant, Les .xii. pers qi t’estoient garant; Il n’a paien jusq’en Inde la grant Ne desq’a Meqe, ne dusq’a Garillant, Ne un ne altre qe arme soit portant, N’i soit venuz a esperon brochant; De toi oncire sunt forment desirant.» Quant l’oï Karles, si mua son talant; A voiz escrie, mot a le cuer joiant: «Mar i entrerent li gloton mescreant! Anqi avront un eschac issi grant, Nus rois en terre n’ot onqes si vaillant! Or tost as armes, chevalier et serjant! Alez encontre, n’alez pas demorant! N’en i ait un coart ni esmaiant! Qi ci mosra, loier en ara grant; O les apostres sera toz jorz manant.»

 Li enperere s’adobe sanz tardance; Il vest l’osberc, lace la conoissance; 4920. C: issi with the s in superscript very lightly and faintly marked.

75.b

336

4930

4935

4940

4945

La Chanson de Roland (Critical Text and Notes)

En son chief lace un eume de Valance; Escu au col, en son puig tint la lance; E ceinst Joiose, qi fist mainte pesance As Tur de Cople et a ces de Briance; N’ot tel espée deci qe en Provance; De maint paien fist de l’arme sevrance. Cel jor dormi Mahomet sanz falance; Trop soulement se misent en balance Entre François, qi grant honor avance. Karles monta, qi fu en grant esrance, Sor un destrier qi ot la crope blance; Fist un eslais dont il fu grant parlance; Quant il l’ot fait, si parla en oïance; A voiz escrie au barnage de France: «Alonz encontre per mot voire creance!» Dient François, «Ci a bone atendance! En tel seignor doit hon avoir fiance.»

76.a

 Par toz les cans descendent li François; Plus de .c.m. s’en adobent manois. Garnimenz ont blois et baucenz et noirs, Chevaus coranz a sorselles d’orfrois,

4930. C: puig, also found in C 2940, 7966, and also soig, C 4877. V7 5044 reads: poing. 4931. C: ceist, with missing ~; V7 5045: ceint. 4934. C: seurace. Insertion of ~ is an obvious emendation, but Foerster reads seurance in both C and V7, i.e. seürance, ‘protection’, whilst Mortier and Duggan read sevrance, ‘severance, separation’. V7 5048 begins the line with De maint paien…, which makes much better sense than C’s Devant paien (an obvious scribal error) and confirms sevrance as the more appropriate reading. 4935. C: Mahons sanz falance, with Mahons written in full, but this gives a hypometric second hemistich. V7 5049 reads: Mahomet sanz falance. 4936. C: soulement, presumably meaning ‘on their own’, ‘without support of allies’, but V7 5050 reads: follement, ‘madly’, ‘rashly’, so C’s reading looks like a scribal error, due to confusion of s and f. C’s misent is the Northern French variant of mistrent (found here in V7): cf. Introduction, 75 (viii), and Pope (1934, 490, § 1320, xxx). 4944. V7 5058 reads: Ci a bone entendance and ‘this is a good thing to hear!’ fits the context well; so C’s atendance may be a scribal error – but ‘this is a good expectation/prospect’, also makes good sense.

337

La Chanson de Roland (Critical Text and Notes)

4950

4955

4960

Escuz vermols, haubers deplentin noirs. Espees ceignent a bon brans vienois; De plene terre saillent sor les morois, Es roides lances les confenons desplois; Contre solel reluisent li plus frois. «Deus,» ce dist Karles, «qi sor toz as pöois, Bien est seürs qi de tel gent est rois! Seinte Marie, essaucez hui nos lois! Lai moi vengier Oliver le cortois, Rollant le preu et mes autres Franzois! Se je nes venge, je ne vesrai un mois. Meus amerai destruire ces Irois Q’en douce France boivre vin orlenois.»



4965

Charllemene apele Rabel et Guinemanz Et dist li rois, «Baron, venez avant! Per amistié et per bien vos comant, Soiez en leu Oliver et Rollant: L’uns port l’espee et l’autre l’olifant. Si chevauchez el primer chef devant; Ensanble o vos .xx. mille parissant,

76.b

4950. Foerster and Mortier both read this as: de Plentinois, but Langois does not identify Plentinois as a place-name and the name has no significance in respect of armour; Duggan reads V7 5064 as du Plentinois without further identification. On the other hand, deplentinois may arise from an accidental merging of deplentin and nois, with an n omitted in error: deplentin being a graphy of doplentin, ‘of double-thickness chain mail’, an adjective used to describe hauberks in C 1679 and 7939, and nois being a graphy of noirs (cf. garirs, C 1594, and note to that line), the ‘black’ of the hauberk mail being contrasted with the ‘red’ of the shields earlier in the line. 4952. C’s saillent sor morois is hypometric; V7 5066 reads: saillent sor les morois. 4954. frois is needed for the rhyme, but this is a graphy of freis, ‘freshly made’, thus here ‘the newest’. V7 5068 ends the line: lor corrois. 4960. C: je vesrai un moi, but both metre and sense require je ne vesrai…, as in V7 5074 (je ne verrai). The future 1 of veïr is surprising here; one expects vivrai, I shall not live…’. It seems to be used in the sense ‘to see out’ a period of time, but is perhaps simply a scribal error. 4963. This is an alexandrine in C, with Charllemene written, unusually, in full. V7 5077 simply reads: Karlles apelle. 4965. C: Pe amistié, but V7 5079 reads: Par amistié. 4969. C: Ensanble o lui o vos .xx. preïsant, with a hypometric second hemistich; but it is clear that more than 20 men must be involved. V7 5083 reads: .xx.mille parissant, P 3423: .xx.mille Franc, O 3019: .xv.milie…, so mille has been inserted. C’s preïsant, ‘presumptuous’ or ‘arrogant’

338 4970

4975

4980

La Chanson de Roland (Critical Text and Notes)

Tuit baceler et nobile conquerant. Et aprés vos en ira autretant, Qi avant aus iront tot detranzant; Sis guiera dan Richer le Normant; Ceste bataille n’ira pas cohardant. Ogier et Naym, en qi je me fi tant, Feront la terce, car preu sunt et vaillant.» Ses autres genz va mot bien ordenant. Se trovent ou, bataille feront grant. Nostre François en ont mot grant talant, De la bataille sunt forment desirant: Vengier voldroient Oliver et Rollant, E l’arcivesqe qi:s aloit semonant, Les .xii. pers q’il laisserent gisant En Rencivals, quant il furent passant.

 4985

4990

4995

Charlles li Maignes o la hardie chere Fu en Espeigne sor la gent pautonere; E .c.m. homes ot o lui per proisere, De ceaus de France et de ceaus de Baivere. Le jor i ot meinte riche banere, Meint confenon vermeil come rosere, E meinte lance de fresne tote entere, E meint osberc, meinte targe doplere, Maint bon ceval, meinte riche cropiere D’acer treslice deci c’al estrivere. E Baligans chevauche a grant enpere;

is clearly an error; it might perhaps be read as pre-issant, ‘going on ahead’, but scribal error in the transcription of a barred p seems more likely, so V7’s reading has been adopted. 4972. C: detrazant, but the verb is detranzer (detranchier), ‘to cut to pieces’; V7 5086: detrenchant. 4995–4996. As they stand in the ms., these two lines have a strange metrical pattern: C 4995 has a 6-syllable first hemistich, with cæsura after chevauche, then a 5-syllable second hemistich, a mot grant enpere (through confusion with enperere, so often written enpere with stroke through the p?); C 4996 has a normal 4-syllable hemistich, followed by another 5-syllable hemistich, vëoit la poudrere. The scribe appears to have been confused by the abbreviation .B. for Baligant, forgetting that this stood for a 3-syllable word and therefore mis-placing the cæsura. If one reads E Baligant as the first hemistich, then the second is easily emended by omitting mot; C then matches V7 5109: Et Balligans chevauce od grant enpiere. V7 5110 contin-

339

La Chanson de Roland (Critical Text and Notes)

De .xv. lieues vëoit hon la poudrere. Les oz s’aprosment, n’i a mester priere; Ainz q’il departent, i aura meinte biere.

 5000

5005

5010

5015

Carlle li rois trois batailles a jostees. Por conbatre sunt bien richement atornees D’osberc et d’eume et de broignes safrees, D’escuz vermeus, de lances peinturees. Les .vii. escheles ne sunt pas obliees: Namon comande qe soient devisees. «Volenters, sire, ja n’erent trestornees. Trois en avez richement ordenees; Ja vos seront ci les autres nomees. Faisons la quarte d’une gent desfaees; D’outre les monz les avez amenees; Alemant sunt, si ont les barbes lees; Tuit sunt locu, grans paus ont afublees. Cil fesront bien de lor trenchant espees, Deci as elz seront ensanglentees, En cui prendront paien tante colees, A duel moront per cez amples valees.»

77.a

 Naymes li dus apelle Joserant: «Traiez vos ça, si orrez mon talant: ues: De .xv. lieues veit hom la poldriere, and the insertion of the subject, hon in C’s usual spelling, emends C’s metre. 5002. C: peiturees, with ~ omitted. V7 5115 reads: penturees. 5003. The reference is to the remaining 7 divisions, since 3 of the 10 have already been organised (as is confirmed by C 5006). 5005. ‘They will never be turned back’, i.e. ‘they will never be defeated’. 5008. The word desfaees agrees in gender with une gent which, as a collective noun, is treated as plural. 5011. C: sunt grans paus afublees, with õt inserted in superscript before afublees, but no barring or expuncting of sunt, which presumably was repeated from earlier in the line in error. The sequence sunt… ont… matches the pattern of the previous line. 5013. C: ensangletees, with n or ~ omitted; V7 5127: ensanglentees, written in full. 5014. C: prenõt = prenont, but the verbs in this sentence are all in the future: fesront (C 5012), seront (C 5014) and moront (C 5015). The V7 5128 reading, prendront, has been adopted.

340

5020

5025

5030

La Chanson de Roland (Critical Text and Notes)

La quinte eschelle forniront li Normant; Fier s’i puet l’enperere puissant: Vint mile sunt, ardi et conbatant; Ja por morir n’en ira uns fuiant; Mot sunt prodome et vasal conquerant. Li sire d’eus va toz jor deffendant; Ja ne venra en estor si pesant, Ne s’en vost outre, qi q’en plor ne qi chant. Iceste eschille n’ira pas cohardant; Fier ci puez com pere en son enfant. Ce dist la geste, se:l trove l’en lisant, Ne fu tel genz des le tens Moïsant.» Nostre chançons va toz tens amendans; Jamais joclere de meillor ne vos chant.



5035

5040

La siste eschelle ont faite de Bretons; Trente .m. sunt, forz et fier as archons. Es roides lances ferment les confenons; Escuz ont painz de diverses façons, De lonz reluist li teinz et li blasons; Odes lor sire pere fu Salemons. Humais amende por voir nostre chançons; Puis qe Jhesus suscita Lazarons – Ce trovent clerc en livre de sermons – Ne vos fu dite plus veraie chançons.

77.b

5018. C: La quarte eschelle, but the composition of the fourth division has already been given in the preceding laisse (C 5008–5010). As laisse 284 deals with la siste eschelle, it is clear that the subject of the current laisse should be la quinte. V7 5132, P 3456 and O 3045 all read quinte here. 5024. C: passant, but the sense requires pesant, ‘weighty’, ‘grievous’, ‘bitter’, as in V7 5138. 5025. vost = vaist, pres.subj. 3, aler, thus ‘without going through’, ‘without winning through’. 5028. C: se:l trove l’on lisant, but the construction ‘so one finds it, reading’ is syntactically poor; V7 5142 has l’en lisant, ‘on reading it’, a standard grammatical construction. This may be Franco-Italian confusion of vowels, but ideally (though hypermetrically) l’on en lisant is required. 5033. ‘in the saddle’; archons is a graphy of arçons. Cf. C 1859, 2651, 2789. 5037. Both C and V7 5152 read: Odes est lor sire, a hypermetric hemistich due to insertion of est in error, when fu in the second hemistich is the finite verb. 5040. C: e livre; V7 5155: en livre.

341

La Chanson de Roland (Critical Text and Notes)



5045

5050

5055

Li rois de France se hasta durement; De sis eschelles a ordené ses genz. La seme eschele fera a son talent. Naymon apele tost et hastivement: «De cels d’Alvergne me ferez un present: A Baligant, qe li cors Deu crevent, Bien vengeront Rollant mien escïent Et Oliver a la dure talent, Les .xii. pers dont ai le cuer dolent. Cil m’en achat hui cest jor venjament Qi conpanssa el ciel lo firmament Et as apostres dona confortament Quant au terz jor leva del monument! Ce est Jocainne et Josue de Clarvent; Iceste eschile menrez vos bellement Sor cele gent qi ne croient noient Qe Deus eüst por nos batizement.» Karles les seigne de Deu omnipotent.

 5060

Aprés la seme ont l’oitisme bastie De ceaus de Flandre, de Frise la garnie. Chevalier sunt de mot grant seignorie; .xl.m. en ont en lor bailie; Ja devers aus n’ert batalle guerpie.

78.a

5043. C: De set eschelles, ‘seven divisions’, followed by L’otisme, ‘the eighth’, appears to indicate that a whole laisse has been omitted, for only 6 divisions have so far been organised. However, the next laisse begins, C 5060: Aprés la seme ont l’oitisme bastie; and V7 5158 here reads set, too, implying error in the common source. The reference must thus be to the sis divisions already organised; and in C 5043, C’s L’otisme (V7 5159: Lutine) must be read as a scribal error for La seme. Cf. O 3060–3061: Li emperere ad .vi. escheles faites,/Naimes li dux puis establist la sedme; P 3485: la sepme in a similar context. 5052. V7 5167 reads: compassa, the normal graphy, without C’s inorganic n (a Franco-Italian feature). 5061. C: De ceaus de France…, but V7 5176 reads Flandres, which fits better with Frise; O 3069 has de Flamengs, V4 3259: de Flandre and P 3506: de Flamains.

342

La Chanson de Roland (Critical Text and Notes)

5065

Ce dist li rois, «Cist ne targeront mie De bien ferir sor celle gent ahie, Que ne croient en Deu, le fil Marie, Ne q’il eüst por nos et mort et vie, Ne s’aünbrast en la virge Marie. Entre Ranbalt et Nemon de Galie Les conduiront per grant chivalerie, En cui verrez batalle ben fornie. Deci qe Meqe en la Mahomerie En ploreront li Turc d’Esclavonie, Quant en aront voire novelle oïe.»

5065a(V7) 5065b(V7)

5070

 5075

5080

5085

Entre Naymon et Joseran d’Elogne La nueme eschele ont faite sanz menzoigne O .xxx.m. chevalier de Sansoigne: Iront o Karle en icele besoigne. N’i a celui qi n’ait escu o broigne, E bone espée de l’ovre de Cologne. A plen s’adobent, or n’i a nulle aloigne; Desirant sunt d’aler en la besoigne Sor Baligant, qe Deus otroit vergoigne! Lors i vint Karles sor le vair de Boloigne; Si:n apela Girart de Terascoigne; Cil ert proudon, n’i valoit meins d’un mogne: «Conduisez cez, gardez n’i ait ensoine. Si ira o vos li dus Terris d’Argoigne.»

 La disme escelle ert des baron de France; Vint millers sunt a une conoisance.

5065. After this line, there is clearly an omission in C, for C 5066 refers back to God. V7 5181– 5182 fit the context exactly. The passage is found only in C and V7. It clearly derives from the CV7 model. 5070. C: verrrez, with extra r due to scribal error; V7 5186 reads: vesrez. 5087. The cæsura marked in C’s usual fashion after vos indicates that the 2 consecutive i are elided.

343

La Chanson de Roland (Critical Text and Notes)

5090

5095

5100

Cors ont ben fait et fiere contenance, Les chés floriz, meinte barbe ot blance; Haubers vestuz de l’ovre de Maiance, Espees ceintes ou il ont grant fiance, Espiés forbiz, li fer sunt de Durance; Lor s’arment tuit per mot grant aïrance; Es chevals montent, n’i ot plus de tarjance; Cil mostreront sor paiens lor puisance. «Monjoie!» escrient, chascuns forment s’avance; De la bataille ont mot grant desirance, En cui prendront paien male fermance E Baligans ert mis en tel balance Q’il maldira Mahon et sa creance.

78.b



5105

5110

5115

Li enperere de son cheval descent, Sor l’erbe vert s’est cochez esrament; Torna son vis tot droit vers orient, Recleme Deu mot escordablement: «Voire paterne, hui cest jor me defent, Qi garesis Jonas tot voirement De la balene et del mortel torment Qi:l transgloti, je’l sai a escïent: Trés jor i fu, se la letre ne ment; Tu l’en getas, sire, tot sauvement; Il s’en issi per ton comandement; Et espargnas lo roi de Ninivent Et sa cité et trestote sa gent,

5093. C reads ou ont g. f, which is hypometric due to omission of il. V7 5209 reads: o il ont… 5096. C: ni ot plus tarjance, which is hypometric; V7 5212’s: plus de tarjance has been adopted; but C often uses a single i where ii (ji) is needed, so this may stand for (tri-syllabic) tarjiance. 5099. C: De la balle, which makes no sense and is hypometric because the scribe has omitted a syllable; V7’s bataille is the obvious reading (V7 5215), although C often has the graphy batalle. 5100. C: prendont; V7 5216: prendront. 5114. C: Bonivent, but this is the Bible story of Jonah (found in the Old Testament book of the same name), which refers to Nineveh. V7 5230 reads: Niniment, O 3103: Niniven and V4 3292: Ninevent. Further details of the story are given in subsequent lines.

344

5120

5125

5130

La Chanson de Roland (Critical Text and Notes)

Et Danïel del lion al grant dent; Les trois enfanz garis del feu ardent; La toie amors me soit hui en present! Per ta merci, se toi plast, me consent Qe mon neveo puisse venger, Rollent, Et Oliver dont ai le cuer dolent.» Lors se dreça li rois cui France apent, Seigne son chief de Deu omnipotent. Monte li rois per grant aïrement Sor un cheval qi de corre n’est lent; L’ester li tint Nemes per bon talent. Prist son escu et sa lance ensement; Gent ot le cors, cler vis et fier talent, E puis chevauche mult orgoillosement. Sonent cil grasle et cil cor durement; Grant fu la noisse et l’esclartissement. Paien l’oïrent, cui li cors Deu crevent! Dist l’uns a l’autre, «C’est nostre jugement! Karles vient ci o trestote sa gent.»

79.a

 5135

Li rois de France ne se velt plus targer; Bien fu armez, cors ot grant et legier;

5116–5117. Two more popular Old Testament stories, both from the Book of Daniel, are referred to here: the casting of Daniel into the lion’s den, from which he came out unharmed (Chapter 4), and the casting of Shadrach, Meshach and Abed-nego (who were not, of course, children) into the fiery furnace, from which they emerged unscathed. Cf. note to C 4161 re the use of enfant in this context. 5118. C: La toie amors ne soit…, but the negative ruins the sense. V7 5234 and P 3586 read: me. 5120. neveo is a variant graphy of neveu (C 527, 2128, 3113, etc.), oblique form, singular: ‘that I may avenge my nephew’. .R. has been transcribed here as Rollent to suit the rhyme. 5122. C: Frace; V7 5238: France. 5124. C: Mote, with ~ omitted in error; V7 5240, P 3591 and V4 3300 read Monta, T 2704: Monte. 5131. The reading esclartissement is also found in V7 5247, but the normal word is esclarissement (‘brightness’). There has perhaps been scribal confusion with the word clarté (‘brightness’) or even, given the context in which violence and noise are apposite, with the the verb esclater (‘to splinter’) and the substantive esclateïs (‘loud noise’, ‘tumult’). 5132. C: Paient for paien, the scribal error arising from proximity of oïrent and the sequence of -ent rhymes in a long laisse. V7 5248 reads paien.

La Chanson de Roland (Critical Text and Notes)

5140

5145

5150

5155

345

Mot fu preudon, si ot visage fier. Par la ventaille fait les cordals sachier De sa grant barbe qe il ot fait trecer; Autretel firent des autres .vii. meiller: Meus s’en conoissent Alemant et Baiver. Li rois chevauche, q’il ne se volt targer. La veïssez tant paissons esrager, Tant paveillons, tantes tentes ploier, Et tant aucube et trosser et cherger. D’un val issi li rois et si princer; Mot ot grant duel, ne fait a merveiller, Trop a perdu, n’i ot nul recovrer. Desiranz est de son neveu venger Plus qe n’est dame de son ami baisser. Passent meint tertre et passent meint rocher, Passent les porz, chascuns sor son destrier, Et meint destroit qi tot furent plaisser. En un plein vindrent Franzois, li bon guesrer; La veïssez mainte tente drecer, Lo tré lo roi a l’aigle d’or saier.

5141. With so large an army, made up of men from distant parts of Charlemagne’s empire, there would always be the danger of attacking men on one’s own side, due to lack of recognition. Thus, identification of ‘friends’ was simplified by every Christian displaying his beard through the part of the hauberk which covered the lower face. 5145. C reads: cherger, V7 5261: charger. Franco-Italian confusion of e and a in C. 5153. The whole emphasis of this laisse is on the speed and urgency of Charlemagne’s return with his main army – and this is a major factor in making sense of the second hemistich. Having camped on the banks of the R. Ebro, the army was heading down-stream towards Saragossa, through a mountainous area whose gorges were all plaisser. This appears to be a graphy of the substantive, placer (sm), translated as ‘flat ground(?)’ in Godefroy – that is, as they travelled quickly, the narrow gorges all gave way to the plains. However, it may very well be a scribal error for laissiez (‘left behind’) or passez – or even placiaus, ‘peaceful’– with ending modified to suit the rhyme. The word seems to have posed problems of comprehension even in the thirteenth century, for V7 5269 reads: que Turc furent blassiez, ‘where the Turks were cut to pieces’. It is very tempting to read plaisser as a scribal error for parlaissiez, ‘left far behind’, with bar omitted from the p; to scan, the line must then either omit tot (‘completely’ or scribal error for tost, ‘soon’?) and read: …destroit qi furent parlaissier, the ending again modified by poetic licence – or retain tot and opt for an alexandrine with a first hemistich: Et meint destroit qi tot…, moving the cæsura. 5156. C: saier or sajer, probably a graphy of seoir, seier, ‘be situated’, ‘stand’ in this context. V7 5272 reads: rager, which makes no sense in this graphy; but if one relates the two readings, taking C’s initial s as a scribal error for r, then raier (‘to shine’, ‘to radiate light’) would make sense: the king’s tent with its gold eagle on top sparkles in the sun.

346

5160

La Chanson de Roland (Critical Text and Notes)

De l’ost Charllon torna un latiner – Sarazins ert, Deus li doinst enconbrier! Mist s’a la voie tot un chemin pleigner: Quant se fu mis per le plus droit senter, Ne s’i tenist en corant un levrer.



5165

5170

Li Sarazins de l’aler mot s’avance, Descauz, nuz piez, vermeus come garance. A Baligan est venuz sanz tardance; Il le salue de la soie creance: «Mahomet, sire, vos doigne grant puissance! Novelles sai, or oiez ma sanblance: Sor vos eschape l’enperere de Franche. Il a passé les destroiz de Bruance; En la montaigne sunt longié sanz falance. Faites armer vos genz por conoisance, Car la bataille arez vos a fiance; Il vient venger son duel et sa pesance.»

79.b

5161. ‘A hare would not have kept up with him in running’, i.e. ‘he ran faster than a hare’. C reads: tenis; V7 5277: tenist, the normal imperfect subjunctive 3. 5168–5173. The Saracen spy’s report in these lines is very confusing. On the one hand he says: Sor vos eschape l’enperere… (‘the emperor is escaping from you’), on the other, in C 5172–5173, ‘You will certainly have to do battle (with him); he comes to take revenge for his grief’. It seems one must read C 5168 as meaning ‘the emperor . is eluding you’, with sor vos in its usual sense of ‘against your will’ or ‘against your wishes’, i.e. ‘which is not what you want’. (Unfortunately, it has not proved possible to identify les destroiz de Bruance. If it is a reference to Braganza in N-E Portugal, it shows a better knowledge of the mountainous terrain there than of the distance involved.) This episode is given much more succinctly in O 3130–3134, but it is not found in PTL. In C and V7, it is designed to provide an opportunity for the poet to describe Baligant’s preparations and the ordering of his army, i.e. a set-piece description to match that of Charlemagne’s army (laisses 280–288), his speech (289) and action (290). 5170. C’s ms reading, sunt lor gré, appears to mean ‘in the hills are their pleasures’ i.e. ‘. their preferred positions’. Reading lor gré as a scribal error for longié (V7 5286), ‘they have gone away into the mountains’, may also indicate a tactical move to higher ground. Alternatively, the CV7 source reading could have been logié (V7’s longié with inorganic n removed): ‘they have taken up position on the high ground’ warns of a real threat to Baligant’s army. Or one might read lor guie, ‘their guides’ or ‘their scouts’ (‘their outriders?’) as the subject of sunt – an equally serious threat. 5173. C: Il vient vient venger.

347

La Chanson de Roland (Critical Text and Notes)

5175

Ceste parole mis paiens en esrance; Al plus hardi en mua la scenblance.



5180

5185

5190

5195

5200

Par tote l’ost font lor grailles soner, Cors et busines qe l’on öoit si cler; Paien s’en tornent por lor cors adober. Li amiranz n’i volt plus demorer: Vest une brogne qi fu faite sor mer, Puis lace un elme qi mot fist a proiser. L’espee ceint dont li poins reluist cler; Per son orguel, li volst un non trover, Por la Charllon dont il oï parler; La soie fist ‘Preciose’ apeler: Ce est s’enseigne en bataille chanper. Ses chevaliers a toz faiz conréer, Et lui meïsme, car mot se vel haster. A son col pent un fort escu bocler; D’or est la bocle et belle a esgarder; La guige en fist de paille ben friser. Prist son espié q’il apela ‘Mater’; L’anste fu grosse, ne vos sai deviser, Car sol le fer ne puet un hon porter. El destrer salt, mot se velt aloser; L’estrier li tindrent Merguileis et Romer. Mot ot gent cors et le viaire cler. Deus, or vos puis bone chanzon chanter D’une bataille qi tant fist a doter! Li rois de France ne se velt arester; A plen chevauche, avec lui sunt si per; Tant sunt venu, por voir le puis conter, Q’il s’entrevoient as confenons ben cler;

80.a

5174. Once again, the past definite 3, mis, without the final t, intermittently a feature of C: cf. Conquis, line 3, and further examples in lines 3231, 3400, 3408. V7 5290 reads: mist. 5180. The byrnie seems to have been made ‘overseas’: sor mer; but this may well be a scribal error for d’or mer, ‘of pure gold’ – which would stress the wealth of Baligant. Although V7 5296 also reads sor mer, P 3634–3635 supports this hypothesis to some extent: …qui moult fist a loer:/Trestouz les pans ot fait a or ouvrer. 5193. C: L’auste, C’s usual graphy of anste; V7 5309 reads: L’aste.

348

5205

La Chanson de Roland (Critical Text and Notes)

Nus hon de char nes poïst pas nonbrer. Je ne croi mie q’il en poissent turner Q’en soi la perte legere a recovrer.



5210

5215

5220

Li amiraus resanbla ben baron: Blanche ot la barbe ausi com auqeton. Nus de sa loi ne fu se sages hon; Et en batalles hardi come lion. Ses fius Malprimes i ot mot clere façon; Grant fu et forz, mot sist bien sor l’arzon; Dist a son pere, «Chevauchez a bandon! Grant poür ai qe ne trussez Karllon.» Dist Baligans, «Mot est Karles proudon. En plusors gestes est de lui grant raison. Il n’a or mie de Rollant li baron Ne d’Oliver, son hardi conpeignon; Il a perdu Berenger et Athon, Si a perdu et Ivoire et Ivon; Qen q’il poet faire, ne pris pas un boton.»



5225

«Baus fius Malprimes,» ce li dist Baligant, «Paien ont mort le preu conte Rollant Et Oliver, lo hardi conbatant, Les .xii. pers qi Karles amoit tant; Tot lor esforz ne pris je mie un gant. Li enperere est ja ici devant;

5204. C: nobrer, with ~ omitted in error. 5208. C: ausi com un auqeton, which is hypermetric, but the un is superfluous. V7 5324 reads: ausi come aqueton, O 3173: ensement cume flur and T 2762: autrecy come flour (although P 3669 has ausiz com une flor). 5209. C: se sages non, but V7 5325 reads: hon, O 3176: hom and V4 3360: om – ‘such a wise man’.

349

La Chanson de Roland (Critical Text and Notes)

5230

5235

Je voi la noisse, si com vont hon disant. Dis grant escheles a faites de sa jant; Ce me dist ore uns mens garzons esrant. Ben sunt .c.m. qe tuit sunt d’un scenblant, Tuit bacheler, preu sunt et conqerant. Baus fius Marprimes, forment m’en espoant; Grant poor ai, por mon deu Tervigant; Trestuit li menbre m’en vont ja fremisant.» Ce dist Malprimes, «Mar doterez noiant! Demein arez un eschac issi grant; Ainc Sarazins n’ot onqes tant vaillant.» De la bataille le primer cop demant.

80.b

 5240

5245

5250

«Baus fius Malprimes, Mahons vos beneïe! Mult estes proz et sages sanz folie, Beaus et cortois, plens de chivalerie. Je vos otroi le don de l’envaïe, Le primer colp de la grant ost banie. O vos menrez tant de ma conpagnie, Mar crembrez home qi encor soit en vie. O vos ira rois Tulles de Persie E Capamor, l’amiralt de Lerie. L’orguel Karllon a la barbe florie Amatirez ainz l’ore de conplie, Et l’olifant, qi a si grant bondie, Qe Karles ot en la söe bailie,

5228. The verb veïr is once again used with ‘noise’ (as in C 3068) – perhaps with the meaning ‘I witness the noise’. It would be possible to emend this to: J’en oi la noisse, assuming n misread as u by the copyist. V7 5346 has a different reading: J’en oi…si corz vont bondisant: ‘I hear the noise of it; his horns go ringing out’; but this also implies a mis-reading in C or the CV7 model of cors as com and bondisant as hon disant. Nevertheless, voi la noisse seems to be a standard colloquial phrase, and the fact that, in C, it is followed by ‘as people have a habit of saying’ gives this view validity, for it shows that the scribe (via Baligant) recognizes the oxymoron inherent in the vulgar idiom using voi. Doubt must therefore remain as to whether V7’s line was the source reading; it could very well be the V7’s copyist’s logical emendment of what he saw as incomprehensibly idiomatic or a scribal error. 5233–5235. These lines directly contradict C 5221 and 5226, in which Baligant says he is not at all concerned about Charlemagne and his army.

350

5255

La Chanson de Roland (Critical Text and Notes)

Conqerez vos, Mahomez vus l’otrie! Je vus donrai un pan de paienie, Terre au roi Floire, qi tant a seignorie.» Son destre gant enz en la mein le plie; Cil le reçoit et forment le mercie; A tel eür l’a iloc recollie, C’onqes nul jor n’en joï en sa vie.

 5260

5265

Li amirals chevauche per ses sorz; Ses fius le suit, qi mot a grant esforz. Li rois Turleus et li rois Capamorz Granz .xxx. eschilles establissent as porz, Forz et pleigneres, qi josterent as noz; En la menor .xv.m. per moz.

81.a

5253. C: Mahomez le vus otrie, a hypermetric second hemistich. This might be due to scribal error in writing Mahomez rather than Mahons, but it may readily be reduced to 6 syllables by reading Mahomez vus l’otrie, changing the order of pronouns on the model of V7 5371: Mahomet vos le die. 5257. C: Cil le reçoit with insertion mark beneath the superscript re. 5258–5259. ‘By such (good) fortune…’, ‘By such fate…’ or ‘Under such circumstances did he then receive it, that he never did enjoy it any day of his life’, i.e. Malprimes never lived long enough to enjoy this territory, freshly given to him, for he was killed that very day. Cf. O 3212: A itel ore…’. 5260. Both C and V7 5378 read: ses sorz; O 3214 reads: cez oz, ‘these armies’. P 3734 has: par effors, although P’s version differs considerably from C, V7 and O in this laisse. Similarity of s lunga and f makes emendation to s’esforz, ‘his army’, a simple and logical step, but the fact that the same words occur in C 5281, V7 5399 may indicate that there is no scribal error. Also, C 5261 ends with esforz (there meaning ‘strength’), making the use of the same word here unlikely. As sorz is found twice in both C and V7, it has been retained as either a graphy of soürz, seürz, meaning ‘loyal (warriors)’, ‘trusted men’ or possibly a reference to the reddishbrown skin-colour of his soldiers (i.e. the shade known as ‘sorrel’, normally applied only to horses). Uncertainty remains, for this could also be a scribal error for cez oz (O 3214) or possibly for ses forz (‘his strong men/mighty warriors’). 5263. C: Granz .xxx. lieues eschilles, showing how easily copyists slipped into familiar phrases. 5265. Both C and V7 5383 read per moz, C using the symbol p with bar through tail for per. Per may be the preposition, meaning ‘by’, or the adverb intensifying the force of the following adjective. Here, moz has been read as the plural of mot/molt/moult, ‘many’, so that per moz has been taken to mean ‘extremely numerous’ or ‘in great throngs’. However, if moz is read as the plural of mot, ‘word’, the implication may be ‘according to report’. O 3219 reads: En la menur .l. milie en out, so scribal mis-reading or the requirements of rhyme may have led to en out > mout > mot > moz. V4 3404 ends the matching line: e os, which Beretta sees as the result

La Chanson de Roland (Critical Text and Notes)

5270

5275

5280

351

La primere est de cels de Boteroz, Dont Judas fu, qi fel estoit et oz, Qi Deu vendi, mot fu musarz et soz. E l’autre aprés de Mont Nigre les Torz, Qi les eschines orent permi les dos; Grondissant vont come porcel en vos. Et la terce est de Nubles et de Bloz; Et la quarte est d’Escoz et d’Esclavoz; La quinte est des Sarazins de Goz Et la .vi. est de cez Ermines forz. Bone est la seme de cels de Ificoz; L’uitime d’Anage et la nueme de Noz; E la .x. est de Baile et de Gloz – C’est une gent, ja ne fuira poroz, Anz esseront et detrencé et morz. Li amiras en a juré ses sorz De Mahomet les vertuz e les moz: «Carlle de France chevauche come soz.

of similar corruption. However, os (found also in C 7436) is a Northern variant of els, eus, ‘them’. Thus perm’oz, with os modified for the rhyme, would be an elided form of permi oz, ‘amongst them’. 5270. ‘Who had their spines through the middle of their backs’ seems to imply some kind of protruding ridge of bone down the back; but the CV7 source has probably omitted a line here, for V7 5388 is identical but V4 3408–3409 reads: Sur li escine ch’i à in mé li dos/Si à sede altersi come pors, ‘On the spines which are in the middle of their backs, there are bristles just as pigs have’. 5271. C: e nos (or et nos, since the abbreviation is used); V7 5389 reads: en vos (= en vois, modified for the rhyme), ‘in voice’/’in their voices’. This points to C’s omission of u = v from the combination en os and mal-comprehension. (However, reading evos, adj., ‘of the water’, ‘of the sea’, gives the correct rhyme and would give porcel evos, ‘sea pig’ or ‘porpoise’ as an alternative source reading). 5279–5280. ‘They are a race who will never flee in fear; sooner they will be cut to pieces and slaughtered.’ 5281. C: ses sorz, as in 5260, but in a different context: juré ses sorz. This is apparently the indirect object, ‘swore to his trusty warriors…’. The same reading is found in V7 5399; and the only corresponding line is O 3232: Li amiralz en juret quanqu’il poet (De Mahumet…), i.e. Marsile affirming to his men with an oath the power of Mahomet, identified as their god. 5283. C: Calle; V7 5401: Karle. C’s second hemistich reads: chevauche sorz, with sor apparently emended by the scribe to soz, but still left hypometric. V7 reads: chevauche come soz, ‘rides like a fool’’.

352

5285

La Chanson de Roland (Critical Text and Notes)

Se il m’atent, et moi et mon esforz, Jamais si home n’aront per lui conforz.»



5290

5295

5300

Grant dis eschilles establirent as prés: La primiere est des Orquenois irés, La seconde des Solteins et des Rés, La terce est des Properté divers, La quarte est d’Orvalois les engrés, Et la quinte est de la gent Samués; La siste eschelle est au roi de Rohés Et la .vii. est au roi de Mont Pantés; L’uime d’Orbrise, la nueme d’Esclavés, Et la .x. est d’Olchéan des desers; C’est une gent qi est pire qe Sers; De plus felons n’orrez parler jamés; Durs ont les cuirs autresi come fers; Ja n’aront soing ne d’escu ne d’osbers; En la batalle sunt felon e engrés; Cil les maudie qi dona Moïsés La seinte loi dont nos tenons la pés!

81.b

 Baligans fu de grant esforcement; Mot ot o lui paiens estrangement. 5284. The meaning is not clear here: if atent is read as present 3, atendre, then ‘if he waits for me’ implies that Charlemagne is running away – and there is confusion earlier in the text (cf. notes to C 5168 and C 5170) over this idea; on the other hand, atent could also be read as a graphy of ateint, present 3, ateindre, ‘to reach’, ‘to attack’. 5286. C: aspres, with the first s in superscript. The scribe first wrote aprés, ‘afterwards’, then inserted s in the same ink, so presumably this is his own emendation to as prés, ‘in the fields’. V7 5403: aprés. 5294. As Mortier points out, the reading may be d’Esclavés, de Sclavés or des Clavés. 5295. C: d’Olchan des desers, which is hypometric. V7 5413 reads: d’Olcéans des desers, so the extra syllable has been inserted. 5299. C: sog; V7 5417 reads: soing. 5302. The Old Testament book of Exodus, xx, recounts how God gave Moses the Ten Commandments. 5304. C’s reading estragement may well be a graphy of estraiement (sm) ‘learning’, but V7 5422 reads: estrangement (written in full), a graphy of estraignement – either the substantive

La Chanson de Roland (Critical Text and Notes)

5305

5310

5313a(V4) 5315

5320

353

Dis grant eschilles a stabli esrament: La primere est d’une hardie gent – Il sunt Jaiant de Valproissié la grent; L’autre de Huns, la terce de Hugrent, Et la .iiii. est d’Albanie et de Kent; Et la .v. est de cels de Val Bruient, Et la .vi. est de Marmoise et d’Aiglent, Et la .vii. est des Nors d’otre orient; L’uime d’Erabe, la nueme d’Abilent E la .x. e de barun de Valfent – C’est une terre o solauz ne s’estent. Ne Deu ne croient, lo Paire omnipotent; Ne pas ne qident qe il aünbrement Preïst en feme por sauver bone gent. Li amiraus chevauche liément O tot .c.m. qe li cors Deu crevent! Contre Charllon qi el mont les atent. Cil cor bondisent et menu et sovent; Les oz s’aprosment, n’i a demorement. Ne remandra, se la geste ne ment, Qe .c.m. homes ne soient ainz sanglent.

(m) ‘serried ranks’ or ‘grinding or banging noise’, or the adverb, ‘extraordinarily’ – ‘in extraordinary numbers’ or ‘of extraordinary appearance’. This last seems the most apposite to the context. 5313. After this, a line is obviously omitted in C and V7 (and in P, L and T), since the tenth squadron is not listed. O 3260 here reads: E la disme est des barbez de Fronde and V4 3430: Et la .x.a e de barun de Val Fonde, the place-name in each case providing a suitable antecedent to which C 5314 can relate. V4’s spelling of Fonde has been modified slightly to suit C’s -ent rhyme. 5314. The word s’estent is a graphy of s’esteint, present 3, esteindre, verb reflex., ‘to extinguish itself’- i.e. this is a land where the sun does not set. This is confirmed by V7 5432 reading: descent. 5315–5217. Ne… Ne…: ‘They neither believe in God… Nor that he became incarnate…’. The first Ne may, however, be a scribal error for En (.believe in God) arising from anticipation of the second Ne immediately below; V7 5433 reads Ne croient en Deu… As the reading stands, the sense requires aünbre nïent and the scribe appears to have written aünbrement in error, but V7 5434–5435 read: …que il aümbrement/Preïst en feme por salver bone gent. C 5317 has Puis eïs a enfer por sauver…, the copyist completely misunderstanding the sentence and confusing the Incarnation of Christ with His Crucifixion, followed by the Harrowing of Hell (cf. note to C 3810–3811). The latter has been the pattern in all the prayers and invocations earlier in the text. 5320. C: mot, with ~ omitted in error; mont, as in V7 5438, is clearly needed. 5324. C: homes ne soient.

354

La Chanson de Roland (Critical Text and Notes)

 5325

5330

5335

Baligans fu de mot fiere raison; Mot per fu beaus et de clere façon. Dedevant soi fait porter un dragon – E l’estendart Tervagan et Mahon – Et une image d’Apolin le felon; Trente chamels chevauchent environ. Mot autement escrie en son sermon: «Qi par nos deus velt avor garison, Il prit et serve per grant afliccion!» Hui en mosront .xx.m. en cel sablon. Li nostres Deus soit garant a Karllon! Ceste bataille fu voée en son non.

82.b



5340

5345

Li amiraus fu mot de grant savoir; Deus rois apele et son fil q’ist son oir. Dist Baligans, «Chevauchez a pöoir, Et mes conpaignes guiez per grant savoir! Mais des mellors retendrai jo l’espoir, Trois grant eschilles qi feront lor pöoir: L’une ert de Turs et l’autre ert d’Ormanois, Et la terce est des paiens de Brunsoir; Li sire d’als ert isdeus et mot noir, Cil d’Ocean qi ne crent mal espoir. Cil josteront anzois demein a soir

5334–5336. These lines in C (which V7 matches exactly) make sense if read as a scribal comment, but it appears that they were not originally intended as such. C and V7 omit 4 lines, found in very similar form in O (O 3273–3276), V4 (V4 3443–3446) and P (P 3786–3789). O’s reading is: Paien i baissent lur chefs e lur mentun,/Lor helmes clers i suzclinent enbrunc./Dient Franceis: «Sempres murrez, glutun./De vos seit hoi male confusïun!/Li nostre Deu, guarantisez Carlun!/Ceste bataille seit vüee en sun num.» – making the final two lines part of the speech by the French and not the writer’s comment. In this ‘Baligant episode’, CV7 consistently derive from a different remaniement from that of OV4PTL. 5338. C: Deus rois apele…; this is followed later by qist… as an abridged form of qi est, ‘who is’. 5341. C: les pöoir, which makes the 2nd hemistich hypermetric; pöoir is found just above (a pöoir, end of C 5339) and also immediately below, at the end of C 5342. The word needed is probably l’espoir (‘the expectation/eagerness’), as found in V7 5459.

355

La Chanson de Roland (Critical Text and Notes)

A l’ost Karllon, qi est venuz ardoir Tote ma terre, ce qide ben por voir.»

 5350

5355

5360

5365

Grant sunt les oz de la paiene gent Et Baligant chevauche fierement, O ses eschelles mot ordenéëment; Contre Franzois chevauche iréëment. («Por Mahomet, ferez i durement! Mot sui honi, se il ne s’en repent!») Et Karllemene et sa gent les atent. Je:l di por voir, se l’estoire ne ment, Ainz q’il departent, i ara meint dolent. Prés a prés sunt; qe Deus ait nostre gent! E Baligans s’escrie hautement: «Mi Sarazin, chevauchez liément Contre Karllon, o douce Franche apent. Por Mahomet, ferez i durement! Mot sui honi, se il ne s’en repent! M’enseigne port Auberis fer-talent. Neiz d’Oliferne, mot a le cuer valent.» Cil prist l’enseigne, ne se tarze nïent; Paien reclement «Preciose!» sovent; Des or chevauchent tot arotéëment.

82.b

5354–5355. These two lines are inserted here through scribal error, then repeated at the correct place, C 5363–5364. V7 does not make the error of inserting them too early. 5356. C: Et .Kll’n.’, which has to be read as Et Karllemene to obtain a regular 4-syllable hemistich. This implies that .K. in lines 7962, 7981, 8000, 8019, 8035 and 8132 should also be transcribed as Karllemene in order to obtain a 6:6 metre in these lines, which occur in a long passage of alexandrines (cf. Introduction, 49–50, 57–58 and Appendix B, 97, and notes to C 4209, 4371, 4864). 5366. C: mot a le cuer dolent, but this is a surprising comment, relative to Auberis’ being given the honour of carrying the banner into battle, a source of pride rather than sadness. (C 5586 and 5961 read mot ot le cuer dolant, though 2181 has …cuer loial and 4918 mot a le cuer joiant.) V7 5466 reads: mult est proz e valent, P 3826: s’en ot le cuer joiant, but T 2876: n’en eut le cueur dolent is closest to C’s reading. These readings imply that a scribal error (dolent) in the source has been emended in a variety of ways. V7’s valent offers the simplest emendation. In this case, neiz is a graphy of nais, nom.sing., naïf (sm) ‘inhabitant’, the diphthong being contracted into a single syllable: cf. traïtre, Introduction, 92–93, and Pope (1934, 109, § 240). 5369. C’s Desoz could possibly relate to ‘under the banner’ (C 5367), but the intervening line makes this less likely and V7 5485 reads: Des or, ‘from now on’, which has been adopted.

356 5370

5375

La Chanson de Roland (Critical Text and Notes)

Meint confenon contre le vent s’estent. Franzois les virent, qe Deus gart de torment! Dist l’uns a l’autre, «Grant joie nos atent.» Li rois de France fait soner esrament Plus de mil grasles por enhaiter sa gent; Contre paiens chevauche fierement. Ne remandra, se cil qe dist ne ment, Liquel qe soit ne s’en partent dolent.



5380

5385

5390

5395

5400

Grant fu la pleigne et large la contree E li paien chevauchent la feree; Luisent li haume, mot fu grant l’asanblee: De Sarazins, de cels de Val Troblee, Cent .m. furent en la primere hee. La veïssez meinte enseigne levee, Meint confenon, meinte sele doree. Li amirals fu de grant renomee; Son frere apelle, Carmilleu de Dorree – Rois ert de Turre, d’une terre esfree; Cil tint la terre deci en Valsevree. L’enseigne Karle li a al doi mostree: «Vez la l’orguel de France la loee! Mot fierement chevauchent per la pree; Fors des ventalles meinte barbes ot getee, Autresi blanche come nos sor gelee; Meint cop de lance i feront et d’espee; Batalle avrons et fort et aduree, Unqes mais princes ne vit tel ajostee!» Plus c’on ne trait une fleche enpenee A Baligant sa conpaigne sevree; Une raison li a dite et mostree: «Venez, paien, car je sui en l’estree!» De son espié a la hanste levee,

83 a

5396. C: tel aduree, but this makes little sense and aduree occurs in the same position in the preceding line, which makes it suspect; V7 5512 reads: ajostee and O 3322: ajustee, ‘joined’ (of battle).

La Chanson de Roland (Critical Text and Notes)

357

Contre Karllon a la more tornee: Ce senefie q’il l’en donra colee.

 5405

5410

Charlles li rois, quant il vit l’amiralt E le dragon q’il porterent en alt – La grant enseigne ou ot peint un girfalt – Et cels d’Arabe et les forz de Duralt, Karles s’escrie, si fist avant un salt. Naymon apele et le ber Clarenbalt: «Veez, seignor, prés somes de l’asalt! Tote lor lois un seul diner ne valt. S’il ont grant gent, dehaiz ait qi en calt! Qi or n’i vient, de desoz moi s’en alt!» Franzois escrient, «Nus de nos ne vos falt!»

 5415

5420

Beaus est li jors e li solaz luisanz; Li oz sunt beles e les conpeignes granz. L’eschelle joste al vasal Guinemanz Li cons Ranbaus, qi prouz est et vaillanz. De l’autre part chevauchot Baliganz, Enscanble o lui .xxx.m. Persanz. Prés sunt de joindre, Deus soit as nos aidanz! Laschent les resnes des bons chevaus coranz: La veïssez tant confenons pendanz, Tantes enseignes contre venz baloianz;

5407. C: le forz, with final s omitted from the article; V7 5523: les forz. 5410. C: asaut with l added superscript between a and u, but asalut produces hypermetry. V7 has: asaut, but C needs asalt for the rhyme. The scribe or revisor has realized this, but omitted to expunct the u. 5413. Both C and V7 (5529) read: de desor , but ‘from above/over me’ does not make sense when Naymon is speaking – and there is often confusion of desor and desoz in both C and V7. That he is referring to those under his command is shown by the reply of the French lower ranks in the next line. 5419. C: chevauchot .B., the -ot verb-ending probably being a form of the imperfect 3 -oit. Alternatively, this may be due to careless copying of chevauche. V7 5535 simply reads: chevauche.

358 5425

La Chanson de Roland (Critical Text and Notes)

Les chevaus brochent des esperons tranchanz, ‘Monjoie!’ escrient de nos les plus vaillanz; Ses vont ferir de lor espiez luisanz.

83 b



5430

5435

5440

Li cons Rembauz fu chevaliers hardis, Sist el cheval c’on apelle Tardis: Plus pöoit cosre per puis et per lairis Qe li Rollant qi ot nom Valentis – En Renchivals fu desoz lui ocis Et il avec, Karles en fu pensis Et douce France en fu povre a toz dis. Ranbiaus le broche; icil li soit aidis Qi d’enfer a geté les siens amis! Fiert en l’escu Turleus, un roi persis; L’escu li piece, l’osberc li a malmis, L’espié aor enz el cors li a mis, Qe mort l’abat, ne li pot faire pis. Ranbiaus escrie, «Monjoie saint Donis! Ferez, Franzois, ja n’en ira uns vis!»



5445

5450

Guinemanz joste a un roi de Leurie; Fert l’en l’escu o li ors reflanbie, Desor la bocle de la targe florie; Li est la broigne ronpue et desarcie; Tote s’enseigne permi le cors li guie, Qe mort l’abat, qi q’en plor ne qi rie. A icel cop, li nostre rois s’escrie, «Ferez, baron, ne vos atargez mie! Je ai grant droit; tort a la paienie. Il m’ont tolu tant de ma conpaignie, Dont douce France est lasse et apovrie.

5436. The cæsura is marked after enfer, i.e. after 3 syllables, leaving a 7-syllable second hemistich. To end the first hemistich after a would entail splitting the compound verb (a: geté). V7 5552 has exactly the same line as C. The allusion is again to the Harrowing of Hell (cf. note to C 3810–3811).

359

La Chanson de Roland (Critical Text and Notes)

5455

Rollant ont mort, mon neveu, per envie, Et Oliver a la chiere hardie. Ce m’a fait Guenes, qe Damedeus maudie! Hé, douce France, jamais n’avrez aïe Des doçe pers qi ont mort recolie!»

 5460

5465

5470

Malprimes sist sor un cheval corant, Sa gent conduit, la presse va ronpant; D’ores en altres i va grans cols donant, L’un mort sor l’autre per terre trestornant. Toz primereins s’encoche Baligant: «Li mien baron, Mahons vos soit garant! Vez la mon fil a l’orguellos talant! Il va Karllon por la presse qerant. Mot li siet bien li haubers jacerant Et cil escu et cil eume luisant; N’i a François qi ne s’en espoant. Mellor vasal ne sai de lui vivant. Alez aprés, baron, je:l vos comant!» Dur sunt li cop et li fer sunt trenchant; La bataille est merveilouse et pesant; Ne fu si forte des le tens Moïsant.

84.a

 5475

Grant sunt les oz de la gent desliee; Baligans joste ses genz et sa masnee. N’i a eschelle ne soit bien adrecee; Franc et paien sofrent mot grant haschee. Deus! Tante lance i ot le jor brisee,

5463. C reads: se choche and V7 5579: se colce, but it is clear from the context that Baligant does not lie down. O 3373 has: s’escriet to introduce Baligant’s speech, and Duggan emends V7 to s’escrie; but a verb of speaking is not essential. However, if one assumes a missing ~ in both C and V7, then s’enchoche and s’encolce may possibly be read as variant graphies of present 3, s’encalcer/s’encaucher (the graphy enchauch- is found in C 4809), here meaning ‘presses forward eagerly’, ‘spurs forward’, which fits the context nicely. 5478. C: Fanc in error for Franc; hascee with h added superscript over the c.

360 5480

5485

La Chanson de Roland (Critical Text and Notes)

Tant mort gesir dont la terre est jonchee; L’erbe del canp, qi ert verte et dolgee, Des sans des Turs estoit envermellee. Li amirals escrie a sa masnee: «Ferez, baron, sor cele gent changee! Se ge puis vencre, ma lois ert esaucee! Tote m’amor vos est ci otriee.»



5490

5495

Li enperere reclama ses Franzois: «Qe faites vos, Normanz et Hurepois? Mot voi baié cele gent d’Orlenois! Ferez i bien, traiez des ars turcois! En Rencevals gisent mort mi Franzois; Paien ont tort, n’en sui mie en esrois.» Quant Franzois l’oent, donc ferent demanois En la grant presse, ou trovent plus espois. ‘Monjoie!’ escrient Angevin et Tiois, Et cil de France qi sunt preu et cortois. Qi vasaus fu, n’i valt sa lance un pois, Tost en a fait ou deus tronchons ou trois; Lor fu li chaples as espees manois:

84.b

5480. The infinitive gesir (found also in V7 5598) is here the gerund, so meaning ‘so many lying dead’. 5481. C: verte dolgee, but V7 5599 reads: verde e delgee and O 3389: verte e delgee. 5484. V7 5602: gent cangee, supports C’s gent changee, but P 4095 and T 3015 both read: la guerre est commencie. Certainly changee poses a problem. It should probably be taken as meaning ‘made different’, i.e. ‘different from us’, ‘alien’. On the other hand, V4 3559 reads: sor la gent cristienté – perhaps a scribal emendation of a difficult source reading, but possibly a clarification of the meaning implicit in changee here: that is, ‘changed’ in the sense of ‘converted’, thus ‘Christian’ or ‘redeemed’. 5492. C: n’en suz mie…, but sui, ‘I am’, is clearly intended and found in V7 5610. Both C and V7 end the line: en espois, but this makes no sense. As espois is found at the end of C 5494, only 2 lines below, here it is probably a scribal error. If it is read as a graphy of espoir, ‘hope’ or ‘judgement’, the statement becomes contradictory. The line is not found in PTL, but O 3413 reads: Ja savez vos, cuntre paiens ai dreit, and V4 3580 echoes this. In the context, the correct reading appears to be en esrois, i.e. ‘I am not at all in error’ or ‘… in doubt’, an affirmation of Charlemagne’s faith. The alternative reading, en esfrois, ‘frightened’ or ‘troubled’, with fr misread as p, is less convincing. 5493. C: Quat with ~ omitted from C’s normal spelling of quant.

La Chanson de Roland (Critical Text and Notes)

5500

361

Sor Sarazins en torne li sordois. Qarante et .c. entre princes et rois En gisent mort permi cez Jaünois.



5505

5510

5515

Li proz Malprimes sist sor un ros cheval; Plus cort par pui c’autres ne fait per val; S’enseigne escrie a guise de vasal: «Ferez, paien, q’il ne vos tort a mal!» Lors se desroie le travers d’un costal; La lance droite, ainc ne vost prendre estal. Naimes lo vit, qi lo cuer a loial, Vait le ferir en l’escu a esmal. Desoz la bocle un pertus li fait tal, Paser i puet une piere pognal; L’auberc li trance et le pan de cendal, Permi le cors, c’onqes n’i ot estal, Li mist la lance au fer juridical;

5502. C: jaunois, read as jaünois to avoid an hypometric hemistich; V7 5620 reads: joanois, also tri-syllabic. This is confirmed by the graphy Johannois or Johenois in Foucon de Candie which Moisan (1986) cites alongside the C-V7 readings and identifies as ‘a Saracen people’. The name may be derived from Gehene, meaning ‘Gehenna’, either as the geographical location (Hebrew Ge Hinnom, the valley of Hinnom, a deep narrow gorge a few miles SW of Jerusalem) or as a synonym for Hell (because human sacrifices had been made to Moloch there). The -ois termination would imply ‘inhabitants of’. The graphy must have been found in an early source text, for it puzzled other copyists: P and T paraphrase and modify this passage, referring only to ‘7,000 French of goodly youth’ (bel jouvente, P 4135; belle juvente, T 3044). No recognizable common noun corresponds to jaünois/joanois unless this is a graphy of jovencel/jovanceus or, as Duggan suggests, javelois, ‘piles’. 5515. Niticoral, graphy of niticorax, ‘night-jar’ or ‘night-heron’, makes no sense here, even as the name of the lance. Duggan (2005, 31, § 8) has modified the second hemistich to au fer jusqu’al coral, but this does not form a logical sense unit – and it is unsatisfactory to split the line 6:4 in an effort to accommodate it: the final 4 syllables need to be adjectival, applying to the fer, not the lance. It should be noted that, in both C and V7 5633, the preceding line has n’i in precisely the same position, giving grounds for mis-reading. More significantly, confusion of -in and -ui or -iu is found in both mss., through the inaccurate marking of the -i-: cf. pin and pui, footnote to C 4139 and Introduction, 71–72. Since n and u are often difficult to distinguish, the initial syllable (ni-) may equally well be iu- (ju-); and the second syllable (-ti-) may be -ji-, -ri-, -di- or even -ist-: thus jujimental, juridical, judicial or justicial. In all these cases, the meaning is clear: it is the sharp metal point which is the death-dealing part of the lance, empowered to pronounce the death sentence, as the embodiment of justice.

362

La Chanson de Roland (Critical Text and Notes)

Li cors chaï e l’arme va a mal Dedenz enfer, el plus puant ostal; Ne:l pot tenir ne estriers ne poitral; Entire fiert l’eume jusq’al nasal.

 5520

5525

5530

5534a(V7) 5535

Rois Canabés freres fu l’amirant; De son neveu fu dolanz, Malpriant, Qe:l voit gesir a la terre sanglant. Des esperons a broché l’auferant, Traite a l’espee qi fu au roi Galant; Naymon esgarde per mot fier mautalant; Fiert le sor l’eume qi est resplendisant; L’une moitié li va jus abatant Al brant d’acer, qi vait les laz trenchant. Li chapeliers ne li valut un gant; Trenche la coife deci ou vis devant, L’os et la char et de l’oreille tant, Il nel vousist por .c. livres d’argant. Granz fu li coup, merveillos et pesant; Estonez fu Nemes li conbatant; Ja chaïst jus, par le mien esciant, Quant il se prist au col del auferant.

85.a

5519. The line begins: En t7re…, the standard abbreviation for en terre (as in C 5522), but the formula en terre fiert, ‘he strikes to the ground’, does not fit the context; it must surely be ‘he strikes entirely/right through the helmet’, i.e. a graphy of entiere/entire, fem. to agree with eume, with adjective used adverbially. The graphy Entire has been used to show this, although the V7 5637 reading is: en terre. 5532. ‘He (Nayme) would not have wanted (to receive) it for 100 pounds of gold.’ 5534. Estornez is found in the C ms. and in V7 5652, both here and in C 5542 (V7 5662), but it is clear from the next few lines that Naymes clings to his horse’s neck and does not fall. Although the verb estorner has the prime meaning ‘to fell to the ground’, it has a secondary sense, ‘to shake’; but the reading estornez, ‘shaken’ is supported only by V4 3605: le dux s’en estornal; and ‘shaken’ plays down the gravity of Naymes’ plight. O 3438 reads: li dux en estonat, P 4203: le duc a estonné and T 3081: Naimes a estonné. Thus, C’s reading is a scribal error for estonez, graphy of estonnez, ‘stunned’. At this point, C omits a line. V7 5653, inserted here, is supported by O 3439: Sempres caïst, se Deus ne li aidast and also by V4 3606 (first hemistich). The insertion of ‘He would already have fallen to the ground, to my knowledge.’ makes good sense of the next line ‘When he hung on to…’, i.e. ‘if he had not hung on to the horse’s neck’.

La Chanson de Roland (Critical Text and Notes)

5540

363

S’un autre cop li donast meintenant, Jamais en France n’oïst messe cantant. Mais Carllomene en ot le cuer dolant; Le cheval broche, cele part vint pongant, Secosra lo, mot en est desirant.



5545

5550

5555

Naymes li dus fu mot en grant freor Estonez fu desus le missoudor. Il fu navrez, si ot mot grant poor; Ainc ne fu mais tant grevé en estor, Car Canabeus o la fiere vigor Le haste mot o son brant de color; Ja l’eüst mort anfin sanz nul retor, Quant i vint Karles qi tint Terre Major. Il li escrie, «Mauvais cuvert traïtor, Mar le ballastes, per Deu lo creator!» Traite ot Joiose, q’il a porté meint jor, Fiert Canabeu sor l’eume q’est a flor – Ne li valt pas une foille d’aubor, Permi le trenche li rois de grant valor. Paien le voient, mot en ont grant poor; Li plus hardi vousist estre en Sator –

5538. V7 adds line 5658: Quant il le vit baillier si malement, but this line is not essential and may well be an addition by V7, not an omission by C. 5542. C: Estornez, but see note to C 5534 above. 5549. The second hemistich is hypermetric if traïtor is read in the normal way, as tri-syllabic. This word poses problems of metre in C from time to time (cf. C 8168 and Introduction, 92– 93), but here there is the further complication that, as this is an epithet addressed to Canabeus, the 2-syllable subject form, traïtre, is needed. Only poetic licence justifies a direct object, traitor. The same subject form for the addressee (i.e. vocative) singular, traitor, is found in C 8175. V7 5669 corrects the metre by reading mal (adj.), ‘evil’, instead of mauvais, but otherwise matches C; and cuvert traïtre/cuvert traïtor is a standard formula, whilst traïtor may be a Franco-Italian form (cf. traditore). 5555. C: ot, with ~ omitted, since the present 6, ont, is needed, unless the implication is that each one has great fear. V7 5675: ont. 5556. In both C and V7 5676, the final words of this line are en sa tor, ‘in his tower’, which reads rather oddly unless one takes tor to mean either ‘home’ or ‘protective structure’ in a general sense. No other MS has this line, but it would be logical to assume a faulty worddivision here and to read the phrase either as ens ator, ‘within armour’/’within protective equipment’, or, more credibly, as es ator (with ~ or n inserted in error) since ‘out of the battle’

364

La Chanson de Roland (Critical Text and Notes)

C’est une terre onqes n’i ot amor – Dist l’uns a l’autre, «Cist fiert ben per vigor! Onqes Rollant cop n’i dona meillor!»

85.b

 5560

5565

5570

5575

Mot ot li rois grant duel de son baron: Del cef li cort li sans jusc’al talon. Lors ot grant duel l’enperaör Karllon, A poi n’i qiet jus pasmez de l’arzon. Isnelement vint poignant a Naymon. «Garez en vos, gentis fius a baron!» «Oïl,» dist Naymes, «bien arai garison. Mot sui blecez per cel qivert felon.» Dist Karllemene, «Chevauchez a bandon! Dorenavant serons nos conpeignon. Del Sarazin est remés la tenzon; De ceste espee qi me pent au giron Li ai donee si grant confession Qe je’l fendi deci q’enz el poumon.» Quant l’oï Nemes, si dreza le menton. Lors point li rois per grant aërison, Ensanble o lui Franzois et Bergognon; Ja:l conpara le lignage Mahon.



5580

Encontre ont le paien Balligant; Porte Maté, son bon espi tranchant; Si vait ferir le conte Guinemant. L’eume li trenche et l’osberc jacerant;

is more appropriate to the context. However, C 5557 continues with: C’est une terre onqes n’i ot amor, ‘It is a land (where) there never was love’ – in which terre must relate back to a place-name, thus en Sator. Line 5557 is found only in C and would appear to have been inserted as a result of scribal mis-reading of en sa tor. Even so, ‘there never was love’ seems singularly inappropriate; it might be read as onqes n’i ot la mort, ‘.there never was death’, or perhaps as onqes n’i ot amort (abridging the infinitive, amorter, and using it substantively, ‘putting to death’, but the most tempting reading is onqes n’i ot tamor (assuming one t omitted through scribal error), giving the meaning ‘he was never afraid’. 5577. C: le lignage Karllon, but this completely spoils the sense. V7 5696 reads: Mahon.

365

La Chanson de Roland (Critical Text and Notes)

5585

Les deus moitiez li vait dou bu sevrant, Qe mort l’abat de son cheval corant. Puis a ocis Guibelin et Morant, Al autre cop Dan Richart le Normant. Quant ce voit Karles, mot ot le cuer dolant. Paien s’escrient, «Preciose est vallant! Ferez, baron, nos i avrons garant!»

 5590

5595

5600

Quant Baligans ot tot fait son pleisir De nos Franzois, qe il ot fait morir, Forment se prissent paien a esbaudir. De lor espees i vont grant cols ferir, Et li Franzois les rechoivent d’ahir. La veïssez un estor mentenir, L’un mort sor l’autre et verser et cheïr, En trois .c. lius ‘Monjoie!’ resbaudir, Per tote l’ost les poïst hon oïr. Ja hon coharz n’i poïst avenir; Qi icil jor velt bien les fes sofrir, En aventure se mist del chief tolir.

86.a



5605

Mot ferent bien François et Arabi; Froissent cez lances et cez espiez forbi. Meint riche escu i ot le jor croissi Et meint osberc desrot et desarci. La oïssiez si grant noisse et tel cri, Qe .c.m. Turc se sunt espoënti. Mot per fu forz li amirauz persi;

5585. This is the sole instance in the ms. of Dan written with a capital D. 5587. C: Preciose et vallant, with the standard symbol used for et or e; but the sense here requires est. The scribe may have used the symbol for e here to represent a Franco-Italian é. 5596. With regard to the spelling and meaning of resbaudir in this context, see note to C 2103. 5601. C: paien et Arabi, and V7 5720 has the same reading, although one would expect a reference to both armies; indeed, it is clear from the allusion (C 5606) to the Turks being terrified, that the French are involved here. For paien, both O 3481 and V4 3650 read Franceis and P 4254 has: Fransois.

366

5610

5615

5620

La Chanson de Roland (Critical Text and Notes)

Mahon recleme, Tervagan altresi: «Mi damedeu, je vos ai mot servi, Toz vos images fis faire a or bruni; Contre Karllon me soiez hui ami!» Atant es vos un paien Gelmafi: «Males noveles vos aport, je.l vos di: Baligans, sire, hui cest jor es huni! Perdu avez Malprime, vostre fi, Et Canabeus, vostre frere, autresi. Li rois de France qi le poil a flori Ocist ton frere, onqes n’en ot merci.» Baligans l’ot, a poi mort ne chaï; Dreza son chief, mas mult fu esbahi; Si apela Anglés, un son ami – N’ot plus sage home dusc’al port Sein Leri.



5625

5630

5635

Mot fu dolanz li paiens Baliganz Por Canabeus, q’est morz et recreanz, E por son fil, q’il ot nosri tant anz; Por ses amis fu forment dementanz. «Baus fis Malprimes, Mahons vos soit garanz! Hé, Canabeus, com estïez vaillanz! Vostre arme gart Jupins et Tervagans! Qi vos a mort n’est pas mes benvoillanz. Se ne vos venz, ne me pris deus besanz.» Il en apele Jafer de Val Dormanz: «Vos estes pruz et cointes et sachanz. Ne me celez vos sens ne vos talanz, Q’il vos sanble de Frans et de Persanz: Se nos vencrons ou nos serons fuianz.» Cil li respont, «Morz estes, Balliganz! Ja vostre deus ne vos sera garanz.

86.b

5630. ‘Whoever killed you is not my well-wisher’, i.e. ‘Whoever killed you does not wish me well!’ 5631. C: vent, for venz, variant graphy of present 1, venjier/vengier. V7 5750: venz. 5632. Unusually for C, the cæsura is mis-placed, after Jafer instead of after apele. 5636. C: ou serons fuianz, an hypometric second hemistich. V7 5755: ou nos serons f…

367

La Chanson de Roland (Critical Text and Notes)

5640

Charlles est fiers et proz et conqueranz; Dur sunt si home plus qe n’est fers qisanz; Ne crement darz ne espee tranchanz. Mais reclamez les barons d’Océanz, Turs et Enfrus et trestoz les Jaianz; La perte iert vostre, bien le soiez sazanz.»

 5645

5650

5655

5660

Li amiraus o la chiere hardie Mot per fu proz, s’ot bone conpeignie. Bien fu armez a la loi de Persie: Escu au col et l’eume qi verdie. Prist sa busine, qi au col li pendie, Trois foiz la sone, si lonz en va l’oïe Paien l’oïrent d’une lieue et demie; Per tot le canp sa conpaigne ralie. Cil d’Océan braient a une hie; Cil d’Argüel, meint en glatist et crie; Franzois requirent per mot grant estoutie. El plus espés ont la presse acollie; De trois cent Frans en fu France apovrie; Des Turs de Cople et de ceus de Hongrie, Plus de deus mille en remestrent sanz vie; Deci qe Meqe en la Mahomerie En fu la terre des paiens estormie.

87.a



5665

Li cons Ogiers qi fu proz et senez, Meillor vasal de lui pas ne savez; De mauvastiez ne fu onqes retez; N’avile mie par lui ses parentez: Vasaus doit estre qi de vasaus est nez. Mot fu dolanz quant voit Franzois rusez, Les rens peciez, en trois lius estroëz;

5664. C: retrez, which makes no sense. This is a scribal error for retez, ‘accused’, as in V7 5783. 5668. C: luis, with the i plainly marked, but this can only be a scribal error for lius, ‘places’. (Cf. Introduction, 71–72). V7 5787 reads: leus.

368

5670

5675

5680

5685

La Chanson de Roland (Critical Text and Notes)

Tel duel en a, a poi ne chiet pasmez. Tieri apele, qi d’Ardene fu nez, Jofroi d’Angou, Joseran q’est senez: «Venez o moi, gardez ne demorez!» A Karllemene en sunt pognant alez. Mot fierement fu d’els araisonez: «Sire enperere, por Deu, car esgardez! Veez ci Baligant qi s’est a vos meslez! Mot a nos homes as premiers cols grevez; Toz les ocient a duel et a viltez. Ja Deu ne place qe corone portez En douce France, dont les onors tenez, Se vos vus homes mot tost ne secorez!» Li rois de France a ses grenons levez; Le cheval broche, q’est de cosre abrivez; Met soi estens, de ferir aprestez; .xx.m. Franc sunt aprés lui tornez. Ne puet mais estre si granz oz desevrez C’om n’i gaaint ou on n’i perde asez.



5690

Mot i fiert bien Karllemene li bons rois, Nayme li dus et Ogier li Denois Jofroi d’Anjou et Tieris d’Ardenois. Alborion gete mort de son sois; Lors chiet s’enseigne a terre en un chemois.

87.b

5677. The C ms. reading is apparently nos (although vos is just possible). V7 5796, O 3537 and P 4302 read: vos; the sense will admit either reading. 5678. C: vitez, but viltez in V7 5797. 5684. C’s reading seems to be estens, i.e. nom.sing. estant, ‘upright in stance’ – and this fits the context better than es rens, ‘in the ranks’ (V7 5803), since he will presumably be at the front, as leader. 5691. Instead of de son sois, V7 5810 has: de son pois, ‘against his will’ (although the usual phrase is sor son pois, as in C 5699), but O 3549 reads: devant sei, V4 3714: davant si and P 4312: devant soi, ‘in front of him’. However, Godefroy lists soir as a graphy of sor, ‘chestnut horse’, and this fits well with the action of ‘casts him down dead’; moreover, the scribe is generally prepared to modify an ending to meet the full requirements of the rhyme (cf. lines 1504, 3218, 3226, 3457, 3463, 3464, 4950). Therefore, sois has been retained as a graphy of soir/sor, rather than choosing between V7’s reading and that of the other versions.

La Chanson de Roland (Critical Text and Notes)

5695

5700

369

Quant Baligans le vit, si devient e plus noirs Qe poiz demise ne carbon de Jarrois. Lors ferent bien Angevin et Tioirs, Et li Normanz, Poier et Hurepois; E dist Karllon, «Or i ferez manois! Hui vengerons Oliver le cortois E mon neveu, q’est ocis sor mon pois.» Lors recomencent la noisse et le befois; Ne fu tel chaples de gent de nulle lois.



5705

5710

5715

Passe li jors, si vint a la vespree; Cel jor i ot meint cop feru d’espee E tante targe pecee et estroee, Et mainte brogne ronpue et desafree, Et meint cheval dont la sele est versee. Lors recomence li cris et la huee. Li amiraus a s’enseigne escriee, C’est ‘Preciose!’, qi paiens mot agree; Karles ‘Monjoie!’, qi tant fu renomee. Li rois de France a sa teste tornee, Vit Baligant et Brum de Val Fondee Devant les autres demie arbalestree. Ot Baligans sa banere fermee; Karllon connuit a l’enseigne fresee, Et Karlles lui, n’i a mester celee; La’n ert la joste de .c.m. esgardee.

5693. The copyist made two errors, omitting si, which had to be inserted superscript, and then writing mors instead of noirs, so that he had to bar mors and add noirs at the end of the line. 5694. C: poiz remise, V7 5813: peinz remise, but the usual term is peiz demise/poiz demise, ‘melted pitch’. C ends the line with: de iarrois (graphy of jarrie, ‘heath’, or Jarrige); Foerster here read de Iarrois and Duggan reads jarrois, but this could also be de larrois, with either lower-case l or upper-case L with vestigial foot, i.e. ‘heath’, ‘waste-land’ or a Proper Name (cf. entry Jarrois). Carbon de jarrois/Jarrois thus means ‘charcoal (or charcoal-burner) from heath or from Jarrois’ (or possibly ‘from Larrois’). 5705. C: ropue, with ~ omitted; V7 5824 reads: rompue (written in full). 5713. C: abarestree; V7 5832: arbalestree. 5717. La’n ert stands for La en ert, ‘There was .’, the elision being required for the metre.

370

5720

5725

La Chanson de Roland (Critical Text and Notes)

Vont soi ferir per ire defrenee; N’i remest targe, tant soit de cuir fermee, Ne soit fendue et pecee et quasee; Il n’i a broine ne soit desclavelee. La lance Karle est lés le Turc passee, La Balligan i est permi troee; Ne se font mal a icele asanblee: Jus s’entrebatent des chevaus en la pree. Ceste batalle est issi atornee; Ja sanz mort d’ome ne la verrez finee.

88.a



5730

5735

5740

Li rois de France ot mot le cuer loial; Vilté li sanble, quant est jus del cheval Por Sarazin ne por Turc desloial. Les deus piez juint, si salt en son estal, Son escu drece, trait l’espee roial, Dont ot feni meint grant estor chanpal. E Baligans, qi ot cuer de vasal – Gros ot le ventre plus qe piere pognal – Contre Karllon est venuz tot engal; Il ne:l redote ne ne crent un chapal. Lor s’entreferent li dui roi natural Amont es elmes, o luisent li cristal; Froissent les pieres, li teinz et li esmal; Toz les abatent jusqe sor le nasal. Ceste batalle ne puet partir sanz mal; Liqel qe soit i fera son jornal.

5729. The letter e with line over to indicate est was added superscript after quant; V7 5848 reads: fu. 5735. ‘His belly was larger than a stone one can hold in one’s fist’ is neither an attractive detail, at least to modern eyes, nor an effective simile. V7 6854 reads: Grant ot le cors e chiere mult crual. The line is peculiar to C and V7. 5742. The scribe initially omitted puet, but added it at the line-end, marking its place of insertion.

371

La Chanson de Roland (Critical Text and Notes)

 5745

5750

5755

Dist Baligans, «Rois, mot es de puit aire! Trop per es fous, si as mot put viaire! Mot te fermie li sans en cel nazaire. Cuides tu donc, per l’arme de ton paire, Avoir Espegne, Tortelose et Balcaire? Ainz qe tu l’aies, aras asez a faire! Deven mes hon: je te ferai doaire D’une grant terre, qi fu au roi Orsaire; Ven moi servir, per raison le doiz faire.» Et dist Karllon, «Je te ferai ainz traire, D’agollons poindre et angoiser et braire! Car pren batisme, je:l te di sanz contraire; Je t’amerai plus c’ainc ne fist ta maire.» Dist Baligans, «Bien savez sermon traire. Je nel feroie por tot Monte Calvaire, Ne por l’avoir qi fu au roi Dotaire.»

88.b

 5760

Li amirauz fu mot de grant vertu; Il fiert Karllon desor son eume agu, Desor la teste li a frait et fendu

5745. C: viare, but the normal graphy viaire is needed for the rhyme. V7 5864 reads: chaire. 5746. The identification of C’s reading, fermie, as a graphy of formie (present 3, formier, ‘to boil’, ‘to froth’) is confirmed by formie in V7 5865. The word nazaire is not found in any dictionaries; it appears to be poetic licence for either nazart or nasal, both substantives (m) meaning ‘the part of the helmet which protects the nose’; but here it seems to mean ‘the front piece of the helmet’ or even simply ‘the helmet’. V7 ends 5865: en cel viaire, ‘in that face’, having read mult puite chaire in V7 5864. If C simply switched the final words in C 5745, 5746 in error, then the V7 readings suggest that C’s en cel nazaire should read: en celle chaire; but there is no obvious link between nazaire and chaire. 5753. C: taire, followed by a list of other punishments in the next line, so that taire = ‘be silent’ seems unlikely here; V7 5872 reads: ferai…traire, giving the meaning ‘have you dragged along (presumably by galloping horses)’ or possibly ‘have you torn apart’. 5754. C: angoiser, the i being inserted superscript. 5756. C and V7 5875 both read: ma maire, but ta is clearly needed here. 5759. King Dotaire has not been identified. It is tempting to read Dotaire as a corruption of Clotaire, the name of 4 early Frankish kings – thus preserving a folk memory of the tribute traditionally amassed at Soissons by Clotaire I.

372

5765

5770

5775

La Chanson de Roland (Critical Text and Notes)

Et le hauberc desmaillé et ronpu; Tresq’es chevols li est li brans coru; La char li trenche, li os sunt remest nu; Grant plene paume a terre en est cheü. Karles canchele, mot pert de sa vertu, Per un petit ne chiet tot estendu. Mais ne volst Deus q’il fust mort ne vancu; Sains Gabriel est a lei desendu, Se li a dit, «Rois magnes, qe fais tu?» Quant Karles l’ot, onqes plus liez ne fu: Son chief dreza, ainc ne soit ou il fu; Escordement a reclamé Yhesu: «Beaus sire Deus, per ta seinte vertu, Garde n’i soie ni mort ne retenu.»



5780

Mot per fu liez li gentius rois de France, Quant Deus li mostre et a fet tal senblance, Et per son angle li a fait demostrance; Et de son cors ne puet avoir dotance. Remenbre li des nons de sa creance Qe il aprist quant fu norriz d’enfance: Trois en noma, qi sunt de grant puissance;

5764. C: chevals, but Baligant has struck Charlemagne on the head, cutting through the helmet and the upper part of the hauberk lying beneath it, right down to the hair: chevols in V7 5883 and chevels in O 3605. He has sliced down to the bone, cutting off a piece of flesh as big as the palm of one’s hand (C 5765–5766). 5767. C: .K. chevauche, a scribal error: V7 5886 reads: canchele and O 3608: cancelet, ‘reels’. 5773. C: ainc ne soit…, and V7 5892 also reads soit; this is past definite 3, savoir: ‘he did not know…’, ‘he had no idea.’. – a regional variant of sot/sout. Similarly, C 1182, 1318 and 1742 read oit, a graphy of ot, past definite 3, aver. The imperfect indicative was little used in Old French; avoit is found only in C 4825 and 4879, avoient only in C 4871, savoit in 7697 and savoient in 2444. This is the only instance of soit used with this meaning; elsewhere in C it is always present subjunctive 3, estre. 5775. The word sire was inserted superscript. 5779. C: Et per son angele, an hypermetric first hemistich; V7 5898 reads: angle, the standard graphy, which corrects the metre. 5782. C: noiriz de France, a scribal error for norriz d’enfance, ‘brought up as a child’; in noiriz the first r has been mis-read as i, then ~ has been omitted so that d’enfance > defance > de France. V7 5901 reads: nori d’enfance.

373

La Chanson de Roland (Critical Text and Notes)

5785

5790

Lors li revint et vigors et menbrance. Traite ot Joiose ou il ot grant fiance; Fiert Baligant sor l’eume de Valance, Trenche les laz, del ben ferir s’avance, Et tot le vis jusq’en la barbe blance, Qe mort l’abat sanz nule recovrance. ‘Monjoie!’ escrie li gentius rois de France. La i vint Nemes et Fochiers de Vilance, Li dus Ogiers qi fu de grant bobance. Li paien tornent, car il sunt en balance, Et li Franzois les suigent sanz dotance.

89.a

 5795

5800

5805

Paien s’en tornent, Franc les ont corochez Por Baligant q’est morz et detranchez; E d’els meïsmes i a mot daumajez. Li rois de France vet aprés eslaissez, Et dist Karllon, «Baron, or vos vengiez! Esclairons nos des qivert renoiez Qi de nos homes nos ont si apovrez! De ceaus vos menbre qe nos avons laissez: En Rencevals en a .xx.m. cochez, Por moi secorre n’en levera uns piez.» Lors i ferirent de lor trenchanz espiez. Iloc ont Franc grant avoir gaagnez, Ainc n’en ot tant rois qi fust batizez.

 Granz ert la nivole, la poudrere est levee; Paien s’enfuient, qi héent la meslee;

5784. C: li revint et secors…, but ‘help’ would not return, coupled with memory. V7 5903 reads: vigors, O 3614: vigur and V4 3777: vigor. 5799. The word or was added superscript. 5808. C’s nivole produces a 5-syllable hemistich so, although ‘mist’ or ‘cloud’ (formed by the dust) makes good sense, this should be read as the graphy niule, read as a monosyllable. V7 5927 reads: nuisse, ‘noise’, which is less apt in the context and reads like the copyist’s emendation.

374 5810

5815

5820

5825

5830

5835

La Chanson de Roland (Critical Text and Notes)

Vers Saragoze ont lor voie tornee; Tresc’a la porte est la chace duree; Tant en ont mort li Franc com lor agree. Sus en la tor Brasmimonde ert montee – O li si clerc de sa loi mesalee E li chanoine, noir sunt comme pevree – El voit sa gent q’est vencue et matee. Uns escuers va poignant la feree, Qi li escrie en la tor crenelee Qe Balligans a la teste coupee. Quant ele l’oït, a poi ne chiet pasmee. Droit en sa chanbre en est corant alee, Au roi Marsille la novelle a contee. «Sire,» fait ele, «Espegne est desertee! Ja par paiens ne sera recovree! Balligans gist la jus en la valee, Toz detrenchez, sanglant, gole baee.» Marsille l’ot, s’a la teste levee; Li cuers li faut, s’a la boche seree; L’arme s’en part, n’i pot avor duree; Bien .xxx. diable l’ont en enfer portee, Devant lor mestre l’ont le jor presentee; Morz fu de duel, l’arme est a mort livree. Et Franzois ont lor bataille afinee; Auquant garisent qi sevent la contree; A Saragoze vindrent a la vespree.

89.b

 Paien sunt mort, la bataille est vancue; A Saragoze est tote l’ost venue.

5814. There are 2 verbs mesaler. One means ‘to lose one’s way’, ‘to stray’, ‘to fail’ or ‘to make an error’ – so that loi mesalee could be ‘mistaken or erroneous religion’. The other mesaler is a graphy of meseler, ‘to be leprous’, with the past participle used as an adjective meaning ‘corrupt’, ‘putrid’, ‘rotten’. It is debatable how strongly the thirteenth century writer would express his view of Islam. V7 5933 reads: sa meisnie privee, but Duggan replaces this with the C reading. 5816. C: Ele, with insertion mark for the superscript e, which would make the hemistich hypermetric. Cf. the use of Il for a feminine subject pronoun in C 4838, 4839 and 7121. It seems the scribe used El to signify ‘She’ either as an accepted alternative form or as poetic licence for

375

La Chanson de Roland (Critical Text and Notes)

5840

5845

La maistre porte a Karles abatue; El ne poet estre encontre lui tenue; La cité prenent cil de Terre Asolue. Icele nuit i est nostre ost geüe, Tresc’al demein qe l’aube est creüe. Bers est li rois o la barbe chenue; Et Braismimonde est devant lui venue, La maistre tor li a iloc rendue. Per la cité en ot meinte estendue, Plus de .l. dont chascuns est fondue; Mot ovre bien cui Damedeus aïue!

 5850

5855

Passe li jors, la nuiz est aserie; Cler luist la lune per la cité antie, La sus el ciel meinte estoile flanbie. Li enperere Saragoze a saisie, A sa gent rueve q’il ne s’arestent mie. « La chité cerchent, q’est d’avoir replenie! N’i ait citerne, ne grant mahomerie,

90.a

the sake of the metre, then a revisor emended this to the correct form, Ele, producing an extra syllable. V7 5935 reads: Et. 5840. C: alsolue, with extra l inserted in error; V7 5959 reads: asolue, the normal graphy. 5842. C: qe l’aube est creue – but one would expect to read the final word as crevé (preferably crevee), ‘when dawn has broken’. For the rhyme, the reading must be creüe, past participle (f) of either creïre, ‘to believe’ (impossible in the context) or creistre, ‘to grow’; the latter is not ideal, but it may perhaps imply that dawn has spread across the sky. It seems likely that the writer has confused the two past participles, crevé and creüe, because of their identical graphy. V7 5961 reads parue, ‘appeared’; but in the (roughly) matching laisse P 4391–4393 read: Prent la cité, la gens est confondue;/Et par lor force, que Deus lor a creüe,/Nostre Fransois…, which may indicate that creüe really belongs to a line which C omitted in error (although C and V7 rarely match P in this part of the text, Section B). 5854. The reading chité cerchent seems to arise from a mis-placed h; elsewhere in C, the noun is always cité or cit. The line may be a statement but cerchent makes good sense as a Jussive subjunctive, ‘let them search…’. See also note to C 5855 below. 5855. The present subjunctive, ait – followed by soit, C 5857 – is used because these are Charlemagne’s orders to his troops: ‘Let there not be…’. C’s reading citerne is matched by cisterne (‘tank’) in V7 5975, but this makes no sense unless one assumes that Roman Baths (Latin thermæ) or sophisticated Moorish water-supply systems were identified in error as pagan temples. The error may also lie in the CV7 source, for O 3662 and V4 3825 read: sinagogues and P 4401: synagües; if sinagogues were changed to cis temple, then perhaps cisterne/ citerne might arise from scribal error. The only other solution would be to read this as 2 words:

376

5860

5865

La Chanson de Roland (Critical Text and Notes)

Ne liu oscur, ne grant herbergerie, Qi tot ne soit torné a desertie! » Et noz Franzois, qi d’avoir ont envie, I sunt coru, ne s’aseürent mie, Froissent et brisent, q’il ne demorent mie. Marsile l’ot asanblé an sa vie. Mahomet trovent et Jupin or dalie; Franzois les ferent, qi Deus port garantie; Cil qi ot aze en ot gregnor partie. Karles croit Deu, lo fil Sainte Marie; Meint Sarazin i ont lor loi guerpie, Plus de .xx.m. la nostre ont recollie. Qi ne:l vost faire, ne pot avor aïe

ci terne, reading terne as a graphy of tertre, ‘hill’, standing in contrast with liu oscur, ‘hidden place’, in C 5856. 5856. C: lui, where liu (‘place’) is clearly needed. Cf. notes to C 5668, 5878, 6096 and 6199. 5858. C: n’ot envie but ont, present 6, would seem more likely. V7 5978 reads d’avoir ont envie (a more credible reaction), but with a blot (of ink?) between avoir and ont, so that the clause might possibly be negative. C’s reading may result from omission of the symbol ~, i.e. n’ont, but the positive ont resulting from metathesis of o and n seems better suited to the context. 5861. This line occurs in the same position in V7, but in C it appears to be out of context: there is no antecedent for the direct object, l’, to indicate what Marsile had assembled – unless it is an elliptical explanation of il, plural, the things which no longer remain, C 5860. V7 offers an antecedent, for it ends the preceding line (V7 5980): tut grant manantie, ‘every great house (or possession)’. C 5860 may thus have a faulty second hemistich (perhaps through omitting 2 separate half-lines) – or C 5861 may be in the wrong position (probably in the source document), since it would fit much better after C 5854, referring back to all the goods which filled the city. The present position of this line is suspect because it interrupts to no clear purpose the trail of destruction carried out by the French (C 5855–5864). 5862. C: et Jupin ordalie. V7 5982: et Jupin et Dalie, with Dalie taken as the name of a Saracen god. In both texts, this line refers to Mahomet as an idol and then, it would appear, to a statue of Jupiter (variant graphy, Jupin). The earlier references to Jupin (C 4959 and 5629) and the initial mention of Mahomet would condition the copyist to expect the name of another god. Both C and V7 therefore read this as Jupin and V7 modifies the end of the hemistich to make some sense. However, ordalie is a precise term meaning ‘trial by ordeal’, which would be familiar to most people of the time. There existed the verbs juper or hoper/huper, ‘to shout’ with implications of mockery. It is therefore likely that the source reading was jupent/hupent, the French mocking Mahomet with calls for trial by ordeal – and in fact proceeding to beat and chop the idol under God’s protection and therefore without fear (C 5863–5864), thus proving that faith in Mahomet was unfounded. Alternatively (and rather persuasively), reading or dalie as or dolie (< dolir, graphy of doloir, ‘to cause to suffer’) would give: ‘now made to suffer’, which is apt in the circumstances. 5863. The reading les in this line supports the identification of 2 gods in C 5862, 3 gods in V7 5982.

377

La Chanson de Roland (Critical Text and Notes)

5870

Ne fust penduz ainz la nuit aserie. Mais Braimimonde ne la volst prendre mie; Li rois de France ne l’en esforza mie; En doce France l’enmena sanz fallie, Tot per amor prendra la loi saintie, Deu servira dedenz une abaïe.

 5875

5880

5885

Passe la nuiz, si aparist li jors; De Saragoze Karles garnist les tors: .M. chevaliers i laisse pognëors; Gardent la vile, les lius et les onors. Monte li rois, o lui ses vavasors, Ogiers et Nemes et Jufroiz l’amoros, E la roïne q’est blanche come flors, C’est Braimimonde, plene de grant valors; Li rois l’enmeine, qi mot li fait onors. Repeire s’en o duel et o tristors En Rencevals, ou fu la grant dolors, Ou sunt ocis li prince et li contors.

 Granz fu li deus la nuit en Rencevaus; Il n’i fist joie ne cheveluz ni chaus;

90.b

5878. C: luis, with the i clearly marked, scribal error for lius. V7 5998: leus. Cf. Introduction, 71–72. 5884. O 3682 reads: Repairez sunt a joie e a baldur and V4 3843 echoes this, but both have already dealt with Charlemagne’s return to Roncevaux to seek out and mourn over the French dead. They now continue immediately with his journey back into France. Indeed, O 3683 reads: Passent Nerbone par force e par vigur, and from this (and the faulty identification of Nerbone, as J S Crompton pointed in his (University of Manchester, 1951) thesis, «The Châteauroux version of the Chanson de Roland: a study in the literary technique of the Old French epic») springs V4’s long digression on the taking of Narbonne. Up to this point, C and V7 have offered a version generally similar to that of OV4, although often diluted and modified. From here on, their similarity to O is far less: line for line concordance becomes very much rarer; only the general outline remains the same; incidents are greatly developed (eg. that of La Belle Aude) and new ones inserted (eg. the near-escapes of Ganelon). After the Narbonne digression (not found in C or V7), V4’s text is much closer to that of C and V7 than to O’s. See also Introduction, 57 and 60–61.

378

5890

5895

5900

5905

5910

La Chanson de Roland (Critical Text and Notes)

Ne n’i menja palefroiz ne chivaus, S’erbe sanglante ne pot per cez terraus. Morz fu Roillanz, li nobiles vasaus. Sor Oliver gist li cons de Frondaus: Cil fu ses oncles et ses amis carnaus; Il en fait duel, onqes on ne fist taus: «Niés Oliver, de vostre mort m’est maus! Laisié m’avez veslez et jovenaus. Meus en sera messire Sainz Marzaus: Toz ses mostiers ert refaiz de qarraus; Li ors des tables ne sera mië faus, Bien iert asis a pieres de cristaus; A clos d’argent ert mis li esmeraus; Qatre cent moines i metrai generaus, Qi tuit aront et miches et meraus, Qi chanteront les messes mortuaus; Si ferai faire deus meissons moniaus. En son la tor reluira li pomaus; Deu mile povres i metrai comunaus: Cil proieront por les barons loiaus Nostre Seignor, qe il les face saus En paradis, qi tant est clers et biauus;

5890. S’erbe reads rather strangely, but stands for se erbe, se meaning ‘so’, ‘therefore’. V7 6010 reads: L’erbe, but emendation seems unnecessary. The verb, pot, is past definite 3, paistre, ‘to graze on’, and the subject is ‘neither palfrey nor horse’ in the preceding line. 5896. C: veslez e jovenaus (subject form, adjective singular, jovenal); V7 6016: vaslez e juvençaus; P 4430 and T 3415: guennez et jovenceaux. ‘You have left me (whilst you were still) a young squire’. Strictly, veslez, a graphy of vaslez, already implies youth, but addition of jovenaus simply stresses this. 5897–5911. These lines specify the embellishments that the Count of Frondaus will make to St. Marsaus’s monasteries: the extra 400 monks and 2000 poor people for whom he will provide, so that they may pray for the souls of the ‘loyal barons’; the vessels on the altar tables made of pure gold (C 5899), with precious stones set in silver. Duggan (2005, 506s.) identifies this with the altar decoration of St. Martial de Limoges in the note to V7 6017. 5900. C: Bien asis, which is hypometric; but V7 6020 reads: Bien iert assis. 5902. C: Qe trecent…, but the copyist has mis-read Qatre cent in this way earlier, in C 1037. V7 6022 reads: Quatre c. 5903. The merel or meral was a jeton or token, given to priests as a reward for their presence at religious services. It could be exchanged for food and other necessities. Here the tokens are coupled with small loaves of bread, miches, in a standard phrase. The implication is that, provided they attend services regularly, the monks will have all basic necessities provided for them, each token entitling them to a due share in that provision. (Cf. also note to C 8118.)

379

La Chanson de Roland (Critical Text and Notes)

5915

5920

5925

La les conduie as angles spiritaus.» Charlle regarde les plens et les egaus; Morz vit gesir les contes naturaus; Tel duel en fait l’enperere roiaus, De son mantel en desront les tasiaus; Puis vint avant, desoz deus arbresiaus. Aflubé ot unes martrines piaus; Ogier apele et le conte d’Evraus: «Baron,» dist Karles, «cist deus nos est mot maus! Faisons deus bieres de verges et de paus, Sis acoplons deus et deus as chevaus. Demain a l’aube, quant chantera li jaus Et reluiront les estoilles jornaus, Si voiderons la terre et les boscauus; Oster voudrai cest duel de Rencevaus.»



5930

5935

5940

Nostre enperere va forment sospirant; Duel ot et ire de son neveu Rollant; «Beaus niés,» fait il, «com mar vos amai tant! Ahi, fel Guene, com vus m’alez trichant, Mon grant damage per mon cors recerchant! Si m’avez point el cor perfondemant, Trait en avez le meillor garnimant Qi me faisoit dormir seüremant!» Tel duel ot Karles, ne pot aler avant; Sor une piere est tornez en seant. Andeus ses poinz va li rois detordant, Sa blance barbe mot fierement tirant; Li enperere va tel duel demenant, Tuit li baron li sunt venu devant, Mot doucement li vont araisonant: «Sire,» font il, «qe vos dementez tant! Perdu avez le prou conte Rollant Et Oliver, le hardi conbatant,

5911. The word as was added superscript. 5925. The word de was added superscript, turned slightly on its side. 5941. C: demetez, with ~ omitted; V7 6061 reads: dementez.

91.a

380

5945

5950

5955

5960

5965

5970

La Chanson de Roland (Critical Text and Notes)

Les .xii. pers qi vos amez itant, Par Guenellon, le qivert sosduiant, Qi:s a venduz a la gent mescreant. Livrez nos, sire, le gloton malfaisant! Nus en ferons la justise si grant, Jamais n’ert jors a tot vostre vivant Qe n’en parolent escuer et serjant.» «Baron,» dist Karles, «tot a vostre comant.» Aprés le duel va li rois somellant; Lors le coucherent desoz un boguerant; Au chef li metent un mantel aufricant; Desor lui tendent un vert paveillon grant, Por la calor qi l’aloit aprochant. Li rois se couche, mais ne li valt niant, Car la dolors le va si angoissant. Remenbre li de son neveu Rollant: Entre les morz va son neveu qerant; Quant il nel trove, mot ot le cuer dolant. Il s’agenolle söef et bonemant, Fist sa proiere, si la dist en plorant: «Damedeus pere, qi formastes Adant, Eve feïstes, sa moillier la vaillant, D’une des costes de l’ome voiremant – Por ce est ele en son comandemant – Terre feïstes et lo ciel ensemant, Solel et lune, et des estelles tant; Feïs occire Seint Michel le serpant; Les trois enfanz garis de feu ardant. E de la Virgine nasquis en Baliant; Vostre miracle furent aparissant: De .v. pains d’orge, de deu poissons noiant

91.b

5961. The word ot was inserted superscript, having been first omitted in error, due to the repetition inherent in the phrase mot ot. 5970. This is a reference to the incident in the Book of Revelation xii.7–9. 5971. Cf. notes to C 4161 and 5117 regarding the source of this story and use of the term ‘children’. 5972. E de la Virgine is a 5-syllable first hemistich, but cf. Qi de la Virgine, C 4157; this may reflect Italian pronunciation of the final e of Virgine or (ecclesiastical) Latin ablative: either would convert this line to an alexandrine. V7 6092 reads: Virge. 5974. C: …deu pois poissons…

381

La Chanson de Roland (Critical Text and Notes)

5975

5978a (V7) 5980

5985

5990

5995

6000

6005

Rasaciastes .iiii. meillers de jant; Et en la croiz vos pendirent tyrant; Et el sepoucre cochez et repossant Resusitastes par le vostre comant; Enfer alastes sanz nul retenemant, Toz vos amis traissistes de tormant: Damedeus pere, tot issi voiremant, Com ce est voirs et je le sui creant, Ce me donez qe je desire tant, Ce est le cors de mon neveu Rollant! Deus en ait l’arme por son comandemant!» Vint al destrier, si monte isnelemant; L’estrier li tint Naymes et Joserant. Sonent cil grasle et derier et devant; Sor toz les autres bondist li olifant; Et François plorent por la mort de Rollant. Karles esgarde vers Espeigne la grant; Desoz un arbre ramu et verdoiant, La vit le cors qe il parama tant. Li enperere li vait esperonant; Mist jambe a terre del bon destrier corant; Il s’agenoille söef et bonemant. Devers les piez lo prist primieremant: Enz en sa boche li mist l’ortel plus grant, Per grant dolçor la li va estregnant. «Beas niés,» dist Karles, «com mar vos amai tant! La vostre mort me va si angoissant, Per un petit qe li cuers ne me fant.» Charlles li Maignes, l’enperere puissant, Iloc se pasme voiant tote sa jant; Por soie amor en pasmerent ben çant. Et «Deus!» dist Otes, «Or voi je duel d’enfant!

92.a

5978. Cf. P 4524: Anfer brisastez sans nul demorement, supported by T 3494, L 1842. 5986. C: tint, past definite 3, tenir, although two subjects are named, Naymes and Joserant; this was common practice and is found elsewhere in C. V7 6107 and O 3113 read: tindrent, past definite 6. 5997–5998. The insertion of such tasteless, macabre details, not found in O, may reveal a coarser, more superstitious, populist outlook. V7 6118 matches C, but P 4541 refers to doi (doigt), L 1858 to talon. However, Duggan (2005, 507s.) cites evidence of a practical explanation for this action: verification that no sign of life remains.

382

6010

La Chanson de Roland (Critical Text and Notes)

Pur coi t’ocis, bons rois, nos els voiant? Ja voiz qel gent te traient a garant: Se il te perdent, qe feront il dolant?» «Baron,» dist Karles, «vus parlez de niant. Ja veez vos bien le damage si grant; N’arai mais joie a trestot mon vivant.»



6015

6020

6025

«Baron,» dist Karles, «por Dé lo fil Marie, Tant sui dolenz et pleins de desverie, A poi qe l’arme n’en est del cors partie, Quant j’ai perdu ma riche baronie, Mes niés Rollant qe la mort a saisie; Et Oliver ne puis oblier mie; Les .xii. pers qi mort ont recollie: Lor armes soient en la celeste vie! Ce m’a fait Guenes qi sa foi m’a mentie, Qi les vendi a la gent de Persie.» Karles apele Richart de Normandie E puis Naymer, li sires de Pavie: «Gardez moi Guene q’il ne vos escanp mie. Il sera ars et la pouldre guerpie. Sa traïson li sera bien merie.» Quant li rois ot sa parole fenie, Un poi ploran, li cors li asoplie;

92.b

6020. C: Ce fa m’a fait .G… 6022. Richard of Normandy, killed by Baligant in C 5585, re-appears several times in Section C. A demonstration of the inconsistency resulting from the episodic nature of the remaniements. 6025. Both C and V7 6145 read: guergie, but it is implausible to link this to gorgïer, ‘to swallow’, It must surely be a scribal error for guerpie, meaning ‘rejected’, ‘discarded’, ‘thrown away’; Charlemagne would have no wish to keep Ganelon’s ashes. This line is not found in any other version. 6028. C: sa parole li asoplie, an 8-syllable hemistich. It is noticeable that the words sa parole are found in the same position in the preceding line (6027) and that parole occurs immediately below in 6029 – a ready source of error. V7 6147 (= C 6027) reads parole, V7 6148 (= C 6028) has cuers and V7 6149 (= C 6029) reads parole again; it thus reduces this hemistich to normal length. However, this would mean that cuers in the second hemistich of C 6028 was followed by cuers in the first hemistich of C 6029. P 4613 reads sa raison definie, followed by car li cuers li souplie (P 4614) and then li cuers… la parole (P 4615), confirming this sequence – or rather this error, for cuers is surely here confused with cors, ‘body’: ‘his body becomes weak’. The

La Chanson de Roland (Critical Text and Notes)

6030

383

Li cuers li faut, la parole li lie; Desor Naymon a la barbe florie S’est acochez voiant sa baronie.



6035

6040

6045

Duel ot li rois quant s’apuie a Namon; Devant lui gisent li .xii. conpeignon, Chascuns covert d’un vermeil ciglaton. «Baron,» dist Karles, «por Deu, qel la feron? Mal m’a balli li orgueus Guenellon! Li siens servises me torne a mespreson; Jesus li rende mot aspre gueresdon! Si fera il, se loingement vivon. Faisons le bien, nos amis enterron, Qe nes menjucent ne liupart ne lion, Cil ors sauvage, ne sangler ne brohon.» Et cil respondent, «Coment les sevresron? Tant i sunt mort paien et Esclavon, Turc et Persant et Bedöin felon.» Et respont Karles, «Vus dites voir, baron. Or vos cochez trestoz a orison: Cascuns prit Deu selonc sa entencion Qe nos en face voire division;

verb asoplie is a graphy of asoploie, present 3, asoploier, ‘to become weak’, the final syllable being modified to suit the rhyme. 6035. ‘What shall we do there?’, ‘What shall we do in these circumstances?’ The graphy quel le, quel la for que le, que la is very common, but the reading here may be: Q’el…, ‘what more?’ or ‘whatever?’. 6038. C: mortas pre g…, through scribal error; V7 6158: mout aspre g… indicates the correct reading. 6043. C: coment les sa sevesron. This may well indicate that the copyist first anticipated writing the future 4, savoir, i.e. «How shall we know them?», «How shall we distinguish between French and Saracen corpses?» The deletion of sa indicates that such a reading is incorrect; sevesron is therefore future 4, severer (variant graphy of sevrer, ‘to separate’), the -sr- representing the -rr- of severron. This could be a scribal error for the standard form sevreron (V7 6163’s reading) with metathesis of re to er and mutation of the resultant -err- into -esr-. (P 4631 reads: averons, T 3557: choisiron.) 6049. C: evoire divison, omitting an i from division and reducing its 4 syllables to 3, probably because of writing evoire in error for voire (thinking of avoir division?) – although the graphy devison is also found in C 6054 (but division, C 1712, 2691, 3134). V7 6169 reads: voire division,

384 6050

6055

6060

6065

La Chanson de Roland (Critical Text and Notes)

En son servise sunt ocis mi baron.» Atant se mist cascuns a genuillon; Li rois meisme refist ses leccion: «Damedeu pere, per vostre digne non, Terre feïstes et mer per devison, Et le saint ciel per vostre esliction; Et herberjastes en l’ostel chés Symon Quant a Marie feïstes le pardon, Qe mist ses els sor vos piez a bandon; Iloc plora per bone entencion; Vus l’en levastes amont per le menton – De son servise en ot bon gueredon En tes sainz ceus, en ta susjeccion: Sainz Espiriz, si com nos ce créon, Faites sevrance, qe ma gent conoison Ainz qe m’en parte, q’enterrer les puison.»

93.a

 Nostre enperere se jut vers terre enclin, Et li viel home et li jone meschin, Et li Normant, Mansel et Angevin, ‘a true separation’, ‘an accurate separation’. P 4637 has: voire demonstrison, L 1909: veray demostrison. 6054–6055. These lines are a reference to Genesis, i.6–10. 6056. C: l’ostel saint Symon, with saint barred and ches added after the full stop at the lineend. The reference is to the Pharisee with whom Jesus dined (Luke vii, 36) and whose name, Simon, is given in Luke vii, 40 and 43 – the man at whose house the incident related in C 6057–6062 took place. 6058. This is once again a reference to Mary Magdalene (mentioned by name in Luke, viii.2), who has traditionlly been identified as the un-named woman in Luke vii.37–48 (cf. note to C 3788–3789). The Gospel narrative refers to the woman washing Christ’s feet with her tears before pouring ointment on them, but reading els as ‘eyes’ requires mist ses els… a bandon to be taken as an elliptical phrase meaning ‘placed (the contents of) her eyes unrestrainedly’, i.e. ‘poured out her tears’. This reading is apparently supported by three other mss. (V7 6178 reads: euz, P 4647: ieus and T 3568: yeuls), but P’s and T’s readings may be due to scribal error, whilst V7’s euz (and C’s els) could well be a graphy of ues/euls (sm), meaning ‘what one has made’, ‘what she has made’, i.e. ‘the ointment’, or indeed a graphy of or scribal error for oils/uils (sm.pl.), ‘oils’. Since C 6059 very plainly deals with Mary Magdalene’s weeping, in C 6058 els has been read as meaning ‘oils’/’ointment’. The line is not found in O, V4 or L. 6060. The word len (= l’en) was added superscript at the appropriate point. 6067. C: mesclin, a scribal error for meschin (V7 6187).

La Chanson de Roland (Critical Text and Notes)

6070

6075

385

De maintes terres li comte palazin. Sor toz les altres ora le fil Pepin: «Damedeu pere, vos sai a voir divin: Le jor feïstes et la nui del serin, Et mer et terre deci en Marmorin: Et convertistes Saint Pol et Saint Fremin. Je sui vostre hon, Sire, jusq’a la fin;

6071. C: auor (for aïuor, ‘help’, or autor, ‘Creator’?); V7 6191 reads: autior with 1st half of u expuncted and slash over 2nd half to indicate i, to produce aitior (possibly for aiütor, ‘help’, as correction of autor (Italian autore), written initially in error?). Since V4 l871 reads: e’ voire devin, P 4662: vrais, T 3581: vrai, L 1919: veray, in C avor has been taken as a graphy of a voir, ‘in truth’, although the more usual phrase is de voir, en voir, par voir or por voir. 6072. del serin appears to stand for des le serin: ‘as from the evening’, ‘following the evening’. 6074. For details of St Paul’s conversion on the road to Damascus, see Acts of the Apostles, ix.1–22. St Fremin (also written Fermin and Firmin) is found in several different guises, three as bishop of Amiens; full details given in the ‘Bibliotheca sanctorum’ published by the Instituto Giovanni XXIII della Pontificia Università Lateranense, may be summarized as follows: (1) St Firminus of Amiens, abbot and saint, listed in the ‘Martirologio Romano’ as having a feastday on 11 March (but no further details given) may well be the product of confusion with (2) St Fermano, a tenth-century Italian abbot born in Fermo nelle Marche, whose feast-day is definitely 11 March, wrongly identified with the two Firminus of Amiens. These last are welldocumented bishops of Amiens: (3) St Firminus I, the Martyr, lived in the second half of the third century. Intriguingly, he was the eldest son of a noble but pagan family in Pampluna, the Spanish city historically linked with Charlemagne’s foray into Spain and with the attack at Roncevaux. After his parents were converted by St Saturnino, bishop of Toulouse, he received a Christian education, became a priest, then a bishop, and at last set out to evangelize Aquitaine, Auvergne, Anjou and finally North-Eastern Gaul, where he settled in Amiens and was martyred by decapitation on 25 September (kept as his feast-day), probably in 290 AD, possibly a few years later. There is testimony to a solemn cult of St Firminus in Pampluna in 1186 and further evidence of an altar and special Offices dedicated to him during the next few centuries, but there his feast was first celebrated on 10 October, then changed to 7 July in 1590. He is recorded in a mediæval litany of saints. Finally, (4) St Firminus II, the Confessor, became bishop of Amiens in the second half of the fourth century and died in Amiens on 1 September 390 (that date being celebrated as his feast-day). During his lifetime, he had had a church built over the tomb of his predecessor (3). Since the text here refers specifically to the converting of St Fermin, it seems likely that the allusion is to the pagan-born Fermin I. The well-documented transfer of some of his relics across France, from Amiens to Pampluna, in 1186 may well have made his story well-known and popular less than 100 years later. His continuing popularity in North-Eastern France is attested by the fact that his statue forms the trumeau of the left-hand portal on the West end of Amiens cathedral (building commenced 1220) and that his life is recounted on the fifteenth–sixteenth-century marble friezes around the choir. Furthermore, legend identifies Fermin II as a descendant of the Christian convert, Faustiniarus, a senator of Amiens who retrieved and buried the body of Fermin I after his martyrdom; so Fermin II is likely to have been born into a Christian family and thus not qualify as ‘converted’.

386

6080

La Chanson de Roland (Critical Text and Notes)

Faites sevrance de la gent Apolin, Qe Cristïen ne torgent a declin.» La voiz de ciel i tramist Saint Martin, Qi en sevra le lignage Caïn – Car sor chascun fist crostre un albespin: Encor les voient li gentil pelerin Qi a Seint Jaqe en vont le droit chemin. Karles se drece, si tint le chief enclin; Ses els essue al pan de son hermin.

 6085

6090

6095

6100

Charlles se drece, si a fait un sospir; Vit tote l’ost a une voiz fremir. Des Sarazins ne pot un seul cosir, Car Deus les fist espines devenir, Pognanz et aspres, si ne poent florir. «Baron,» dist Karles, «ben devons Deu servir. Nus ne se doit esmaier de morir. Nos amis voi a la terre gesir; Franc chevalier, pensez del enfoïr.» Et cil si firent, qe nes volsent guerpir: A peus aguz font les charners ovrir, Sisante o .c. en un seul lius covrir. «A! Deus,» dist Karles, «or devroie morir, Quant teus barons voi en terre covrir, Per cui je suel repouser et dormir, Qi me soloient mes granz oz esbaudir

93.b

6077. torgent is present subjunctive 6, torner, ‘to turn’. For the occurrence of -g- in the present subjunctive in Old French, cf. Pope (1934, 344, § 910). 6079. C: Qi e sevra, with ~ omitted; V7 6199 and P 4670: Qui en sevra. ‘The descendants of Cain’, ie descendants of the first murderer (Genesis iv, 1–15), is here a pejorative term for the Saracens. 6085. C: KCharles, with K extending over 2 lines and C nearly erased. The small c in the margin was disregarded, probably because the copyist had already, in error, written a standard-sized C at the beginning of the laisse. 6086. Both C and V7 6206 read: a une voiz instead of the more usual graphy, a une veiz, meaning ‘at one and the same time’, ‘as one’ or ‘simultaneously’: veiz = feiz = ‘time’. 6088. C: en spines, with intrusive Franco-Italian n, but the preposition is unnecessary; V7 6208: espines. 6096. Once again, luis as a scribal error for lius; cf. note to C 5668 and 6199.

387

La Chanson de Roland (Critical Text and Notes)

6105

6110

6115

E mes batalles sor Sarazin fornir Et la loi Deu esaucer et tenir! Or m’estora les grant penes sofrir, Porter mes armes et mon escu tenir! Mot ai dur cuer, quant il ne puet partir, Ne nostre Sire nel me velt consentir. Mon boisëor faites avant venir; Voiant vos toz le ferai ja saisir, En grant chäaines enserer et tenir, Q’il ne s’en puisse escaper ne föir. Ne nul de vos ne devroit consentir; Mais per la barbe dont voi le poil florir, N’i a si riche, s’il s’en ouse aatir De:l respitier ne del plus consentir, Q’ensanble o lui nel estoïsse morir; Sa traïson li volrai ben merir!»



6120

Nostre enperere se dreza en estant; Lo ciel regarde per merveillos talant; Vit la clarté et les angles chantant, Q’il vont les armes a grant joie menant. Les morz enterrent Borguegnon et Normant; Ainz q’il se partent, i fist Deus vertu grant: Permi les fosses vont les coudres croissant,

94.a

6103. C: estora for estovra, future 3 (impersonal), estoveir. The same graphy is found in C 7301. V7 6223 reads: Or mes stora, P 4695: Or m’estevra. 6108. C: les ferai, but this refers back to mon boisëor, C 6102, so the singular, le, is needed. V7 6228 and P 4700 read: le ferai. 6109. C: chaines, but 2 syllables are needed to avoid a hypometric hemistich and the normal graphy (cf. C 3205) is chaënes/chäaines. This implies omission here of one a. Cf. note to C 3143. 6115. C: esstoïsse, imperfect subjunctive 3 (impersonal), estoveir – with the ss of the verb ending leading the copyist to insert an extra s before the t. 6119. The C ms reading, la clarté des angles, makes good sense but leaves the second hemistich hypometric; V7 6239, P 4710, T 3520 and L 1945 agree in reading: la clarté et les angles. 6123. C: vont les tonbes croissant, but ‘tombs’ cannot grow ‘fresh and green’. V7 6243 reads coudres (‘hazel trees’); P 4714 has Parmi les tombes vont les corres naissant, T 3623 Parmy les tombes vont cordes croissant, L 1947 Permi les tonbes vont les codres levant – implying a confused source reading as the origin of C’s error.

388

6125

6130

La Chanson de Roland (Critical Text and Notes)

Beles et droites, fresches et verdoiant, Qi a toz jors i sunt aparissant. Colpent cez perches chevalier et serjant As bieres faire, dont trestot sunt dolant; Desus leverent Oliver et Rollant. A forz somiers les font porter avant; De Rencevals issirent a itant. Lors plora Karles et Nemes le vaillant Et li Danois, qi lo cuer ot dolant. Iloc refu la criée si grant, Nus hom de char n’i oïst Deu tonant.

 6135

6140

6145

6150

Quant l’enperere ot sa gent enterree Et la conpaigne qi fu bone eüree, Qi a duel fu en Renchevals livree, Chascune biere fu mot bien atornee, Sor les somiers et trosée et levee, Des .xii. pers dont France est adolee. Plore li rois o la barbe meslee, Sovent se pasme sor la sele doree. Les porz passerent un poi ainz l’ajornee; A Seint Johan vindrent a la vespree; La herbergerent la gent qi ert lassee. N’i veïssiez ne terre ne valee Ne combe oscure qi tant soit ennublee, O feu n’eüst ou chandele alumee, Ou cerge espris, ou lanterne enbrasee. Entor les bieres fu mot grant la criee, Par tote l’ost est la noisse alee.

94.b

6146. The symbol 7, standing for er or re, has been added over the final e of veïssiez, perhaps in a mistaken attempt to convert the (perfectly acceptable) imperfect subjunctive 5, veir, into the conditional 5. The extra syllable would make the hemistich hypermetric. 6150. C: Entor les cerges, but the word cerge occurs immediately above, in C 6149, as one of a list of lights: fires, candles, lanterns. Here it seems to be a scribal error for bieres, the reading in V7 6270.

389

La Chanson de Roland (Critical Text and Notes)



6155

6160

6165

6170

6175

6180

La nuit i jut nostre enpereres ber; Li autres poeples pense del osteler. En son les lances font cerges alumer. Iloc fist Karles un moster estorer, Por son nevou San Johan apeler, As piez des porz, si com l’on doit passer. Meint bon vasal i veïssiez plorer, Lor meins detordre et lor chevols tirer, Lor dras de soie desronpre et desramer! Li enperere ne puet nul conforter; N’est pas mervelle, ja n’en doit nus parler, Quant cels voit morz per cui doit reposser: Ce est Rollant e Oliver li ber, Qi suelent faire terre soz als tranbler, Paiene gent batizer et lever, Trestot le monde envers als acliner. Quant Karles vit au maitin ajorner, Nevelon fait devant soi apeler, Gerart de Blois et Gui de Seint Omer, Joifroi d’Anjou qe il pot tant amer: «Baron,» dist Karlles, «je vos ai fait mander; Un mien servise vos voudrai comander: .C. chevaliers me faites conraer; Par la gastine vos convient a paser, Droit a Viene, qe tient Girart li ber. Dites au duc c’a moi viegne a parler; Se m’ament Aude qi tant a le vis cler, Et si li dites qe la vorai marier. Jo la cuidai a mon nevou doner, Mais Guenelons nos a fait desevrer.» Li rois se pasme, ne poit sor piez ester; Naymes l’a fait desor soi acliner, Desor son cors un petit repouser.

95.a

6156–6157. St Jean-Pied-de-Port was razed by Richard the Lionheart in 1177, then rebuilt on a different site a short distance away. 6168. As Mortier notes, V4 (laisse 316) takes up the thread of the story again at this point, after the Narbonne digression, henceforward in a version almost identical to that of C and V7.

390 6185

6190

6192a(P)

6195

La Chanson de Roland (Critical Text and Notes)

Li rois revint, cels prist a esgarder: «Baron,» dist il, «puis moi en vos fier?» Il lor a fait a toz sor sainz jurer – Et a toz cels qi devoient aler – Qe il feront cele dolor celer «Tant qe je puisse au duc Gerart parler. Et la bele Aude volrai reconforter. Se li pöoie sol icest duel enbler, Plus en avroie le cuer et sain et cler. S’ensi nel faites, ja n’i porai parler; Ainz li vesrai le cuer el cors crever.» Lors fist li rois reliqes aporter; Quant ont juré, si se vont repouser; Es chevals montent; n’i ot qe del esrer. Et Karles fist ses grasles soner; Tot droit a Blaives fist les berres torner.

 6200

Quant l’enperere revint de pasmison, Del conforter se penent li baron Et duc et conte qi furent environ. Karles apele Basin le Bergoignon, Garin d’Alvergne, Mile de Besençon,

6185. C: Li rois revint co cels…, the last word possibly being written cals (c’als?). 6187. C: soz in error for sor, perhaps due to the close proximity of toz. 6192. P’s reading (P 4781) is necessary to complete the sense of this line and is supported by V4 4439. Both C and V7 omit the line. 6198. In C, the cæsura is placed after fist, with the first hemistich Et .K. fist – but this makes the second hemistich hypometric. Similarly, T 3684 reads: Et Charles fait, followed by a 5syllable second hemistich; but V7 6319 reads: E l’emperere fist ses grailes soner; as do P 4787 and L 1994. If .K. is transcribed as Charllemeine with cæsura immediately following, the second hemistich has 6 syllables. 6199. In the C ms., the reading is Blauies with the i clearly marked; Foerster, Mortier and Duggan all transcribe it thus, making the first hemistich hypermetric. The -iu- combination is also found apparently transposed in luis for lius, ‘places’, in C 5668, 5856, 5878 and 6096, and in sauies = savies or saives (saujes?) in C 6787, 6801 and 6827; and the same problem with Blaives/Blavies/Blaujes recurs in C 6482, 6483, 6522, 6826, 6835, 6985. The simplest explanation is scribal error in the careless placing of the ‘slash’ of the i over the final vertical stroke of the u. In all cases, reading -ui- produces faulty metre and sometimes nonsense. It is here (and later) emended to Blaives. Cf. Introduction, 71–72.

391

La Chanson de Roland (Critical Text and Notes)

6205

6210

Gui de Nevers et Girart d’Orion: «Baron,» dist Karles, «entendez ma raison! De mon servise vos pri toz et semon: Vus m’en irez a la citez de Mascon Por ma suer Berte o la riche fazon. Celle fu dame au riche duc Millon, Puis la donai al conte Guenelon. Rendu m’en a mot malvai guisredon.» Et cil respondent, «Fallir ne vos devon.»

 6215

6218a(L) 6220

«Franc chevalier, encor vos dirai al: En chascun leu o vos prendez estal, Celez mot bien le damage mortal. Dites ma suer, qe je teing a loial, De franc corage, qi ainc ne pensa mal, Que vigne a moi, nel lait per rien mortal. Je la donai a cil traïtor mortal Qi m’a tolu tant nobile vasal Et a bien proef ma corone roial –

95.b

6208. Vus m’en irez involves the ethic dative (‘for me’, ‘on my behalf’), very common in Old French. In the second hemistich, V7 6328 reads: a la cité de Mascon, hypermetric as in C– but Duggan reduces V7’s cité to cit, on the model of P 4798: a la cit de Mascon. However, T 3695 has: en la cité Mascon, dropping de. Both cit and citez (or cité) are viable alternatives, but the choice of emendation is difficult in view of this variety; so C’s line has been left hypermetric. Is it possible that a la, pronounced with a very liquid l, was reduced to a single syllable, al? (Cf. C 6788, where the same problem occurs). 6210. As Foerster and Mortier note, the C ms. first read: Celle fille…, but the final letters of fille were then erased, leaving only fu and a gap in the line; dame was then added at the end of the line, with a mark for its insertion after fu. 6214. C: vos donrai al, but V7 6334 and V4 4458 read dirai, which is what the sense calls for. 6217–6218. As it stands, this is an incomplete sentence: we never learn what the messengers are to tell Charlemagne’s sister. V7, V4 and P have the same reading as C and clearly omit a line, which is supplied by L 2009, as inserted above. 6219. C’s reading, a cil traïtor mortal, apparently produces a 7-syllable hemistich, but see Introduction, 92–93 regarding the problems of metre involving the word traïtor. Although V7 6339 reads al trahitor, stressing the 2 syllables in -ahi-, P 4810 has: au traïtor and T 3707: au traïtour. C’s reading has been retained because it provides evidence of the fluctuating value of aï in the thirteenth century.

392

La Chanson de Roland (Critical Text and Notes)

Jesus l’en rende un si fier batistal Qe toz li monde l’en esgard comunal!»

 6225

6230

Li .v. baron qe rois Karles semont Congié demandent, puis montent, si s’en vont. Grant duel ot Karles, quant mot ne lor respont; Mot a fier cuer quant de dolor n’en font. Et cil chevauchent qi de l’ost parti sunt; Il ne redotent ne pui ne val ne mont. Prochenement arere revenront; La suer Karllon avec aus amenront. A! Deus! Qel duel quant il asanblaront, Quant Aude et Berte ensanble ajosteront!

 6235

6240

Li rois chevauche, qi s’est mis el retor Par la gastine envers Terre Major. Franzois regretent durement lor seignor: «He! Rollant, sire, qel duel et qel iror! Qi donra meis ne chastel ne honors, Ne bones armes, ne destrier misoldor!» Vindrent a Sorges, o herbergent le jor

6222–6223. C, V4 4466 and P 4813 read: batistal, V7 6242: batestal. The meaning ‘noise (of combat or pursuit)’ does not fit the context, but Tobler-Lammasch also gives ‘fight’, ‘battle’; even better, in his 1995 edition of V4, Beretta (1995, Glossary, 408) cites a secondary meaning, ‘punishment’. To complete the sentence, C 6223 reads: ‘that all the world, all together, judges him over it’. Here, C has toz li, plural, with the singular substantive, monde – and tout le monde = ‘everyone’ is supposed to be a later development, although V7 6343 has: tot li mondes, P 4814: touz li mons, T 3710: tout le mont. P and T, like V4 4467 (tote li pople), support the rest of C’s reading. Whilst this might be an early instance of tout le monde = ‘everybody’, it reads better as a scribal error (in an early, common source) for Qe toz l’imondent en esgard comunal: ‘That all condemn him in common judgement’. 6228. The Latin sunt contaminates the reading here (despite the -ont rhyme); but sont is found only twice in C: in C 6419 and in the form son, C 2552, contaminated by son (‘sound’) in the same line. 6232. The word il has been added in superscript. 6240. C: herbergerent, which, in mid-hemistich, has 4 syllables, making the hemistich hypermetric. V7 6360 reads: la erbergent le jor, using the (historic) present instead of the past definite, and this is supported by P 4832 and T 3727.

393

La Chanson de Roland (Critical Text and Notes)

6245

6250

6255

Sor la rivere, qi fu de grant laor. La crut au roi grant duel et fort iror: Eschapa li son felon boisëor, Guenes li fel, qi mut la grant dolor, Dont doce France torna en grant freor. Guenes s’adobe com on de grant feror; El dester monte Garin de Mont Arsor; En fuie torne, n’ot cure de sejor. Quant la novele en vint l’enperaör, «A! Deus!» dist Karles,«S’or pert mon traïtor, Jamais n’arai ne joie ne baudor! Or i parra se j’ai nul pugneor Qe:l me rendra – je li crostrai s’onor Et a sa vie ert saissi de m’amor.» Franzois entendent le cleim de lor seignor; Mil en i sallent par force et par vigor; Avant fu traiz meint destrier misoldor; Deci qel vespre ne prist nus d’al retor.

96.a

 6260

Fuit s’en fel Guenes sor un destrier morois; Aler s’en quide en son reigne aspois, Ou a Tolete o as altres manois. Per grant vertu i sallirent Franzois, Plus de .x.m. i sallent demanois.

6253. Qe:l me rendra: ‘Who will bring him back to me’. 6259. The initial capital letter of this laisse resembles H rather than F, but f is indicated in the margin. 6260. C reads: a espois, possibly for a espoir (s added by poetic licence and assimilation of r, as in manois = manoirs, C 6261) ‘hopefully’, but this is tautologeous after quide. Reading a Espois means this is an unidentified locality, but P 4853 has enz ou regne espaignois, T 3743: eu regne en Espaignois and Michelant Fragment No.16 en son regne Espaignour. V7 6380 reads a spois with the e inserted superscript by a revisor. If this emendation is faulty, V7 6380 may be read as ending: en son regne aspois – i.e. ‘into his lands in the Aspe region’, the Col d’Aspe being the Pyrenean pass already mentioned as Aspre in C 1250 and 1475. Reading aspois as the adjective deriving from the (River) Aspe means that, for the metre, there is no elision between reigne and aspois in C or V7. 6261. The word as is added in superscript. The final word should of course be manoirs, but the r has been omitted for the sake of the rhyme – another example of pseudo-rhyme or poetic licence.

394

6265

La Chanson de Roland (Critical Text and Notes)

Danz Otes sist sor un dester norois, Toz eslaissiez s’en va devant Franzois. Forment l’enchaucent permi les vals d’Orcois, O li esclos de son cheval sunt frois.



6270

6275

6280

6285

6290

Par la gaudine va Guenellon fuiant Vers Saragoze, la terre a l’amirant. Passa un tertre et une aive corant, Enz el chemin vit une gent esrant: Marchëant sunt, si vont foire qerant. Il les salue toz primereins avant; Cil li demandent de qel part vient esrant Et de chemins, com il sunt aquitant. «Seignor, mot bien,» ce dist li sosduiant. «N’a home en terre n’en cest siegle vivant Qi pas vos tolle un seul denier vallant. Ci venent genz qi tos me vont chaçant; Ocis lor ai un lor ami vaillant – Je n’en poi mais, je:l fis moi deffendant. Les vos merciz, dites lor en itant, Bien puis ja estre .v. lieues la avant.» Icil respondent, «Tot a vostre comant.» Otre s’en passe sor le destrier corant Et cil s’en tornent tot lor chemin esrant; L’aive passerent et le tertre pendant. Oton encontrent desor un auferant. Il lor demande, «Bau seignor marchëant, Veïssiez vos un chevalier esrant?» Et cil respondent, «Folie alez qerant. S’ensi se tient com il fait le samblant, N’ert mais balliez per nul home vivant. .v. lieues larges, sanz menzoigne disant,

96.b

6266. C: vas; V7 6386 reads: valis with l inserted in superscript; P 4858 reads: vauls. 6276. C: les sosduiant, but this refers to Ganelon, so the subject form singular, li, is needed. The initial s of sosduiant appears to have contaminated the preceding article. 6278. The verb tolle is present subjunctive 3, toldre; thus ‘there is not a man… who would rob you of a single penny’.

395

La Chanson de Roland (Critical Text and Notes)

6295

6300

A ja passé la grant aive corant.» «A! Deus,» dist Otes, «Qi fus en Biauliant, Si solouz basse qi se vait declinant, N’iert mais bailliez, je:l sai a esciant.» Arere torne a cels qe:l vont sivant, Si lor conta le dit del merchäant; Cil s’en retornent, si vont l’enchaz laissant.



6305

6307a(P)

6310

6315

6320

Cil s’en retornent qi ont l’enchals laissié; Tot droit a l’ost sunt Franzois repairié. «Baron,» fait Karles, «com avez esploitié? Avez vos pris Guenelon et lié?» «Sire», dist Otes, «mot l’avons enchacié. Se Deus m’aït, ne l’avons pas baillié. Enquis avonz forment dou renoié A merchëanz de Limojes le fié, Qe encontrames a un tertre puié. Cil nos ont dit, conté et enseigné Qe bien puet estre .v. lieues eslongié; Enz en un bos s’est li fel enbuschié.» «A! Deus!» dist Karles, «Qel duel et qel pechié! Sainte Marie, com m’avez oblié, Qe je ne:l teing, tant q’en fuse vengié! Otes,» dist Karles, «vers moi avez boissié, Come coharz avez le canp laissié. Alez vos en; je vos doins le congié, Car coarz hon ne doit estre esaucié, Ne de seignor tenir castel ne fié.»

97.b

6297. The graphy solouz, a variant of the normal subject form singular, solauz, shows FrancoItalian confusion of vowels. 6298. In C, mais, initially omitted, was added at the end of the line with insertion mark before bailliez. 6299. C: suiant, but this should be sivant, present participle, sivre, as Mortier points out. V7 6419 and T 3785 read: sivant, P 4897: sievant. Cf. note to C 6199 regarding -ui- and -iu-. 6308. In order to make sense of this line and to explain the indirect object, A marcheanz, with which it begins, another line is needed after C 6307. V7 6428 has a reading identical to C’s (indicating an error in the common source) and V4 has a much more concise text here, but P 4907 and L insert an appropriate line. (L 2060: Enquis avons assez dou renoié.)

396

La Chanson de Roland (Critical Text and Notes)



6325

6330

6335

6340

Dolanz fu Karles, irez et abosmez, Por Guenelon qi li est eschapez. De ce fist Otes qe chevaliers menbrez: Ist de la cort sanz congiez demandez; Tresc’a sa tente s’en est mot tost alez; O lui ala Sanses et Ysorez, Deus barons riches qi de lui sunt privez. «Baron,» dist Otes, «mot doi estre adolez. Jamais en terre ne serai honorez, Por Guenelon qi nos est eschapez. Alez, as armes, chevalier, si montez! Mes cuers me dist qe ja sera trovez!» Li dui baron ont bien lor cors armez; Otes meïsmes s’est mot bien conréez; Monte en Morel qi fu l’amustadez; Issent del ost, n’i furent arestez. La lune luïst, qi lor dona clartez. Trestote nuit ont les chevals hastez, Toz les esclos qi Guenes est alez. Passerent l’aigue et les porz et les guez; Uns païssanz les avoit encontrez; Li pautoners vit les vassaus armez, Mot bonement les a araisonez: «Je n’ai avoir ne diners moënez.

6329. C: coronez, but this is quite inappropriate; there is never question of Otes being ‘crowned’. V7 6449 reads: honorez and V4 4567: honoreç. 6335. Morel may be the horse’s name or it may be used as an ordinary substantive, meaning ‘horse’. Both C and V7 6455 read: l’amustadez, a graphy not found in any dictionary. The closest approximations are amustal and amustant, titles for which ‘Governor’ is the nearest equivalent. V4 4575 has que fu a l’amirez and P 4934 qui fu al amiré (graphies of amiral, ‘emir’), so the horse belonged to a Moorish ruler. (C and V7 imply possession.) Thus amustadez could well be a graphy of (or error for) amurafle, found in O as a Saracen title of honour. Given the mediæval penchant for allotting meaningful names, it is interesting that the verb amoustarder means ‘to dupe’, ‘to deceive’. If this is the origin, then ‘deceived (in his belief)’ implies a Moslem – thus Morel is an Arab stallion. 6339. There is typical confusion here of qi and qe. 6342. C: vaissiaus, but the reading should clearly be vassaus, as in V7 6462. Godefroy lists vassal as a graphy of vaissel (‘dish’, ‘coffin’ or ‘ship’), so a reverse connection may have been made by a confused copyist: vaissel/vaissail/vaissial as a graphy of vassal.

397

La Chanson de Roland (Critical Text and Notes)

6345

6350

6355

6360

Uns colers sui, meint fardaus ai portez; De deus lasrons ai esté desrobez.» «Amis,» dist Otes, «bien es aseürez! Mais d’une chose me diras veritez: Se par ci passe uns chevalier armez?» «Nenil, bau sire, por saintes loiautez, Fors d’une chose, me sui je porpansez: Esgardez, sire, cez grant arbres ramez Qe vos veez ensor cel pui plantez – Iloques dort uns chevalier armez! Ses bons escuz est sor son dos tornez Et ses destriers a son braz aresnez, La crope fauve et toz blans le costez.» Et «Deus,» dist Otes, «per les toies bontez, C’est li traïtes, je:l sai c’est veritez!» Il point Morel des esperons dorez – N’est nule beste ne cers tant esfréez, Qi se tenist deus arpenz mesurez.

97.b



6365

6370

Ote esperone le pendant d’un costal, Car mot redote icil traïtor mortal; Se il s’esveille o se rent al cheval, N’ert mais bailliez par nul home charnal; Et de ses armes le seit tant a vasal, Se il fu si com autre home loial, N’eüst mellor en France la roial. Li bons destriers vit venir le vasal,

6364. C’s second hemistich is apparently hypermetric, but much depends on whether the -aïof traïtor counts as one or 2 syllables (cf. Introduction, 92–93). It is always disyllabic in V7 and here V7 6484 reads simply: le traïtor (4 syllables), perhaps correcting the CV7 model. C may have read this as icil traitor (4 syllables) or made a scribal error in writing icil for cil (i.e. cil traïtor in his model) . 6365. C: se ret al cheval, but V7 6485 and T 3852 have: se prent, V4 4600: se prende and P 4965: se prend. In fact, allowing for a missing ~, se rent makes better sense than se prent: ‘gets on horseback again’ rather than ‘joins his horse’ or ‘turns to his horse’. 6367. C: seit with ~ added in error, producing seint with inorganic n. V7 6487 reads: set in an otherwise identical line. (cf. V4 4602: Car de ses armes il est prou e vasal; P 4967: Car de ses armes i a si prou vasal).

398

La Chanson de Roland (Critical Text and Notes)

Si henist cler, tot fait tentir le val: Per grant freor saut sus le desloial Et trait l’espee, tint l’escu comunal.

 6375

6380

6385

Grant fierté mene le destrier misoldor; Cil saut en piez, qi de mort ot pöur, L’escu enbrace com on de grant vigor. «Ahi, traïtes,» dist Otes per iror, «Vos n’alez mie com on de grant valor! Je vus comant de per l’enperaör, Tornez arere, si croistra vostre honor. Mot se fioit en vostre grant vigor, Quant al mesage vos tramist l’autre jor; Au roi Marsille feïstes tel felor, Dont doce France est tornée a dolor; Et la roïne vos i dona s’amor. Vos en pendrez en son cel pui auzor!» Guenes l’entent, trait le brant de color, L’escu enbraze com on de grant valor.

98.a

 6390

6395

«Otes,» dist Guenes, «vient en i meis qe vos? De moi laidir sanblez mot orguellos! Faites le bien, com hon chevaleros: Monter me lasse desor mon destrier ros; Conbaton nos per nos cors anbesdous; Ja mostrerons de nos lances les trous.» Ce respont Otes, «Trop estes engeignos! Se estïez el destriers rabinos, Ja fuiriez permi cez vals erbos; Je en serai vers le roi vergognos.

6394. These trous or tros may be the holes in their armour or wounds in their bodies made by the lances, but trous/tros is also a graphy of tors (sm), meaning ‘fragments’ or ‘splinters’, implying that the two combatants would strike one another with such great force as to shatter their lances.

399

La Chanson de Roland (Critical Text and Notes)

6400

Mal dehait ait ja en ert cremetos! Prenez l’escu, si montez sor le ros! Se vos fuiez, trop ferez qe boisous. De l’enchaucer serai maltalentos; Avant Morel ne puet garir le ros.»

 6405

6407a(V4)

6410

6415

6420

Por le congié s’est Guenes esbaudi; L’escu enbraze, lo destrer a saisi, Car de ses armes fu bien amenevi. Se il n’eüst vers son seignor guenchi, N’i a si bon civaler in France la centils. Danz Otes a hardiment recolli, Broche Morel des esperons feri. Otes li cons a son espié brandi; Dist au felon, «Vasal, je vos desfi! Ersoir au vespre fui mot per vus laidi.» Ce respont Guene, «Vers moi t’es aati, Quant je me sui de mes armes saisi Et sui montez sor Fauvel l’arabi; Desfié m’as, tot autretel te di, Mot corte joie en aront ti ami.» Lors se requirent li mortel enemi; Lor escu sont et fendi et parti Et chascuns d’aus fu el cors mot blesmi; Otes trebuce et Guenellons chaï; Del redricer ne sunt mie esbahi.

98.b

 Li dui baron pensent del relever; Si s’entresferent come chevalier ber 6399. ‘Cursed be the man who is afraid of this!’. This is Otes’ exclamation as he decides to take the risk of Ganelon’s escaping. V7 6519 is identical to C’s reading, which leaves the relative pronoun implicit. P 5001 and T 3882 insert qui but modify the line slightly. 6405. Because of a fault in the parchment, this line is inset slightly. 6407. C, V7 and P have the same reading here, omitting the line needed to complete the sense of C 6407 (as Mortier noted in his edition). V4 4642 provides the additional line. 6412. C: vui, possibly due to contamination by fui earlier in the line or influenced by the -i of the rhyme. V7 6532 reads: vos, but vus is frequently found in C and is closer to the ms. reading.

400 6425

6430

6435

6440

6445

La Chanson de Roland (Critical Text and Notes)

Qe les escuz se font des cols voler Et des verz elmes font les pieres qasser, Plus fierement qe liupart ne sengler, Quant li dui autre comencent a crier: «Cuivert traïtes, ne savez o aler!» Quant voit fel Guene ne lor puet escaper, Per nul engin son destrer recovrer, S’espée rent, merci prist a crier. Senpres le font devant als desarmer, Sor un destrier isnelement monter – Sor le plus lent qe il porent trover – Et deus osberz derere lui troser Et un escu li font al col porter; Tot droit a l’ost pensent del retorner, O l’enperere fait les beres garder. Quant il vit Guene, si comence a plorer: «Otes,» dist Karles, «gentius estes et ber! O repeüstes mon traïtor trover? A mot grant tort vos fis ersoir blasmer. Tenez mon gaje por le droit presenter.» «Sire,» dist Otes, «ice laissez ester! Vos estes rois, je sui un bacheler; Quant vos plaira, sel posrez amender.»

99.a



6450

6455

«Droit enperere,» dist Otes a Karllon, «Je vos ren pris le conte Guenellon. Por lui ateindre sui mis en grant randon.» «Otes,» dist Charlles, «cuer avez de baron; Rendre vos qit mot riche guesredon.» Ogier apele et le conte Sanson, Foqe del Mans et d’Estanpes Aton, Et le visdame qi fu de Chaëlon: «Je vos comant cest traïtor felon. Gardez lo bien per tel condicion:

6442. ‘Where were you able to find my traitor again?’ with the prefix re- (‘again’) attached to peüstes, past definite 5, pooir rather than to trover, for the sake of the 4:6 metre. 6448. C: eperere, with ~ omitted over the first e.

401

La Chanson de Roland (Critical Text and Notes)

6460

6465

S’il vos eschape, ja n’arez raënzon! Por tant c’on puisse chaucer mon esperon, Je n’en prendrai nul rachat dou felon, Qe je nel pende en alt come lasron. Quant dant Gerart vendra et si baron, Et la bele Aude o la chiëre fazon, Fiere venjance en prendrai del gloton; Sel jugeront tuit mi mellor baron.» Et cil respondent, «Mot bien le garderon, Dusc’a cele eure qe nos le vos rendron.» Li rois se pasme sor l’espaule Naymon; Ogier l’enlieve et tuit si conpeignon.

 6470

6475

6480

Quant li rois fu de pasmison venu – N’est pas mervelle se il pert la vertu, Per un petit q’il n’a le cuer fendu – Rollant regarde soz le paille o il fu. «Beaus niés,» dist il, «com or vos ai perdu! Et la bele Aude, qi de vos fist son dru! Des noces faire me sui je trop tenu. Guene, traïtre, qel duel m’avez meü, De mon neveu q’a Marsille as vendu! Las, g’en serai si sofraitos et mu. N’en puis el faire, car del tot l’ai perdu!» Gascogne passe a force et a vertu; Deci qu’a Blaives sunt ensanble venu.

99.b

 A Blaives fu li rois o son enpire; La oïssiez si fait duel et tel ire,

6472. q’il n’a…, a pleonastic negative. 6482, 6483. In both lines, the graphy is Blauies, but this must be read as Blaives (or Blaujes) to obtain the correct metre in C 6482. Cf. note to C 6199 regarding C’s repeated scribal error in marking the i.

402 6485

6490

La Chanson de Roland (Critical Text and Notes)

Sonent cez sainz et font cez messes dire, Chanter vegiles, faire cez sauters lire! Plore li rois, sa blance barbe tire, Son bon blishalt en desront et dessire. Des mesagers vos doi mais hui ben dire, Qe a Viene ot envoié lor sire: Il furent .c., mas mot furent plen d’ire Por les barons qi sunt mort a martire; En Renchevaus les desconfist Marsire; Ja n’ert mais jor qe France n’en soit pire.

 6495

6500

Li cent mesage ont mot bien esploitié: Les tertres passent, forment sunt travaillié, Tant q’il vinrent a Viene le fié; Icele nuit sunt mot bien herbergié. Lors d’Alemegne fu Girart repeirié, Quant vit les mes, mot per ot le cuer lié; Asez les a acoilé et basié. Puis lor demande, n’i a gaires targié: «Qe fait mesire? Laissastes le haitié? A il mot bers en Espegne estié?»

 6505

6510

«Sire Gerart,» ce dient li mesage, «Tote a Espaigne conquis per vasalage Et d’Almarie reçut le treüsage. Li rois Marsille pensa mot grant otrage, Cui eschéoit l’onor et l’eritage: En Rencevals lo volst faire domage,

6485–6486. ‘They ring these bells and have these Masses said, Offices for the dead sung’ poses no problems, since the infinitives dire and chanter depend on font; but faire ces sauters lire, must also depend on font, which becomes rather complicated: ‘they have someone have the psalters read’. V7 6606–6607 has the same reading as C, but P 5295–5296 changes the infinitives in C 6486 to finite verbs: Chantent… et font… (reading et vont la messe dire in the previous line). T and L have only the first line. 6499. C: Girat (and also in C 6520), but Girart on all other occasions in C and Girart in V7 6620.

La Chanson de Roland (Critical Text and Notes)

6515

Sor lui torna li deus por son folage; Ocis i furent li Sarazin evage; De tote Espeigne reçut le treüage. Bers ert li rois, ben velt qe l’on le sage, Forment vos aime, vos et vostre lignage; Croistre vos velt d’onor et de bernage.»

403

100.a



6520

6525

6530

«Sire Gerart,» dient li mesager, «Saluz vos mande Karllemene au vis fier; Forment vos aime, en son cuer vos a chier.» («Deus,» dist Gerart, «toi puisse gracïer.») «Ses genz a faiz d’Espeigne repairier Et sor Geronde a Blaive herbergier; Iloc se fait ventoser et segner, Et les malades reposer et beigner; Le duc Rollant et le conte Oliver, Cil vont sovent en bois e en gibier. Per nos vos mande l’enperere au vis fier Q’alez a lui, pensez de l’esploiter. Et la bele Aude, qi suer est Oliver, Al duc Rollant la donra a moillier; Vos la vesrez a Blaives nozoier; Mot fier doaire li a fait otrïer: Trestote Espegne li velt Karles laiser.

6512. The term evage was often coupled with Sarazin to indicate the Arab pirates who attacked the pilgrims travelling by sea to the Holy Land. (Cf. the thirteenth-century Roman de Jourdain de Blaye (1991, 2693) at the point when the eponymous hero is captured by Arab pirates to be sold as a slave; lines 2695–2699 tell of pilgrims being taken and killed.) As an adjective meaning ‘inhabiting the banks (of rivers)’ or ‘inhabiting the shores (of seas)’, it may well designate Arabs from all around the Mediterranean, from the Dardanelles round through North Africa, thus covering all Marsile’s allies. It retains the pejorative sense of ‘piratical’ in many cases. 6518. C: .K’., which must be read as Karllemene in order to obtain a 6-syllable hemistich. 6520. C is alone (and in error) in inserting this line at this point, completely out of context. It occurs again, with correct spelling of Girart, in its proper place as C 6535, ending the laisse (cf. V7 6656). 6522, 6531. C: Blauie for Blavie, but see note to C 6199 and Introduction, 71–72 regarding the scribe’s graphy of -iu-. Emendation is necessary in both cases, to avoid a hypermetric second hemistich.

404

6535

La Chanson de Roland (Critical Text and Notes)

For ment vos qide et crestre et esaucier.» «Deus,» dist Gerart, «toi puisse gracïer.»



6540 6540a(V7) 6540b(V7) (V7)

Quant Gerart ot qe Karles l’ot mandé, Andeus ses meins a tendues vers Dé: «A! Rois de Gloire, tu soies mercié! Ier ere riches, or sui plus asauzé! Et quant mesire [a fet sa volunté, Dame Giborc, oiez com grant fierté, Quant mis lignages] sera al sien meslé. Ou est belle Aude al gent cors honoré?» Respont Guibor, «Bon or fust Karles né! Por bon seignor est vasal redoté; Cist tient henor et mot grant richeté.»

 6545

6550

Dame Gibor a la novele oïe; Del fier mesage est forment esjoïe; Vint en sa chanbre, ne s’en atarja mie: «Bele niéce Aude, or vos crost segnorie Del mellor conte qi onqes fust en vie: Ce est Rollant, cui vos estes amie, Q’a vos reqert la soie baronie; Cist mariajes ne poet demorer mie.» Lors l’a Guibor conréee et vestie: Ele est plus blanche qe n’est rose espanie;

100.b

6534. C: Fort vos qide, hypometric due to scribal error. V7 6655 reads: Forment vos cuide. 6539. asauzé is a graphy of esaucié; but V7 6660, second hemistich, reads: or sui je plus assé. 6540. C: Et quant mesire sera al sien meslé, which makes no sense. V7’s reading in lines 6661– 6663 shows that the scribe has omitted 2 lines between the first and second hemistichs of C 6540. Their re-introduction serves to explain Respont Guibor, C 6542, and is supported by P 5356–5358 and T 4180–4182. 6542. The second half of this line, which is nearly at the bottom of the page (line 30 of folio 100 recto) was left completely blank in the ms., but V7 6665 completes it. V4 4762 reads: en bon ora fu né; P 5362: Buer fust donc Rollans nés and T 4186: Le bon fust Charles né. One would in fact expect a bon or or buer (as in P) here.

La Chanson de Roland (Critical Text and Notes)

6555

6560

405

La pelice est del visanz de Nubie: Ovree fu el regne des Claudie Et perdesus de volsure garnie. Li dux Girarz l’achata a Pavie; Qi ce vendi, mot ot grant manantie. Mot fu ben Aude conréee et vestie.



6565

6570

Mot est belle Aude quant el fu acesmee; La granz clartez li fu el vis montee; Soz ciel n’a rose qi si soit coloree Qe sa bautez n’ait tote trepasee. Dame Gibors l’a el palais menee: Tote la sale en fu enlumenee. Franzois regardent, chascun l’a saluee. Jofroiz d’Anjou l’en a araisonee; «Franche poucelle, bien serez mariee! Del meillor home serez vos esposee Qi portast arme ne ferist cop d’espee.»

6555. C reads: de luisanz de N…(with the first de inserted superscript), but luisanz exists only as an adjective, ‘shining’; can this be some shining or gleaming fabric? V7 omits this line and it seems to have posed problems for other copyists: V4 4771–4772 reads: In dos li vest un pallie d’aumarie/Ce fu ovreç in l’islie de Nubie; P 5373–5374: Elle ot chemise de soie d’Aumarie/Et par desuz un paile de Pavie and T 4204: La pellice yert de l’uissart de Rossie. T’s reading provides the clue: it agrees with C in referring to a fur-lined cloak (pelice) and, if T’s uissart (= vissant?/ visant?) is a mediæval graphy of the modern vison (‘mink’) – and ‘from Russia’ would support this – then C’s de luisanz must be read as del visanz, again meaning ‘mink’. (In the recent refurbishment of Dover Castle by English Heritage, care was taken to obtain Russian squirrel fur, as used originally, to line covers on Henry II’s bed.) 6556. Langlois lists both Esclaudie and (des) Claudie as a Saracen country, but gives no further identification. Perhaps the writer was familiar with Roman history and alludes to Italy as the ‘kingdom of the Claudii’. There were two Roman emperors of this name: Tiberius Claudius Drusus Nero, reigning 41–54 AD as Claudius I, and Marcus Aurelius Claudius, reigning as Claudius II 268–270 AD. Foerster and Mortier read: des Claudie and mention of Pavia in Italy, C 6558, supports this; but Duggan reads (V7 6679) d’Esclaudie.. 6558. C: achete, present 3, but V7 6681: acheta, past definite 3, makes better sense in the context. 6560. C: conree, but this is the past participle of conréer, ‘to adorn’ or ‘to dress’, and should also be feminine. In a sequence of 4 e: conréee e…, haplography is understandable. 6563. C: sor ciel, which is clearly an error; V7 6686 reads: soz. 6568. C: Ajou with ~ omitted in error.

406

6575

La Chanson de Roland (Critical Text and Notes)

«Oncle Girart,» dist Aude la senee, «Quant vos plait, sire, qe m’avez mariee, Congié demant a Guibor la senee, Qi m’a norrie en sa chambre pavee Com s’el m’eüst dedenz son cors portee.»



6580

6585

«Belle niëce Aude,» dist Gerart le guesrer, «Cest mariage ne vel entrelaisier. Vos fustes fille al bon conte Reiner E niëce Hernaut de Baulande le fier, Si estes suer al bon conte Oliver – En nulle terre n’a meillor chevalier. Il et Rollant me firent apaier Et m’acorderent a Karllon au vis fier; Des icel tens vos i fis acointer. Metez vos seles, pensez de l’esploiter.» Lors ont grant duel tuit li cent chevalier.

101.a



6590

6595

Girart apele Bernart et Amaugis, Bos de Lions e Gui de Montamis: «Faites la gent monter de mon païs: Cent chevaliers de cels de Comarcis. N’en i ait nul qi n’ait peliçon gris Et chier mantel et bons cheval de pris! Car mandé m’a li rois de Seint Donis; Li sien mesage m’ont richement reqis. Et ma niëce Aude, qi tant a cler lo vis, Chevauchera le mul qe fu Claris: Desoz Baulande en un ost le conquis; Le Sarazin a m’espee en ocis;

6583–6584. These lines imply that Bertrand de Bar-sur-Aube’s thirteenth-century chanson de geste, Girart de Viane, was already written – or that the story it relates was already familiar to C’s readers/hearers. 6591. C: Cent chevaliers et de cels. 6595. C: mot for m’ont, the symbol ~ being omitted in error; V7 6718 reads: m’ont.

407

La Chanson de Roland (Critical Text and Notes)

6600

Li muls est blans plus qe n’est flor de lis. Conduira nos Floires, li fius Paris; Cil set mot bien les destors del païs. Quant nos venrons au roi de Seint Donis, Mot voluntiers viesrons les nos amis.»

 6605

6610

6615

6620

Li dus Gerart ne s’i atarze mie; L’aindemein monte per son l’aube esclarie; Ist de Viene o riche conpeignie. Et Aude sist sor le mul de Surie; Vestue fu d’un paile d’Aumarie; Plus belle dame ne fu onqes en vie. En sa mein destre, qe a amanuïe, Ot un anel o durement se fie, Qe li dona Rollant por druerie: Quant li livra, sa foi li a plevie. Mot i eüst cuens Rollant bone amie, Se il durast o eüst longe vie. Las! Qel amor a duel est departie En Rencevals entre la gent ahie! Mais la belle Aude n’en set encore mie; El li rendra mot loial conpegnie. Li dus Girart, qi söef l’a nosrie, La tint as resnes, et Gerart d’Espanie.

101.b



6625

Gerart chevauche, le hardi conbatant, A Charlemene, l’enperaör vaillant; La trovera le damage pesant:

6604. The word les was inserted superscript. 6611. C: amaneïe, variant graphy of amanevie (past participle, f, amenevir), meaning ‘prepared’, ‘ready’ or ‘dexterous’ – but this gives a less satisfactory reading than V7 6734: amanuïe (amanuïr meaning ‘to make small’ or ‘to polish’). The intention must surely be to describe Aude’s hand as ‘tiny’ or ‘smooth’. 6614. V7 adds line 6738: Qu’il la prendroit se Deus li donoit vie, found also as V4 4788 and T 4269, but not in P.

408

6630

6635

6640

6645

6650 (V7)

La Chanson de Roland (Critical Text and Notes)

La rote est fiere et l’avalee grant Et Aude sist sor un mulet amblant. Girart l’i balle, Berart et Guinemant: Li uns fu dux, li autres cuens puissant; De plusors coses vont tote jor parlant Et la belle Aude, mot sovent sospirant. Gerart apele, si li dist en plorant, «Oncle,» fait ele, «mot ai le cuer dolant Et cest mien cors tressue en tranblant. Maistre Amaugin m’amenez ci devant. Enuit sonjai un songe si pesant, N’oï mais tel nus hon qi soit vivant.» Li clers i vint sor un mulet amblant: «Bau sire clers,» dist Aude la vaillant, «Or escotez un poi de mon sanblant, Qe il m’avint anuit en mon dormant: Primer me vint un fauconet volant, Qi ert plus blanc qi n’est nois qi espant, Et s’ert asis sor moi en mon devant. Les gez des piés furent mot avenant; En poi de terme fu merveillos pesant. Entre ses piez me saissi maintenant, Si m’enporta en son un pui volant; La me guerpi ainc puis n’en vi sanblant. Enprés ice m’avint autres plus grant, Car tote Espegne ert vers moi apendant; De Saragoze venoit le dus Rollant Et Oliver mon frere, lo vaillant;

102.a

6626. C’s reading lo damage grant, immediately below le damage in C 6625, may point to a scribal error and ‘The defeat is terrible and the loss (or grief) great’ fits very clumsily into the context, for C 6627 continues: Et Aude sist sor un mulet amblant. V7 6749 and 6751, V4 4796– 4797, T 4292–4293 and L all have lines corresponding to C 6625 and 6627, but omit C 6626. Only P 5442–5444 is equivalent to C 6625–6627; but here P reads: Fiere est la roche, et la valee grans, describing the landscape through which Aude and Girart journey. This leads much better into C 6627, especially if the irrelevant la valee is read as l’avalee, the ‘slope’ or ‘descent’ from the Massif Central. C’s rote is therefore taken with the alternative meaning ‘road’ or ‘way’ and lo damage emended. (Duggan here inserts C 6626 as V7 6750.) 6651. C: Car tote Espegne en un bois verdoiant, the second hemistich being manifestly a scribal error, for it is found later in its proper place, as C 6654. V7 6775 provides the correct reading, supported by P 5467: ert vers lui apendans, V4 4819: ert a moi declinant, and T 4314: yert vers ly aclinant.

La Chanson de Roland (Critical Text and Notes)

6655

6660

6665

6670

6675

409

Chacer alerent en un bois verdoiant. Murent des porz, paoros et corant; Ils les chacerent contreval un pendant; Lez une roche, joste un pré verdoiant; Lors retornerent tant arere et avant, Plus de .xx. hors lor furent secorant, Qe toz lor chens lor vont si depeçant, N’en eschapa ne mais un seul vivant. Uns lions fiers vint vers Rollant esrant, De lui mengier fist merveillos sanblant; Et Rollant trast Durendal la tranchant, Le destre pié li treinche maintenant; Bien l’eüst mort, quant s’en torna fuiant. Ja me cren je q’il i ait perze grant; Forment me greve, Deu en trai a garant, De Guenellon, le felon sosduiant, Qi le mesage porta per maltalant Al roi Marsile, qi Deu ne croit niant; Venduz les a, per lo mien esciant! Il en a pris roge or et blanc argant, Se vint somiers chargiez a maintenant. Ja jusqe la n’arai mes cuer joiant, Ou qe j’arai d’eus mesages si creant, Qe je sarai ou il sunt sejornant.»

 Gerart chevauche et sa gent onoree, Et Aude sist el mul qe li agree.

102.b

6659. C: plus de de hors, which seems to arise from repetition of de and omission of .xx., rather than from faulty copying of plus de dis… V7 6783 reads: plus de .xx. ors, which is also found in P 5475, T 4322 and V4 4827. 6663. C: vengier, but this has been taken as a scribal error for mengier, the reading in V7 6787, P 5479 and T 4326, since this makes much better sense. 6665. A scribal error: direct object singular, le, for indirect object singular, li. 6667. The C ms. reading perze (or perçe) is a graphy of perte. Cf. C 6711. 6671. C: Al tre Marsile, with tre a scribal error for rei/roi. V7 6795 and P 5487 read: Au roi, V4 4836: al roi and T 4333: au roy. 6675. C: me cuer, where V7 6799 and P 5491 read: mon cuer, but T 4336 has: le cuer, the normal usage. Although men is found as the possessive adjective, masculine, in Northern France, here me reads better as abridgement through scribal error of mes, adverb: n’arai mes: ‘I shall nevermore have’.

410 6680

6685

6690

6695

6700

La Chanson de Roland (Critical Text and Notes)

Sor la sanbue est un poi acoudee, Envers le clerc est un poi anclinee: «Sire,» fait ele, «je sui si adolee Et por cest songe travellee et penee. Ce ne vit mais feme de mere nee. Se je en son le pui en fui portee Et li fauz m’ot et guerpie et lassee, Puis vint une aigle hidose et enpenee; Sor moi s’asist, si m’a acoverclee Cum se je fusse entre le mer entree. Quant m’a guerpie, si m’a si mal menee Qe ma memele senestre en a portee; Pois retorna, s’en a la destre ostee. Je remes lasse, dolante et esgaree, Quant Charllemene o la barbe meslee I vint pongnant, si m’a iloc trovee. Entre ses braz m’en a sus relevee: Il me venjast, se l’aigle fust trovee. Aprés me dist qe ne fuse adolee, Il a sa gent en France retornee. Devers Espeigne revint une nuee, Qi plus ert noire qe n’est une isuee; En Rencevals, une terre esfree, La a sa gent si fort acoverclee Come s’il fust entre le mer entree.

6686. In the C ms., lines 6685 and 6686 both end with portee, which is suspicious – and portee makes no sense when used with guerpie. V7 6811 reads levee, which does not fit the context, nor does P 5503: posee; and T 4347 has: portee as in C. Emendation has therefore been based on V4 4848: lassee (‘left’), which reinforces the sense of guerpie. 6687. C: un angle, here and in C 6697. The word ‘angel’ seems dissonant with the description ‘hideous and feathered’, and un aigle is the reading in V7 6812, with une aigle in V4 4849 and T 4348, une aygle in P 5504. 6697. C: angle; V7 6822 reads argle and P 5514: aygle. V. note to C 6687 above. 6700. C: revint nuee, omission of the article leaving the hemistich hypometric. V7 6825 reads: sailli une nuee; V4, P and T all insert une before nuee. 6701. C: Qi plus ert noire sues qe… The graphy isuee is not found in the dictionaries, although eissue (‘exit’ or ‘outlet’) exists. This could be another case of pseudo-rhyme in C, with the word-ending modified for the rhyme. V7 6826, V4 4861, P 5518 and T 4355 all compare the blackness to that of fumee, perhaps mis-read as issuee by C. Alternatively the redactor could here be thinking specifically of the blackness of a sooty chimney, rather than of any ‘outlet’ in general.

La Chanson de Roland (Critical Text and Notes)

6705

6708a(P) 6710

6715

Desoz lor piez ert la terre crevee, Tant en perist, jamais n’ert recovree. A feu grezois vi la terre alumee, Qi tot ardoit deci en mer salee. A Karlemaine a la barbe meslee L’espaule destre li vi del cors sevree O toz le braz esraigé et sevree. La qit je, lasse, q’il ont perce encontree! J’en ai poor por cele remenbree Qe li fel Guenes ala en la contree Al roi Marsille d’Espegne la desvee; Parole i ot qi mot fu porpensee; Li .xii. per l’ont ja chier conparee; La traïson ne poet estre celee: Je l’ai songié, si est vertez provee.»

411

103.a

 6720

6725

Aude la belle fist forment a priser; Mot fu cortoise, n’i ot qe enseigner; Mais le fier songe la fist fort esmaier. Grant poor ot, ja n’en estuet plaidier, Por son grant frere qe ele avoit si chier E por Rollant, son ami le guesrer, Qi en Espeigne estoient ostoier. Le clerc apele qi ben sot son mester; Aude sospire, qi ne s’en pot tarzer: «Bau sire clers, mot me puis merveller

6709. An extra line is essential, prior to this one, to explain to whom li refers. V7 also lacks this line, but P inserts line 5526, quoted above in the text, and the sense of this is supported by V4 4869: L’espaulle destre de Carlo le vit del cors ostee. 6711. C: perce, a graphy of perte; cf. C 6667. Confusion of t, z and c is occasionally found in C. 6714. The reading Espagne la desvee is supported by V7 6839, P 5531 and T 4366, but desvee, past part.feminine, desver, ‘to go mad’, seems inappropriate here. This could be a variant graphy of the past participle of desveïer, here taken as meaning as ‘strayed’ or ‘led astray’ (from Christianity), thus ‘sinful’, ‘apostate’, ‘wicked’. 6721. C: la fist esmaier, a hypometric second hemistich, but V7 6846 reads: la fist fort esmaier; P 5539 and L 2293 read: moult esmaier. 6723. The word si was added in superscript, with insertion mark at the appropriate place.

412

6730

6735

6740

6745

La Chanson de Roland (Critical Text and Notes)

Si sui pesant, qe ne me puis aidier; A une nuit me crut grant enconbrier: Avis m’estoit qe j’ere en un terrer, En un grant val, desoz un aiglenter; Tote nue ere dedesoz l’aiglenter, Fors ma chemise que ne vel despoillier. Lors vi des hors qi me volsent mangier; Il me prenoient et devant et derier, De trente parz me fasoient segner. Donc oï cor soner et un grasler; Li ors fuirent, ne me vousent laiser; Lors vi venir un veillart pautoner, Qi me levot el col de son destrier, Si m’enportoit en son un grant rochier; La avoit moines en un petit moster; Illoc chantoient por amor Deu proier. Quant o aus fuï, si laisai l’averser; Dales l’autel, dejoste un oliver,

103.b

6729. me was added in superscript, rather cramped and with arrow below, at the appropriate place. 6732, 6733. The almost identical second hemistichs in these two lines must make one reading suspect. However, they make sense, so both readings have been retained. V7 6858 reads: desoz un sentier (presumably intending desor) in the first of these lines, keeping C’s reading in the second; and P 5551–5552 has …delez un sentier, followed by …delez le vergier. On this evidence the correct ending for C 6732 may be either desor un sentier or delez un sentier. 6737. C: .parz q me…, q being partly written, then barred. 6738. There is a scribal error in the text in that the cæsura is marked after soner, giving 6:4 scansion, instead of after cor, which would give a standard decasyllabic line. The word grasler appears to be the diminutive of grasle/graile, ‘bugle’, with the -er ending pseudo-rhyme for -et. 6741. The graphy levot appears to represent imperfect 3, levoit as in V7 6866, rather than the expected past definite 3, levat, since the imperfect, m’enportoit, is used in C 6742 and V7 6867. This is again the typical Northern French and Franco-Italian reduction of oi to o. 6745. C reads: laverser, V7 6870: laversier; it is not found in any other version. In both C and V7 it is preceded by si laisai, the verb being past definite 1. Mortier reads this as la verser: ‘I let (myself) fall down there’ or ‘I let (myself) turn round there’ – which makes sense in that Aude then sees Roland and Oliver; but there is no reflexive pronoun in either C or V. If one reads l’averser, this may mean ‘anything horrifying’, either ‘devil’, ‘adversary/creature attacking me’ or (most relevantly) ‘nightmare’. 6746. An olive-tree seems distinctly out of place inside a monastery, beside the altar – and may result from anticipation of the name Oliver 2 lines below; but P 5562 reads: delez un ayglentier, which is equally strange. Both phrases are standard epic formulæ, liable to be used inappropriately. Moreover, this is a dream, so wild inconsistencies may be acceptable. V7 6871

La Chanson de Roland (Critical Text and Notes)

6750

6755

6760

6765

6770

6775

413

Gisoient mort doi mot bel chevalier; Ce m’ert avis qe ce ert Oliver, Dejoste lui dans Rollant le guesrer. Sire Gerart,» ce dist Aude au vis cler, «Nus hon de char ne se doit merveiller Se je m’esmai, car li songe sunt fier. Uncle Girart, quant me dui esveillier, Le mauvas songe et guerpir et laissier, Lores m’estut de chief recomencier. Nostre enperere Karllemene au vis fier M’avoit menée a mon frere Oliver Et a Rollant le duc, qe j’ai si chier; Enz en un bos erent alez chacier; Un cerf leverent merveillos et pleigner Et il laiserent aler lo lïevrer. Si com il fu entré en un senter, En une spoisse, dedesoz un lorer, La me guerpi l’enperere au vis fier. En halte voiz comence a hucher, ‹Oliver, frere, volez m’i vos laisier? Sire Rollant, car me venez aidier!› Outre passerent enprés li lïevrier, Mais onqes moi ne volsent arasnier. Desoz Rollant trebuche son destrier, Desoz mon frere Rondel q’il ot tant chier; Soz aus vi fondre la terre et rochier, Qe l’uns des contes ne pot al autre aidier. Quant m’esvellai, si fui en un mostier; La vi Rollant soz un paile chocher

has the more suitable delez un chandelier (candles would of course be lit beside the dead bodies). Later, C 7054 has Prent le suaire de desus l’estajier (= trestle for the bier?), where it is a matter of an actual event, not a dream. This suggests an alternative reading in the model, misread by C. 6749. C: dans .R. les guesrer, with the plural definite article through scribal error. 6753. The graphy Uncle provides one of the rare examples of Franco-Italian u replacing closed tonic o before a nasal. 6756. Once again, .K. needs to be transcribed as Karllemene to obtain a metrically correct line. 6763. C: une spoisse, with s impura lacking the prosthetic e, very common in North-Eastern French and Franco-Italian. V7 6888: une espeisse. 6765. The verb comence must here be present 1, i.e. its subject is the speaker, Aude, since the words in the next line are addressed to Oliver as her brother.

414

6780

La Chanson de Roland (Critical Text and Notes)

Et joste lui mon chier frere Oliver; Andeus les contes vi ensanble enbracher, Mais onqes moi n’i volsent arasnier; Quant de la boche m’issi un espirver: Il prist son vol q’il ne s’en pot aidier, (La mort des contes i vit et la lorer.) Desus Rollant et desor Oliver. Lors m’esveillai, si lasai lo songer; Je me cren mot q’il n’i ait enconbrier.»

104.a

 6785

6790

«Deus,» dist Gerart, «ci a fere dotance!» Dien li mes, «N’i aiez mescreance!» Li clers fu saives des q’il issi de France Et fu Normant de la cité de Costance; Astronomie sot bien e nigromance. Il prist un livre, si i list sanz falance: La mort des contes i vit e la pesance, Et come Guenes les vendi en balance

6779. C: in issi, apparently a scribal error for m’issi, as in P 5586. V7 6904 reads: me issi, V4 4924: m’ensi and T 4421: m’iessit. 6781. This line is not found in this position in any of the other versions and it is clearly out of place in C. Without it, C 6780 annd 6782 together constitute a perfectly good sentence, the sparrowhawk taking wing perforce over the bodies of Roland and Oliver. C 6781 is found in its correct position in the next laisse, as C 6791, where it has the correct ending, pesance, to match the rhyme there. In C 6781, la lorer may, at best, be a misreading of l’aloier (infinitive used as substantive), ‘alliance’ or ‘close friendship’, which supplies an appropriate rhyme in -er. It relates to the clerk who looks in his book for the interpretation of Aude’s dream. Both lines are found on folio 104 recto and in each case the preceding line (C 6780 and 6790) begins with the same words: Il prist. Having copied most of C 6791, the scribe realised his error and tried to correct it as simply as possible. 6787. The C ms. reading appears to be savies (also found in C 6801 and 6827), but the metre calls for a single syllable. For confusion of -ui- and -iu- graphies, Cf. C 4018 and 7083, uit = jut, and C 6826, 6835, etc. in respect of Blaive. Cf. Introduction, 71–72. 6788. V7 6912 and T 4431 also read: de la cité de Costance, a 7-syllable second hemistich. This is comparable to a la citez de Mascon, C 6208. Cf. note to C 6208 regarding emendation of cité to cit. 6791. C: L’amor, ‘the love’, but the other rhymed versions agree in reading La mort, which makes much better sense (cf. V7 6915, V4 4933, P 5596, T 4433, L 2326). L’amor is also found in C 6781 and it occurs as a scribal error for la mort in C 2445 and 2456 (cf. notes to these lines).

La Chanson de Roland (Critical Text and Notes)

6795

6800

415

Au roi Marsire, qi en Deu n’ot creance. De vint mellers, n’en ala pié en France. Li clers ot duel, nel tint pas a enfance, Mais por belle Aude fist gente contenance. Dist a Gerart, «Chevauchez a fiance! De vos amis ne soiez en eranche: Ainz demein none, en vesrez tel senblance, Don il sera dolzor et joie en France.»



6805

6810

6815

Li clers fu saives, qi la dolor cela; Cum ainz il pot, son livre en esconsa. Forment sospire et au cor angossa. Mais por belle Aude gentement se cela; En altre sen le songe trestorna. «Dame,» fait il, «ne vos esmaiez ja, Car por le songe neguns mals ne venra. Öez por voir et traiez vos en ça: Le fauconet qi el poing vos porta, C’est Charllemene qi ersor vos manda; Et l’aigle fier qi illoc vos trova, Qi les mamelles del cors vos desevra, C’est une dame qe Rollant conqerra; Fieres batalles li dus endurera; Icele dame Rollant iloc prendra,

104.b

6796. C: en fist gente contenance, with cæsura marked after Aude. The hypermetric hemistich is readily corrected by omitting en. Strictly, Auden, oblique form, is required after the preposition por, so en could have strayed into the wrong hemistich. But the graphy Auden is found only in C 7298; and the 4-syllable formula, Mais por belle Aude, is repeated in C 6804 and also found in P 5601, L 2328 and T 4438. P continues with fist bele contenance, although L and T modify the second hemistich to accommodate en. This line is not found in V7. 6799. C: ne vesrez…, but the negative ruins the sense. V7 6923 has: venrez vos…, presumably in error for vesrez vos… (and emended by Duggan); but V4 4940 has: en vereç tel s.. In C, the ne probably results simply from repetition, none en becoming none ne through scribal error, although it may be an example of Franco-Italian, i.e. the Italian ne equivalent to French en. P 5604 reads: en orrez la s…, confirming en as the correct reading. 6800. C: dolzor, with z inserted superscript, apparently in the same ink and script. 6801. As in C 6787, the ms. graphy is sauies, making the hemistich hypermetric. Cf. Introduction, 71–72 regarding scribal carelessness in writing iu. 6802. The word il was added superscript at the appropriate place.

416

6820

6825

La Chanson de Roland (Critical Text and Notes)

Por sa bauté laisser vus en voldra. Et l’esprever qi del cors vos vola, Ce est uns enfans, se Deu plas, qi’n istra; Girart vostre oncle mot ben le norira, E Karllemene grant honor li fera.» Respont belle Aude, «Si ert cum Deu plaira.» Girart escote, ferement chevaucha; De ce q’il ot, li cuers li engroissa; Pois icele eure noient ne se tarza – E la bele Aude forment esperona Deci c’a Blaive, o sa dolor trova.



6830

6835

6840

Saive est li clers qi le duel ot celé; En altre sen a le songe atorné. Gerart chevauche, et o lui si privé; Elas, qel duel quant seront asanblé! Si com il sunt permi Sene passé, Jofroiz d’Anjou ne s’est pas oblié: Deus barons a de la rote sevré; Sor les mellors chevaus q’il ont trové, Tot droit a Blaive en sunt avant alé. A pié desendent, el palais sunt monté, Le rois saluent, aprés li ont conté: «Ci vient Girart et o lui si privé, Et la belle Aude o le cors honoré. Mais lo fer duel lor avon si celé

6818. C: qin istra, i.e. qi’n istra for qi en istra, meaning ‘which will come out of this’; or possibly qi nistra for qi naistra, ‘which will be born’. The underlying sense is the same, so either reading is possible. 6826. C: Blavie, but a monosyllable is needed. In V7 6950 Duggan reads Blaive (though Foerster reads Blavie); in Mortier’s editions, V4 4966 reads: Blavia, P 5632: Blaivies, T 4469: Blevez, L 2349: Blaives. Cf. note to C 6199 and 6787 above and Introduction, 71–72 re faulty marking of -iu- as -ui-. 6827. C: Savie (Sauje), exactly analogous with Blavie above. Cf. Introduction, 71–72. 6831. V7 6956 reads: Saintes, making the identification of Sene easier – and this town in Charente-Maritime would be on the route from Vienne, just south of Lyon, to Blaye on the Gironde. 6835. C: Blavie, although a monosyllable is needed. Cf. C 6787 and 6826 and Introduction, 71–72.

La Chanson de Roland (Critical Text and Notes)

6845

6850

Qe il ne:l sevent per home qi soit né.» «Baron,» dist Karles, «mot avez ben esré. De si grant duel somes asez usé, Dont nos avons tantes larmes ploré. Bien le saront ainz qe soit avespré. Mot me merveil com je l’ai enduré, Qe mon dur cuer n’ai el ventre crevé!» A cez paroles chaï entr’als pasmé. D’une grant piëce n’en a il mot soné; Quant il revint, si a un poi parlé. Naymon apele, q’il tint por son privé: «Hé, Nayme, sire, de bon consel menbré, Per quel engin serons nos porpensé, Coment il soient de cest duel conforté?»

417

105.a

 6855

6860

«Bau sire Neme,» ce dist Karles li ber, «Por amor Deu, alez mot tost monter: Permi cel ost faites un ban crier; Les dames faites dancer et caroler, Et les enfanz per cez rues joër. Et je irai al duc Gerart parler, Por la bele Aude qe voldrai conforter; Mot li ert fort cest deus a trepaser. Deus! Si grant duel qi porra trepaser?

6845. C: Bien le sarons, but it is not clear what ‘we shall know’. V7 6970 reads: Bien le sauront, indicating that Girart and Aude will know of the grievous death of Roland and Oliver, and this is supported by P 5648: Que il nel saichent and T 4498: Ilz le saron. The copyist has been led astray by somes, present 4, C 6843, and avons, present 4, C 6844, to write the future 4 here. 6856. C: moter, written mot7 with ~ omitted; V7 6981 reads: monter. 6857. After this line, V7 adds line 6983: Facent tuit joie, laissent le duel ester, and the same line is found in V4 (4971) and P (5656). When, from C 6979 on, Girart arrives at Blaye, C does include a line (C 6983 = V7 7118): Grant joie i trovent, n’ont pas duel demené, recalling V7’s earlier line, so the omission after C 6857 may be a scribal error. The line is not essential to the text. 6859. Once again, V7 inserts an extra line (6986): Les chevaliers por les champ baorder, found in V4 as 4974 and in P as 5659. The fact that, in laisse 378, C lists les dames, li enfant and then, in C 6986, E li baron par les chans bohordé gives credibility to the line’s having been omitted through scribal error.

418

6865

La Chanson de Roland (Critical Text and Notes)

Je ai perdu la flor de mon barner Et g’en serai legiers a afoler. France en ert povre, ben le pot hon jurer.»



6870

6875

6880

6883a(V7) 6885

Naymes a fait per l’ost le duel laissier: Les dames font charoler et treschier, Et les enfanz pertot esbanoier. Et l’enperere se vait apareller; Si est montez sor un corant destrier Et o soi mene le bon Denois Ogier Et d’Aspremont le bon conte Richer, Et tant des altres q’i sunt .c. chevalier. De l’ost issirent, n’i volsent plus targier; Pluisor i poinstrent sor lor corant destrier, Por ce q’il volsent Girart eslecéer. L’emperere chevauche tot le chief primier; Granz noef lieues pense del chivauchier Ainz q’il veïst dan Gerart lo guesrer. Lés un boschet en un valet pleigner, La desendirent et laissent refroider; Gerart et Aude venoient tot primer. Girart descent, si vait le roi baisier; Aprés lui erent et Ogier et Reger. Charlle Mene vit Aude, si la va enbracer, Et ele lui, sel comence a baisier. «Sire,» fait ele, «ou est mes frere Oliver, Li dus Rollant qi me doit noçoier? Mot m’avra fait en Viene lassier.

105.b

6878. C’s alexandrine becomes regular decasyllabic metre in V7 7005 (Li emperere chevauce tot primiers), P 5677 (Nostre enpereres chevauche touz premiers) and V4 4991. 6881. C: et un valet, using the normal symbol for et, so that the preposition lés is applicable to both substantives. V7 7008 and T 5680 have: en, which gives a much better reading: ‘alongside a grove in a valley’; but est would also fit well here. This is simply a scribal error in C. 6884. Since the preceding lines indicate that Charlemagne is leading the way, with Ogier and other barons following, the masculine singular lui in this line must refer to Charlemagne (le roi in V7’s line), not Girart. The V7 7011 reading is confirmed by V4 4997 and by P, where it is the second of 2 additional lines, 5683 and 5684. C’s omission of this line may arise from the fact that, like C 6883, it begins with the name Gerart/Girart.

La Chanson de Roland (Critical Text and Notes)

6890

419

Quant je nel voi, mot me pois merveller; Ja croi je mot q’il n’i ait enconbrier.» Karles s’estut, si esgarda Ogier; Un poi plora, q’il ne se pot celer.

 6895

6900

6905

6910

6915

Li rois ot duel, je ne me mervel mie: «Per foi, belle Aude, bien est qe je:l vos die: Il sunt de moi parti per felonie; Les ai laissiez el reigne d’Aumarie O il estoient contre la gent haïe; Li dux Rollant a feme recollie, Fille Florent, le roi de Val Serie; Por sa bauté a la vostre guerpie. Et Olivers li est en bone haïe; Il l’eme tant, sa foi li a plevie, Ne li faudra en trestote sa vie; S’ara son reigne et s’onor recollie. Por Dé, belle Aude, ne vos en post il mie; D’andeus les contes laissez la druerie. Je vos donrai le duc de Normandie: Riches hon est, de grant chevalarie; Demie France vos lairai en bailie.» «Sire,» dist Aude, «ice ne croi je mie. S’uns autres hon deïst tel legerie, Je le tenise a mot grant estoutie. Sire enperere, nel me celez vos mie: O est Oliver o la chiere hardie, Li dus Rollant qi de moi fist s’amie? Se je nes voi, n’en porterai la vie. Il sunt alé en la grant ost banie;

106.a

6891. C: Ja croie mot, but the present 1 is simply croi. However, haplography is an occasional feature of C, particularly where there should be -ii- for -ij- or -ji-; and croi je makes very good sense. The subject pronoun is found in V7 7020 (Je criem), P 5693 (Je mescroi) and T 4560 (Je me craing). The same graphy croie occurs again in C 7156 and 7165. 6915. C: O est Oliver (also in V7 7044 and V4) gives a hypermetric first hemistich, unless the O est is elided into a single syllable (or Oliver > Olivier, giving a 6-syllable hemistich). P omits this line, but T 4585 has Ou est Olivier; L 2390–2391 modifies the text to read: Ou est Rollant…/ E Oliviers a la…?

420

6920

La Chanson de Roland (Critical Text and Notes)

Li pesanz songes qi m’a espoërie Me dira voir, ainz l’ore de conplie.» Donc plore Karles o la barbe florie Et .c. des autres; mais ele nel vet mie.



6925

6930

6935

6940

«Damesele Aude,» dist Karles le vaillant, «Laisez ester les amor de Rollant Et d’Oliver, le hardi conbatant. Il sunt de moi parti par maltalant; Je:s ai laissez en Espeigne la grant O il ostoient contre gent mescreant En Aumarie, une cité vaillant. Prim jor de mai, par som avril issant, En Babiloine en iront ostoiant. Cuens Olivers en sera amirant Et prendra feme la seror Baligant; Une pucele a pris li dus Rollant, Fille Florent, un roi de Val Dormant; Por sa bauté vait la vostre laissant.» «Merci, beau sire,» dist Aude la vaillant. «Por amor Deu, n’en alez delaiant. N’a feme en terre, n’en cest siecle vivant, Qe puïsse mais partir mes amor de Rollant. Je l’ai perdu, je:l sai a esciant; Qi q’en ait joie, g’en ai le cuer dolant.»

106.b

 Quant Karles vit qe ja ne:l celera, Ne la bele Aude conforter ne porra,

6922. C: mais ele ne sevet mie, the second hemistich being hypermetric unless one reads sevet (imperfect 3, saveir) as a scribal error for seut, the Northern French form of the past definite 3 (cf. Pope 1934, 488, § 1320, viii): ‘but she was not aware of it’. V7 7053 reads: ele nel voit mie and T 4592: mes elle ne les vit mie; but P 5712 reads: mais Aude nel seit mie, reducing the verb to the monosyllabic present 3, saveïr. Following V7 and T and taking the s of sevet as a scribal mis-reading of l, ne has been emended to nel and vet read as present 3, veïr (cf. veit, C 1393, 1961).

La Chanson de Roland (Critical Text and Notes)

6945

6950

6955

421

Del cor sospire e un petit plora; Al bon Denois la pucele livra. Vint a Gerart, envers lui s’aclina; Un seul petit des autres s’eslongna; Ne pot parler li rois, ainz se pasma; Lors sot Gerart qe grant daumage i a. Li rois revint, al duc merci cria; Tot le damage et le duel li conta: La traïson qe Guenes fait li a. Girart l’entent, a poi ne forsena; Per vaselage son corage cela Et son seignor mot bien reconforta.



6960

6965

Granz fu le deus en aprés l’asanblee; Or voit li rois n’i a mester celee. Gui de Nevers a sa raison contee: «Ici vient Berte, vostre suer, l’anoree.» Li rois l’entent, sa color a muee; Aude la bele a au Denois livree; Vait li encontre aval permi la pree, Sor un mulet qi mot bien li agree. Baiser la velt, quant il l’ot saluee: «Bele suer Berte, savez vos la dolee? Morz est Rollant, n’i a mester celee; Traï l’a Guenes, cui fustes esposee. Si m’a tolu lo meus de ma contree.»

6946. C: la pucele plora livra, plora being first copied from the previous line-ending in error. 6959. V7 here has 2 lines, 7090–7091: Gui de Nevers vint poignant la feree,/A K’meine a la raison contee, and so does P (lines 5748–5749), the second being identical to V7 7091. Howsoever, V4 5071, T 4628 and L 2402 have only one line here, like C. Thus the abrupt transition to Berte may occur in an early source, of which the V7-P reading would then be a copyist’s emendation. 6966–6969. Charlemagne’s abrupt announcement to Berte of her son’s death and her husband’s treachery contrasts strongly with his procrastination and considerate attitude towards Aude. This is partly explained by C 6958: ‘Now the king sees that it is of no use to conceal anything’. Probably also the writer did not wish to spoil the effect he was building up in the story of Aude’s mourning and its outcome by giving any prominence to Berte and her sorrow. Today, the tragic situation of the latter makes her the more interesting figure, but one is only told briefly of her collapse and grief.

422 6970

La Chanson de Roland (Critical Text and Notes)

Berte l’entent, de dolor chiet pasmee; Mais Karllemene l’en a sus relevee, Mot doucement l’en a reconfortee.



6975

Berte se pasme, sa dolor demena; Li enperere ben la reconforta. Torna arere o belle Aude laissa; Aude vit Berte, encontre li ala; Puis s’entrebaissent, mais chascune plora; Jamais nus hon plus grant duel ne vesra.

107.a

 6980

6985

6990

Li dus Gerart fu forment abosmé; Par vaselage a son cor recovré Et Charllomene l’a mot reconforté. Venent a Blaive, laienz en sunt entré; Grant joie i trovent, n’ont pas duel demené; Les dames ont tresché et carolé, Et li enfant par les rues joé, E li baron par les chans bohordé; Ne as mostriers n’ot onqes sainz soné, Si come Karlle l’ot Naymon comandé. Li rois descent del mulet afoltré; Lui et Gerart sunt el palais monté; Et conte et duc et li prince chasé Vindrent encontre, Gerart ont salué, Et la bele Aude o le cors acesmé.

6982. C: Venent a Blivie, which is hypometric, but the graphy Blaive is found in C 6522 and Blaives in C 6199, 6482 and 6483. V7 7117 reads: Vienent a Blaives and L 2423 also has Blaives, but P 5786 has Blaivies and T 4650 Blevez (with the verb in these cases monosyllabic). 6988. C: Si come .Klle’. l’ot Naymon a commandé , an alexandrine with the cæsura marked after l’ot. To make sense, comandé must be understood with both verbs: ‘Just as Charles had (commanded) it, Naymes has commanded’, but, given the oblique form Naymon, it could equally well be read as: ‘Just as Charles had commanded Naymes’, dropping the a and assuming a mis-placed cæsura. Indeed, since Naymon is oblique form only, Si come Charlles/l’ot Naymon commandé is the better reading, coinciding with that in V7 7123 and P 5792; V4 5105 and T 4656 omit l’ot and read a Naymon comandé.

423

La Chanson de Roland (Critical Text and Notes)

6995

7000

7005

Mais d’Oliver n’i ont mië trové, Ne de Rollant, lo vasal aduré. Plore belle Aude, li roi a regardé: «Sire enperere, merci, por amor Dé, Car plus vos aim qe home qi soit né! De ceste lasse vos preeigne pieté! Del duc Rollant me dites verité Et d’Oliver, mon frere, l’alosé.» «Bele,» fait Karles, «ne puet estre celé: Andoi sunt mort li vasal aduré; Il m’ont guerpi en mot grant orfenté, Et moi et vos ont del tot oblié.»



7010

7015

«Damisele Aude, se:l vos osasse dire, Mort est Rollant, mes niés et vostre sire, Et Olivers li cons de franche orine, Li doce per a duel et a martire; Soz ciel n’a home qi ait talant de rire.» Aude se pasme, mot durement sospire; Quant ele revint, plus est jane qe cire; Deu reclama, qi de tot ben est sire, Qe mort li dogne, car li cors li enpire. «Oncle Girart, ci a mot fort remire: Ce sunt mes noces, vos en arez grant ire.»

107.b

 «Damisele Aude, mas celer ne:l pöon: Trestoz sunt morz li doze conpeignon; 7004. C: orfeté, with symbol ~ omitted; V7 7139: orfenté. 7008. Most unusually for Section C, this line ends in assonance (orine); V7 7143 changes this hemistich completely: dont avrai grant enpire, maintaining the -ire rhyme throughout the laisse, and V4 5125 reads: dont dolce France enpire. However, in the same laisse (348) V4 later has -i- assonance in 5131–5132: Deo reclame, li filz saincte Marie,/Che li duni mort ançi che la vive, equivalent to C 7013–7014. This confirms the Palumbo argument (cf. Introduction, 18 and 58) that the Section C remaniement was written between 1180 and 1195, when assonance was still widely used. 7012. C and V7 7150 both read the grammatically correct ele which makes the hemistich hypermetric – but Duggan emends this to el.

424

7020

7025

7030

La Chanson de Roland (Critical Text and Notes)

En Rencevals les traï Guenelon, Quant les vendi al roi Marsilion Mot grant avoir, cum encrisme felon. En orfenté m’ont guerpi mi baron, Si cum la beste fait el bos le foön, Quant l’a ocise o liupart o lion.» Aude l’entent, ne dist ne o ne non; Sovent se pasme entre les braz Charllon. Li rois la tint, si fu agenuillon; Li cors li tranble, norci li la fazon; Mot estut Aude en longe pasmison – Nus hon de char ne puet avoir raison, Ne clerc ne prestre doner confession.



7035

7040

Charllemene tint Aude entre ses braz issi, Q’el ne parole ne les els n’en ovri; Li cuers li tranble, la color li norci, Les els ennublés, le front enpaloï. Quant ele revint, si a fait un grant cri: «Sire enperere, por amor Dé, merci! Car me mostrez le cors de mon ami Et d’Oliver, mon frere lo hardi. Li cons Rollant por moi sa foi plevi, Q’il me prendroit, et je lui a mari; Ainz m’ert ancui le cuers el cors parti, Si m’en irai ensanble o mon ami Et o mon frere, qi la dolor sofri.»

108.a

7021. C 7021 and V4 5138 are in accord in reading mot grant avoir as dependent on vendi (‘sold them for very large sums’), the same structure as that later found in C 7072. V7 7159 reads: Mout grant avoir en prist come fellon and the same verb, prendre, is found in P 5826 and T 4693. 7031. C: preste, for prestre, ‘priest’. The symbol 7, for r, has probably been omitted in C. V7 7169 and T 4703 read: prestre, P 5834: prestres, but V4 5145 reads: preste, like C. 7035. Godefroy quotes a verb, paloïr, ‘to grow pale’, as well as the form paleïr, although enpalir is more common. Thus enpaloï (with 4 syllables) is a graphy of the past participle, rather than a graphy of the adjective empaleis (3 syllables). 7041. V4 5152, an alexandrine, makes the sense more explicit: Ch’el me prendoit a feme. V7 matches C.

La Chanson de Roland (Critical Text and Notes)

7045

425

«Bele,» dist Karles, «tot i avez failli; Et moi et vos ont il mis en obli.»



7050

7055

7060

7065

Charlle prent Aude, qi tant a lo vis cler, Li et Girart fait al mostier aler. Aude regarde per dales un piler, Mot vit chandeles et cerges alumer Et vit les bieres, si comence a plorer: Ce fu Rollant et Oliver li ber. Aude se pasme, qi ne se pot celer; Prent le suaire de desus l’estajier, A soi le trast, si comence a plorer. Et vit le sanc e la dolor morter Qe il sofrirent por Sarazin tuer. Ele comence la crine a detirer, Sa tendre face o ses doiz esgrater; Per trente lius en fist lo sanc voler. Vers Oliver prist Aude a retorner; Trestoz les dras fist de sor als oster, Et les suaire, por lor cors esgarder, Qi sunt blesmi por les cous endurer Et camoissié por lor armes porter. Desor la boche se prist a acliner; Aude le baisse, si comence a crier: «Oliver, frere, volez moi oblier?

7047. C: pret, but prist in V7 7185. As all the verbs in this passage in C are in the present tense, it seems better to take this as present 3 with ~ omitted through scribal error, rather than a graphy of the past definite 3. 7054. C: le stau7: l’estaver or l’estauer, and V7 7192 has l’estaver (‘candle’), but l’estajier appears to be the best reading here, the u being a scribal mis-reading of ii. This may be a graphy of estagiere, meaning ‘scaffolding’ or ‘framework’, since a temporary structure or trestle had presumably been erected to support the bodies, with the shroud draped over them. However, the shroud is the individual winding-sheet and the word estagier exists in its own right, meaning ‘a vassal required to remain in the king’s house in time of war’ and thus, by extension, ‘an inhabitant’, ‘a person residing therein’. In the latter sense, Aude would pluck the shroud from ‘the person residing therein’ or ‘… enclosed therein’, ‘the person occupying it’ – a much more striking (and indeed disturbing) picture. 7060. C: luis, with the i clearly marked, but lius is needed. Cf. earlier examples of transposition of i and u in C and V7, plus Introduction, 71–72.

426

7070

7075

La Chanson de Roland (Critical Text and Notes)

Ja sui je Aude, qi soliez amer, Sor tote ren senblant d’amor mostrer. Ne pois vos els veïr ne esgarder, Ne vostre boche a la moie parler?» Lores se pasme, ne puet en piez ester; Li rois la va entre ses braz lever Et fait les cors covrir et conréer.

108.b



7080

Aude revint et mot fierement crie; Desus Rollant a sa chiere guenchie; Leva le paile de soie d’Aumarie Et le suaire de cendal d’Aumarie. Vit la char noire et la color blesmie, La bele boche dont la levre ert partie; N’est pas mervelle s’ele est descolorie.

7071. The verb pois is present 2, pooir, because it is addressed to Oliver: ‘are you unable (with) the eyes to see … and (with) the mouth to speak…?’; but one would really expect the present 6, poent or puent. On the other hand, it could be a rather strange present 3 with the final t omitted; ‘eyes’ is treated as a collective noun and followed by the present 3 in C 1336. Exactly the same structure is found here in V7 7209 (Ne puis vos euz…), V4 5177 (pois) and P 5879 (puis). 7075. The copyist wrote courer (= covrer), then changed the e to an i with the usual mark above. 7079. The repetition of d’Aumarie (already found in the previous line) is not found in the other versions. V7 7217 reads: cendal detosie (de Cosie according to Foerster, but this is one word in the ms., with the definitely lower case letter resembling t, and Duggan reads de Tosie – a place not identified); V4 5186 reads: telle de Rusie; P 5887 reads: qui fu fais an Nubie (Nubia being noted for its fabrics). The variety of alternatives points to a difficult source reading. If this is a Proper Name, V7’s detosie may be a mis-reading of d’Etopie (poetic licence for Etiope), ‘Ethiopia’. No verb is recognizable in detosie and there is no reason for a feminine agreement. Given the writer’s occasional play on words – and allowing for poetic licence (daumatie) – C’s d’Aumarie might result from the mis-reading of an adjective, daumatique, relating to the ‘dalmatic’, a tunic worn by deacons under the chasuable during the Mass. As there is no obvious alternative available, C’s reading has been retained. 7071. Between this line and C 7082, V7 adds 8 extra lines and 2 additional laisses (V7 7220 – 7261). The 8 lines read: N’est pas merveille, car grant soi ot soffrie/En Rencesvals entre la gent haïe./Aude la belle mult fierement s’escrie:/«Sire Rollant, je sui vostre amie!/Franc chevalier, avez moi vos guerpie?/Je me fi bien el fiz Saint Marie,/Que vos tendrai mult leial compaignie.»/ Lores se pasme, si est esvanoïe;/N’est pas merveille s’ele est descolorie. V4 5189–5197 is in complete accord with this and so is P 5890–5897; L 2470–2474 has a shorter version of the same lines. The repetition of N’est pas merveille (V7 7220 and 7228) makes it all the more

La Chanson de Roland (Critical Text and Notes)

427



7085

7090

Desor Rollant se jut bele Aude encline, Plore des els e sa face esgratine: Li sans li chiet sor la tendre poitrine, Qele est plus blanche qe n’est flors d’aubepine. «Sire Rollant,» dist Aude la meschine, «Parlez o moi, frans quns de bone orine! Ne devez pas avoir ver moi nulle haïne! Oliver frere, com ore sui frarine! Com mar vi ore ceste gent Sarazine!» A icest mot, bele Aude s’acline; Les duel des omes ot cil de la meschine Fist lor plorer meint fil de Palatine.

 7095

7100

Grant duel ot Aude, le cuer prist a changer; Entre ses braz la tint li dus Ogier. Envers le roi se prist a abaisser; Au roi parole, mot per ot le cuer fier; Öez qe dist la fille au duc Reiner: «Sire emperere, por Deu vos vel prier Qe me facez delivrer cest moster, Qe n’i remegne ne clerc ne chevalier: Sole i vel estre, por amor de prier.

certain that C has omitted these lines through a saut du même au même. Nevertheless, they have not been inserted in the text because they are not strictly essential to it and one cannot prove that they were not omitted deliberately, however unlikely that may be. V7 7220 is the first line of folio 119 verso and V7’s passage fills one page and 12 lines. The 2 extra laisses, 377 and 378 in Foerster’s edition, 374 and 375 in Duggan’s, are not found in any other version. (Although T has one page missing at this point, the evidence points to its having a close match with C.) These laisses, totalling 33 lines, are discussed in the Introduction, 58. They may have been written as a stylised ‘triplication’ of Aude’s laments to Roland and Oliver. 7083. In both C and V7 7262, se uit is written with the i clearly marked, but the sense requires se iut = se jut, as in P 5898. L 2475 reads: se gist. Cf. note to C 4018 and Introduction, 71–72. 7085. C: La sans, but this is the wrong gender; V7 7264 and P 5913 both read: Li… 7086. Qele, interrogative pronoun, feminine, is here used as the relative pronoun, ‘which’. V7 7265 and P 5901 read: Qui. 7093. V7 7274, P 5909 and T 4715 read: et, but ot in C is a valid graphy of od, ‘together with’.

428

7105 7105a(V7)

7110

La Chanson de Roland (Critical Text and Notes)

Si parlerai mon frere Oliver Et a Rollant, le duc qe j’ai si chier; James dolor ne me pora tocher; Ainz m’ara fait cest mien cuer esclarier.» Et respont Karles, «Tal m’estuet otrier.» Il en apele et Naymon et Ogier: «Faites moi tost delivrer cest moster.» Et il si firent, ne se volsent tarzer.

109.a



7115

7120

7125

Grant joie ot Aude quant ot l’otroiement; Envers le roi s’acline belement. Li enperere entre ses braz la prent: «Or faites, bele, vostre comandement.» «Sire,» fait ele, «grant merci vos en rent.» Li rois de France s’en ist primerement, Et tuit li autre trestot comunalment; Aude remest el moster seulement. Ferma les huis et sesra durement; Unqes mais feme ne fist tel hardiment: El vint as bieres, entre ses braz les prent, En lor séant les dreza belement Et afaita per tel devisament Qe nus n’encline, ne tant ne quant ne pent. A orisons belle Aude s’estent; Puis bat sa cope et a Yesu se rent; Sovent recleme le voir omnipotent: «Glorious Sire, qe formas tota gent

7105a. This line, V7 7287, exactly matched by V4 5223 and T 4728, approximating to P 5856, makes the value of Ainz, C 7106, clearer. 7121. C: Il vint…, but this is Aude. V7 7303 reads: Puis, as do T and L; V4 5237 has simply: E; and P 5922: Quant. The same use of il to designate a feminine subject was found in C 4498, 4499, 4838 and 4839 in relation to Bramimonde. It seems to show the derivation from Latin ille, illa: cf. Pope (1934, 320, § 826, and 322, § 836(2)), but it does not appear to be a feature of Franco-Italian (no mention in Bogdanow’s comprehensive list). As in the lines cited above, il has been emended to el. 7125. C: Aude sestent, with the se inserted in the same ink, with a finer pen. 7128. The C ms. reading is either fomas or fornas, but the context requires formas, past definite 2, former, as in V7 7310.

La Chanson de Roland (Critical Text and Notes)

7130

7135

7140

Et mer et terre, la sus el firmament; De qatre abismes fesis issir le vent, Qi par le mont cort par divisement; Tu tiens lo ciel per ton engegnement; Tote bauté la desus i resplent. Ni ja traïtor n’i prendront chasement, Ne li felon n’avront herbergement Qi de vos, sire, firent tel traïment Q’il vos pendirent per lor enchantement Enz en la croiz sanz point de vestiment. Longis i fist mot fer enforcement, Qi de la lance vos feri durement; Il ert avogles, sanz point de veïment, Quant senti l’eve et le sanc qi descent:

429

109.b

7130. C: fessis e issir…, with an extra (and superfluous) s apparently inserted, perhaps intended for an i to give the alternative graphy feisis. It may simply be a revisor’s error. The subsequent e is barred, so issir, not eissir. The allusion is obscure; it is not biblical, but may originate in Greek mythology: the story of Zeus casting the monster Typhœus down into Tartarus, where he was a source of ‘fierce, rain-blowing winds’ bringing calamity on men (cf. Hesiod, Theogony, ll.822–880) or the fact that Boreas, the North wind, was reputed to dwell in a cave on Mt.Haemus in Thrace. 7133. C: la de desus t’atent, ‘all beauty awaits you there from above’, but this is addressed to God, not to Aude. There is general agreement on making the final word resplent: V7 7315 has: la desus i resplent, V4 5248: de lasus i resplent and T 4756: la dessus moult resplent – meaning either ‘all beauty shines from above’ or ‘all beauty shines there above’ (i.e. ‘in Heaven’). Although dedesus might possibly be read as a single word, ‘above’, the repeated de looks like a scribal error, arising from the mis-reading t’atent; the V7-V4 reading, i resplent, has therefore been adopted. 7134. The first hemistich is hypermetric, unless one reads traïtor as dissyllabic (cf. Introduction, 92–93, regarding the problems of metre involving the -aï- diphthong) – or unless the initial Ni has been inserted through scribal error in anticipation of Ne in the next line; there is certainly a plethora of negatives. V7 7316 reads: Ja trahitor and T 4757: Ja traïtres (reading traïtres as tri-syllabic), but P 5932 has: Ne ja traïtres. The problem is complicated by the fact that, strictly speaking, the subject form plural, traïtor (cf. Pope 1934, 312, § 800), is required here, but (cf. Pope 1934, 110, § 242, and 313, § 806) the declension system disintegrated in the thirteenth century and, unusually, the subject form singular, traïtre, became generalised. At the same time, the word lost a syllable as the -aï- diphthong reduced to -ai-. 7139. The reference is to the legendary Roman soldier, Longis, who pierced Christ’s side with his lance and whose blindness was then cured by the water and blood which streamed down. This was clearly originally allegorical, with the ‘blindness’ not a physical defect but a lack of recognition of Christ. The Gospels attest to the presence of a Roman centurion at the Crucifixionm (cf. Matthew xxvii.54 and Mark xv.39) and to the reaction: «Truly this was the Son of God» (cf. Luke xxiii.47). John xix.34 recounts that blood and water flowed from Christ’s side.

430

7145

7150

7155

7160

La Chanson de Roland (Critical Text and Notes)

Il terst ses els, si ot alumement, Merci cria per voir entendiment. Sainz Josep fist reche demandament, Por ses soudées ne volst autre present For soul ton cors, ce:l reçut dignement, Mist le el sepulcre, icel reposement; Les trois Maries te qisent loignement, Qi t’aporterent lo seint cher unguement; Enprés tenis le voir aparlement, Quant as apostles donas confortement: Chascuns avoit por toi le cuer dolent. Deus, qi te sert, fer guesredon atent En paradis o sunt li inocent: Ce croi je, lasse, sanz nul repentiment. Fai venir, Deus, alcun demostrement A la chaitive qi el moster t’atent, Qe Oliver me die son talent!» A cez paroles li sainz angles descent, Qi nostre sire i tramist doucement; De la clarté toz li mostiers resplent.

 Grant joie ot Aude, la clarté ot coisie. Encor n’ot pas sa raison afinie:

7145. C’s reading reche is an unusual graphy (scribal error?) of riche, the normal graphy found in V7 7327, P 5947 and T 4769. This is a reference to Joseph of Arimathea (Matthew xxvii.57– 61, Mark xv.43–46, Luke xxiii.50–53, John xix.38–42) a wealthy disciple who begged Pilate for Christ’s body and put it into the tomb he had bought for himself (which would be a considerable expense). 7149. C: reqisent loigement; but re- is a scribal error for te, found in V7 7331 (V4 has: vos). This leaves qisent as a Northern regional variant of quistrent, past definite 6, qerre – cf. misent (C 1510, 1990, 4936, 7736), prisent (C 1943, 3735, 3746) and fisent (C 1533, 2865, 7289, 7358, 7823), also Introduction, 75 (viii), and Pope (1934, 490, § 1320, xxx). The graphy loigement has been emended to loignement as a scribal error (omission of symbol ~), but the letter g may well represent gn. V7 7331 reads: te quistrent longement and V4 5262: vos querent longement. Cf. note to C 3883 re les trois Maries. 7152. C: apostoles, which makes the first hemistich hypermetric. The use of this graphy, rather than the equally valid variant apostle, may show Franco-Italian influence (Latin: apostolum). 7156. C: croie, as in C 6891 (cf. note). Once again, this is a graphy of croi je (‘I believe’), the -ii- being reduced to i through haplography. Exactly the same graphy is found in C 7165, but

La Chanson de Roland (Critical Text and Notes)

7165

7170

7175

7180

«Deus, ja croi je qe fustes fius Marie Et suscitas Lazare en Betanie, Qi de quart jor avoit la char porrie, La pecherise tornas a ta baillie Qi t’oïnst les piez, de grace replanie. Judas estoit de vostre conpagnie, Quant te vendi, mot fist grant vilenie, Trente diners desoz Jessemanie. De ton cler sanc fu la croiz esclarie; Quant cil i vindrent de la toie partie Ta seinte char fu morte et sepelie. Si voirement, com tu en Galilee A ta masnee, qi por toi est smarie, Apareüs loiaument sanz bosdie: Metez, beau sire, en Oliver la vie, Tant qe il m’ait sa volenté gehie!» Li verais Deus sa pucele n’oblie,

431

110.a

the correct graphy croi je occurs in C 6911. V7 7338 reads: Ce croi je, T 5783: Ceu croi ge, P 5961: Et gel croi. 7165. C: croie for croi je. V7 7348: je croi, P 5970 and L 2511: croi je. Cf. C 7156 and 6891. 7166. C: lo Lazare en Betanie, but lo is superfluous (probably a scribal error) and makes the hemistich hypermetric. No lo is found in V7 7349, V4 5279, P 5971 or T 4793. 7168–7169. Again a reference to Mary Magdalene; cf. notes to C lines 1607, 3789, 6057. 7170–7172. A reference to the betrayal of Christ by Judas Iscariot, for which he received 30 pieces of silver (Matthew xxvi.14–16, Luke xxii.3–6). 7173. C: croiz crois esclarie. 7177. C: est smarie. The verb is either esmarir (‘to trouble’, ‘to distress), s’esmarir (‘to distress oneself’), marir (‘to grieve’, ‘to afflict’) or s’esmaïer (‘to be troubled’). There are 3 possibilities: (1) aphæresis of initial e and particularly of prosthetic e before s impura, a common feature of Franco-Italian – and it is useful here to lose one syllable for metrical accuracy; (2) that est smarie is a scribal error for s’esmaïe; or (3) that the s has simply been misplaced: smarie for masrie, graphy of marrie. P 5981 reads: fu marrie, T 4802: yert moult marrie and V7 7360: masrie with omission of est (ert?) or fu through scribal error. In view of the Franco-Italian element occasionally found in C, it seems unnecessary to emend the reading; hypothesis (1) is therefore taken as valid. 7178. After this line, V7 adds line 7362: Jel croi bien, lasse, sanz point de tricherie and V4 5289 and P 5983 have the same line, but it is not essential to the sense. 7181. C: ta pucele nobile, but a finite verb is needed. Whilst nobile (adj.) is a common graphy of noble, here it is a scribal error for noblie (due to metathesis of i and l); cf. V7 7365: sa pucele n’oblie and V4 5292: sa pucele no oblie. Emendation also improves the rhyme. The question remains as to whether oblie is imperative and part of Aude’s prayer (as C’s ta implies) or whether this is a return to the narrative (as indicated by sa in V7 and V4 and la pucele in P 5986). Since li sanz angles can only be subject case singular (unless li is an error for les, direct

432

7185

7190

7195

La Chanson de Roland (Critical Text and Notes)

Car li sanz angles a sa voiz eshaudie. Joste Oliver s’apoia, lés l’oïe; Lores parole come si fust en vie: «Bele suer Aude, ne vos esmaiez mie, Ne remandrez ne soule ne mendie! O moi venrez en la Deu conpeignie, Amont el ciel, o joie est esclarcie! Tote bauté nos est amoneïe; Poi aproiseriez la terïene vie – El ne valt pas la moité d’une alie, Or ne argent une pome porie. Cil qi Deu sert conqiert grant manantie Ensanble as angles o ja n’avras bosdie. Levez vos en, si soiez esbaudie! Je n’ai congié qe plus raison vos die.» L’angle s’en vait et Aude est sus sallie: «A! Deus,» fait ele, «com ore sui garie! Jamais por duel ne serai esbaïe!»

object plural), Car in C 7182 must be the conjunction, ‘for’, linking the two lines in a statement: ‘The true God does not forget his/the maiden, For the holy angel has listened to her voice’ (cf. note below). The reference to sa voix supports this reading, as do V7, V4 and P, so ta (C 7181) must be emended to sa and these two lines read as narrative. 7182. V7 7366 reads: esbaudie, as do T 4808 and L 2515, giving the meaning: ‘for the holy angel rejoiced at her voice’. Duggan makes li sanz angles the subject of s’apoia, C 7183, with the second hemistich of C 7182 an adjectival phrase, ‘delighted by her voice’. However, the C ms. reading eshaudie, < eshaudir, graphy of exhaudir/exoïr, ‘listen to’ or ‘hear’, makes very good sense: ‘the holy angel has listened to her voice’. It is then the angel who rests his head by Oliver’s ear. 7184. C: paroles, but taking this as a substantive gives a very abrupt reading; a verb is really essential. V7 7368 reads: parole (< parler – cf. Pope 1934, 348s., § 925) for this variant of present 3). V4 5294 has: parloe, but P 5989 and T 4810 read: parla. The final s has been dropped to match V7’s reading. 7192. C: porine, but V7 7376 reads porrie without the inorganic n. 7194. C reads: n’avras, future 2, because Oliver is here addressing Aude, his sister; V7 7378 and V4 5300 read: n’avra, future 3, because they recognize that this sentence is a generalisation: ‘The person who serves God… will never have…’. 7197. C: L’angele s’en vait, which is hypermetric; but elsewhere in the text – apart from C 5779 where the same graphy has to be emended to give the correct metre – the graphy angle (for ‘angel’) is used consistently. V7 7381 reads: Li angle, V4 5303: L’angle.

433

La Chanson de Roland (Critical Text and Notes)

 7200

7205

7210

7215

Aude se lieve et s’orison fina Et les deus cors doucement racouza, Les deus boens pailes gentement raloa Et del mostier toz les huis desferma. Et Charllemene et sa gent i entra. Gerarz li dus sa niece i trova: «Belle niece Aude, ne vos esmaiez ja, Car Charllemene granz honors vos donra.» «Sire,» fait ele, «de ce ne parlez ja. Ja, se Deu plast, nul jor si grant n’avra Ceste chaitive, qi de son duel mosra. Oncle Girart, qel daumage ci a! Jamais en France si grant duel n’en ara.» Enver le roi un petit s’aclina; Confession la belle demanda, Et respont Karles, «Ja plus ne demora.» Un archevesqe qe li rois mot ama A confeser li rois li comenda; Et cil prist Aude, une part la mena En une croute o ses pechiez conta.

110.b

7201. C’s ms. reading is ra9za, which would normally transliterate as raconza, but there are a number of occasions in C when the symbol 9 is used (particularly before an m, but also before other consonants) to stand for -co-. This is probably the case here: racoza for racouza. On the other hand, V7 7385 reads reconcha, written in full. This might possibly be a variant graphy of resconser, ‘to hide’, implying that Aude conceals the two bodies again under the shroud; but, in view of C 7122–7124 in which she sits the bodies up, it is more logical to have her lay them down again. This would then confirm the reading as a variant graphy of the verb recolchier/recoucher/recouzer with features typical of Franco-Italian. (V4 5307 reads: reseta.) 7209. C: Ja, se Deu plast, se nul jor si grant n’avra; but the second se, inserted in error, makes the second hemistich hypermetric. Foerster and Mortier agree in reading si grant (written grt + ~ in a standard abbreviation) and it makes excellent sense, referring back to the grant honor in C 7207. However, it is not found in any other version: V7 7392 reads: nul jor seignor n’ara, V4 5314: nul segnor no avra and T 4835: mes nul segnour n’ara. These readings may reveal the original text (cf. C 6908) or derive from a mis-reading of si grant; when this is taken in conjunction with C’s hons in C 7207, it is impossible to determine where the earlier error may lie. 7215: Ja plus ne demora, taking demora as graphy of demorra, future 3, reads either as a kind of semi-reported speech: ‘he will not delay any more’, or an impersonal use of the verb in direct speech: «It will not delay any more», i.e. «There shall be no further delay» – but demora reads better as present subjunctive 3: «Let there be no further delay», with Italian -a substituted for French -e ending.

434 7220

7225

La Chanson de Roland (Critical Text and Notes)

Cil li enjoint selonc ce qe fait a La penitance qe la bele fera. Un seul petit Aude s’agenoila, Bati sa coupe et Yesu reclama; Le Segnor proie, qe tot le mont fait a, Qe mort li dogne, qe mot la desira: A Oliver son frere s’en ira, Et a Rollant, qi mervelles ama. Aude se seigne et sa raison fina, Puis torne arere, son duel recomenza.

 7230

7235

7240

7245 7246a(V7)

Aude est confesse, sa raison a finee; Pois torne arere come couse adolee: «Frere Oliver, com dure destinee! Sire Rollant, aviez moi juree, Se Deu pleüst, qe fuse mariee, Sor totes dames fuse de vos privee. Oncle Girart, n’i a plus demoree; La mort me vint, qe tant ai desiree! Sor totes dames soit Gibor saluee, La meltre dame qi soit de mere nee, Qe me nosri en sa chambre privee Com s’el m’eüst dedenz son cors portee» A icest mot chiet belle Aude pasmee. Li rois la prent, ver soi l’a retornee Et de ses paus un petit afublee. Li cuers li part, n’i ot longe duree, Entrals s’en est l’arme del cors sevree. Angle l’enportent qui l’ont tres bien gardee, Devant Yhesu l’en ont la sus portee.

111.a

7232. C: Fre Oliver; V7 7415: Frere Oliver. Fre might perhaps show the influence of Italian Fra, but it could readily arise from omission of the symbol 7. 7233. C: Sire .R. ne aviez… , the barred ne having ~ over the e. 7242. C: be Aude, and bel Aude in V7 7425, but C’s usual graphy is belle Aude. 7246. After this, C has clearly omitted a line: the text as it stands gives no indication as to the subject of ont in C 7247. V7 7430 is matched by V4 5343, P 6019, T 4874 and L 2547, although V7 reads the present 3, ot, whereas the other versions have present 6, ont, as needed to match the verb in C 7247.

La Chanson de Roland (Critical Text and Notes)

7250

435

Le rois l’esgarde, qide qe soit pasmee; Quant la redrice, si l’a morte trovee. Lors recomence li deus et la criee; Ne fu ainc plente por feme qi soit nee.



7255

7260

Or est li duel de Karllon enforcié; Aude vit morte, dont a le cuer irié; Encontre soi a le cors enbracié. Tant fierement a lo roi angossié, Per un petit ne l’a tot trebucié. «Sire Girart,» dist Karles l’ensegnié, «Veïstes mais tal duel ne tal pechié? Ahi, belle Aude, com m’avez trechié! Et mi baron, qi tant erent proisié! Jamais por als ne serai essaucié!» Doi arcevesqe ont le cors preseigné, Entre deus bieres soz un paille dolgié La ont le cors de belle Aude couchié.

 7265

7270

Morte est belle Aude, mot est grant la dolor; Ne fu mais dame qi morist por amor. Dui arcivesqe la conroient le jor, Qi l’ont envolse d’un paille de color. Joste Oliver, dales le pogneor, La font cochier el nun del Creator. Plorent li prince et li magne contor,

7251. plente is here a graphy of plainte. 7252. In the ms., the word de has been added superscript between duel and .K. 7255. C ms. reads: roi et angossié. 7266. ‘There was never lady who died more of love’ or ‘No other lady ever died so completely of love’. This line is rather cryptic, but the same structure is found in V7 7450, V4 5359, P 6037 and L 2356, although T 4890 has the imperfect subjunctive 3, mourust, ‘would die’. 7270. C: e nun al creator, but V7 7454 and V4 5364 read: al nom del Criator, P 6041: el non dou Criator. The standard formula has been adopted, whilst keeping as close as possible to C’s reading; but C’s error may arise from omission of ~ over e (> en) or from transposition of the prepositions.

436

7275

7280

La Chanson de Roland (Critical Text and Notes)

Et povre gent, et clerc et vavasor; En la cité orent tel tenebror, L’un ne vit l’autre, tant fu grant la dolor; Ne il ne sevent s’il fu o nuit o jor. Fier fu li deus del riche enperaör; Tire sa barbe plus blanche qe n’est flor. «Sire,» dist Naymes, «frans rois de grant valor, Veez qel gent vos tenent por seignor. Rois, ne t’ocire, anz recovre vigor; Si les conforte a loi de pogneor.»

111.b



7285

7290

7295

«Droiz enperere,» ce dist Gerart li ber, «Por amor Deu, laissez le duel ester; Nus n’i pöon nule rien recovrer. Faites les cors des transiz enterrer, Car cist enpires nel pot plus endurer.» Et il si firent, n’i volsent demorer: Il font les oz des transiz enterrer; Les doze pers fisent bien enbasmer Et en sarqeus pouser et aloer. Sor toz les altres font Rollant conréer Et Oliver et Audein au vis cler; Oliver firent a un moster porter, Prés de Rollant ne:l vousent enterrer. La veïssiez un duel renoveler, Chevols detraire et tant baron pasmer! Et dan Gerart de Viene li ber Auden sa niece comence a regreter,

7273. Initially omitted, tel was added at the end of the line with a mark after orent for its insertion. 7289. The form fisent as past definite 6, faire, is vouched for by Pope (1934, 490, §1320, xxx) as a Northern variant and by Fouché (1967, 288 § 146 (d)) as typical of North-Eastern France. It occurs 5 times in C;misent 4 times, prisent 3 times and qisent once; cf. note to C 1510 and Introduction, 75 (viii), Features peculiar to Northern and North-Eastern France: Morphology. 7290. C: Et en sarqes, which appears hypometric, but the normal graphy is sarqueu or sarcou. V7 7474 reads: Et en sarques, also omitting the second u – but Duggan emends this to sarqueus. 7298. Auden, a graphy of Audain, the object form of Aude is rarely used in C.

437

La Chanson de Roland (Critical Text and Notes)

7300

7305

7310

7315

Et Oliver qi tant fist a loër: «Beas niés,» fait il, «bien devroie desver. Ahi! Con grant duel m’en estora raler! Bele niece Aude, qe porai ge conter A la duchesse qi tant vos selt amer? Cest mariage ne li poet on celer! Ja Deu ne place, ainz q’il doie avesprer, Qe cist miens cuers anz me puïsse crever!» Qi donc veïst lo duc Girart pasmer, Bien li poïst de grant dolor menbrer! Karles ne pot si grant duel endurer; Dedenz sa chanbre se fait mot tost mener: Deus jors i fu nostre enperere ber, Onqes n’i pot a nul home parler; Quant au tierz jor se prist a apenser, Dés ore velt ses barons conforter. Et fait ses oz partir et deviser; Per tote l’ost fait hucher et crier Q’en douce France pensent del retorner, A ses barons le hontage mostrer, Por la venjance de Guenellon parler.

112.a

 7320

«Droit enperere,» dist Gerart le guesrer, «Cil qi sunt mort, ne se poent aider. J’ai mis en terre mon nevo Oliver Et ma niece Aude, la fille au duc Reiner.

7301. The C ms. reading Ahi seems to produce an hypermetric first hemistich but, despite the h, it must surely constitute a single syllable. This is again an example of the aï diphthong and its dubious metrical value (see Introduction, 92–93), but here the difference in sound between V7’s A! and C’s Ahi! is surely important: the greater sharpness of the latter expresses Girart’s anguish far more effectively. 7302. C: pora ge, presumably arising from porai ie/porai je > pora ie > pora ge, with earlier haplography of one i; V7 7486 reads: porai je. 7313. The change in Charlemagne’s attitude and mood ‘on the third day’ is, of course, an oblique reference to Christ’s Resurrection. A certain ‘religiosity’ is typical of this text, with long invocations of God listing the most popular elements of mediæval biblical teaching. 7322. C: Lai mis…, with lai, imperative 2, laier, but ‘Leave buried…!’ accords ill with the possessive adjectives, mon… ma…, which follow, and V7 7506 reads: Je ai, P 6091: J’ai.

438

7325

7330

7335

La Chanson de Roland (Critical Text and Notes)

Ber, car chevauche, si fai le duel laisser! En douce France en ont mot grant mester; A Mont Löon, sus el palais plegnier, Venront les dames, ne:l vosront pas lasser, (En doce France en ont mot grant mester) Cels demander qi vos firent chargier. Mostrez li Guene, qi vus fist l’enconbrier! Se loiaument ne se puet derasnier, Voiant auz toz, le faites depecier! Sifaitement les en porez venger.» «Ahi, Gerart,» dist Karles au vis fier, «Unqes en terre n’ot meillor chevalier Por son seignor loiaument conseillier! Li tuen conseil vuel je tot otroier.»



7340

7345

Va s’en li rois, en terre lait Rollant, Les doze pers don i a damage grant. Karles apele dan Richart le Normant: «Frans chevaliers, esperonez avant! Faites crier mon ban et mon comant: Qe s’en retornent mot grant joie menant; Laissent ester le damage pesant. De Guenelon les proi toz et comant, Qe il le gardent por merveillos sanblant.

112.b

7328. This line repeats C 7325 in error (possibly through accidental reference back to laisser, the last word of C 7324, instead of lasser at the end of C 7327). V7 does not make this mistake. 7330. C’s use of li instead of the plural lor (found in V7 7513) is either grammatical error or to indicate that it is ‘to each one’. Cf. C 7838. 7340. C: Gerart le Normant, but it is confusing to have another Gerart when Gerart de Viene has figured so much in the text from C 6505 on. This appears to be a scribal error, since V7 7523 has: Richart, V4 5426: Riçard, P 6109 and T 4964: Richart, although L 2579 reads: Gautier. C 5585 implies that a Richart le Normant was killed by Baligan, but a man of this name continues to play a part in Section C (cf. C 6022 and 7640: Richart de Normandie; C 7837 and 7843 simply: Richart). This probably arises from grafting on a new ‘final section’ to the poem – unless he was struck but not killed earlier. If the copyist had recalled at this point that Richard was dead and deliberately changed the name, one would have expected him to choose a different name from Gerart (as L does) to avoid confusion. The C ms. reading has been emended to Richart.

La Chanson de Roland (Critical Text and Notes)

439

Quant nos vendrons en France la vaillant, Se:l jujeront le plus saive puissant.»

 7350

7355

Va s’en li rois o sa grant conpegnie; D’or et d’argent orent grant manantie, Chevaus d’Espegne et mulez de Surie. Passent Anjou et Toregne et Valie, Tot droit vers Chartres ont lor voie acollie. A Boneval, une riche abaïe, Herberja Karles enmi la préerie.



7360

7365

Al matinet, quant il dut ajorner, A icele ore qe il durent monter, De meintes parz fisent grasles soner, Tabors et tinbres et bugleraus corner. La messe est dite, si pensent del monter. A nostre roi font son mul amener; L’estrier li tint Gerart de Seint Omer. Le duc Naymon fait Karles apeler: «Faites ma gent en cel canp arester, Car je volrai a mes barons parler, De Guenelon grant consel demander; A lor talant m’en volrai acorder De cel traïtor, qe si mal sot ovrer.»

7348. The occasional use of le in place of li, the normal masculine subject form plural of the definite article, is a feature of Franco-Italian. C here uses the graphy saive with the i correctly marked. 7355. C: emi; P 6122: en mi, T 4981: enmie (V7 7538 changes this to: a la barbe florie). 7359. C: tinbes, with r or its symbol omitted; V7 7543 reads: timbres. 7368. Once again, a hemistich containing the word traïtor appears hypermetric. V7 7552 reads: Del trahitor (written in full) and V4 5452: Del traïtor, which solves the problem. P’s Sa traïson and T’s Com de traïtre retain the disyllabic diphthong. Cf. notes to C 1319 and 7134 and Introduction, 92–93, for further details of this problem.

440

La Chanson de Roland (Critical Text and Notes)

 7370

7375

7380

Nostre enperere son grant dol n’oblia, Mais en son cuer fierement le cela, Et rit et gabe et sa gent conforta. Naymes chevauche, pois q’il le comanda; Desor Morel mot tost esperona; En une place tote l’ost aresta. Li enperere ses barons apela, De Guenelon conseil lor demanda; Et sachent bien, nul ne departira Jusc’a cele oure qe il jugiez sera Et de sa boche trestot reconostra La traïson, si com la porpensa.

113.a



7385

7390

«Frans chevaliers,» dist Karles au fier vis, «Servi m’avez en estranje païs, Et meinte terre avrai per vos conquis, Citez et bors, et meint bon chastel pris. Retorné somes el reigne Seint Donis; Demain seron a Monmartre a Paris. Por Dé vos pri, confortez vos amis. Mais de Guenon n’en avez ren apris, Qi le daumage et lo mal vos aquis, Por qoi ma gent furent issi malmis, En Rencevals detrenchez et ocis. A Mont Leon le menron, ce m’est vis; La jujeront la gent de mon païs; Tolu lor a le meus de lor amis.»

 7395

«Droiz enperere,» ce li dist Salomon, «Vus avez pris le conte Guenelon

7382. C: estraie or estraje with missing n; V7 7566 reads: estrange. 7383. The word avrai must surely be pluperfect 1, like averai, C 7618, and not future 1. V7 7567 reads simply: je ai… conquis; V4 5466 has: e mant ai… c.; but T 5007 has: aroi… c.

441

La Chanson de Roland (Critical Text and Notes)

7400

7405

Et se:l retez de si grant mesprison, Q’il porchaza la fere traïson, Don furent mort li .xii. conpegnon. Menez lo, sire, deci a Mont Leon! De maintes terres vienront li baron; S’il ne s’en puet defendre par raison, Voiant aus toz en prenez vengneson; Et s’il vos plast, nos le vos jujeron Si autement com deviser savron.»

113.b



7410

Va s’en li rois por prendre son juïse; Ainc ne fina deci a Seint Denise: Chers est li lius et mot riche l’eglise; Mot riche ofrande a desus l’autel mise. De Guenellon, qi sa gent a maumise, A Mont Löon ert faite la justise, Si qe:l vesront la gent de mainte guise, Cil de Braibant, de Flandres et de Frise.

 7415

Li rois chevauche sanz nul demorament; Del conforter ne se fist mie lent Barons et dames et enfanz ensement. Ja lo sien cors n’ara repousement Jusqe cele ore q’ara pris vengement De Guenelon, qe:s mena malement.

7397. The C ms. reading seems to be: Et sel retraez, since retez is written with symbol like inverted w over the te; this would be present 5, retraire, ‘to reproach’, but it produces a 5syllable hemistich. The verb here is simply retez, present 5, reter, ‘to accuse’; V7 7581 has: retez and V4 5481 reads: raitez. 7408. C: luis for lius, cf. Introduction, 71–72. V7 7592 reads: lous, P 6147: lieus, T 5032: lieu. C ends the line with la vile, which spoils the -ise rhyme. V7 reads: la glise, P 6147 and T 5032: l’eglise – confirmed by allusion to an altar in C 7409. 7412. C’s ms. reading is gise but it seems best to insert u for clarity. V7 7597 reads: guise. 7415. The subject of fist, past definite 3, faire, must surely be each ‘baron, lady and child’ (C 7416), barons, dames and enfans all being in the nominative singular; the use of ensement, ‘equally’, also stresses the idea of ‘each, individually’.

442 7420

La Chanson de Roland (Critical Text and Notes)

Ainz en ert Karles corozos et dolent, Car li traïtres a meint riche parent Qi per esforz venent a jugiment; De lui defendre se metront en present.

 7425

7430

Li rois de France ne s’est pas demoré; Cum hons gentius a sa gent conforté. Enfanz et femes ot asez encontré, Car chascun a de s’ami demandé; Cil a grant joie qi lo sien a trové. A Mont Löon se sunt tot aresté; Et li traïtes est de grant parenté Et il les a privéëment mandé, Car mot se dote qe il est afolé.



7435

7440

A Mont Leon est li rois de Paris, Et conte et duc et demene et marchis. Li enperere qi tante reigne a conquis A la cort d’os s’apoie, ce m’est vis. Blanche ot la barbe plus qe n’est flor de lis, Vairs ot les els, et mot ot fier lo vis; Afublé ot un mantel qi fu gris. «Baron,» fait il, «conseil vos ai reqis, Qe:l me donez por amor Seint Donis. Conqesre alai Espegne et lo païs:

114.a

7434. C: et demene marchis. The word demene may be read as an adjective, ‘seignorial’, ‘royal’ or ‘landed’, used with the substantive marchis. However, in view of the repeated et… et in C’s first hemistich, it is preferable to follow V7 7621: e demeine e marchis. L 2601 reads: E duc e conte e prince e meschis in support of this. The line is not found in V4 or T, and P 6172 offers a greatly modified version. 7436. In the phrase, a la cort d’os, os is a variant of the stressed (disjunctive) pronoun, masculine plural, eus. C 7434 has listed the nobles gathered at Laon and in C 7435–7436 the writer declares that the emperor relies on the court of justice they constitute to judge their peer, Ganelon. Charles himself is the plaintiff and accuser of Ganelon, so it must be the high nobles who judge the case – and he has to rely on these men, most of whom did not accompany him to Spain and so are not aware of all the circumstances, to deliver the correct verdict.

La Chanson de Roland (Critical Text and Notes)

7445

443

Dusc’a Seint Jame ai les chemins conquis; Jamais nul jor n’i faudront, ce m’est vis. Vus me carjastes vos fiz et vos amis; Asez en ai et perdu et maumis Per Guenelon, qi malvas plait m’a mis.» Quant cil l’entendent qi es rens sunt asis, Chascuns plora des baus els de son vis.

 7450

7455

7460

7465

«Danz enperere,» dist li cons Otoier, Qi tint Amiens et Bologne sor Mer, «Guenelon faites devant vos amener Et si osrons son dit et son penser! S’il ce conoist qe vos oï ci conter, Del respiter ne doit ja nul parler, Mais del livrer le cors a turmenter. L’on ne doit pas son baron afoler Se l’on ne:l poet per grant raison prover O a defendre per ses armes porter.» Dist Guandeboes, «Ja ne sera tant ber. Por tant com puisse sor mon cheval monter, Se il le nie, je sui proz de prover.» Et dist Karllon, «Faites le me amener!» Et cil i corent qe:l devoient garder; Sus el palais le font avant aler.

 Quant Guenes fu el palés en estant, Totes ses gardes l’amenerent avant; Encontre lui sunt levé si parant,

7459. The verb poet (present 3, pooir) in C 7458 is followed by an infinitive, prover, directly dependent upon it; but the structure poet a + infinitive is also found – and used here in C 7459 with a defendre. V7 7646: ou au defendre confirms this. 7462. C’s reading, proz, is apparently subject form singular, preu, ‘well-qualified’, but it reads better as prez, ‘fully prepared’, ‘ready’. V7 7649 reads: prez. Cf. note to C 758, re C’s use of proz for prez.

444

7470

7475

La Chanson de Roland (Critical Text and Notes)

Qi le saluent belament en riant. Primers parla dan Richart le Normant: «Per ma foi, Guene, vos avez blasme grant. Li rois vos rete de son nevo Rollant.» «Sire,» fait Guenes, «il dira son talant! Vez ci mon gaje, s’est qi en aut avant De traïson, dont je mon cors garant: Ne dons n’en oi, ne argent n’i demant.» Dist Gandebues, «Taisez vos en atant!»

114.b



7480

7485

«Entendez moi, franc chevalier vallant! Quant nos venimes en la batalles grant, En Rencevals per som les porz passant, De l’ost Marsire preïmes un Persant Qi nos conta lo fait et lo sanblant, Come fel Guenes nos vendi l’amirant. Se il ce noie, sire, tenez mon gant! Ainz q’il soit vespre ne le solel cochant, Le qit je rendre et mort et recreant.» Dient Franzois, «C’est parole avenant.»

7469. C: Qil le saluent, with extra l inserted in error. V7 7656 and T 5044 read: Qui le saluent, V4 5511: Che le saluent. 7470. C: parla, with la inserted in superscript after par. 7473. After this line, V7 adds line 7661: Je sui del tot por voir en son comant; V4 5515 and P 6207 have a similar reading. C may have omitted this line, but it is not essential to the sense. 7474–7475. ‘Here is my gage of battle, if there is anyone who speaks up to make any accusation of treason, against which I stand guarantor of my body (i.e. in denial of which I am prepared to fight).’ The proving of innocence by armed (judicial) combat was accepted practice in the Middle Ages – and doubtless Ganelon here sought to plant doubt in his hearers’ minds by sounding so confident (and thus prevent their challenging him). In C 7474, avant is present subjunctive 3, avancer, here meaning ‘to put forward an opinion’ and en aut the adverbial phrase, ‘in a loud voice’. 7476. ‘Neither did I have gifts from them, nor ask for silver (money) there’, but the second hemistich reads rather strangely and demand, present 1, is rather surprising, so immediately following oi, past definite 1. The other versions have a variety of readings: V7 7664: Que je n’en oi ne argent ne bessant (bessant being a Byzantine coin); V4 5521: Che jeo non prist or ni arçant; and T 5052: Ne je n’en ai ne or fin ne besant. In each, the second hemistich specifies what ‘gifts’ Ganelon did not receive. C’s reading may therefore be faulty: n’i demant would then derive from mis-read n’ademant, ‘nor diamond’ 7486. Mortier maintains that it is illogical to read et mort et recreant and, if one takes recreant as meaning ‘ready to confess defeat’, this is indisputable; so V7 7674, reading, ou mort ou

La Chanson de Roland (Critical Text and Notes)

445



7490

7495

Li rois de France estut en son estaje Et Guenelon li a tendu son gaje, Vers Gandebuef de Frise la salvage. Nostre enperere fait bien garder le gaje Et cel i mist cels de son parentage; Car prou le sevent et mot de fer corage. Armer l’enmenent la gent de son lignage; Ja n’en gasra li traïtes en eage, Qe n’en reçoive un si fer treüage, Dont parleront et li fol et li sage.



7500

Armer l’enmenent li parent Guenelon, Aval el borc, ches son oste Simon. Chauches de fer, blanches com auqueton, Li ont lacées les jambes environ;

recreant , should then be adopted. (It is also found in V4 5525.) However, recreant can also mean ‘treacherous’ or ‘renegade’, epithets which may validly be applied to someone after death: perhaps here ‘recognized as a traitor’. In C, although the phrase o mort o recreant is found in C 4608, the clause q’est morz et recreanz is applied to Canabex in C 5624. C’s reading has therefore been left unaltered. 7490. In the C ms., la salautage has a rounded mark added to the top of the t, apparently to correct this to la salaufage (Foerster’s and Mortier’s reading) – but no tail is added below the line. Even so, the reading is unsatisfactory – and is hypermetric. V7 7678 reads: la salvage; V4 5529: la salvace; P 6226: la sauvaige; T 5066 and L 2630: la sauvage. C’s reading appears to be a strange amalgam of salvage and sauvage – or salvage contaminated by the preceding la: la sala… Indeed, the superfluous a after the l is very faint, as if it had been partially erased by the scribe. 7495. C: traïtes evage, with evage, adjective, ‘inhabiting the banks of a river (or the shore of a sea)’, possibly applied to Ganelon because his domains lie around Mâcon (where his wife Berte was living) – i.e. ‘from the banks of the Saône’ – but this adjective is often applied to Arab sailors with the meaning ‘piratical’, as in C 6512 (cf. footnote). V7 7683 reads: Ja ne garra li trahites en evage, with the second hemistich hypermetric and the meaning of the final phrase unclear: Duggan excises en and reads evage as ‘piratical’. Since C often confuses n and u (cf. hauste, hanste), the reading has been emended to en eage, meaning ‘for the length of his life’, i.e. ‘for as long as he lives’. 7496. C reads recovie, with the usual ui for iu (cf. Introduction, 71–72); this must be read as reçoive, as in V7 7684. The word treüage, ‘recompense’, also acquired an unnecessary i, giving treüaige and spoiling the rhyme. V7 7684 has: treuaje, T 5072: truage, V4 5533: treusage.

446

7505

7510

7515

7520

La Chanson de Roland (Critical Text and Notes)

De fin or furent andui li esperon; Fort fu l’osberc, blanc come fremellon; Vestu li ont un hermin pelizon; Lacent li l’eume Clarenbalt de Mascon. Cil sire le destruie qi confonde Mahon! Escu li baillent ou ot paint un lion; Puis li amenent un bon destrier gascon; Per son l’estrer est montez le felon. D’armes porter resanble ben baron, Mais de conbatre n’atendra ja raison; Or s’i porpense q’il qesra garison: Il s’enfuira, cui q’en poist ne qi non. En une place fu devant la maison; Sanblance fist d’asager l’asragon; Tuit lo regardent et serjant et garzon; Puis s’en torna, fuiant a esperon; Parmi la porte, si qe ben lo vit hon, Les grant galoz s’en ist de Monleon. Quant il fu fors, lait aler a bandon; Ne l’atensist sagete ne bouzon. A voiz escrie Gondeboef le Frison,

115.a

7503. The adjective fremelon, ‘shining’ or ‘gleaming’, is regularly used in standard descriptions of hauberks (cf. C 1707), but the phrase blanc come fremellon poses problems: the adjective used substantively would simply give ‘white as a shining thing’. The other versions have avoided this in a variety of ways. V4 5538 reads: Vestu li a un auberg fremillon; V7 7691 has l’auberc qui fu au viel Doon and P 6239: l’auberc Clarembaut de Mascon; T 5080 makes Fremillon a person: le haubert que forja Fremillon. C’s fremellon may represent a scribal error in the source or the fault may lie in blanc come, possibly a misreading of blancoï e/blanchoï e (blancheï e) ?, ie ‘white and shining’ or ‘cleaned and shining’. If one assumes corruption, the standard phrase fres esmolu, ‘freshly made’, is an attractive option; but fres maillon, ‘freshly made chain mail’, may be most suitable. 7506. V7, V4 and T all omit this line, but P 6241 here reads: Qu’il li donna quant fist la traïson. In C, after this line, 5 lines are left blank in the ms., but nothing appears to have been omitted, according to the other versions; nor is the parchment unduly thin. 7517. C: Et puis s’en torna, which is hypermetric; but both V7 7705 and V4 5547 omit the initial Et. Alternatively, torna may be a scribal error for torne (historic present). P, T and L lack this line. 7522–7524. Of these three lines, the middle one (C 7523) is clearly out of place, since it simply repeats C 7520 out of context. Instead of C 7522, V7 7710 reads: Ce fut la joie Gondebof le Frison, supported by V4 5550 and T 5093; in each case, this is the last line of the laisse (but Duggan then inserts C 7524, which V7 lacks). V7 7710 is identical to the first 7 syllables of C 7538, the second hemistich in each case beginning with the same Proper Name; and V7 7710 (beginning Ce fu la joie) offers a far better link to C 7524 than C 7522 does. In place of C 7522,

447

La Chanson de Roland (Critical Text and Notes)

(Quant il fu fors, lait aler abandon.) Donc dist al roi, «J’ai conquis lo felon!»

 7525

7530

7535

7540

Quant li traïtes fu fors de la cité, En fuie torné sor le destrer armé, Come li om qi fu de mal pensé, Et son ostage a lassié enconbré, Meus velt q’il soit ocis et afolé Q’il ne conduie son cors a salveté. Cil qi lo gardent en ont lo cri levé Et la tumulte levé per la cité. A bien .c.m. i sunt Franzois monté. «A! Deus!» dist Karles, «com ai le cuer grevé! S’ensi s’en vait, le cuer avrai crevé! Qe:l posra prendre, mot m’ara achaté: Mil besant d’or li seront ja doné. Ce fu la joie Gondebuef l’alosé: «Sire,» fait il, «non aiez mal pensé! Je:l vos rendrai ainz qe soit avespré! Je le vos di, de fi en verité. Ancui vesre Maltalant esprové: N’a mellor beste en la Crestïenté, Qe por grant cors nel vi onqes lassé.»

115.b

P 6250 has an alexandrine: Quant le voit Gondebuef, a poi d’ire ne font, which is also the last line of the laisse. In this line, there may have been confusion based on C’s voiz, P’s voit and the V7V4T reading joie; and it is tempting to follow V7V4T. Nevertheless, C 7523 may represent a line (of direct speech) missing from C, since C 7524 does not follow smoothly from C 7522. But C 7524 is relevant here: Gondebuef’s exultant cry of victory is justified because Ganelon, by galloping away rather than face Charlemagne’s champion, has admitted defeat – and, by implication, guilt. 7527. Since C’s reading is li om, presumably two separate words were written intentionally; thus ‘like the man who was of evil intent’. V7 7714 reads only one word, lion, but comparison of a man on horseback to a ‘lion of evil intent’ (fleeing away) does not produce an effective simile. T 5095 offers the only other version of this line and it reads: ly hom. 7542. C: vesre (perhaps due to the repeated -é ending of the rhyme) for vesrai, future 1, veïr’ (or vesrez, future 5?). (V7 7729 reads: avrai… prové.) See Pope (1934, 197–199, §§ 528, 529 (ii), 533 regarding ai > e, plus 365, § 966 for this usage in future 1; although § 966 implies that this was typical of South-Central and Western France, page 495, § 1322, xxiv, speaks (in brackets) of the early levelling of ai to e as a typical South-Eastern trait – which would fit in very well with the Franco-Italian features of C.

448

La Chanson de Roland (Critical Text and Notes)

 7545

7550

7555

7560

Fuiz s’en est fel Guene, grant poür a de soi; Toz ses ostages en a guerpi au roi. Es vos levé et lo cri et l’esfroi! Meint bon vasal en ont pris le conroi; Aprés chevauchent les gent de bone foi. Mais Gondeboef, qi de Frise fu roi, Son Mautalant li a mis ademoi, Car li esclos de son cheval sunt froi. Si prés li vent Gondeboef ad esfroi, Per un petit qe nel fiert per dedroi, Mais tot tens fuit com traïtres renoi. Et Gondeboef li proz et li cortoi Li escria, «Qivert de puite loi, Car joste a moi, si feras qe cortoi! Per cortoisie si fai le tor franzoi!» Respont fel Guenes, «De zo sui ben aproi, Mais vostre force vos suit la endroi!»

7551. C: ademoi. Having changed the rhyme to -eis, V7 7738 reads: Sor Maltalent se mist tot ademeis (written ademeis), the last word being either a graphy of ademis, past participle, ademettre, ‘launched forward with head down’ – or the adverb, ‘hurriedly’, cited by Duggan. The other versions have a variety of readings roughly paraphrasing CV7: V4 5571: a esfrois and T 5118: en tiel effroi (but C reads ad esfroi, C 7553, just 2 lines later) and P 6273: les esclois (but C 7552 contains li esclos). 7553. This line is not found in any other version. Reading esfroi as a graphy of esfrei, ‘noise’, ‘uproar’; gives the meaning ‘with noise’, ‘noisily’, ie ‘with a thundering of hooves’ (or possibly ‘with great clatter (of arms)’). V4 5571 (equivalent to C 7551) ends with a esfrois, which Beretta (1995, 484) translates in this way. Alternatively, this could be read a desfroi, meaning ‘with trouble’, ‘with an expenditure of effort’, or possibly even as a desroi, ‘at speed’, if the s has been repeated as f through scribal error. 7554. C’s reading per dedroi is confirmed by V4 5572: per dedrois, ‘from behind’; V7 7741 has: par detreis, L 2650 (fiert) detrois, T 5124 de destrois, but P 6275 demanois. In C 7558–7559, Gondeboef calls on Ganelon to turn and face him in combat, so that he may avoid attacking him from behind. 7558. C: josta, when the imperative singular, joste (as in V7 7750 and L 2653), is needed. The fact that V4 5583 reads: josta may indicate an Italian influence here. 7561. The MS reading, la dendroi, is found only in C. This is either dedroi (as in C 7554) with pleonastic n, therefore ‘behind (you)’ or the common phrase la endroit, ‘over there’ with roughly the same meaning. The second hemistich of this line varies from ms. to ms.: V7 7753 reads: vos vient grand part detreis, P 6280: vos vient et si vos croist, T 5133: vous suit moult grans destrois. The verb suit is disyllabic (süit according to Pope (1934, 134s., § 328) and may well be a scribal error for siut (earlier standard form) due to C’s confusion of -ui- and -iu- (cf.

449

La Chanson de Roland (Critical Text and Notes)

7565

Dist Gondeboef, «Encor sunt el valoi. Ainz avrons fait q’i venent li guerroi.» Lors s’entreferent es escuz vienoi, Plenes les hanstes s’abatent el desroi. Ja fust li chaples as brans sarazinoi, Quant s’escria Richart de Vermendoi, Hué del Mans et cons Gui d’Aminoi.

116.a

 7570

7575

Or sunt a pié li chevalier hardi; Bien se requirent as brans qi sunt forbi, Quant escria et dan Richart et Gui, Hué del Mans et li cons Naimeri: «Lere traïtres, petit avez fuï! Teus chevaliers vos avot acolli, Ne vos laisast demein devant midi. Gondeboef, sire, avez mester de mi?» Il li respont, «Trop tost somes parti.» Et Guenes dist, «Anzois vos a gari Et j’en arai mon cors forment laidi.»

 7580

7585

Tuit li baron sunt a pié descendu Et Guenelon ont pris et retenu. Gondebuef prist Maltalant lo crenu: «Baron,» dist il, «trop tost estes venu.» Guenes respont, qi de grant engeing fu, «Vos ont gari et moi ont confondu.»

notes to C 6199, 6787, 6826, 7408). There are Northern and N-Eastern variants, siut and seut/ sieut: cf. Pope (1934, 134s., § 328, 488, § 1320, vi, and 492, § 1321, xii). 7563. C: gerroi, but the graphy with gu- is found in all other instances in C. V7 7755 reads: Franceis, having a different rhyme. 7574. C: avot for avoit, imperfect 3, aveir – sometimes cited as a feature of Western French, but Pope (1934, 488, § 1320, vii) lists oi > o as typical of Northern French; V7 7766 reads: avoit. 7577. Both here and in line 7583, C’s reading tos must, in the context, stand for tost, adverb, ‘soon’ or ‘early’, not the adjective meaning ‘all’. V7 7769 and T 5149 read: tost here; instead of the C 7583 reading, V7 7775 and P 6292 read: tost. 7583. C: tos; cf. note to C 7577.

450

7590

La Chanson de Roland (Critical Text and Notes)

De totes parz sunt li baron venu. A pié desent Garins de Mont Agu. Desor Fesrant, qi auqes lassez fu, Firent monter le gloton mescreü. Cil prist lo sor qi de Gascoigne fu. Lors s’en retornent tot lo chemin batu.



7595

7600

Pris fu fel Guenes qi mut la grant dolor, Dont furent mort tant noble pugneor; Cil le conduient qi l’ont mis el retor. A Mont Leon sus el palais auzor, La lo rendirent al riche enperaör Qi en son cuer avoit mot grant dolor. Toz ses ostages avoit mis en la tor. «Baron,» dist Karles, «mi duc et mi contor, Jugiez moi tost icel felon traïtor, Car ma venjanche viel prendre hui cest jor.»

116.b

 Li rois comande Guenelon a juger; Atant es vos pognant un escuer Sor un cheval q’il ot fait angoiser! 7589. C: Puis firent monter, a hypermetric first hemistich, but V7, L and T lack the initial Puis: V7 7781: ont remonté; L 2665: Firont monter; T 5156: Monter y firent. So Puis has been excised. The punctuation here is vital: in order to explain the introduction by name of Garins de Mont Agu (C 7587–7588), it must be understood that it is his horse, Fesrant, which is ‘somewhat weary’ after the chase, that is used as a mount for Ganelon – just in case he attempts to escape again. 7591. The C ms reads: tost. Having omitted the t from this word in C 7577 and 7583, the scribe now inserts s in error in tot. The adverb, tost, makes limited sense here but sits awkwardly after the cæsura – and V7 7783 reads tot, which fits well in the phrase: tot lo chemin batu, ‘the whole road they have travelled (in seeking Ganelon)’. 7594. C: Ce lui conduient or Celui conduient, but neither ce nor celui matches the verb in number. V7 7786 reads: Cil le conduient… 7600. C’s use of icel rather than cel reinforces the argument for reading traïtor as disyllabic. In this, C is more advanced, in terms of the developing pronunciation of the ai diphthong, than V7 7792 (which reads: cel) or P 6308 and T 5167, which both have: cest. Cf. Introduction, 92–93.

La Chanson de Roland (Critical Text and Notes)

7605

7610

7615

451

Isnelement descendi del destrier, Puis est montez sus el palais pleigner O les barons q’il mult fist a priser. Devant Guenon se vait agenoller, Tot soavet li prist a conseiller; Puis li a dit, «Ne vos chaut esmaier. Secors vos vient del meillor chevalier Qi soit en Franche por ses armes baillier: C’est Pinabel de Florence au vis fier. En ceste cort vos vent al roi pleger; Dist q’il n’i a si hardi chevalier, S’a cort vos juge, ne le conper mot chier.» Guenes parole, si respont l’esqïer: «Per foi, amis, je l’averai mot chier; De totes parz m’oëz ci blastenger.»

 7620

Guenes parole, qi se volt delivrer; Al roi demene lait l’escuer aler,

7607. O les barons must here be read as ‘amongst the barons’, rather than ‘with…’, and the second hemistich taken to mean ‘whom he there greatly impressed’ (caused to value him) – the ‘he’ in question being the squire (C 7603). No other version has this line, although for C 7604, P 6312 reads: Sor .i. cheval qui moult fist a priser (which V4 5605 echoes). (P then has: A pié descent dou bon corrant destrier, whereas V7, V4 and T all correspond to C 7605; V7 and T also match C 7604.) The confusion probably arose in the source text; and the faulty line (7607) was retained by C; Duggan here emends the second hemistic of C 7607 to qi mult font a priser and inserts this line as V7 7799. 7613. Florence (found also in T 5179) would seem to be a scribal error for Sorence, the (unidentified) place associated with Pinabel everywhere else. Since Fiorenza, rather than Firenze, was the mediæval name of the city, this may be further evidence of Italian influence. 7616. P 6326 has the same reading: S’a cort…, Pinabel (C 7613) apparently threatening any noble prepared to act as judge, i.e. pass judgement on Ganelon. This may well be a scribal error for S’a tort, ‘If (he condemns you) unjustly’, as in V7 7808. V4 5615 reads: Se a tort…, and L 2679: Si a tort.. in support of V7. C’s ne le conper… is present subjunctive 3, conperer. 7620. C: qi soi aviler. This is hypometric, lacks a finite verb and makes no sense as it stands. It may be intended as a scribal aside: qi soit avilé, ‘may he be brought low!’, but this remains hypometric and ceases to rhyme, whilst qi soit a aviler makes little sense. V7 7812 reads: qui se velt delivrer, P 6330: qui se volt delivrer, a plausible source of C’s scribal error. If, however, C’s model had: qi soi velt aïuder copied by C with omission of velt through scribal error, -iuwritten -ui- and d mis-read as l, then qi soi velt auiler is explicable. Since V7 and P agree here, V7’s reading has been adopted.

452

7625

7630

7635

La Chanson de Roland (Critical Text and Notes)

Donc se li dist son cuer et son penser: «Per ma foi, sire, trop mi poez blasmer, Car mot vos suel et servir et amer. Je sui toz prez desor es sainz jurer Qe n’ai talant de fuir ne d’aler; Per qe laissasse mes amis enconbrier? Anzois volrai mon cheval esprover; A ma batalle voloie retorner; Fait m’avez prendre, si me dot mot peser.» De ce q’il ot, son chief prist a croler; Toz ses barons comence a esgarder; «Franc chevalier,» ce dist Karles li ber, «Tan grant menzoigne ou pot onqes trover? De tot se ment, ben le pöez prover!»

117.a



7640

Charlles apele dan Gerart lo guesrer, Ogier lo duc et Naymon lo Baiver, Odon de Lengres et Basin et Garnier, De Lonbardie lo riche duc Gaifer, De Normandie dan Richart lo guesrer, Joifroi d’Anjou et Salemon lo fier:

7625. C: proz, subject form singular, preu, ‘worthy’ or ‘well-qualified’, but ‘ready’ is needed here and V7 7817 reads: prez, P 6332: pres and V4 5619: prest. The same graphy is found in C 758 and C 7462 (cf. relevant notes). 7627. ‘Why would I leave my friends to bear the blame?’, a rhetorical question in which enconbrier has the meaning ‘lay the burden of guilt on’. Ganelon’s kindred, as his guarantors, would be executed if he ran away. 7630. C: dot, for doit, present 3, devoir (doit in V7 7822, P 6337, T 5191, L 2686). The graphy may be a Western or Northern French feature (cf. note to C 7574: avoit) – or simply arise from scribal anticipation of mot, immediately following, or of ot in C 7631. 7631. ‘At what he hears…’; the il is of course Charlemagne, as .K. in C 7633 eventually makes clear. 7634. Where C reads: on pot onqes trover!, V7 7826 has: o pot cist gloz trover? and P 6339:ou puet cil or trouver?, and the rhetorical question (‘where can he ever find?’) reads better than the exclamation (‘one can ever find!’) because it relates this comment directly to Ganelon. Since on was written in full in the ms., it has been emended as a scribal mis-reading of u as n. 7636–7643. This is the 3rd consecutive laisse with the rhyme in -er. V7 treats all three as a single laisse.

453

La Chanson de Roland (Critical Text and Notes)

«Baron,» dist Karles, «nobile chevalier, Je vos comant Guenelon a jujer.»

 7645

7650

Li jugement fu mis sor meins baron; Selonc les ovres en dient les raison Del jugement a l’encrisme felon. Ainz qe soit faiz li diz de Guenelon, Crostra le duel a nostre roi Charllon, Car Pinabeaus descendi al peron. Ensanble o lui vint Hues de Lion, Et avec aus quatre cent conpeignon. Cil sunt venu al conte Guenelon; Fier et hardi, armé d’une fazon, Per lui defendre se metront a bandon.

 7655

7660

7665

Del jugement se sunt cil aséné; Li enperere lor a dit et prové: «Gardez, baron, qe il soit ben jugé!» Et si croi ben q’il sera respité, Car Pinabaus est descendu a pié; Ensanble o lui ot son cosin Hervié, Qi por Guenon se sunt mot travaillé. Ensanble vont li baron tot rengié; De beles armes furent aparellié; Per grant orguel sunt el palais puié. Et Pinabaus fu mot bien en arché:

117.b

7648. C reads le duel for li duel, subject form, the error probably arising from verb-subject inversion. 7651. C: qinze, but V7 7842 reads: quatre – and so subsequently does C 7685: .cccc. 7665. C’s reading, en arché, could be a graphy of en archie, but it is inappropriate for Pinabel to be ‘very well within bow-shot’ within the palace. The basic meaning of the adjective arché or enarché was ‘curved like a bow’, often applied in complimentary fashion to eyebrows: ‘with well-arched brows’ meaning that a man was handsome. There is, however, no evidence of enarché on its own having this meaning. V7 7856 reads: enforchié, signifying ‘strongly built’ or a general term of approbation relating particularly to a man’s legs and whole stance. This would imply omission of f in C. On the other hand, the adjective encharné, ‘well-fleshed’, ‘strongly built’, was also regularly used in descriptions and it is appropriate here – and also

454

7670

7675

7680

La Chanson de Roland (Critical Text and Notes)

Un petit ot son mantel deslacié; A or bendé sunt li chevol trecié; Ceinte a l’espee al pon d’or entalié. Devant lo roi s’en vint lo chief drecié; Pas nel salue, ainz l’a contralié. Al roi parole, si li a anoncié: «Per ma foi, sire, tort faites et pechié! Mon oncle avez et pris et justisié. Il ne sera per home or jugié, Huimais ne puis, per les els de mon chié! N’a un baron en France tant prisié, S’il voloit dire q’il vos eüst bossié, Ne l’en defende a ceval et a pié.» Franzois en sunt per les rens enbronchié; Qi n’a ami, mot se puet faire irié, Et qi l’a bon, mot s’en doit faire lié; Qe per un home furent toz acoisié, En la cort Karle tot li plus essaucié.

 7685

7690

Pinabaus fu enz el palais plegner; Aprés lui vont .cccc. chevalier. Mot per fu baus, si ot visage fier; Devant lo roi se prist a aficher; Dist tel parole qi mot fist a priser: «Trop vos volez, enperere, abasser, Qi tenez pris lo meillor chevalier Qi soit en France por ses armes ballier.

a feasible source of scribal error via metathesis of -char- to -arch-. T 5225 reads: adrecié, ‘upright in stance’, ‘shapely’ . 7675. Both C and V7 7866 read: Huimais se puis, ‘Now, if I can’, implying: ‘Now, if I can do anything about it’ in a rather weak, elliptical reading. The standard phrase, Huimais ne puis, ‘now or hereafter’, gives a stronger reading and involvea only small, plausible scribal error, se instead of ne. 7679. The C ms. reading, enbuschié (‘hidden’, ‘concealed’), seems to imply that the French nobles draw back from Pinabel’s challenge and conceal themselves in the ranks (hiding behind one another?). V7 7870 reads: embronchiez, ‘with heads bowed’ implying ‘dismayed’, and P 6392 has: embronchiez, T 5237: embronchié; since C 7705 has Charlemagne comment: je voi si mes barons enbroncher, C’s scribal error has been emended.

La Chanson de Roland (Critical Text and Notes)

7695

7700

7705

7710

7715

De fuir n’ot ne talant ne mester: Ainz s’en issi essaier son destrier, Por sa batalle ou voloit repairer. Sus li corrurent et Franzois et Bauver; Se il fui, ne fist a merveiller, Car ne savoit c’on le deüst chacer. Fait l’avez prendre et fortment laidenger; Il n’a en France si hardi chevalier, S’il voloit dire q’il vos volsist boissier, Ne q’il feïst honte ne enconbrier Les doze pers, Rollant et Oliver, Ne l’en defende o m’espée d’acer.» «Deus,» ce dist Karlles, «or n’ai je mas mester, Quant je voi si mes barons enbroncher, Ne ne se muevent por mon droit derasnier.» Aprés cest mot, es vos levé Ogier, Et d’Aspremont lo prou conte Richer, Et de Viane dan Gerart lo guesrer: Chascuns d’aus offre son gaje a desplier, Quant uns vaslez les fist toz desrengier. Tieris ot non et si fu escuer Al bon Rollant, qi le tenoit mot chier. «Franc chevalier, ne vos doi anuier Se je sui vaslez, vus estes chevalier.

455

118.a

7692. C: moster, but mester, ‘need’, is the obvious reading. V7 7885 and T 5247 read: mestier. 7698. C: laideger, with omission of n; V7 7891 reads: laidier, but P 6388: laidengié. 7701. C: Ne q’il feïst herten honte ne enconbrier. This might be read as an alexandrine, but the cæsura is indicated after feïst. Since herten honte ne is a strange combination of words, herten may simply be a scribal error for honte ne, mis-spelt and immediately corrected, but the error left unbarred. Emendation has been made on this basis, since it produces a comprehensible, well-balanced and metrically accurate line, and also fits well with C 7702. V7, V4, P and L omit this line; but T 5254 reads: Ne vous querir perte ne encombrier, substituting perte for C’s herten, omitting C’s honte and, by inserting vous, the link to C 7702. T’s perte offers an acceptable correction of herten, if one assumes honte to be inserted in error by C, but one would expect honte to be the correction in C’s sequence, herten honte. Duggan inserts into his Critical Text as V7 7894 (lacking in V7) a line based mainly on C but reading perte from T. V7 7895 differs from C 7702, which there is no reason to modify. 7704. C: or na ie mas, with a missing i, either due to haplography (reduction of ai je) or as a reduction of diphthong ai to a, a feature of Franco-Italian and Northern and North-Eastern French: cf. Pope (1934, 488, § 1320, vii). V7 7898 reads: or n’ai je mais mestier, confirmed by P 6406, T 5256 and L 2705.

456

7720

La Chanson de Roland (Critical Text and Notes)

Un petitet me laissez derasnier: Cil qi ne seit, coment poit droit juger? Meis je doi bien lo mien gaje ploier. Por ce lo vuel doner et avancier, Qe je sai bien coment mut l’engonbrier, Coment vendi mon seignor droiturer, Lo duc Rollant, qi tant fist a priser; Mais ja menace ne lor ara mester, Qe ne l’en face son guesredon paier.»

 7725

7730

7735

Pinabaus fu el palais en estant Quant li vaslez li est salli avant. Terris ot non, cors ot mot avenant; Escuers fu lo preu conte Rollant. Mot parla bien et si dist avenant: «Tenez mon gaje, enperere puissant, Per vostre droit, qe vel mostrer avant, Qe lerre est Guene, et traïtre a esciant. En Saragoze vendi lo duc Rollant Al roi Marsile, qi en Deu n’est creant.» A ces paroles, tient Pinabel son gant; Por les ostages s’i misent si parant. Et Pinabaus jura qe tuit l’entant: «Mar lo parlastes, lechiere sosduiant. Ferai vos en vencu et recreant.»

118.b

 7740

Lor gages donent, si sunt en piez levé. Pinabaus a ses ostages doné:

7717. poit is a variant of puet, present 3, poeir. 7727. There are several holes in the vellum; indeed, avenant is written ave nãt around one hole. At the foot of the page, outside the written area, a hole has in the past been stitched together: the stitch holes are still visible. 7732. The metrical value of the -aï- diphthong in C comes again into question in the second hemistich, which the et appears to make hypermetric. V7 7924 reads: Que Ganes est traït[r]e e soduiant, P 6434: Que Ganes est traïtres soduians and T 5279: Que lerre est Guenne, traïtre e soudeant, this last the closest to C. However, the fact that the scribe has inserted et (which is

457

La Chanson de Roland (Critical Text and Notes)

7745

7750

7755

Trestot manois as gardes sunt livré Et si sunt bien a garder comandé. Pinabaus a son oncle demandé; Li rois en a sa corone juré, Q’il l’ara ainz per lo canp aqité. Dist Pinabel, «Lor en aiez malgré!» Per maltalant sunt de la cort torné. Pinabel ont isnelement armé. Li enperere a Terri apelé, Mot doucement l’a el vis esgardé; Jone lo vit et de petit aé, Mais fier cuer ot et proëze et beauté. «Seignor,» dist Karles, «soiez aseüré, Ancui vesrons nostre grant droit mostré, L’orguel arons de Pinabel maté.» «Fius,» dist Joifroi, «tu soies comandé A Yhesu Crist, lo Roi de majesté Qi en croiz fu por nos pechez pené!»

 7760

7765

7770

Terris fu baus et de gente fazon; A lui armer ot maint riche baron. Chauces de fer de bone forgison Li ont lacées es janbes environ; Et Salemons li chauce li esperon. L’osberc li vestent al Marchis de Lion: Il l’ot vestu quant fist pais a Charllon; Puis fu doné Gerart de Rosellon. Lacent li l’eume Gondeboef le Frison: Dedenz fu faiz de costé d’un poisson; Soz ciel n’a elme ne brant d’acer si bon, Qi ja l’enpirt la monte d’un boton.

119.a

not essential) implies once again that he reads the -aï- of traïtre as a single syllable. Cf. Introduction, 92–93. 7753. The cæsura was first written after cuer in error, so a second cæsura was inserted after ot.

458

La Chanson de Roland (Critical Text and Notes)



7775

7780

7785

Terri armerent sus el palais plegner; Entor lui ot meint gentil chevalier; Et Charllomene li ceinst lo brant d’acer, Cortein la clere, qi est au duc Ogier: Onqes ne pot en batalle enpirer. Et bon cheval li font apareiller: Ferrant ont pris, qi fu au duc Reiner. En nule terre n’en ot cheval tant fier, Car se il voit un armé chevalier, Sore li cort por son cors depecer; Dedevant Cordes le conquist Oliver. Terris i monte, qi lo velt esaier; Sor les esters se prist a aficher, Grant demi pié les a fait alongier. «Baron,» dist il, «ja mellor ne vos qier.» Ce dist li rois, «Deus te doinst bien aidier, Qe Pinabel me puïsses abaisser Lo sien orguel, q’il demene tant fier.»

 7790

7795

7800

Pinabaus fu armez tost et isnelament Aval el borc, chés son oste Florent, Ches un borzois – riches fu et manent, Qi herberjoit Pinabel et sa gent. Chauces de fer qi sunt a son talent Li ont lacié es janbes maintenent; Ses esperons li chauce Gui d’Aiguent, Fius sa seror, ben sunt andoi parent; Lacent li l’eume Conubre de Clarvent; Puis ceinst l’espee dont li ponz fu d’argent. Deus, qel baron, s’il amast nostre gent!

119.b

7787. C: doist for doinst, present subjunctive 3, doner with ~ omitted in error. 7797. C omits all mention of the hauberk, a vital item, in this list (although it is included in Tierris’ armour, C 7765–7767); V7 here inserts an extra line (7990): Il vest l’osberc qui fu mult avenent, matched by P 6498: Il vest l’auberc dont la maille resplent – and V4 5728, with the same general sense, is inserted (with an extra line) after C 7798. These additional lines are not found in T or L. 7799. C: ceist, with ~ omitted as in C 4931; V7 7792 has: ceinst, as found in C 1575, 1813, 7774.

La Chanson de Roland (Critical Text and Notes)

459



7805

7810

7815

Ceint a l’espee quant l’orent adobé; Herver li a son escu aporté, En une lance son confenon fermé, Qe mot fu roide et li fers aceré. Un vair destrier li orent apresté, Bien l’ont covert et richement armé; Par son l’estrier est Pinabel monté. Fist un eslais mot bel et acesmé Et mot adroit est arer retorné. Desoz la sale avoit un mot bel pré Et un vergier qi fu mot bien gardé Qe on véoit des murs de la cité. Karles i a son chanpion mené. Des plus saintes reliqes de la Crestïenté Li aporterent por lor faire juré. Et Charllomene lor a ben afié, Qi s’esperjure, mais n’en ert onoré, Ainz ert honi ainz q’il soit avespré. Environ ot meint bon clerc ordené.

 7820

7825

Li dus Gerart se sot ben apenser, Et li dus Naymes qi fu gentius et ber, Et li Denois qi mot fist a loër: Charllon meïsme fisent mot tost armer; Mil chevaliers i ont fait conréer. Gart soi Herver de son orguel mostrer, Car Pinabel porroit tost conparer! Li enpereres a fait son ban crier; Granz saintuaires fist el pré aporter; Qi se parjure ne puet longes durer,

120.a

7817. Qi s’esperjure here but Qi se parjure, C 7829. Both graphies exist. V7 8011: Qui se parjure. 7818. C: qi soit, for q’i = q’il: ‘before it was evening’. V7 8012 and P 6512: q’il soit, but T 5339 also reads qi. Cf. Pope (1934, 324, § 841) regarding effacement of præ-consonantal l in the pronoun, il.

460 7830

7835

La Chanson de Roland (Critical Text and Notes)

Ne li conveigne del cors a meschever. Et Pinabel comence a crier: «Ma batalle ofre, qi q’en doie peser.» Li rois respont, «Je sui proz d’escouter.» De Loereigne ot fait le duc mander, Et dan Bovon et le conte Guinemer, Et dan Basin et le conte Otoier, Le duc Richart et Salemon li ber; Puis li comande la bataille garder.

 7840

Li enpereres fait son ban et son cri. Nus ne remeigne, ja n’i ait si hardi, Ne mais qe li set conte qi tant sunt segnori. Cil desendent a pié qi le canp ont oni:

7830. ‘without his having to suffer damage to his body’, ‘without his having to suffer serious wounds’. 7833. Once again, proz, normally subject form singular, preu, but better read as a scribal error for prez, as in V7 8027. C never uses the graphy prez. Cf. C 758, 7462. 7836. C here reads: et le conte Otemer, but this is the sole mention of this name; in C 7843, the name cited is Otes but C 8134 has Otoier. V7 8030 reads: Otoier and omits le, although in both V7 and C elision of conte Otemer/Otoier gives a regular 6-syllable hemistich. V7 8029 similarly omits et to obtain the correct metre whereas C 7835 has a hypermetric second hemistich: et le conte Guinemer, whose last 2 syllables (-emer) may explain C’s graphy Otemer. The name has been written as Otoier in C 7836 and in C 7843 for the sake of consistency and clarity. It is also found earlier, in C 7450 and V7 7637, both versions agreeing on Otoier. 7838. The word li was inserted superscript at the appropriate point. It is either used in preference to lor, plural, to imply ‘to each one’, as in C 7330 or is a scribal error; V7 8032 reads lor. 7841. Beginning with this line and continuing up to and including C 8133, the verse is in alexandrines; the only exceptions are lines 7843, 7851, 7891 and 7899, all of which are decasyllabic – plus C 7854, 7981, 8000, 8019 and 8132, if .K. is read in the normal way as Karlles, rather than Karllemene. The change in metre may indicate that a ‘set piece’ battle scene has been inserted here, or it may be intended to give added weight to this ‘judgemental’ single combat between champions. It certainly makes it easier for the writer to insert the names of the contestants with a greater variety of actions. Duggan comments that , whereas seven Counts are referred to here, only six are listed at the end of the previous laisse – but it is surely splitting hairs to discount the Duke of Lorraine because he is not a Count. 7842–7844. C 7842 reads: oni, a graphy of uni, past participle, unir, with secondary meaning ‘levelled’; a flat area would be needed for the combat. Instead of this line, V7 8036 reads: Qu’il ne viegne la bataille veir which Duggan excises, substituting C 7842. V7 8037 then matches C 7843, both mss. naming Otes where Otemer (appointed by Charlemagne in C 7836) or Otoier (as in V7 8030) is needed for both accurate identification and alexandrine. After this, V7 replaces C 7844 with 3 lines: Et tuit li autre qui li chanp ont garni/Il canpion qui le chanp dont

La Chanson de Roland (Critical Text and Notes)

7845

7850

7855

461

Richart et Otoier ont lor chevaus saissi. Andoi li chanpion furent amenevi; Les reliqes aorent et chascuns i ofri Trente .v. besanz d’or, qe uns clerc recolli. Salemons de Bretegne, qi lo cuer ot hardi, Lor a le serament devant toz eschevi; Lors dist al damisel, «Alez avant, Terri! Vus jurerez premiers de cest canp arami, Si vos ait Deus et li sainz qi sunt ci, Qe li cons Guenelons si a Rollant traï, Lui e ses conpeignons a Marsille vendi; Si boissa Karllemene e sa foi li menti.» «Ensi l’otroi, beau sire,» Terris li respondi. «Danz gloz,» dist Pinabeaus, «tot i avez menti. Vus estes perjurez, ja vos vesrez honi.»

 Pinabaus s’agenolle et Terris s’en leva; Salemons de Bretegne lo sairement dira, garni/Descendent des destres arabi – in which the repeated second hemistich and scribal error indicate corruption. Duggan excises all three lines, substituting C 7844. 7845. C: et chascuns d’aus i ofri, a 7-syllable hemistich unless i ofri is elided. V7 8039 and P 6537 agree in reading: e chascuns i offri; d’aus, here part of a common formula, is not essential to the sense. 7848, 7850. To read arami twice so close together in one laisse makes this suspect. V7 reads arami in line 8044 (= C 7848), but omits the second hemistich of C 7850, combining the first 2 words of C 7850 with the second hemistich of C 7851, introduced by sor instead of et. P has both lines, P 6540 ending…devant euls eschavi and P 6542 . champ qu’as arami. L has only the first line, 2754 ending devant toz l’eschevi; T has only the second, 5367 confirming the C and P readings with de cest champ arami. Since eschevi, past participle, eschevir, fits well here, it has been substituted in C 7848; C 7850 then retains de cest canp arami, ‘in relation to this solemnly sworn (trial by) combat’. 7854. If .K. is transcribed as Karlle, the line becomes decasyllabic in a long passage of alexandrines. Reading Karllemene has already been necessary in C 6756 to obtain standard decasyllabic metre. As Leslie Brook (1990) argues in The Alexandrine Sequence in the post-Oxford Roland manuscripts, the general metrical pattern of the lines (particularly those introducing speech) calls for reading .K. as Karlle Mene (or Karllemene, the usual graphy in C) . 7859. The second hemistich, lo sairement dira, is almost identical to that of C 7860, lo sairement mostra, which makes it suspect. Moreover, the future 3, dira, reads oddly amidst a sequence of past definite 3 verbs, although it simply indicates that Salomon is about to administer the oath. V7 8053 ends this line with a standard phrase: qui li rois mult ama. V4 5762 reads: lo sacramento dit a, P 6555: le sairement dit a, T 5375: le serment visa and L 2762: le

462 7860

7865

7870

La Chanson de Roland (Critical Text and Notes)

Encontre Pinabel lo sairement mostra. Belement le li dist, sa raison afaita: «Vos jurez, Pinabel, sor les sainz qi sunt la, Qe Guenelons vostre oncle a Karllon ne boissa, Ne ne qist traïson, ne ne la porpensa, Ne Rollant ne vendi, ne avoir pris n’en a.» Lor jura Pinabel et cil li devisa; Baiser volst les reliqes, mais onqes n’en toça: N’en plot a Damedeu; tot tranblant se leva. Il a prise sa lance, sor son cheval monta. Terris lo damisals vers lo ciel regarda, Reclama le Seignor qi lo mont estora Qe li doinst hui honor, selonc ce qe droit a; Puis monte sor Ferant, qe plus ne demora.

120.b

 7875

7880

7885

Montez est sor Ferant Terris al cor loial, Tint l’escu as enarmes, si broche le cheval, Pinabel en apele a loi d’ome vasal; Et se li dist, «Vasal, nos somes en igal. Ge vos desfi des ore com home deslial, Quant vers moi defendez lo traïtor mortal.» Lors respont Pinabel, «Vos parlerez tot al.» Lors s’entrelaissent cosre permi le font d’un val. Mervellos cous se donent es escuz comunal, Q’a la terre en abatent et l’azur et l’esmal; N’i pot fauser clavel, ne n’i chaï cheval; Fort furent anbedui, mot sunt proz li vasal.

seyriment leva; but all 4 texts then proceed immediately with Salomon’s speech as set out in C 7862, omitting C 7860–7861. 7861. C: Belement li dist, a 5-syllable hemistich. V7 reads: Bellement dist, but here has decasyllabic metre. In this passage of alexandrines, accidental omission of le (= le sairement, C 7859) is most likely. 7879. The C ms. has desloial as the final word of this line, almost exactly duplicating deslial at the end of the line above and making the second hemistich hypermetric. This arouses suspicions of a scribal error, confirmed by the V7 8073 reading: le trahitor mortal, by traïtor mortal in P 6573 and L 2770, and by traïtre mortal in T 5393. Nevertheless, this again poses the problem of the syllabic value of the -aï- diphthong (cf. Introduction, 92–93).

463

La Chanson de Roland (Critical Text and Notes)



7890

7895

7900

Chascuns torna sa resne et son tor a repris. «Terri,» dist Pinabel, «mot per m’as bien requis. Mar i fu ta jovente, quant tel cose as enpris. Car guerpis la batalle, fui t’en toi ademis! Je te donrai Serence, Besenzon te guerpis; Si pren ma fille, qi tant a cler lo vis, Ne qit q’il ait si belle deci a Mont Cenis.» «Dahait ait tel moiler,» ce respondi Terris. «Mot me fi en Celui qi en la crois fu mis. Je venjerai Rollant et mes altres amis, Et Oliver lo conte, qi a tort fu ocis, Qi furent per vostre oncle coroceus et malmis. S’en ara Karles joie, qi sire est de Paris.» Lors s’entrelaissent cosre les chevaus ademis; Merveillos cous se donent es escuz de Paris, Qe li fer passent outre et li fust sunt malmis. Lo cheval Pinabel le jor en ot lo pis: Li cous li pezoia, desoz lui fu ocis; Sor l’auferant d’Espegne s’en passe outre Terris.

121.a

 7905

Pinabel s’en va outre, si a perdu Morel; Quant il lo vit morir, ne li fu mie bel.

7899. According to Godefroy, s’entrelaissier means ‘to leave one another’, ‘to separate’, but it is clear from the context that the two knights on horseback gallop towards one another. This is confirmed by P 6593: lors laissent corre (repeating P 6576) and T 5410: s’entrelessent courre (repeating T 5394), the earlier lines cited matching C 7881 (s’entrelaissent cosre). V7 8075 has s’entrelaissent, omitting cosre in both instances, but it is surely essential, as in P and T. In C 7899, it is the knights who ‘give free rein to the horses’ to gallop towards one another, since the account of blows follows immediately. 7903. ‘The blow cut it to pieces’, a hyperbolic expression for ‘cut it down’. 7904. C ms.: s’en passe outre t7re = terre; but Terris, which was written t7ris in 7893, is required by both sense and rhyme. V4 5785 reads: Tieris pas oltre, P 6601: oultre s’en va Thierris and T 5416: s’en passe outre Tierris, confirming the reading. Duggan inserts C 7904 as V7 8098, which V7 lacks. 7905–7920. At the end of the previous laisse, the two knights have charged forward and struck at each other, Thierry’s blow killing Pinabel’s horse, Morel. The force of the charge carried both forward, but Pinabel landed on his feet as his horse collapsed, drawing his sword and holding his shield sideways. Thierry turned his horse and came back to strike Pinabel

464

7910

7915

7920

La Chanson de Roland (Critical Text and Notes)

Il a traite l’espee don trenche li coutel, Per vertut s’est tornez, tint l’escu en cantel. Terris lo trestorna per desus un moncel, Per mot grant maltalent va ferir Pinabel. {Lo cheval aconsut, qi fu fort et isnel.} Tot envers l’abati devant lui el praël. Mais quant il fu cheüz, mot releva isnel, Per vertu se trestorne, se revint au doncel, Si volst ferir Terri lo gentil damisel. Lo destrier aconsiut per desoz le penel; De sa bone armeüre sunt fausé li clavel, Qe la jambe li tranche jusc’a l’os del noël; Lors trebuche Feranz desoz un arbresel Et Terris fu a pié, tint l’escu en cantel.



7925

Or est des deus chevaus la bataille finee. Terri salli en piez desor l’erbe en la pree, Si ot la bone targe devant son vis tornee; Ja sera la batalle des deus barons privee. Pinabel de Sorence s’espee a recovree; Redrizez fu en piez sor l’erbe ensanglentee; Mais fortment ot la car blesmie et desolee, Et la vertu del cors li fu mot tost alee; Por le faus sairement, la veüe a troblee.

121.b

 7930

Cel jor fist mot caut et li vent fu serin; Ferement se requirent ambedoi li mesquin.

(now on foot) and knocked him down. But Pinabel leapt up again and attacked, from his lower vantage-point being able to strike Thierry’s horse under the saddle, cutting through the armour and cutting deep into its leg. The horse fell and Thierry too was on foot. This sequence is confirmed by V4 5786–5791, P 6603–6615 and T and L. In C, the only point of confusion is the insertion of C 7911 into the sequence; it begins in much the same way as C 7916 and may be corrupted by it. Because the pronoun object in l’abati (C 7912) and the subject, il, in C 7913– 7914 refer to Pinabel (C 7910), not to his horse, C 7911 has been placed within brackets to signify that it is out of position. The copyist’s error may have arisen from his assuming that this laisse repeated the events of the previous laisse. 7930. The first hemistich has 5 syllables, hypometric for an alexandrine. V7 8124 has a 4syllable paraphrase: Bels fu li jor, reading this as a decasyllabic line (since V7 always attempts

La Chanson de Roland (Critical Text and Notes)

7935

7940

7945

465

Les bons chevaus de pris orent ja trait a fin, Et anbedeus les glaives qi furent de sapin. For fu li chaples des bons brans acerin. Mot fu greim Pinabel desoz son eume enclin, E tint traite l’espee dont li pons fu d’or fin; Puis va ferir Terri en l’eume poitevin, Mais nel puet enpirer vaillant un romesin: Li brans cola aval sor l’osberc doplentin, Por force le fausa et trenche lo samin; Lo sanc li trast dou cors prés de dimi bacin. «Terri,» dist Pinabel, «deça vus sui voisin.» «Je m’en sent per mon chief,» ce dist li Angevin, «Car li vileins lo troeve e dist en son latin, ‹Per mauvais voisinage a proudon mal maitin›. Meus volrai enrajer qe ne vus traie a fin.»

to reduce alexandrines). T 5440 reads: Li jour fist moult grant chaust. P 6625: Icel jor fist moult chaut offers a simple emendation, but the capital C in the ms. is confirmed by a c in the margin. Duggan inserts grant into C’s line, in line 8124 of his Critical Text. 7933. The C ms. reading, glavies, makes the hemistich hypermetric, but this is once again a confusion of -iu- and -ui-. Cf. saives (C 6787), Blaives (C 6826 and 6835) and jut (C 4018 and 7083). V7 8127 reads glaives, which has been adopted. Godefroy mentions a transitional form, gladie, closer to the Latin gladium. See also Introduction, 71–72. 7934. C suddenly reverts to 4:6 metre in this line, for no apparent reason (matching V7 8128); in contrast, T 5444 reads: Moult par fu fier le chaple des bons brans acerin, an alexandrine. 7938. The C ms. reads: rovesin, which in the context must mean ‘something of little value; but the dictionary meaning of rovaison is ‘Rogation-tide prayer’ or ‘prayer (in general)’. Both Godefroy and Tobler-Lammasch list biauvaisine (sf) in an identical context as ‘a coin of low value coming from Beauvais’ and, more significantly, romesin or romoisin as a kind of small Roman coin of very low value. The inhabitants of Reims (Remois) may also have had a remoisin. It appears that there has been either scribal error in C, with m mis-read as u, or that the copyist has substituted a familiar word, rovaison, for one he did not recognize. V7 8132 reads: revoisin. 7940. C: Por foce; but P 6634 and T 5451 read: Par force. 7944. Most unusually, e meaning ‘and’ is written as a single letter within the line. 7945. C: Per mauvaisse losenge a proudome mal maitin, and the proverb is found in the same form in V7 8139: Per malveis losenze a prodom mal matin, P 6639: Par felon losengier a on bien mal matin and L 2814: Per mavais losengier a prodonz mal matin; it is slightly modified in T 5456: Par mauvés losenger a perdu maint matin. One would presume therefore that it was familiar to the copyists; but it is not found in Morawski’s ‘Proverbes français antérieurs au quinzième siècle’ or Tobler’s ‘Li proverbe au vilain’. There is thus a strong possibility of scribal error, which is reinforced by the fact that the proverb does not fit the context. Here, Thierry is so dazed by Pinabel’s blow to his helmet (C 7937) and by the loss of blood as Pinabel’s sword pierces his hauberk (C 7939–7941) that he is not aware where Pinabel is. When the

466

La Chanson de Roland (Critical Text and Notes)



7950

7955

7960

Mot fu dolanz Terri quant senti soi navré, Car li sans et le chals l’a durement grevé. Per mot grant vasalage a suen cors recovré; Ne fu mie esperdu, ainz a mot bien pensé. Pouceles lo regretent des murs de la cité, Et Terris les ot bien, sa force en a doplé; Il tresal en travers, mot l’en ont tost loé. Va ferir Pinabel desor l’eume gemé. Mot adricéement li a son cop geté, Entre col et l’escu a son cop avalé; Qatre cent mailles fause del bon oberc safré E lo pug et la guinche a enscanble copé; Tot li cope lo braz, or l’a bien engané. Li escuz chiet a terre, qi forz fu et boclé. Les dames qi lo voient en ont Deu aöré.

122.a

latter warns him that he is close by (C 7942), gloating over his imminent victory, Thierry replies (C 7943) that he knows that from the state of his head – and then quotes the proverb, which must surely relate to proximity, as a riposte to Pinabel’s deça vus sui voisin, ‘that even a good man may have a bad time through proximity to evil’. This is borne out by the Roman de Renart br.XVIII.2: Si est escrit en parchemin/Que cil a sovent mau matin/Qui pres de lui a mau voisin; it is also found in Le manuscrit 19152…de la Bibliothèque Nationale (ed. E Faral) folio 73 verso, column 3, De proverbes et du vilain, lines 24–25: Qi a mal voisin si a mal matin. This indicates that either (1) mal voisinage has been transcribed as malvoise loãge/malvaise loange (with the first i read as e and the second i read as l, and either n mis-read as o and a read as error for ã, or na read as metathesis of an, with o inserted to provide a vowel) or (2) mauvais voisinage has been mis-read, either (a) with vois- being deleted as a duplication of -vais and subsequent modification as indicated in (1), or (b) with mauvais unchanged, voisinage becoming losange through metathesis of o and l (a mis-reading of i) and of n and a as above (and, of course, with e added to show feminine agreement and to correct the metre). C has been emended on this basis, on contextual as well as literary grounds. [The late Wolfgang van Emden (2005, V/450) came to the same conclusions, citing Ebert, 1881, 9 and Hassell, 1982, 252.] The common error throws further light on the relationship between the rhymed versions; V4’s equivalent laisse, 394, has only 6 lines and does not mention any proverb. 7958. The C ms. graphy pug again raises the question of whether, in C, g in this position stands for gn or ng. V7 8152 and P 6648 read: poing. Cf. C 715: poig, 7966: puig, 7984: poig and also soig, C 8009. It may be significant that the scribe here writes escanble for enscanble: perhaps, as the end of his travail approached, he was increasingly careless over the insertion of n.

La Chanson de Roland (Critical Text and Notes)

467

«A! Deus,» dist Karllemene, «per la toe bonté, Garissez hui Terri, q’il ne soit afolé!»

 7965

7970

«Terri,» dist Pinabel, «bien trenche vostre acer; Diable l’esmolurent qi lo firent forger, Quant del puig ne del braz ne me puis plus aider. Se ma vertuz ne faut, vos le conparez chier!» Fierement lo requert, ben se qide vengier, Et va ferir Terri sor son eume d’acer: Tant fu dur et serré, ne li pot enpirer; De l’escu de son col li osta un quarter. Li cols fu mot pesanz et li vasal fu fier; Per un poi ne l’a fait a terre trebucher.

 7975

7980

Li fius Jofroi d’Anjou recovra sa vertu Et de mot bon corage a reclamé Jesu. Vait ferir Pinabel desor son eume agu, Qe lo nasel li trenche o l’escarboncles fu; Et de son fort osberc, qi fu maillez menu, Lo nas et lo menton li a tot porfendu. Cil li qida guenchir, si chiet tot estendu. «A! Deus,» dist Karllemene, «fai miracle et vertu!» Li vaslez li cort sore quant il l’ot abatu.

7962. .K. is transcribed as Karllemene so as to maintain the alexandrine; cf. notes to C 7854, 7981, 8000, 8019 and 8035, and especially that to C 6756. A case can be made for adopting the 10-syllable metre here, because this line interrupts the flow of the narration, but the pattern of interruptions by the king where no abbreviation is involved is consistently dodecasyllabic (cf. C 8046, 8055, 8079, 8081, 8092). Both V7 8156 and T 5466 have standard decasyllabic metre here, but P 6650 is an alexandrine, beginning: Ha! Deus!, dist Karlemaines. 7966. C: puig, for puign. V7 8160 reads: poinz, P 6654: poing. Cf. note to C 7958 above. 7968. C: requeit, scribal error for requert (V7 8162) or requiert (P 6659), pres 3, requerre. 7981. As in C 7854 and 7962 (see notes), .K. is transcribed as Karllemene.

468

La Chanson de Roland (Critical Text and Notes)



7985

7990

Pinabaus trebucha desor l’erbe en la pree Et fors de son poig destre li eschapa l’espee: Si dut il bien faire, falsé li ot l’entree. Li vaslet li cort sore, o la chiere menbree, Mot per lo feri bien sor la broigne safree. Lors vit de la potrine un petit desarmee: Corteïn i apuie, per vertu l’a botee; Jusc’a l’enhaudeüre li a el cors colee. Li cuverz ne dit mot, l’arme s’en est alee. Et Terris se dreza, a lui sacha s’espee, Et les gardes i corrent, la bataille est finee. Pinabel ont saissi qi gist goule baee.

122.b

 7995

Une hart li fait entor le col fermer, La coë del roncin estaqier et noer; Deci en un alt tertre l’en ont fait traïner. Son osberc en son dos ne volent dessachier, Ensi l’ont fait as forches contremont sus lever.

 8000

«Seignor,» dist Karllemene, «baron de grant valance, Li Rois qi nos governe est mot de grant puissance,

7983. C: Pinabaus trebuche, a 5-syllable hemistich. V7 8177 reads: Pinabel chiet and P 6673: Li gloz trebuche, making regular decasyllabic metre; but V4 5848 has: Pinabel trabuchà, T 5491: Pinabel trebucha and L 2836: Pynabel trabucha – indicating the source of C’s error in a matching alexandrine. Note also the past definite 3, eschapa, C 7984. 7984. Again the graphy poig for poign or poing; V7 8178 reads: poinz, P 5574: poing. Cf. C 7958. 7990. C: enhaudeüre, with h inserted in superscript. After cors, a word beginning c… (crevé?) is barred; the script is so worn on this page that this word is illegible. 7995. Provided the h of hart does not prevent elision, this is a regular decasyllabic line amongst the alexandrines; V7 8189 reads: Une harc le firent (either 4 or 5 syllables), which Duggan emends to Une hart li ont fait; but P 6681 has: Un cein li font. 7997. The metre requires traïner to be read as tri-syllabic; this is confirmed by V7 8191: trahiner. 8000. .K. is again transcribed as Karllemene, as in the earlier examples found in C 7854, 7962, 7981.

469

La Chanson de Roland (Critical Text and Notes)

8005

8010

Qi fist et ciel et terre et la mer a fiance. Il seit bien del felon abatre la bobance. Or est morz Pinabel per sa desmesurance, Des q’a traï fel Guenes le bernage de France. N’en poi je mais avor nulle reconuissance Del cuvert traïtor qi m’a fait tel pesance?» Hervier et ses lignages en orent esmaiance: En fuie sunt torné, n’ont soig de remanance; Tes hontes lor avint devant le roi de France, Tant com dura le siegles, en sera reparlance.



8015

8020

8025

Li jor vait a declin et la noit vint serie; Mot per fu grant la joie et la vile esbaudie, Et de dux et de contes et de chivalarie. Karllemene est asis, a la barbe florie; Devant lui fu Terris a la chiere hardie, Mais forment ot la char et la color blesmie. Li rois lo reconforte et vers lui s’umelie. «Baron,» dist Karllemene, «gentius gent segnorie, Demain arez congié, el non Sainte Marie. Si revesra sa feme cui Deus l’ara garie. Teus dames i ara lor segnors n’aront mie, Car Guenes les vendi la puite gent haïe.» Li rois en ot pité, a la barbe blanchie; Un petit sospira, bien va qe ne s’escrie. De Rollant li remenbre a la chiere hardie, D’Audein et d’Oliver, c’ainc ne firent bosdie; A Yhesu les comande, lo fil Sainte Marie.

123.a

 8030

La nuit se jut li rois en son palais alcior, En une chanbre peinte de diverse color;

8009. C: soig; V7 8203: soing. Cf. pug (C 7958), puig (7966), poig (7984) and note to C 7958). 8019. After the graphy .K.mene in C 8015, only 4 lines above, it seems politic to transcribe .K. as Karllemene so as to obtain a 6-syllable first hemistich in a laisse composed of alexandrines. Cf. also notes to C 7854, 7962, 7981 and 8000. V7 8213 reads: .K., which Duggan transcribes as Karlemeine.

470

8035

La Chanson de Roland (Critical Text and Notes)

La vont chocher Terri por bien et por amor. Au maitin per som l’aube, quant aparçut li jor, S’est levez l’enperere qi mot ot grant valor: Venu i sunt li prince, li duc et li contor. Karllemene en apele dan Gerart de Monflor, Le conte de la Marche, Ugon de Val Color: «Baron, amenez moi mon felon boisëor, Qi m’a tolu de France lo bernage et la flor.»

 8040

8045

8050

Li baron s’entornerent, li rois l’ot comandé; Si ont fors de la tor Guenelon amené. Grant ot la forcheüre et le cors ben formé, Gros fu per les espaules, grasles per le baldré; Mult i avoit bel home, si l’ont mot esgardé; Mar i fu sa faiture quant il ne vint a gré! Quant lo vit l’enperere, si a des els ploré: «Fel Guene,» dist li rois, «mot per m’as adolé! Ma seror te donai per bone volunté, Et tu m’en as rendu mot mauvaisse bonté.» «Sire,» ce dist li traïte, «malement ai ovré: Per moi fu mort Rollant, ne puet estre celé. Baron, se je:l vos di, n’en doi estre blasmé: Sor moi mist le mesaje, estre ma volunté;

8032. C reads: aparçut, past definite 3, aparceivre, obviously a scribal error for aparut, past definite 3, aparoir in the standard phrase quant aparut li jor. V7 8226 reads: aparut and P 6712: apparut. 8035. As in C 7854, 7981, 8000 and 8019, .K. has been transcribed as Karllemene so as to keep to the alexandrine pattern of this passage. 8042. A standard epic formula: cf. Studer & Waters ‘Historical French Reader’, No.24 Lyrics in the Popular Style, (a) Chanson de Toile (twelfth century), l. 26: Gros par espaules, greles par lo baudré. 8045. C: eperere, with n or its symbol omitted. 8049. traïte may have the second r (or nota tironia 7) omitted in error, but cf. C 568, 6357, 6377, etc. See also Introduction, 92–93 regarding the syllabic value of aï diphthong: here disyllabic aï would produce a 7-syllable hemistich. V7 normally has standard decasyllabic metre, but V7 8243 here reads: Sire, ce dist Guenes (5 syllables), corrected by Duggan to Sire, dist li traïtre. V4 5897 has: Sire, ço dist Gaino, a 6-syllable first hemistich; P 6724 and T 5581 have normal 4-syllable hemistichs.

La Chanson de Roland (Critical Text and Notes)

471

A Marsille en alai ad enviz o de gré. Mais Rollant volsist ben q’il m’eüst demenbré.»

 8055

8060

8065

«Baron,» dist l’enperere, «tolez lo devant moi. Je ne puis escouter la merveille qe j’oi; De vergogne tressue quant de mes els le voi, Car onqes vers seignor n’envers Deu non ot foi. Il traï mes barons, onqes ne soi por coi; Or et argent en prist, si fist mot grant besloi; Altresi fist Judas, ce sa ge bien et croi, Son conpeignon vendi as Judeus de la loi; Il reconut son tort, si se pendi per soi. Ahi, beaus niés Rollant, qel damage ai de toi Et des altres barons qi erent de ta foi! Del servise Deu faire estoient en esfroi.» Lors dist a ses barons sa parole a desroi: «Des qel mort doit morir, baron, jugez le moi!»

123.b

8059. The C ms. reads sot, as does V7 8253, supported by P 6733 (seit), T 5590 (scet) and L 2869 (sot), but the 3rd person makes little sense: Ganelon certainly knew what he was doing and why. V4 5905 reads the first person: je o n’en soi par choi, which must be the original reading, the i being read as t through scribal error: it is Charlemagne who never knew why. The error implies a common source for all the rhymed versions except V4 in this section. 8061. C: sa ge, with sa = sai, present 1, savoir – due either to verb-subject inversion (cf. note to C 3316) or to haplography based on a source reading sai ie. Re Judas, see Matthew xxvi.15– 16 and 47–49, Mark xiv.10–11 and 43–45, Luke xxii.3–5 and 47–48, John xviii.2–5. 8062. The term Judex/Judeus de la loi refers to the Sanhedrin, the Supreme Council and Court of Justice in ancient Jerusalem, which was to judge Christ; as Judeus is oblique (= indirect object) plural. Cf. St.Mark’s gospel xv.1, confirmed by St Luke xxii.66. Judex is thus used as a precise Latin term for ‘judge’; the ‘law’ referred to is the Mosaic Law governing the Jewish religion. V7 8256 reads: as viels de la loi, but Duggan substitutes Judeus as in C. In view of the (possible, but rejected) ms. reading for C 2842 (q.v.), it is interesting to note that here Judas is cited as the betrayer of Christ, with Judeus in the very next line; there is no scribal confusion between the two.

472

La Chanson de Roland (Critical Text and Notes)

 8070

8075 8075a (V7)

8080

«Seignor,» ce dist le rois, «por Dé vos vel proier De la plus aspre mort qe vos sarez juger Le me faites morir, ge ne vel plus targer.» Aprés lo roi parla danz Girarz au vis fier, Li sire de Viene qi fu oncle Oliver: «Per ma foi, enperere, ben vos sai conseillier: Mot sunt longes vos terres et larges por chercher: Faites le bien en chaëne serrer Et puis mener a pié, com ors o liëmer, Et batre de coriges por son cors angoiser! Et quant venra lo soir, q’il devra herberger, Uns des menbres li faites a l’ostel ostager! Per un e un li faites chascune noit trencher; Por l’escot q’il dovra, se li faites laisser.»

8070. C: De la plus aspreme mort, which is hypermetric. The scribe has mis-read aspre mort, with anticipation and quasi-repetition of mo. Although aspreme is a graphy of the superlative, asprisme, the presence of le plus would make this hyperbolic. V7 8264 reads: aspre mort. 8075. The verb chercher here has the meaning ‘travel through in all directions’, stressing the extensiveness of Charlemagne’s domaines. V7 8269 reads: checher and P 6743: chascer. The lack of a finite verb on which the infinitives in C 8076 and 8077 depend shows that C has omitted a line here. The other rhymed versions accord with V7 in inserting the extra line and on its general sense: eg. P 6744: En .ii. grans cordes le faitez bien lier and V4 5917: En dous perches lo faites liger et atacher. 8076. C: com un ors liemer. Liemier/lienier was a large hound used for hunting deer and boar. This meaning is supported by V7 8271: com un livrer mener, ‘lead him like a dog’ – but it also suggests faulty copying of …li mener as …liemer. The reading ‘bear’ is confirmed by the other versions: V4 5918, cum un ors lanier and P 6745, com ors laniers (‘like a bear on a leash’?) or perhaps l’anier(s), as in T 5601, com hours li anuier. L 2880 has: com un hors leomier. The variations imply a difficult source reading. Since both bear and hound are appropriate here as being ‘led on foot’, ‘walked on a lead’ and in the case of a bear ‘teased’ or ‘tormented’, there is no clear preference between the various alternatives. As the general import is not in doubt , C’s graphy liemer has been retained . 8077. C reads: coriges, in this line and coreges in C 8142; and V7 8272 has cories here, coreges in 8337; these are valid alternative graphies of the more usual corgie/corgee. 8079. Beretta cites ostager (as also found in V7 8274), ostagier (L 2883), ostiger (V4 5921) as meaning ‘to pay for (a night’s) lodging’. This is confirmed by C 8080–8081; cf. note to C 8081 below. 8080. This was first written: Per un li faites, then e un was added at the end of the first hemistich, with a sign for its insertion before li. 8081. Girart de Viene proposes that Ganelon should have his arms and legs chopped off one by one, at each place where he stays, ‘to pay his reckoning’ or ‘to pay what he will owe (for board and lodging)’.

473

La Chanson de Roland (Critical Text and Notes)

8085

«Baron,» dist l’enperere, «ci a jugiment fier! Mais je ne le vel pas si longes respiter. Tant com il vivra plus, ne qit jamais menger, Car trestoz mes amis vel de son cors vengier.»



8090

«Par ma foi, enperere,» dist Bone al cor vaillant, «Je vos ensegnerai un juïse pesant: Comandez a vo homes tost et isnelamant De verz albesespines a faire un feu ardant; Puis i faites jeter le gloton sosduiant, Et environ lui soient trestot vostre serjant!» «Per ma foi,» dist li rois, «ci a mot fort tormant! L’arme en partira per merveillos sanblant! Icestui prendron nos, se ne trovons plus grant.»

124.a

 8095

8100

«Par ma foi, enperere,» dist li cons Salemons, «Uns plus aspre juïse per tens vos eslirons: Or faites jejuner tres ors et deus lions; Al terç jor si lor soit delivrez le glotons. Toz nuz soit despoillez ses cors et sa fazons. Lors ert fait de son cors si grant destrucions, Devorer lo vesroiz per mil devisions; Ne remanra enters cuirs ne os ne brohons;

8083. The C ms. reads: respirer; V7 8278 has: respleter, but P 6751 reads: respitier, which makes much better sense. In C, respitier is found in C 6114, respiter in C 7455, but the same scribal error (r for t) occurs in C 8105: respirons for respitons. This may well be Italian influence: respiro (sm) meaning both ‘breathe’ and ‘pause’, ‘delay’ or ‘respite’. 8088. isnelamant is a Franco-Italian form, with a for e; V7 8283 has: isnellemant. 8091. After this line, V7 adds: Qui le gard bien qu’il n’en soit fors issant – but Duggan excises this. Although it is not found in any other version, V7’s line makes far clearer the purpose served by the sergeants. 8097. C reads: jejuner or jeiuner instead of the normal jeüner, perhaps to stress that it is trisyllabic. See also note to C 4334. 8102. C: entres for enters, adjective, ‘whole’, ‘entire’. This may be another example of Italian influence, but it could well be a simple scribal error (metathesis of e and r) and has been emended as such. Cf. also C 5519, where the ms. reads: ent7re for entire (entrire?).

474

8105

La Chanson de Roland (Critical Text and Notes)

Car ensi doit on faire de tes traïtor felons.» «Seignor,» ce dist li rois, «ce me sanble raisons, Mais je n’ai pas corage qe plus lo respitons.»



8110

8115

8120

Aprés parla Ogier, li bers al cor vasal: «Fere destruccion vus sai del desloial! Sire, faites lo metre en cele tor aval; Ne jor ne nuit n’i ait solauz ne nul ostal, Forz qe sol la vermine qi istra del terral; Tot environ son cors ben ert parti igal. Ja n’i menjust de pein per nul home carnal. Quant venra al terz jor, asez i avra mal; Toz i morra de feim, sofrira grant traval. Lors le feroiz fors traire el palais principal, Se li aparolt on lo menger comunal; Si soit ben conréez de poivre et de sal, Ne ja ne vin ne aive n’i ait qe plein meral; Donqes mosra de soi et d’angoisse mortal, Ensi com fist Rollant lo quns en Renceval.» «A! Deus,» ce dist li rois, «qel esgart de vasal! Mais je nel vel oimais qe je li prest ostal.»

 «Droiz enperere,» dist Neme li Baiver, «Un fier joïse vos sai a enseigner! 8103. C reads: des tes traïtor felons, which is hypermetric (7 syllables) when the -ai- diphthong is read as disyllabic (cf. Introduction, 92–93, and notes to C 1319 and 7134). There are two possible emendations: (1) to reduce des tes to des or de, as in V7 8298: de trahitor felons and P 6767: de traïtor felon, i.e. reading des tes as a accidental repetition, a quasi-duplication, or (2) to retain de tes traitor…, ‘with such traitors…’ with monosyllabic -ai- in traitor, because tes gives greater stress to the phrase. 8105. C: respirons, another example of the verb respirer being used in error for respiter (cf. note to 8083 above). V7 8300 reads: respitons, V4 5943: respiton and L 2890: repitons; P 6769 has: que plus respit li dons. 8116. C reads: aparolt, a graphy of aparelt, present subjunctive 3, apareillier, ‘let people prepare’. This is an example of Franco-Italian confusion of o and e. 8118. It is suggested that Ganelon be allowed only a plein meral of wine or water, i.e. the single token’s worth that a monk would obtain when he exchanged the jetons received for

475

La Chanson de Roland (Critical Text and Notes)

8125

8130

8135

Faites lo, sire, trestot vif eschorcer Et lo cuir en alt pendre, lo cors asoleller! De mel lo faites oindre et devant et derier; Aprés lo faites, sire, a set chevres lechier! Lors lo vesrez destroit et forment fresceller Et por la grant angoisse toz les denz reqigner! Ja de plus aspre mort ne:l poez justisier.» «Per foi,» dist Karllemene «bien fait a otrier! Icestui ferons nos, se no trovons plus fers!» «Droiz enperere,» dist li cons Otoier, «Ja n’est ce mie corseinz a deviser. Se faites ce qe vos sai ensegner, Nus hon de char ne poet meus deviser: Icil traitor me faites amener;

124.b

attendance at divine service for his share of the basic food and drink: a very small measure of drink (cf. note to C 5903). 8130. The usual graphy is: les denz rechigner (‘to clench one’s teeth’); V7 8325 reads: rescrigner. 8132. Once again, .K. has been read as Karllemene to give a 6-syllable first hemistich, because this suits the rhythm of the line and conforms to the general pattern. Although laisse 448 opens with 3 decasyllabic lines (C 8123–8125), lines 8126–8131 and line 8133 are all alexandrines, so it makes sense for C 8132 to be one as well; the remainder of the laisse is then in decasyllabic metre, i.e. from the point at which Otoier begins to speak, proposing the punishment which will be adopted. The ‘fantasy element’ of the suggestions rejected by Charlemagne is reflected in the alexandrines; with Otoier, there is a return to realism – and to the decasyllabic metre. This laisse is unusual in that it contains proposals from two different people, Nayme and Otoier; and the first 3 lines, each of 10 syllables, may reflect the introductory lines of a separate, possibly assonanced, laisse in an earlier version. 8135. C: Ja ne ce…, for Ja n’e ce…, in which the verb e may be either the Franco-Italian form è or represent an abridged form of est, found in C as e with horizontal line above – but here with the line omitted in error. Corseinz (for cors seinz?) probably here means ‘the Host’, the bread broken and distributed at Mass as the Body of Christ, but it could also simply mean ‘holy relics’, i.e. a reference to the mediæval practice of dividing up a saint’s body into many relics, shared out between different churches as a source of status and revenue. Fair distribution of relics could pose problems when dealing with a large number of conflicting claims; but the method of Ganelon’s execution for treachery is easy to decide. The poet apparently makes a play on two meanings of deviser: ‘to share out’ or ‘divide up’ (C 8135) and ‘to devise’ (C 8137). Duggan in the note to V7 8330 reads this as ‘This is not a sacred relic over which to meditate’. 8138. C’s Icil traïtor poses problems over pronunciation and metre, as in C 6364. Since the scribe wrote icil when cil was equally appropriate, presumably he read traïtor as disyllabic: traitor. Cf. Introduction, 92–93 regarding the fluctuating value of the aï diphthong in C. V7 8333 reads: Guenes li fel before a second hemistich identical to that in C 8138.

476

8140

8145

La Chanson de Roland (Critical Text and Notes)

Qatre chivaus m’i faites anseler, Les plus coranz qe l’ons posra trover, Et per desus un pautoner monter; Portent coreges por lor meus esfréer; Les poinz li faites et lier et serer, Et puis as coes des chevaus bien noer; Lors vesrez vos son cors tot dessirer!» «Certes,» dist Karlle, «ne:l doit on refuser! Ja ert ciz faiz conpliz sanz demorer!»



8150

8155

«Seignor,» dist Karles, «franc chevalier loial, Li jugement soient tot perigal; Cestui prendrons, car je n’i voi plus mal. Or i pasra qi a tirant chival! Alez monter, mi duc et mi vasal! Issons nos en, la fors, en cel igal! Se li ferons son juïse mortal: De son servise recevra batistal!» Guenelon prenent prevost et seneschal; Fors de la vile lo menent comunal.



8160

8165

Li baron montent, si ont le cri levé; Karles meïsme sor un mul est monté Et li borzois qi mot l’ont desiré. Chascuns fait joie, les dames a lor tré, De la venjance lor seignor naturé. Guenelon ont de la vile jeté; Fors de la presse en sunt aval alé. Iloqes sunt li cheval apresté Per cel marché qe Guene ot achevé.

125.a

8142. For comments on the graphy coreges, see note to C 8077. 8146. C: .Kll’e., therefore transcribed as Karlle, in contrast to .K. > Karles, C 8148. 8159. C: sor un mulet monte, but this must be the passé composé, est monté, for the rhyme. V7 8354 reads: sor un mul est monté. .K. here and in C 8148 is transcribed as Karles. 8163. C: Guenelont ont, a very obvious scribal error.

La Chanson de Roland (Critical Text and Notes)

8170

8175

8180

477

Sor chascun ot un paotoner monté; Lo traitor ont as coës bien noé, Mot durement et lié et serré. Li pautoner furent tot enragé; Chascuns fiert bien son cheval abrivé: Trestot ont Guene destruit et devoré, En poi de terme l’ont tot decipliné. A la persone li ont toz escrié: «Cuvert traïtre, or avez conparé La traïson qe vos avez mené, La malefoi qe vos avez porté As doce pers, qi mort furent geté! Et por le mal qe vus avez pensé Sera plus vil tot vostre parenté!» Teus n’en sot mot qi en a plus ploré.



8185

«Baron,» dist Karles, «or ai qenqe je vel, Quant cil est morz qi m’a tolu l’orguel, Rollant le conte, per che repousser suel. Les doze pers a mis en mal esvel:

8168. The cæsura is marked after ont, so th pronunciation here is traitor, not traïtor. For the value given to the ai diphthong, see Introduction, 92–93. 8175. Cuvert traitor in C, with the usual problem over syllables, as seen in the note immediately above. Despite the graphy, traitor, this epithet, addressed to Ganelon, must be subject form singular. V7 8372 reads Culverz trahitor, very obviously a 5-syllable hemistich; Duggan emends this to Culverz traïtre. The line is not found in other versions. This is a standard formula and emendation to Cuvert traïtre solves the problem (just as reading Culverz trahitre in V7 corrects the metre). The vocative, traitor, is also found in C 5549 (poetic licence or a graphy of traiteor, ‘leader’?), but may well be due to Italian influence (traditore). It may reflect the Italian scribe’s uncertainty whether to use the subject form for the addressee or the oblique/object form because this is not the subject. 8177. C: qe vos a vos avez comparé, with scribal error in inserting an additional vos a before vos avez. 8179. The word avez is added at the end of the line, with the place for its insertion marked. 8181. ‘Anyone who wept over him any further did not know anything about him’ or ‘… knew nothing of what he had done’. V7 8378 reads: Tels n’en soit rien que puis en a ploré, with soit a scribal error for sot or sout. 8184. ‘Roland the count, by means of whom I was wont to have repose’, i.e. … ‘thanks to whom I was wont to rest with a quiet mind’; suel is strictly present 1, soleir, but is here used as an historic present.

478

La Chanson de Roland (Critical Text and Notes)

Tant com je vive, nel vesrai mas itel! Per als conquis je Rome, Valence et Morel, Palernes lonz, jusq’el val de Siduel.»

 8190

8195

8200

Quant Karles fu en la sale montez, Aprés ice qe il fu retornez De Guenellon qi si fu tormentez, Toz ses barons a devant lui mandez, Mot belement les a araisonez Et doucement les a toz acolez. Congié demandent, et il lor a donez. Li rois sospire, de Rollant s’est menbrez. Et cil s’en tornent, s’avalent les degrez. Et Charll’ remest dolanz et abosmez. De cest romein nus n’en seit plus chantez. Cil vus beneïe q’en la crois fu penez Et au terz jor de mort resuscitez! DEO GRACIAS AMEN

125.b

EXPLICIT RONCISVALI E DE ROLLANT E D’OLIVER E DE AUDE.

8188. Both V4 and P end with this line. 8198. The MS reading is Charll’, to indicate this is a monosyllable. After this line, V7 8397 adds: Chascuns s’en est en son reng entrez – then nothing more. Lines 8199 to the end are unique to C.

III Glossary This is a selective glossary. Words readily recognizable from their Modern French equivalents are omitted, unless there is a problem over ambiguity (e.g. in the case of cors and dis). An infinitive followed by a line reference indicates that the verb is found in this form in the text; when the infinitive is enclosed in [ ], it is an indication that only finite forms of the verb are found in the text. Specific finite forms are listed only where they are irregular, potentially ambiguous in meaning or used in a special context. The persons of the verb are indicated using numbers 1 to 6. Noun and adjective forms are usually listed in the oblique singular, but non-standard subject and feminine forms are identified. Standard abbreviations are used: sm.: substantive masculine; sf.: substantive feminine; subj.: subject; obj.: object; dir.obj.: direct object; ind.obj.: indirect object; sing.: singular; plur.: plural; pron.: pronoun; adj.: adjective; adv.: adverb; prep.: preposition; conj.: conjunction; v.trans.: transitive verb; v.intrans.: intransitive verb; v.impers.: impersonal verb; v.reflex.: reflexive verb (including all pronominal verbs); pres.: present indicative; imperat.: imperative; pres.subj.: present subjunctive; fut.: future; cond.: conditional; imperf.: imperfect indicative; imperf.subj.: imperfect subjunctive; past def.: past definite (perfect, preterite); p.part.: past participle; pres.part.: present participle. Where the text has been emended by either excision or insertion (indicated by italics in the text), the glossary entry shows only the emended version. Where the copyist has exercised poetic licence over the word-ending to meet the requirements of rhyme, the resultant spelling is listed with the annotation (poet.lic.) (cf. I.4 Editorial Policy, Introduction). Where an entry is followed by *, this indicates that further details are given in the notes.

a aage 1936; eage 7495 sm. life 1936; en eage advl.phr. for the duration of his (natural) life 7495 aaitie 239, aatie 758 sf. animosity; challenge 758; par aaitie 239 with unfriendly intent, in enmity; proz sui de l’aatie 758 I am ready to challenge him, I am prepared for the challenge (proz being consistently used by C as a graphy of prez *) aatir 6113 v.reflex. challenge, defy 6113, 6413 aatison sf. impetuosity 2130, 2549 abaïe sf. abbey 5874, 7354 abaisser 7097, 7788; abasser 7689 v.trans. bring low 611, 934, 7788; bring down, reduce 4190; insult, denigrate 7689; v.reflex. bow, make reverence 7097 [abandoner] v.reflex. offer battle 1306 abismes obj. plur. sm. abysses 7130 abosmé 6979; abosmez 1408, 6321, 8198 p.part.used as adj. confounded, taken aback 1408, 6321; overwhelmed with grief, cast down 6979, 8198 abrevez graphy of abrivé q.v. abrivé 8171, abrivez 222,1513; abrevez 3677 adj.(based on p.part) swift, impetuous 222, 1513, 8171; adj.used as adv. headlong, at a great pace 3677 aceré 1035, 7804; acerez 1143, 1521, 1785, 3668; p.part.used as adj. sharp, hard (as steel) acerin adj. of steel; hard, firm, unbreakable, dependable 3445, 7934

480

Glossary

acesmé, acesmez p.part.used as adj. equipped 1145; adorned, beautiful 2023, 6993; all dressed and ready 6561; polished, skilful 7808 acline adj.f. subject (to), in subjection (to) 4669 acliner 6167, 6183, 7066 v.trans. bow 2030, 3569, v.intrans. lean 4174, 6183; bow down 6167; v.reflex. bow, make obeisance, render hommage 1789, 6947, 7112, 7213; bend down 7066; bow down 7092 acoarder 1956 v.intrans. be cowardly acointer 938, 2006, 6585 v.trans. make the acquaintance of (person) 381; acquaint (with), make known, teach (facts) 2006; get to know (person), become acquainted with (one another), ?be betrothed to 6585; v.reflex. become acquainted with 938 acoisiez 309, acoisié 7682 past part.used as adj. silent, put to silence [acoler] v.trans. embrace 6501, 8194 [acolir] v.trans. choose, take up (path, journey, route) 531, 2471, 3064, 4449, 7353; gather, assemble 2840; attack 5656; pursue 7574 [aconpagner] 2617 v.refl. be companion to, link oneself in friendship to [aconsivre] past def.3 aconsut 7911, aconsiut 7916 v.trans. strike, wound acontez 1407 irreg.graphy of infin. aconter (poet.lic.) vb.trans. count, estimate the number of [acoster] v.refl. put oneself beside (someone) 1685, 1722, 1746, 1774, 1819 [acovercler] v.trans. cover 6688, 6703 [acoveter] v.trans. cover, envelope 3657 [acrester] v.refl. draw oneself up haughtily, proudly 1485 ademis adv. forthwith 7889, 7899 adenz adv. face-downwards, prone 1840, 3398, 4025 [adeser] v.trans. touch 3371 adober 2544, 5178 v.trans. equip as a knight 907; arm, put armour on 1138, 1678, 1705, 3179, 5178, 7801; ornament, deck out 2544; v.reflex. arm oneself, prepare for battle 1369, 1370, 1532, 1733, 3111, 3176, 4927, 4947, 5080, 6246 adobez p.part.used as adj. clad in full knightly armour, as a true knight, noble 1869 [adoler] v.trans. cause grief, distress 8046 adolé p.part.used as adj. distressed, grief-stricken, desolate 4046, 6140, 6328, 6682, 6698, 7231 adonc adv. then, next 882, 1658, 1914, 2906, 3465 [adrecer] v.trans. arrange, prepare, draw up 5477 adricéement adv. in a direct line, accurately 7955 aduré 969, 1135, 1629, +; adurez 1146; adurer (poet.lic.) 1667; p.part.used as adj. strong, tough 969, 1135, 1146, 1629; stalward 3467, 6995, 7003; hard, (of battle) fierce, stubborn 1649, 1667, 2317, 2359, 5395 aërison sm. cf. aïreson aëz 410, 448, 808, aé 7752, haé 961 sf. life 410, 448, 961; age 808, 7752 afaitier 52, [afaiter] v.trans. accommodate, satisfy 52; prepare, make ready 3474; arrange, set out 7123, 7861 afaitiez p.part.used as adj. educated, skilled, wise 2293 [afermer] v.trans. hold firm, fix 3407 aficer 1281; aficher 7687 v.trans. affirm, declare, assure 1281, 4614, 4892, 7687; hold firm 1787; v.intrans. be firmly resolved, be strengthened 4815; v.reflex. place oneself firmly 1992, 7784 [afier] v.trans. assure, promise, swear 899, 921, 7816

Glossary

481

[afiner], [afinir] v.trans. finish, bring to an end 3483, 4792, 5833, 7164 affliccion 1094, 1714; afliccion 5333 sf. humility afoler 21 v.trans. wound, kill, put to death 3643, 6865, 7963; harm, oppress 7457; infinitive used as sm. harm, damage 21 afoltré p.part.used as adj. harnessed, saddled 6989 [afubler] v.trans. dress, wrap 389; put on (of clothes) 711, 5011, 7244, 7439 agenoiller 4884; agenoller 7608 v.reflex. kneel 4884, 5962, 5995, 7222, 7608, 7858 agenollons adv. on (her) knees 4500 agollon sm. needle, point, goad 5754 [agreanter] graphy of acreanter v.trans. promise 965 agu adj. pointed 1811, 3667, 4010, 5761, 7976; sharp 2728, 3290, 4068, 6095; (of hill) peaked, pointed 4098 ahant obj.sing. 2410; ahanz obj.pl. 314, 3155 sm. trouble, torment, pain ahi 5065a, 6618 p.part. of ahir used as adj. hated, detested ahir 2583, aïr 2713, 3587; haïr 1853, 2143; sm. courage, strength, spirit; d’ahir with fury 5593 aidier 36; imperat.2 aïes 2799, aïuz 1421; pres.subj.3 ait 5359, 7851, aïut 1839, aïue 5848 v.trans. aid, help; infin.used as sm. help, assistance 7787 aidis adj. of assistance 1500, 3230, 3994, 5435 aiglenter sm. wild rose bush 1659, 3829, 6732, 6733 aigue 559, 1006, 1617, 6340; eigue 3892; aive 3870, 3871, 3895, 6270, 6287, 6295; eve 4272, 4289, 4300, 4303, 4315, 7142 sf. water 1006, 1617, 3892, 4303, 4315, 7142, 8118; stream 3870, 3871, 3895, 4586; river 559, 4272, 4289, 4300, 6270, 6287, 6295, 6340 ainc 171, 413, 715, 929, 949, 2069, 2830, 2832, 2956, 3082, 3408, 4042, 4218, 4336, 4339, 4473, 5238, 5508, 5773, 5807, 6218, 7251, 7407, 8027; ainz 4297 (cf. onques, unques) adv. ever 5756; ne… ainc never 171, 413, 715, 929, 949, 2069, 2830, 2832, 2956, 3082, 3408, 4042, 4218, 4297, 4336, 4339, 4473, 5238, 5508, 5773, 5807, 6218, 7251, 7407, 8027; ainc meis ne never before 2780; ainc ne… mais never before 5544; ainc puis ne never since 6649 aindemein adv. next day 6606 ainz 179, 587,1428, 1449, 1635, 2091, 2229, 3163, 3185, 3275, 3409, 3541, 3632, 3636, 4035, 4227, 4271, 4281, 4306, 4399, 4766, 4786, 4998, 5250, 5358, 5749, 5869, 6065, 6122, 6143, 6799, 6845, 6880, 6920, 7305, 7485, 7540, 7563, 7647, 7818; anz 424, 5280, 7280, 7306; prep. before 3632, 3636, 4227, 4271, 4766, 5250, 5869, 6143, 6799, 6920; ainz le solel cochanz before sunset 3632; ainz conplie sonanz before (the bell rings for) Compline (the final prayers of the day in church) 3636; ainz quatre mois before four months were out, within four months 4227; ainz qatre liues within four leagues 4271; adv. before, before this, beforehand 424, 4297, 5324, 6949, 7420, 7693; rather than this, sooner (… than) 1428, 1449, 1885, 1901, 2445, 3588, 3611, 4899, 4907, 5280, 5753, 6194, 7042, 7106, 7280, 7306, 7670, 7746, 7818, 7950; cum ainz il pot as best he could 6802; ainz qe, anz q’ conj. before 179, 587, 1635, 2091, 2229, 3163, 3185, 3275, 3409, 3541, 4035, 4281, 4306, 4399, 4786, 4998, 5358, 5749, 6065, 6122, 6845, 6880, 7305, 7485, 7540, 7563, 7647, 7818 aïr cf. ahir sm. aïrance sf. anger 5095 aire sf. lineage, origin; qualities, good or bad characteristics of race; de puit aire of vile birth 3551, 5744; de bon aire of noble birth, noble 3554 aïrement sm. impetuosity, eagerness, anger 5124 aïreson 4652, aërison 5575 sm. wrath

482

Glossary

aïrez adj. angry, furious 681 aisiez adj. in a comfortable or convenient position (for doing something) 2310 aïue 1121, haïe 6902; sf. help 1121; en bone haïe as a good help, to render assistance 6902 ajornée sf. morn, dawn 1037, 1903, 6143 ajorner 4580, 6168, 7356 v.intrans. grow light, (of day) dawn 1064, 3638, 3938, 4338, 4580, 6168, 7356 ajostée sf. joined battle 5396 ajoster 1550, 2080 v.trans. join, clasp (hands) 3388; line up in battle order 1367, 1550; come near, join, approach 1795, 2361, 6233; come to blows, join battle 2070, 2080 al adv. in another way, additionally 6214; otherwise, differently 7880; par al otherwise, in any other way 1642 alant sm. one who leaves, deceiver 1119 albesespines cf. aubespin alcior 1391, 8029 auçor 6386, auzor 1234, 1812, 7595 compar.adj. higher; sus en auzor up to the highest point OR higher and higher *; el son auzor right at the top 1812 alcun adj. some 7157 alemele sf. blade (of sword) 2228 alene sf. breath, wind, blast 3077, 3109, 3172 alenée sf. breath 3048; speech, words 1897 alenter 2774 v.intrans. slow down, go more slowly aler 28, 41, 160, 180, 193, 378, 380, 1004, +; pres.indic.1 vois 3452; 2 vois 4908; 3 vait 334, 379, 423, 494, 1117, 1270, 1526, 1558, 1721, 2184, 2234, 2324, 2370, 2733, 2747, 3277, 4141, 4149, 4483, 4884, 5510, 5528, 5580, 5582, 5993, 6297, 6870, 6883a, 6936, 6963, 7197, 7535, 7608, 7976, 8012; va 776, 1028, 1040, 1067, 1564, 2218, 2548, 4175, 4853, 6265, 7338, 7349, 7406, 7905; vet 2546; pres. subj.3 aut 47, 3408, 4616; alt 4736, 4898, 5413; 4 (imperat.) alomes 3234; 6 voisent 3461; v.intrans. go 28, 41, 62, 160, 180, 193, 225, 295, 334, 338, 378, 379, 380, 423, 439, 512, 525, 813, 1004, 1010, 1321, 1324, 1506, 1526, 1558, 1703, 1721, 1881, 1923, 2084, 2184, 2234, 2324, 2372, 2741, 2988, 3122, 3126, 3234, 3277, 3408, 3766, 3795, 3856, 3861, 3871, 3880, 3896, 3975, 4141, 4149, 4152, 4417, 4606, 4636, 4746, 4884, 5081, 5162, 5510, 5528, 5580, 5582, 5673, 5934, 5993, 6188, 6208, 6260, 6297, 6325, 6339, 6429, 6759, 6761, 6870, 6883a, 6936, 6963, 7048, 7465, 7520, 7523, 7608, 7621, 7626, 7976, 8012; go fast 1270; ferir alomes is! Let’s go strike them! 3234; aler sor get the upper hand of 4616; s’en aler v.intrans.& reflex. go away 47, 197, 494, 504, 776, 1117, 2218, 2370, 2548, 2733, 2747, 3461, 4175, 4483, 6265, 7197, 7338, 7406, 7535, 7905; (of day or night) depart, (come to an) end 1028, 1040, 1067, 1564; aut s’en let him go away 47; s’en aler fuiant run away, flee 3452, 4853, 4908 aleüre sm. grant aleüre at a great pace, swiftly 2371 alie sf. sorb-apple, the small, pear-shaped fruit of the service-tree (or possibly garlic?) i.e. something of little value 7191 alne sf. ell (a measure of about 1.1 metres) 3752 aloer 7290 v.trans. place, lodge aloigne sf. delay 5080 [aloser] vb.trans. praise, honour 819, 2928; v.reflex. gain honour for oneself, show one’s worth 5195 alosé 1130, 1445, 7001, 7538, alosez 535, 2368, 2984, 3944 past.p. aloser used as adj. honoured, praised, far-famed, renowned 535, 1130, 1445, 2368, 2984, 3944, 7001, 7538; praise-worthy 3944 alquant cf. auquant

Glossary

483

als cf. aus altegne adj. deep 3 alumement sm. enlightenment 7143 amanuïe past.part.f. amanuïr (= amenuïr - make small; polish) used as adj. ?tiny, ?smooth 6611 amanzor 764, 2862; aumanzor 2172 sm. emir, oriental ruler [amatir] v.trans. conquer, debase, dishonour 5250 ambedui 1466, 2379; ambedoi 7931; anbedui 305, 7885; andui 3384, 4040, 7502; ambedeus 770, 2319, 2460; andeus 2718, 3388, 5936, 6537, 6777, 6907; andos 1788, 4009; adj. and pron. both, two; d’ambedeus parz on both sides 770, 2319, 2460 amender 6447 v.intrans. improve 5030, 5038; v.trans. make reparation for 6447 amendie sf. reparation 794 [amenevir], [amoneïr] v.trans. prepare, make ready 7844, 7189 amenevi past.p. amenevir used as adj. well-trained, skilful 6406 ami obj.sing. 3350, 4217, 5150, 5621, 6280, 6417, 7038, 7043, 7427, 7680; subj. plur. 5611, 6417; amis obj. sing .(for rhyme) 3397; subj. sing. 212a, 927, 954, 1383, 1485, 3864, 4433, 5893, 6347, 7618; obj. plur. 507, 517, 699, 1744, 1882, 2430, 2605, 2914, 3222, 3983, 3995, 5436, 5626, 5979, 6040, 6092, 6604, 6798, 7387, 7394, 7445, 7627, 7895, 8085; amiz subj. sing. 212a; sm. (close) friend, comrade 212a, 507, 699, 927, 954, 1383, 1485, 1744, 1882, 2430, 2914, 3222, 3350, 3397, 3864, 3983, 3995, 4217, 5436, 5611, 5621, 5626, 5979, 6040, 6092, 6280, 6347, 6417, 6604, 6798, 7387, 7394, 7445, 7618, 7680, 8085; beloved, lover 2605, 5150, 6724, 7038, 7043; kinsman 4433, 7627; person one loves (i.e. friend, husband, father or other kinsman) 7427; amis privez close, intimate friend 517; amis carnaus kinsman 5893 amie sf. beloved, lover 3805, 3807, 6650, 6916; close friend, comrade, partner 4082; wife 4461, 6615 amie adj.f. friendly, well-disposed 1335 amiral subj. sing. 101; amirals subj. sing. 1344, 4521, 5260, 5385, 5483; amirauz subj. sing. 3721, 4508, 4588, 5607, 5760; amiraus subj. sing. 4729, 4824, 5207, 5318, 5337, 5708; amiraus subj.sing. 5645; amiranz subj.sing. 5179; amiras subj.sing. 5281; amirant obj.sing. 338, 4757, 5520, 6269, 7483; subj.sing. 4799, 6932; amiralt obj.sing. 1287, 4773, 5404; subj.sing. 5248; amiraut obj.sing. 4777; subj.sing. 4848 sm. emir amont adv. upright 1125; upwards, up 1475, 4401, 6060; up, on high 4078, 7188; upstream 4576; up on, upon 5739 amoraive sm. Moor (used as title), Moorish nobleman 1272, 2202 ample adj. wide 5015 amustadez sm. governor, Saracen title of honour 6335 ancedis sm. ancestor, forefather 3223 ançois 7578, anzois 7628 adv. beforehand, in advance 7628; rather (than this) 7578; ançois ….qe sooner… than 491; anzois 4523, 5347 prep. before; ançois c’ 3858, anzois qe 1010, 2886 conj. before ancui 7042, 7542, 7755 cf. anqi andos, andeus, andui cf. ambedui anel sm. ring 955, 6612 angarde sf. van-guard 832, 1188, 1546, 1579, 1588 angle sm. angel 1219, 4176, 4258, 4395, 4440, 5779, 7197, 7246a angoiser 5754, 7604, 8077; pres.indic.3 angosse 3383; p.def. angossa 6803; pres.part. angosant 3886, angoissant 5958, 6000; past part. angossié 7255 v.trans. torment, cause to

484

Glossary

suffer 3383, 3886, 5754, 5958, 6000, 8077; v.intrans. feel anguish 6803, 7255; gallop at top speed 7604 angosal (graphy of angosos, due to scribal error?*) adj. bitter, grievous, causing distress 1641 [anomer] v.trans. call, name 521 anoncion sf. incarnation 1599; the Annunciation 3915 anoré (graphy of honoré) past part. used as adj. honoured, noble 6960 anqi 1576, 2087, 3012, 4920; ancui (2665?), 7042, 7542, 7755; (encui 2670?); adv. today, soon, then anseler v.trans. saddle 8139 anste 2266, 2796, 5193 cf. hanste anti adj. ancient, of antiquity 4451, 5850 antregne sf. entrails, belly 2678 anuier 7714, 8076 v.trans. annoy, irritate 7714; tease, torment 8076 anuit adv. during the night, this night 4867, 6641 [anvier] v.trans. send 736 anz adv. cf. ainz anz qe conj. cf. ainz qe anzois, anzois qe cf. ançois aor adv. now 4048, 5439 [aorer] v.trans. pray to, worship 899, 1571, 7845 apaier 1657, 6583 v.trans. bring peace to, settle with a peaceful conclusion 1657; reconcile 6583 [aparceivre] past def.3 aparçut 8032; v.trans. see apareilier 2655, [apareiller], [apareiler]; pres.subjun.3 aparelt 8116 v.trans. dress, arm, prepare 506, 4405, 4545, 4635; organise 4553; prepare, set out (terms in document) 737; caparison, put armour on, make ready 7777; v.intrans. take up arms, fight 2655; aparelt let someone prepare 8116 apareüz cf. aparoir aparisanz adj. visible, capable of being seen 3095; made manifest, revealed 3621 aparlement sm. word, Gospel 7151 [aparoir] v.trans. past part. apareüz 3411 show, reveal apartenant 2411 apartenanz 3157 sm.pl. people belonging to (him), thus retainers, supporters, kinsmen apendant pres.part.apendre used as adj. subject (to), dependent (on) 590, 6651 apenser 7313, 7820 v.reflex. think, reflect [apendre] pres.indic.3 apent 3516, 4683, 4849, 4862, 5122, 5362; v.intrans. be subject (to), be under the rule of, depend (on) [apoier] v.reflex. press, make an effort, blow 3526; lean 4559; lean, support oneself, rest one’s head 7183; rely on 7436 aprester 1690; v. trans. make ready, prepare 1690, 1776, 4536, 7805, 8165 aprestez 415, 5684; aprestee 1896 past part.used as adj. ready, prepared aproi… de 7560 adj. ready, eager aprosmer 2088, 3508 v.trans. approach 3508; v.intrans. draw near 4562; v.reflex. approach 718, 3960; draw near to one another 2088, 4997, 5322 aqisson 2706, aquison 2765 sm. (affair, business); sanz nulle autre aqisson without any further ado 2706; sans nulle aquison without more ado, straight-way 2765 aquitant adj. clear, well-guarded 6275

Glossary

485

aqiter v.trans. safe-guard, protect 750; deliver, free 1249; clear of guilt 7746 [aquerir/aquerre] past def.3 aquis 7390 v.trans. cause, draw aquis adj. over-come, exhausted, (badly) wounded 3992 arabi sm. Arab horse 6415 araisoner, arasner cf. arasnier [aramir] vb.trans. swear (oath), solemnly undertake; de ce canp arami in relation to this solemnly sworn combat * 7850 arasnier 6769, 6778, [araisoner], [arasner]; v.trans. address, speak to 463, 3815, 5674, 5940, 6568, 6769, 6778; question 807 arbarestrée 2335, arbalestrée 5713 sf. bow-shot arbresel 7919, obj.pl. arbresiaus 5916 sm. bush arche sf. edifice made of arches and columns, i.e. church or cloister 3032, 3040 archie sf. bow-shot 3070 archier sm. lo trait a un archier a bow-shot 2095, 2101 archier 2093; arçoier 1673 v.intran. rear up on hind legs, shy (of horse) ardiment cf. hardiment ardoir 5348, pres.indic.6 ardent 4407; past part. ars* 25; v.trans. burn, destroy, consume 25, 5348; shine, gleam 4407 arement sm. ink 3314 aresnez adj. attached by the reins 6356 arestison sf. delay, hesitation 3269 [argüer] v.trans. press, pursue 2475; torment 3060, 4128; v.reflex. be eager for 1368; be tormented 4060 arme 530, 594a, 1015, 1792, 1793, 2118, 2189, 2206, 2218, 2650, +; arma 1634 sf. soul arotéement adv. without a halt, swiftly(?) 4861, 5369 aroter v.intrans. take the road, set out, follow 4248 arpent sm. measure of land, approx.½ hectare 3878, 3909, 4114 ars 25 cf. ardoir ars sm. bow 2701, 5490 ars sm.plur. arts 739 artimaisse sf. black magic, sorcery 2390 arzon s.m. saddle 1718, 2375, 2745, 5212, 5563 asaier 1872, asager 7515; asaiez (poetic licence) 2002; v.trans. try out, put to the test asanblée cf. ascenblée asanbler cf. ascenbler asaucé (graphy of esaucié) past part.used as adj. exalted 6539 ascenblée 1451, asanblée 5380, 5724, 6957 sf. assembly 1451, 5380; encounter 5724; meeting 6957 ascenbler 150, [asanbler] v.intrans. assemble 1232, 1404, 4539, 5861; meet 6232, 6830; v.reflex. join, ally onself to 150 ascenbler sm. encounter, battle 1916 aseïr 2579, past.def.6 asistrent 4593, past part. asis 202, 628, 716, 7448, 8015, asiz 204 v.trans. place 628, 716; set 930, 5900; seat, instal 2579, 4593; v.reflex. seat oneself, sit down 132, 641, + [asener] (de) v.trans. assign (to); instruct (about) 7655 [aserir] v.impers. grow dark, fall (of night) 5850, 5869 [aseürer] v.reflex. be sure (of s.thing) 173; take account of one’s own safety, hang back 2104, 2266, 5859

486

Glossary

aseüré 3652, 7754; aseürez 6348 past part.used as adj. sure 3652; safe, in no danger 6348; assured 7754 asier 1076 v.trans. assign asoleller 8126 v.impers. dry in the sun asols 1511; asolu 4074 past part. asoldre used as adj. absolved from sins, blessed 1511, 4074; terre asolue (the holy or sacred land), i.e. France 5840 [asolvir] v.trans., variant asoldre absolve from sins 1504 [asoplier, var.graphy assoploïer] v.intrans. grow weak 6028 aspois 6260 adjective of the Aspe/Aspre region in the Pyrenees . * aspreme adj. bitterest, most agonising 8070 asragon sm. horse from Aragon 7515 asteier, aster cf. haster [astrendre] cf. estreindre atant adv. now, at this point, then 137, 905, 1269, 1719, 1745, 1794, 2315, 3711, 4176, 4434, + atargier 937; atarzier 2953; v.reflex. delay 351, 511, 937, 2953, 3098, 3710, 4182, 5450, 6547 ateindre 6450; pres.indic.3 ateint* 12; pres.subj.1 ateng 4624; imperf.subjunc.3 atensist 7521; past part. atenz 2811; pres.part. atendanz 3610; v.trans. attack, wound 12; overtake, reach 2811, 4272, 4624, 5284, 6450, 7521; strike 3610 atendance sf. expectation, prospect 4944 atendre 1362 v.intrans. consider, weigh the matter [atorner] v.trans. treat 4788, dress in armour 5000; dress, prepare 6138; turn 6828; v.reflex. prepare oneself, put on armour 1514; de bataille atornez ready for battle, decided upon battle 1515; ceste batalle est issi atornée this battle is decided here, set up here 5726 auberc 574, 1406, 1661, 5513, hauberc 1782, 2247, 2744, 3844, 4409, 5763; oberc 1139, 1679, 7957, osberc 1414, 1691, 1707, 1734, 1810, 1854, 2053, 2174, 2187, 2215, 2224, 2237, +; osber 2731; osberz 2520, 6436; osbers (poetic licence) 5299; sm. hauberk, i.e. long coat of mail protecting the trunk aubespin 3899, albespin 6080, aubepine 7086, albesespines 8089 sm. hawthorn aubor sm. willow-tree 5553 auchor (graphy of altor) sf. height 1198 aucube sm. camp bed, small tent 5145 auferant sm. (grey) horse, courser 853,1109, 1717, 2458, 2566, 2790, 5523, 5535, 6288, 7904 auferant adj. highly strung, lively, swift 40, 3265 aufricant adj. African, made in Africa 5954 augalie sm. Caliph 701, 751, 778, 785, 1003, 2496, 3296, 3324, 3335, 3640, 4665, 4792 aumanzor cf. amanzor aumosner sm. inheritor, heir (ie inheritor of eternal life); alms-giver; almoner 3772 [aünbrer] v.reflex. hide oneself, find shelter 5067; become incarnate 5316 aunée sf. assembly 1582; battle, combat 1748 auner 4764 v.trans. assemble auqes 424, 2871, 3511, 3623, 4027, 4388, 4899, 7588; aqes 491; adv. a short time 3511; little 3623; a little, somewhat 4027, 4388, 4899, 7588; auqes (de) a little, a small amount 424; aqes (de) a certain amount 491; auqes de teus a few (more) of such (people) 2871 auqeton 5208, auqueton 7500 sm. a material of a dazzling white colour auquant 128, 5834; alquant 392 pron. some, a certain number 128, 5834; alquant… auquant some… others 392 auquanz sm. the majority 4294

Glossary

487

aus 127, 219, 1152, 1865, 2611, 3322, 3625, 4894, 4972, 5064, 6231, 6420, 6745, 6772, 7403, 7651, 7710; als 1169, 1367, 1938, 2351, 2360, 2877, 2928, 3308, 3515, 4194, 4595, 5345, 6165, 6167, 6433, 6848, 7062, 7261, 8187; els 398, 567, 1379, 2361, 3271, 5674, 5797 pron. them; used reflexively themselves 127 autement adv. nobly, solemnly 7405 autresi 5616, altresi 5608, 8061 adv. likewise; autresi come 5298 as; autresi… come 5393 as… as autretant adv. as many, in like number 4971 autretel sm. the same, likewise 5140; tot autretel te di I tell you as much 6416 auzor adj. cf. alcior auzor adj. used adverbially higher up; el son auzor at the very top 1812 aval adv. down 304, 607, 1618, 1639, 6963, 7499, 7791, 7939, 8164; below 1030; below ground, i.e. in the dungeon 8108 [avaler], [avaller] v.trans. flow down, run down 3355; bring down 7956; go down 8197; v.intrans. come down, descend 1058, 2371, 3661 avancier 7719, [avancer]; pres.subjunct.3 avant 7474; vb.trans. put forward, throw down (gage) 7719; favour, give advantage to 4937; v.intrans. put forward an opinion * 7474; v.reflex. move forward 5098; press forward, hasten 5162, 5787 avant adv. forward 335, 895, 1320, 3880, 4152, 4699, 4964, 5408, 5916, 5934, 6107, 6129, 6257, 7341, 7465, 7467, 7731, 7849; ahead 1579, 3829, 6283, 6658, 6835; first, before (others) 4595, 6273; in front 7726; before all 7731; ne puis en avant I cannot go further, there is nothing more I can say 352; ne puet mas en avant he can no longer go on 4384; son petit pas aloit tot en avant he was tottering forward 3875 avant prep. in front of 4972; ahead of 6493 avel sm. desire, pleasure; faire son avel do his pleasure, gain all he wishes 2389 avenant 2065, 2240, 2783, 6645, 7487, 7727; avinant 1524 adj. apt, pleasing, becoming 1524, 2065, 2240, 7487; becoming, attractive 2783, 6645, 7727 avenant adj.used as adv. becomingly, pleasingly 7729 avenir 5598; fut.3 avenra 1968; past.def.3 avint 4216, 6641, 6650, 8010 v.intrans. befall, happen 1968, 4216, 8010; come (there), do well, have a place 5598; come 6641, 6650 aver cf. avoir averser sm. devil 2206, 2650; devil (?), nightmare(?)* 6745 avesprer 4251, 7305 v.impers. grow late (of day), become evening, become dark (as night falls) 4251, 6845, 7305, 7540, 7818 aveüz past part.used as adj. avowed, determined? * (cf. note) [avier] v.reflex. set out (on a journey), begin 4567 avinant cf. avenant [aviser] v.trans. look closely at 1763; recognise (as a Saracen) 4033 avison sf. dream, vision 4400, 4425 avoé 964, 2037, 3580; avoëz 401 sm. defender, protector, guarantor avoir 1086, 1386, 2972, 3254, 4444, 4945, 5748, 5780, 7030, 7089; avor 1423, 5868; aver 1924; fut.1 avrai 353, 1222, 1984, 4024, 4213, 7383, 7535; arai 4035, 5566, 6011, 6251, 6675, 6676, 7579; 2 aras 958, 3275, 3361, 5749; avras 7194; 3 avra 161, 770, 1181, 1307, 1316, 1489, 1493, 1973, 2007, 3311, 6889, 7209; ara 177, 837, 902, 2229, 2608a, 2945, 3023, 3241, 3262, 3362, 3731, 4342, 4346, 4445, 4925, 5358, 6905, 7106, 7212, 7417, 7418, 7536, 7723, 7746, 7898, 8021, 8022; arra 3908; 4 avrome 1301; arons 1382, 2020, 3552a, 3557, 3658, 3723, 7756; avron 2560; 5 avrez 409, 614, 652, 669, 882, 883, 2558, 2981, 3940, 5457; arez 1253, 1330, 1981, 2241, 4220, 5172, 5237, 6458, 7016, 8020; 6 avront 873, 1194,

488

Glossary

1302, 1326, 2048, 3519, 4920, 7135; aront 1438, 1443, 1800, 2559, 2982, 3520, 3613, 3635, 5073, 5285, 5299, 5903, 6417, 8022; past def.2 eüs 2033; 3 ot 124, 133, 134, 272, 325, 354, 391, 393, 394, 459, 460, 602, 604, 605, 606, 630, 678, 711, 741, 761, 906, 929, +; oit 1182, 1318, 1742; 6 orent 1591, 1686, 2960, 3572, 4303, 4473, 4571, 4657, 5270, 7273, 7350, 7801, 7805, 7932, 8008; imperf.subj.3 eüst 2871, 2995, 3512, 4226, 4551, 4751, 5058, 5066, 5547, 6148, 6369, 6407, +; 4 eüsem 1474; pluperf.1 averai (< habueram) 7618; 2 averas 2276; 3 avera 1974; past part. eü 1244; v. trans. have 3254, 4945; obtain 4444; possess 5748; have, feel 7089; avoir chier hold dear, be grateful to 1086; avoir de pris consider valuable, consider to be of value 1386; avoir mesage send a message 2972; aver ontage be ashamed 1924; avoir dotance doubt, lack confidence 5780; avoir raison do anything about, prevent 7030; je l’averai mot chier I shall appreciate it 7618; v.intrans., impers. be 1423 avoir 151, 220, 614, 958, 1001, 5759, 5806, 5854, 5858, 6344, 7021, 7965 sm. wealth, treasure, possessions avrir v.trans. open 864 aze sf. axe 5864

b baceler 129, 4970, bazaler 1014, bacheler 5232, 6446 sm. young warrior, young knight bacin sm. basin, bowl 7941 [baër] la gole v.trans. open one’s mouth 1297; gole baée 5826, goule baée 7994 openmouthed, with gaping jaws; baié open-mouthed, drop-jawed, aghast 5489 bailie sf. power, keeping, governance; en bailie in his power, under his jurisdiction 361, 368, 746, 868, 1839, 1848, 3009, 4465, 6910; en lor bailie under their command 5063; en la soë bailie under his command 5252 bailier 50, baillier 1088, 1656, [bailler], ballier v.trans. govern 50, 132, 1656, 1954, 2656, 3020; give, hand over 290, 1348, 1997, 7507; carry, bear (arms) 1015, 1525, 3842, 7612, 7690; take, have authority over 1088; lay hands on, catch 5550, 6293, 6298, 6307, 6366; treat, thus mal bailliez soie may I be badly treated, may I be damned/cursed 436 bailir 1861, baillir 2585, [ballir] 6036 v.trans. control, determine the fate of 1861, 2585, 3596; treat 3718, 4793, 6036; mal bailie in a bad state, tormented 2824; bailli malement in a bad state 3722; mal nus avez bailis you have done us an evil turn 701 baillis subj.sing. sm. governor, ruler 629 baïs adj.(subj.form sing.) frightened, surprised, dumb-struck, tongue-tied by fear * 631 balance sf. uncertainty, danger 4936, 5101; en balance in the balance, uncertain 5793 baldré sm. belt, cross-belt 4022; waist 8042 [balier] v.intrans. flutter 2113 (cf. baloier below) [baller] v.trans. conduct, protect during travel 1861 [baloier] v.intrans. flutter 5434 ban sm. proclamation 6857, 7342, 7827, 7839 bandon sm. a bandon in self-surrender 1608, 3914, 6058; without restraint, at speed, impetuously 5213, 5568, 7520, 7523; metre a bandon hand over, cede 613; expose oneself to danger, sacrifice oneself 7654; metre son cors a bandon place oneself in a vulnerable position, expose oneself to danger 2129, 2693 baniere sf. war cry 2141

Glossary

489

[banir] v.trans. summon by proclamation; ost banie feudal levy, army of vassals 238, 253, 874, 1044, 2826; military expedition in which all vassals are summoned to participate 5244, 6918 barbarin adj. Berber, of Barbary; barbarian 1266, 4469 barguegne 2349, 2665; bargaigne 2671 sf. bargain barnaje 420, barnage 446, 550, 818, 4942; bernage 327, 1069, 1130, 1469, 1925, 1933, 3358, 6516, 8005, 8038; sm. barons, assembly of barons, aristocracy 420, 446, 818, 1069, 1925, 4942, 8005; nobility, wisdom, valour (qualities of a baron) 327, 550, 1469, 3358; noble status 1933, 6516 barnez 63, 671, 2993; barné 573, 3566, 3582, 8038; barner 4266, 6864 sm. barons, assembly of barons baronie 6551 sf. entitlement to status of (Aude’s) husband (‘husband’ being a secondary meaning of baron), right to marry her (a wife being seen as a chattel of her husband) bataille sf. battle-formation 4999 batal sm. boat (usually with oars, sometimes with sails, often for river rather than sea journeys) 4727 batellé adj. battlemented 4557 batistal sm. battle, fight; punishment 6222, 8155 baudor sm. delight, pleasure 6251; a baudor 2484, par baudor 2853 with ardour, with enthusiasm bauzens 2816, baucenz 4948 adj.m.pl. piebald (of horse) 2816; black-and-white 4948 befois cf. bofois bendé adj. banded; a or bandé with a gold fillet 7667 beneïçon 3925, benecion 2558 sf. blessing, benediction benvoillanz sm. well-wisher 5630 ber adj. noble, valiant 14, 148, 989, 1188, 1543, 1568, 1909, 1939, 2736, 4268, 5409, 5843, + ber adj. used as sm. noble person 1525, 1746, 1748, 2919, 3124, 3484, 3526, 3866, 4014, 7324, 8106; noble king 2074 bernage cf. barnaje besloi sm. treachery, wrong 8060; a besloi wrongly, off the true line 3328 besoing sm. need 2513*, 3082, 3434; (OR effort, haste 2513 ?*) besoigne sf. commission 364, 421, 5078, 5081 bien sm. what is just, honest, advantageous; par bien according to what is right, according to what is in your interest 4965 bis adj. brown 627 blastenger v.trans. blame, reproach, insult 7619 bliaut 390, 714, 3751, blishalt 6488 sm. under-tunic blicez past part.used as adj. wounded 4421 blishalt cf. bliaut [blesmir] v.trans. give bluish tinge to, bruise 6420, 7064, 7080, 7927, 8017 blois adv. only, simply 7667 blos 1374, blois 3181, 4948 adj.(obj.pl.) bluish or yellowish white bobance sf. pride 5792; pride, presumption 8003 bocle sf. boss, central stud on shield 2186, 2594, 5190, 5445, 5511 bocler 5189, boclé 7960 adj. with a boss on it, i.e. curved, convex 7960; escu bocler buckler 5189 boële sf. entrails, bowels 2230, 3673, 3958 boen adj. good 1974, 7202

490

Glossary

bofois 851, 854, 4748, bofoi 3322; befois 5700 sm. arrogance, pride 851, 854, 3322, 4748; tumult, fighting 5700 boguerant sm. fine buckram 5953 bohorder v.intrans. joust 6986 boins sm. obj.pl. (Roman) baths at Aix-la-Chapelle (modern Aachen) 182 boiseor sm. liar, traitor 6107, 6243, 8037 boisous adj. deceiving, treacherous; trop ferez qe boisous you will behave too treacherously, too dishonourably 6401 boissier 7700; [boisser], [bossier] v.trans, deceive, betray 7677, 7700, 7854; v.intrans. behave dishonourably 6316, 7863 bondie sf. reverberation, echo, sound 2825, 3008, 3061, 5251 bondir 3601 v.intrans. sound, resound, reverberate 3601, 5228, 5321, 5988 bonement adv. in the proper manner, with due respect 3961, 4845, 4847, 6343 bonté obj.sing. 472, 820, 954, 2032, 2554, 7962, 8048; bontez obj.sing. 2507, obj.pl. 1790, 6358 sf. goodness, benevolence, bounty 472, 820, 954, 1790, 2032, 6358, 7962; courage 2507; (ironic) bounty, favour, generosity (of Ganelon towards them) 2554 *; service 8048 borc sm. town 5,1350, 7499, 7791 borzois sm. townsman, citizen 7792, 8160 bos sm. wood 6312, 6759, 7023 boscage obj.sing. 1927; boscaus obj.pl. 5924 sm. wood 1927; bushes, groves 5924 boschet sm. grove 6881 bosdie sf. evil, treachery 242, 2832, 7178, 8027; ja n’avra bosdie will never suffer unkindness (or hurt) 7194 bosine 2082; busine 3550, 5177, 5649 sf. trumpet, bugle [boter] v. trans. press, push in 3752, 7989; bote pié thrusts out his foot 4111 boton sm. button, object of no value 269, 1711, 1725, 2695, 2756, 7771; ne m’est pas un boton I don’t care a fig 2140; ne pris pas un boton I don’t consider him worth worrying about 5221 bouzon sm. arrow, bolt (for cross-bow) 1721, 2701, 7521 braire 3549, 5754 v.intrans. blare, bray 3547; howl 5754; cry out, shout 5653; sanz b. ne crier 2217, sanz b., sanz c. 2746 without howling or crying out, i.e. before he has a chance to cry out * brandir 2098, 2151 v.trans. brandish, shake 759, 1047, 1176, 2098, +; v.intrans. shake 2151 brans 1429, 1521, 2327, 2515, 2521, 2610, 2820, 2905, 3445, 3680, +; branz 1824, 2056, 2358, 3626 sm. blade of sword, sword bregne cf. breheigne breheigne 4233 adj. barren, without children 1293; barren, sterile, profitless 4233 bres obj.sing. 453, subj.sing. 455; bries subj.sing. 736, obj.pl. 484, 501, 742; brief obj.sing. 4519, obj.pl. 773 sm. letter bricon sm. fool 261 brief, bries cf. bres [brisier] v.intrans. break, snap, shatter 1048 [brocher], [brozer] vb.trans.& intrans. spur 510, 1496, 1720, 1788, 1976, 2084, 2130, 2200, + brohon sm. bear-cub, bear 4427, 6042 brohon sm. calf, thigh 4437; fleshy parts of limbs, ligaments(?) 8102 broine 516, 883, 1033, 1767, 2116, 2464, 2795, 2865, 5721; brogne 2136, 2704, 5180, 5705 sf. byrnie (leather tunic, on which are sewn plates or rings of metal), cuirass brolet sm. grove, thicket 1772

Glossary

491

bruel sm. forest 1549 [bruire] pres.indic.6 bruient 3545, brugent 4652 v.intrans. be full of noise 3545; make a din 4652 bruni past.part.brunir used as adj. burnished 5610 [brunir] perf.indic.3 with inorganic e for visual rhyme (cf. note) brunie 4467 v.intrans. be burnished, gleam brunor sf. glint of burnished metal 1395 bu obj.sing. 3493, 5582; buz obj.pl. 2909 sm. trunk (of body) buef subj. pl. 4059 sm. oxen buël subj. pl. 2375 sm. bowels, entrails bugleraus obj.pl. 7359 sm. horns (made from bull’s horn) buz cf. bu

c ça 377, 650, 4753, 6808; za 1485, 2942, 2949, 5017; zo 92, 1490, 2406, 7560; dem.pron.neut. this, that 92, 377, 1490, 6808; adv. here 650, 4753; there 1475; traiez vos za! stand over there! 1485, 2942, 2949, 5017; traiez vos en ça be comforted by this (?) 6808; cui chaut de zo…? what does it matter…? 2406; de ça e de la on all sides 3545 cache sf. gage, recompense 802 caies 799 adj. cf. coiz [caleir] pres.indic.3 calt 2012, 2577, 3126, 3187, 5412; caut 721, 753, 2429; chalt 3114; chaut 2406, 3295, 3605, 7610; chet 4577; v.impers. matter, be important (to); ne vos calt d’esmaier you have no reason to be dismayed 2011; ne nos calt de morir what does it matter if we die? 2577; Cui chalt de ce?, Qi calt de ce?, Cui calt de ce? Who cares about that? 3114, 3126, 3187; de ce qe chaut what does it matter if? 3295; mais nos qe chaut what does it matter to us? 3605; qi en calt anyone to whom that matters 5412; ne vos chet plus la chanson demorer there is no reason for the story recounted to you to be delayed any longer 4577; ne vos chaut esmaier you have no need to be afraid 7610 calor sm. heat 575, 5956 cals dem.pron. those 3942 calt obj.sing. 1385; caut obj.sing. 3523, 7930; chals subj.sing. 3060, 7948 sm. heat camoissié past.part.used as adj. bruised 7065 cange sm. exchange 3953 [canger], cangier 2015, changer 7095 v.intrans. change, alter 2015; falter 1460, 7095; cele gent changée this alien race (?) OR these Christian people (?) * 5484 canp obj.sing. 1301, 1418, 1540, 2369, 2492, 2523, 2612, 3284, 3309, 3494, 3567, 3741, 3747, 3820, 3948, 4680, 4741, 5481, 5652, 6317, 7364, 7746, 7842, 7850; chanz obj.sing. 3436; chans subj. sing. 3770, obj.pl. 2401, 2805, 3090, 4345, 6986; cans obj.pl. 4946 sm. armed combat, battle, fight 1301, 1418, 1540, 2369, 2612, 3284, 3494, 7746, 7850; battlefield 2492, 3436, 3567, 3741, 3747, 3820, 4741, 5481, 5652, 7842; field 2401, 2523, 2805, 3090, 3309, 3770, 3948, 4345, 4680, 4946, 6986, 7364; fight in sense of armed pursuit 6317 canpé 4528, chanper 5186, chanpal 5733 adj. en bataille canpé 4528, en bataille chanper 5186 in pitched battle; estor chanpal pitched battle 5733

492

Glossary

canpion subj. sing. 3924, chanpion obj.sing. 7813, subj. pl. 7844 sm. champion, defender cans cf. canp çant card.num. one hundred 6004 cantel sm. en cantel sideways, on one side 2381, 7908, 7920 canz cf. chant canzons cf. chanzon caple obj.sing. 2802, subj. pl. 1481, 2879; chaples subj.sing. 2318, 2523, 3164, 5499, 5701, 7566, 7934; caples subj.sing. 2906 sm. thrust of sword 1481, 2879, 7934; carnage, massacre 2318, 2523, 2802, 3164; battle, combat 2906, 5499, 5701, 7566 caploier cf. chapler car 352, 1074, 1257, +; char 3378; adv. now 352, 1257; then, just (intensification of imperative) 519, 1074, 1257, 1424, 1432, 1444, 1778, 1890, 2043, 2373, 2551, 2875, 3278, 3285, 3453, 3454, 3457, 3655, 3936, 4415, 4637, 5675, 5755, 6767, 7038, 7324, 7558, 7889; if only (intensification) 2291, 2870 car 1055, 1535, 1766, 1970, 2261, 2835, 3059, 7917; char 5204, 5531, 5765, 6134, 6751, 7030, 7080, 7167, 7175. 8017, 8137; sf. flesh car cf. chier cargié 3942, cargiez 959 cf. charger carnal 8112, carnaus 5893, charnal 6366 adj. of flesh, mortal 6366, 8112; related by blood 5893 caroier cf. charger caroler 6858; charoler 6868 v.intrans. dance caroles (a round dance) 6858, 6868, 6984 carpentiers 1162 cf. cherpenter carrals cf. quarals çase, graphy of chase sf. (humble) house, cot 2447 casez sm. vassal 95, 226, 537 cassé cf. chasez caste adj. deprived, left to live without 2605 casteigne sf. chestnut 4232 catheigne cf. cheveteigne caucha cf. chaucer caut vb. cf. caleir caut sm. cf. calt ceans 799 adv. within these walls ceaus 3332, 3651, 3695, 4988, 5061, 5802; (see also cil) dem.pron.m.pl. these, those cée cf. chaïr [ceindre], ceignuz, irreg.past participle, 1824; v.trans. gird on (used of sword) 515, 1142, 1360, 1575, 1665, 1680, 1695, 1737, 1768, 1784, 1813, 1824, 1843, 2889, + cel, cels cf. cil celée sf. concealment; n’i a mester c. concealment serves no purpose, nothing is gained by concealment 5716, 6958, 6967 celer 31, 798, 1007, 1196, 1700, 6189, 6893, 7017, 7053, 7304; [celir], [cellir] (variant form); pres.indic.4 celison 4668*; infin.used as imperat.5 celer 798; past part.celé 6827, 6840, 7002, 8050; p.p. fem. cellie 246*, celee 1627, 1747, 1754, 6717; v.trans. hide, conceal, keep secret 246, 445, 798, 867, 1007, 1627, 1700, 1747, 1754, 4668, 6189, 6717, 6801, 6914, 6943, 6955, 7304, 7370; v.reflex. conceal one’s views, hide what one thinks 31, 6804; conceal one’s feelings 6893, 7053; n’a tant de cuer qe s’en puïst celer he has not so

Glossary

493

much strength of mind that he would be able to conceal this from himself (ie deceive himself about this) 1196 cels sm.(subj.sing.) sky, heaven 2440, 4105 cenbel sm. fight, battle 2391 cendal sm. silk fabric, similar to modern taffeta 1858, 2183, 5513, 7079 cerf obj.sing. 6760, cers subj. sing. 6361 sm. stag, hart cest, icest cf. cist chaainon obj.sing. 3143, chaënon obj.sing. 3205, chaaines obj.pl. 4427, 6109, chaëne obj.pl. 8075a sm. chain chace sf. attack, pursuit 1657 [chaëler] v.trans. lead, command, direct 1314, 2227 chaïr 2156, cheïr 2719, 5595; pres.indic.2 ches 481; pres.indic.3 qiet 495, 1358, 2458, 2733, 4479, 5563; pres.indic.6 chient 2435; pres.subjunc.3 cée 4564; past def.3 chaï 3279, 3515, 3574, 3879, 3994, 4406, 5516, 5619, 6421, 6848, 7884; chaïs 3408, chaïst 5534a; past part. cheü 458, 3482, 4097, 5766; cheüz 854, 2525 v.intrans. fall chaitis subj. sing. 674; chaitif subj. pl. 2423, 2807, 4660 adj. used as sm. captive, prisoner 674; wretch 2423, 2807, 4660 chaitive adj.used as sf. unhappy wretch 4500, 7158, 6210 [chalonger] v.trans. lay claim to, dispute by force of arms, or defend 3307 chals cf. calt chanper 5186 cf. canpé changé 5484 cf. canger chant obj.sing. 4909, canz obj.pl. 1068 sm. song chanuz 3084 cf. chenuz chanz cf. canp chanzon obj.sing. 3628, 4577, 5198; canzons subj. sing. 1388 sf. song 4577, 5198; male canzons, male chanzon defamatory song, scurrilous song 1388, 3628 chapal sm. (literally) cap, garland; Il ne:l… crent un chapal 5737 He does not fear him a scrap chapeliers subj. sing. 5529 sm. hood, part of the hauberc which lined the helmet and was made of chain-mail (covering the head and neck and spreading over shoulders, arms and body to form a knee-length tunic) chapler 1958, caploier 2016, [capler] 2287, chaplier 2952 v.intrans. fight, give battle 1958, 2016; thrust 2287, 2952 char, graphy of car 2429 conj. for char 1055, 1535, 1766, 5204, 5531, 5765, 6133 sf. cf. car charbon 5694 sm. coal, charcoal; charcoal-burner charger 5145, chargier 7329, [cargier], [carjer], caroier 44 v.trans. load 44, 959, 4548, 5145, 6674; entrust 3942, 7329, 7745 charné adj. of flesh and blood, mortal 3367 charners obj.pl. 6095 sm. charnel-houses, mass graves charoler cf. caroler chasé cf. chasez chasement sm. fief, property, domain; prendre chasement make one’s home, find lodging 7134 chasez 444, 668; cassé 580; chasé 6991 past part.used as adj. endowed with dominion (over land, especially feudal domain), made ruler 444, 580, 668; with feudal domains, landed 6991

494

Glossary

chaucer 6459; past def.3 caucha 908; v.trans. put on (someone’s) foot 908, 6459, 7764, 7796 chauches de fer obj.pl. 7500, chauces de fer obj.pl. 7762, 7794 sf. iron leg-coverings (trousers), iron shoes (items of armour) chaut cf. caleir chaus adj. as sm. bald man 5888 chemois sm. field of stubble 5692 chenuz subj.sing. 272, subj.plur. 825; chanuz subj.sing. 3084; chenue f. 2895, 4070, 5843 adj. white (of hair), hoary 272; white-haired 825, 2895, 3084, 4070, 5843 chenuz subj. sing. 3659, obj.pl. 3422; chanuz subj. sing. 285 adj.used as sm. white-haired (old) man cher cf. quesre cherpenter obj.sing. 3825; carpentiers subj. sing. 1162; sm. one who cuts his enemies to pieces with a sword, a fighter, a warrior (literally, one who cuts and assembles the wooden framework or charpente of a house) ches 481 cf. chaïr chés 5091 sm. cf. chief chesrue sf. plough 4059 cheü cf. chaïr chevaleros adj. valiant, brave, chivalrous 6391 cheveteigne obj.sing. 4228, catheigne subj.pl. 4237 sm. lord, ruler 4228; captains, noble leaders 4237 chief obj.sing. 167, 192, 256, 257, 471, 558, 1308, 1536, 1571, 1592, 1684, 1713, 1783, 1856, +; obj.pl. 251, 749; qief obj.sing. 885; chiez obj.sing. 3477; ches obj.pl. 4657, 5091 sm. head 167, 192, 251, 256, 257, 471, 558, 1308, 1571, 1684, 1713, 1783, 1856, 2025, 2030, +; end 471, 1713, 2115, 2173, 2223; es chiez in the head 3477; el premier chief 1536, el primer chief 1592, 4687; right at the front 1536, 1592; right at the top, first of all 4687; de chief from the beginning 6755; tot le chief primier right at the front 6878 chient 2435 cf. chaïr chier 941, 1086, 1157, 1723, 2473, 2637, 2647, 2657, 2721, 2888, 2917, 3012, 3742, 3760, 3839, 4602, 4871, 4895, 6519, 6723, 6758, 6771, 6776, 7105; car 2142; adj. dear chier 348, 436, 556, 696, 2091, 2886, 2900, 3012, 3427, 4065, 6716, 7616, 7967; qier 475; adv. dear, dearly chiere sf. mien, appearance 234, 605, 872, 1089, 2662, 3000, 3043, 3058, 5455, 5645, 6915, 7986, 8016, 8026; face 3709, 7077 chité sf. city 5854 chivalarie 875, 1334, 2106, 2797, 8014; chevalerie 2829; chivalerie 5069, 5245; chevalarie 6909 sf. knightly prowess 875, 1334, 2797, 5069, 5245, 6909; knightly adventure, combat 2106, 2829; knighthood, assembly of knights 8014 chocer 6775, 8031 v.trans. put to bed 8031; v.intrans. lie, be couched 6775 ciglaton 615, 6034; ciglator (graphy adapted to suit rhyme) 1228 sm. silk cloth, brocade cil subj. sing. 142, 994, 2862, 3173, 3175, 4408, 4693, 5468, 7506; subj.pl. 129, 967, 1405, 1406, 1567, 1937, 2748, 3189, 3190, 3191, 3623, 4586, 4676, 5130, 5321, 5987, 6042; obj.sing. 899, 4350, 6219; icil obj.sing. 5599, 8138; cel obj.sing. 56, 345, 361, 433, 474, 488, 499, 521, 534, 1097, 1112, 1548, 1589, 1818, 2090, 2153, 2280, 2345, 2452, 2612, 2864, 2963, 3436, 3574, 3899, 3990, 4343, 4935, 5334, 5567, 5703, 5746, 6353, 6386, 6857, 7364, 7368, 7930, 8153, 8166; icel obj.sing. 1953, 2408, 2770, 5449, 6585, 7148, 7600; cele obj.sing.f. 1393, 1457, 1567, 1742, 1836, 1849, 1918, 1928, 2133, 3796, 4360, 4546, 5057, 5484, 5489, 5539,

Glossary

495

6189, 6467, 7378, 7418, 8108; subj. sing.f. 2675, 3551; icele obj.sing.f. 4544, 4549, 5077, 5724, 5841, 6498, 6815, 6824, 7357; celes obj.pl.f. 1407; dem.adj. that, those cil subj. sing. 31, 287, 379, 907, 1260, 1267, 1277, 1299, 1333, 1562, 1574, 1575, 1839, 2126, 2546, 2692, 2701, 2876, 3201, 3267, 3280, 3297, 3494, 3585, 3621, 3892, 3920, 3934, 4032, 4277, 4583, 4649, 4692, 5779, 4827, 4851, 5051, 5085, 5257, 5301, 5367, 5376, 5388, 5637, 5864, 5893, 6375, 6602, 7093, 7193, 7218, 7220, 7428, 7590, 7717, 7866, 7980, 8183, 8200; subj. pl. 104, 1319, 1413, 1443, 1541, 1922, 1949, 1952, 2062, 2075, 2680, 2821, 3027, 4125, 4235, 4415, 4639, 4674, 4679, 4710, 5012, 5097, 5346, 5347, 5496, 5653, 5654, 5840, 5908, 6043, 6094, 6213, 6228, 6274, 6286, 6291, 6300, 6302, 6310, 6466, 6526, 7174, 7321, 7413, 7448, 7464, 7531, 7594, 7652, 7655, 7842, 8197; icil subj.sing. 1974, 2617, 5435; subj.pl. 1397, 3203, 3302, 6284; cel obj.sing.m. 29, 2628, 2637, 2674, 3212; neut. 931, 2615; subj.sing. 7492; ice obj.sing.neut. 6445, 6650, 6911, 8190; icel obj.sing.(f) 3513, (m) 7148; celui obj.sing.m. 368, 529, 705, 1554, 1705, 1728, 2386, 2462, 2678, 3184, 3194, 3737, 4028, 4200, 5078, 7055; cels obj.pl. 125, 147, 199, 255, 425, 508, 536, 1193, 1237, 2028, 2147, 2148, 3125, 3218, 3415, 3458, 3549, 3602, 4524, 4547, 5046, 5266, 5276, 5310, 5381, 5407, 6163, 6185, 6188, 6299, 6591, 7329, 7492; subj.pl. 2954; ceus obj.pl.m. 5658; ceaus (= ceaus) obj.pl. 3332, ceaus obj.pl. 3651, 3695, 4988, 5061, 5802; icels obj.pl. 487; cele obj.sing.f. 2892; dem.pron. that, those, that man, those men 31, 125, 368, +; the former 7492; (neuter) that 931, 2615, 6445, 6650, 6911, 8190; celui qi, cil qi he who 368, 379,1839, +; anyone who 3494; n’i a cel ne …, n’i a cel qi ne …. there is no-one who does not … 2674, 3212; por cel for that reason 931 cist subj. sing.m. 334, 717, 736, 2792, 2843, 2869, 2874, 3080, 3108, 3110, 3157, 3663, 4102, 6552, 7306; f. 7286; obj.sing. 702; subj.pl. 2067, 2123, 2194; cest subj.sing.m. 446, 3172, 3498, 6634, 6862; obj.sing.m. 24, 158, 178a, 315, 345, 378, 339, 473, 585, 707, 929, 955, 964, 1162, 1557, 2040, 2160, 2290, 2501, 2871, 2878, 2943, 2968, 2974, 3022, 3360, 3537, 4103, 4138, 4190, 4625, 4627, 4770, 4813, 5051, 5107, 5614, 5925, 6277, 6456, 6578, 6683, 6854, 6939, 7101, 7106, 7109, 7304, 7601, 7707, 7850, 8199; f. 2898; subj.pl.m. 2165; icest obj.sing.m. 1366, 3261, 3286, 3372, 3381, 3428, 3475, 3620, 4756, 6192, 7092, 7242; ceste subj.sing.f. 284, 421, 1026, 1494, 1632, 1754, 4565, 4793, 5174, 5336, 5726, 5742, 7210; obj.sing.f. 172, 239, 294, 307, 321, 670, 802, 843, 1262, 1657, 2146, 2456, 2480, 4738, 4974, 5571, 6999, 7091, 7614; iceste subj.sing.f. 5026; obj.sing.f. 4238, 5056; cez obj.pl.m & f. 794, 1172, 1551, 1615, 1686, 1702, 2696, 2796, 2822, 3041, 3121, 3728, 4193, 4324, 4385, 4407, 4878, 5015, 5275, 5502, 5602, 5890, 6126, 6352, 6397, 6485, 6486, 6848, 6859, 7160; icez subj.pl.f. 4221; dem.adj. this, these cist subj.sing.m. 3826, 5558, 5919, 6544; neut. 774, 2720; neuter sing. as collective + plur.vb. 1472, 5065; cest neut. 1477; ceste obj.sing.f. 267, 425, 1747, 3056; iceste obj.sing.f. 4425; icestui obj.sing., neut. 8094, 8133; cez obj.pl.m. 5086; dem. pron. this, these; this man, this thing, these men, these things clamer 2212; [clemer], [cleimer] v.trans. claim, ask for 158, 1503; proclaim 2212; name, call 3589; v.reflex. proclaim oneself, show oneself 3611; call oneself 4500; qite c. renounce all claim upon 442; c. vos copes recite your mea culpa, confess your sins 1503, c. sa cope confess one’s sins 1572, 4135 [clarier] v.intrans. light up, brighten 1336 clavel sm. link of the hauberk 2382, 7884, 7912 cleim sm. cry, complaint 6255 clergie sf. learning 3814 [cliner] v.intrans. bow 167, 1571; v.reflex. bow down 1093 clos obj.pl. 2550, 5901 sm. claw, clasp

496

Glossary

[cobrer] v.trans. seize, grasp 470, 1770 coë sf. tail 7996 [coharder] v.intrans. act like a coward 4954, 5026 cointe (variant graphy, coint) adj. valiant, brave 3583, 3732; wise 5633 cois sm. choice, what a person prefers; (OR what one accepts as payment in full); a son cois to his satisfaction 847 [coisir], cosir 6087 v.trans. notice, mark, see 4115, 4429, 6087, 7163 [coivrier] 3060 v.trans. torment coiz m. 258, 320; caie(s) f. 799 adj. silent, quiet, calm 258, 320; private, to oneself, unspoken 799 colée sf. violent blow 1575, 4066, 5014, 5403 [coler] v.intrans. flow, run down 1618, 2941, 3105; plunge, run through 2327, 2365, 2789, 3680, 7939, 7990 colers sm. porter 6345 collir 2586 v.trans. receive colorez m. 394; coloris m. 2610; coleriz m.pl. 1181; colorée f. 6563 adj. (of sword-blade) damascened, i.e. inlaid with gold or silver or with watered pattern produced in welding 394, 1181, 2610; coloured 6563 colpe obj.sing. 2074, 3912; cope obj.sing. 4135, 4154, 7126; copes obj.pl. 1503 sf. blame, guilt 2074; clamez vos copes confess your sins 1503; sa cope c. 1572, cleime sa c. 4135, bati sa c. 3912, bat sa c. 4154, 7126 repeat one’s “Mea Culpa”, i.e. confess one’s sins (‘bati’ may imply beating one’s breast) cols obj.pl. 6425 sm. necks cols 1387, 1440, 1646, + cf. cop com 32, 185, 513, 656, 674, +; come 134, 450, 708, 777, 1732, 2047, +; con 3969, 5027; adv. like, as, in the manner of 134, 450, 513, 674, 777, 1514, 1532, 1732, 1733, 1853, 1869, 2047, 2423, 2567, 2807, 2813, 3199, 3219, 4268, 4990, 5027, 5163, 5208, 5210, 5271, 5283, 5298, 5393, 5881, 6246, 6317, 6376, 6368, 6378, 6388, 6391, 6424, 6461, 7231, 7500, 7503, 7527, 7555, 7878, 8076; how (often exclamatory, occasionally interrogative) 767, 1625, 1628, 2313, 3239, 3644, 3803, 4189, 4231, 4274, 4470, 4678, 4684, 5628, 5928, 5929, 5999, 6304, 6314, 6474, 6792, 6846, 7090, 7198, 7259, 7483, 7534; to what degree, whether 6275; interjec. what a …! 745, 2687, 3419, 3807, 4504, 4704, 6540, 6792, 7232, 7483; conj.& rel.adv. as 656, 949, 960, 1101, 1613, 2145, 2859, 3786, 3969, 3988, 4158, 4172, 4546, 5812, 5981, 6292, 7176, 7405; than 4546; when, as (temporal) 3360; se com, si com, si come, si … com, tot issi … con just as, even as, exactly as 32, 708, 902, 1389, 2021, 2682, 3251, 3331, 3786, 3790, 3969, 4271, 5228, 6063, 6157, 6762, 6831, 6988, 7176, 7380; tant com as much as 176; as long as, so long as 185, 829, 1019, 2267, 3183, 3624, 7461, 8011, 8084, 8186; tant … com as much … as, as many … as 960, 5812; com il poït as best he could 3988; ausi com as 5208; com mar in what an evil hour 5928, 5999, 7091; com s’., come s’., come si as if 6576, 6704, 7184, 7241; com vos est convenent? How are you? How are matters with you? 4678; com lui est convenent how matters are with him 4684; ensi com just as 8120 comander 6173 v.trans. entrust 297, 337, 1561, 1574, 1633, 3197, 4044, 6173, 6456, 7643; commend, recommend (soul to God) 3697, 3898, 3982, 3987, 7757, 8028; v.intrans. command 188, 190, 773, 864, 1121, 1509, 2339, 2614, 2615, 4394, 4600, 4621, 4965, 5004, 5471, 6379, 6988, 7345, 7372, 7602, 7743, 7838, 8039, 8088; specify, instruct 3134 comandez sm. vassal 97, 228, 441, 666 comandie sf. command, will 803

Glossary

497

comant obj.sing. 496, 1017, 1323, 4173, 5978, 7342; obj.pl. 596; comanz subj.sing. 323, obj.pl. 1072, 3089; sm. command; tot a vostre comant (I am) entirely at your command, just as you wish 1323, 5951, 6284 comencëor obj.pl. 4712 sm. first of fighters, warrior who begins the battle coment adv. how, by what means 20, 933,4687, 6843, 7720, 7721; interrog.adv. how?, by what means? 6043, 7717; coment qe soient however much they may be 2584 comunal adj. all together, acting together, as a group 2191, 6223, 8157; familiar, usual 6373, 7882; communal 8116; comunaus in a (religious) community 5907 comunalment 2422, 2436, 7117; comunaument 2560, communalment 3261 adv. all together, one and all con 890. 1860, 7301; com (V7) 6540a; prep. with conbele sf. little valley 2226 coneüz past part.conoistre used as adj. renowned, famous 2735, 2915, 4063 confaitement 173, 866, 2972 adv. how, in what way confenon obj.sing. 1527, 1716, 2058, 2248, 2258, 2544, 2550, 2705, 2764, 4990, 5384, 7803; subj.sing. 5370; confenons subj.sing. 1144, 1786; obj.pl. 1374, 3181, 4953, 5034, 5203, 5423; subj.pl. 4407; confanon obj.sing. 2137, 2204; sm. gonfanon, pennon (attached to lance) confenoner sm. standard-bearer 123 confondison sm. destruction 4661 [confondre] v.trans. destroy, throw into disorder, confound 2356, 2474, 2893, 2911, 3240, 4277, 4583, 4837, 7506, 7585 [confreindre] v.trans. break, destroy 2350 conoissance 4928, 5089; conoisance 5171 sf. cognizance, the distinctive colours borne in the armorial badge and on the shield to enable the bearer to be recognized; the shield bearing these colours 4928; a une conoissance with a single armorial badge 5089; por conoisance according to their armorial badges, under their colours 5171 [conoistre] v.trans. know 1118, 3086; recognize 1593, 2780, 4582, 4587; recognize, admit 7454; v.reflex. recognize one another 5141 conpaigne obj.sing. 1209, 3213, 5398, 5652, 6136; subj.sing. 2500; conpeigne obj.sing. 1458, 2342, 4226; subj.sing. 2675; conpeignes subj.pl. 5416; conpegne obj.sing. 1290; conpaignes obj.pl. 5340 sf. company, body of troops 1209, 1290, 1458, 2500, 2675, 3213, 5340, 5416, 5652; company (of angels, prophets, saints, +) 2342; army 4226, 5398, 6136 conpaignie 873, 3015, 3066, 3299, 5452; conpegnie 7349; conpeignie 5646, 6607, 7187; sf. company, body of troops 873, 3066, 3299, 5452, 5646, 7349; companionship, sympathy 3015; company (in general) 6607, 7187 conparer 2740, 7826; [conprer] 475; [conperer] 556, 3636, 4446, 7616; v.trans. buy 696, 1520; pay for 475, 2091, 2740, 3636, 4445, 7616 [conpansser] graphy of conpasser v.trans. order in a regular, symmetrical manner, structure 5052 conpeigne sf. companion, trusty friend (applied to the sword, thus feminine) 2682 conpeignon obj.sing. 1604, 2694, 2759, 3368, 3391, 3774, 3850, 3909, 5218, 8062; subj.pl. 1529, 1940, 2548, 3138, 3792, 5569, 6033, 6469, 7018, 7651, obj.pl.minus final ‘s’ for rhyme 3201, 3274; conpeignons obj.pl. 1863, 2162, 4509, 7853; conpeign subj.sing. 1154, 1318, 1381, 1424, 1432, 1444, 1485, 1544, 1871, 1890, 1905, 1964, 2306, 2499, 2635, 2873, 2949, 2967, 2985, 3003, 3234, 3358, 3374, 3401, 3827, 3840, 3850, 3856; conpaing subj.sing. 2378; conpainz subj.sing. 212a sm. companion, comrade; conpeignon de la coisine scullions 3201–3202

498

Glossary

conplie sf. the hour of Compline (Vespers, the Evening Service in Church); the Compline bell 3636, 3858, 3887, 5250, 6920 conpliz adj. accomplished, finished 8147 conperer, conprer cf. conparer [conquerre] past def.1 conquis 3421, 8187; 3 conquis 3, 6599; conquist 812, 2848, 7782; past part. conquis 551, 575, 2894, 3225, 4012, 4142, 6506, 7383, 7435, 7443, 7524; pres.part. conquirant 598; v.trans. conquer 3, 551, 575, 598, 813, 2894, 3225, 3421, 4012, 4142, 6506, 6599, 7383, 7435, 7443, 7524, 8187; win 2848, 7782 conquister 24, 4260 v.trans.& intrans. conquer 24, 1364,; achieve 4260 conréer 3116, 5187, 7075, 7291, 7824; conraer 6174; [coréer], [conroier] v.trans. fit out, dress, equip 513; prepare for battle, arm 3116, 5187, 6174, 6334; deal with 4035; adorn, apparel 6553, 6560; prepare (ie dress and adorn) for interment 7075, 7267, 7291; make ready, arm 7824; dress (as in cooking) 8117 conroi sm. disposition, order 841, 7548; en ont pris le conroi have taken care of it 7548 [consirer] v.reflex. fast, go without food 4333 [consivre] past part. consuz 3426, conseü 4103, consut 7911, consiut 7915 v.trans. follow up, strike down 3426, 7911, 7915; stricken down 4103 constantis adj. strong, durable, dependable 1372 conte obj.sing. 797, 1614, 1668, 3133, 3196, 3841, 5223, 5580, 5918, 5942, +; subj.sing. 6525, 8184, subj. pl. 565, 1469, 1943, 2192, 4590, 6202, 6991, 7134, 7841; obj.pl. 3898; cons obj.sing. 4869, subj.sing. 212, 234, 376, 384, 389, 456, 512, 536, 540, 592, 637, 640, 686, 717, +; cuens subj.sing. 209, 6615, 6629, 6932; quens subj.sing. 366, 1653 sm. count contenance sf. containment 1213; manner, deportment 5090; containment, restraint, keeping silent, thus hiding (of knowledge) 6796 contenant sm. fashion, style 1242; manner, mien 2782 contençor cf. contenzon contenir 3256 v.reflex. conduct oneself, behave 3256, 3264 contenzon 2134, contençor 1235, contençon 3130 sf. ardour, impetuosity; par mout grant contenzon 2134, par fiere contençor 2134, par fiere contençon 3130 in great impatience, with great eagerness contor subj.pl. 4204, 7271, 7599, 8034; contors subj.pl. 5885 sm. counts, nobles contrarier 3016, [contralier] v.trans. argue with, quarrel with, contradict, oppose 358, 3001; irritate, annoy 7670; v.intrans. quarrel, argue 3016 contre 732, 1049, 1526, 2030, 2730, 3189, 3311, +; prep. towards 1049, 1526, 2030, 3386, 3388, 3398, 3914, 4304, 4623, 5320, 5353, 5362, 5375, 5402, 5736; with head towards 3386; against 2730, 3189, 3677, 3909, 3924, 4756, 4954, 5370, 5424, 5611, 6928; (fighting) against 6898; compared with 3311; about, round-about (in approximations) 4114; contre cheval upon a horse 732; contre mont on high 4104 (cf. contremont) contredit adj. impious la contredite gent 3313 unbelieving race, infidel race contrée 691, 1030, 1773, 1893; contrie 1356 sf. land contremont adv. on high 7999; (cf. contre mont advl.phrase 4114) contrester v.trans. withstand 2214, 2745, 4374 contreval adv. 6656 downhill, down; en contreval 4295 downstream convenant sm. pledge, promise; tenez me convenant give me your word, promise me 903 [convenir]; [covenir] pres. convent 1546, covent 3853; convient 6175; pres.subjunct. covegne 1091, conveigne 7830 v.impers. behove 180, 1091, 1546, 2149, 2163, 2612, 3855, 6175, 7830

Glossary

499

covenent 4678; convenent 4684 adj. satisfactory, agreeable (to), seemly, fitting; com vos est covenent? how are you? How are matters with you? 4678; com lui est convenent how he is, how things are with him 4684 cop obj.sing. 1141, 2213, 2245, 2864, 3372, 3683, 3717, 4038, 5239, 5394, 5449, 5536, 5559, 5585, 5703, 6571, 7955, 7956; subj.pl. 2123, 2393, 2879, 5472; obj.pl. 1957, 2213, 2245, 4310; colp obj.sing. 5244; coup obj.sing. 2019, 2742, 3275; subj.sing. 5533; subj.pl. 1481; cols subj.sing. 1710, 2792, 7972; obj.pl. 1387, 1440, 1646, 1822, 1916, 2044, 2079, 2332, 2420, 2505, 2528, 2657, 2681, 3186, 3204, 3491, 4060, 4287, 5461, 5592, 5677; subj.pl. 2165; cous subj.sing. 7903, obj.pl. 1164, 1428, 7064, 7882, 7900; sm. blow 1141, 1164, 1387, 1428, 1440, 1481, 1646, 1710, 1822, 1916, 2019, 2044, 2079, 2213, 2245, 2332, 2393, +; cut 2681; a cop so as to strike a blow, in order to fight 3717 cope cf. colpe cor 1424, 1459, 1931, 2977, 3074, 3106, 3172, 3525, 3538; corn obj.sing. 3008; sm. horn 1424, 1459, 1931, 2082, 2825, 2977, 2992, 3008, 3035, 3074, 3080, 3106, 3108, 3172, 3178, 3525, 3538, 3705, 3726, 5177, 6738 cor 3512, 4202, + cf. cuer corage sm. inmost thought, decision 326, 1460; pride 402, 544; inmost thought, sentiment 548, 2966, 2975, 6955; mind (heart) 1932; heart, spirit 1952, 4816, 6218; per bon corage heartily 1915, 3963; je n’ai pas corage qe… I cannot bear that… 8105 corant 5161, 5821 pres.part.corre, q.v.; 1523, 1531, 2239, 2315, 2414, 2781, 5459, 5583, 5994, 6270, 6285, 6295, 6871, 6876; corante 1671; corantz 1332; coranz 222, 2051, 3617, 4290, 4949, 5422, 8140; pres.part. used as adj. swift 222, 1523, 1531, 1671, 2051, 2239, 2315, 2414, 2781, 2877, 3617, 4949, 5422, 5994, 6285, 6871, 6876, 8140; galloping 3429, 5583; running 1332; swift-running 6655; swift-flowing 3870, 4290, 6270, 6295 cordals obj.pl. sm. strands of hair, tresses 5138 corée sf. bowels, belly 2326 coréé 4035, coréez 513 cf. conréer (usual Old French graphy of coréer) coreor adj. swift 1396 coreza cf. corochier coriges 8077, coreges 8142 obj.pl. sf. straps, whiplashes cornée sf. sounding of horn 1900 corocement sm. anger 3023 coroceus 107, 349, 388, 502, 855, 1117, 4120, 4396, 7897; corozois 2370; corozos 3193, 7420; corozous 3954; adj. angry, grieved 349, 388, 502, 1117; tormented, afflicted 107, 2370, 3954, 4120, 4396; afflicted 7897 corochier 1655, 3021, 3736; corocer 4889; [corocher], [corezer], [corozer] v.trans. anger 1998, 2295; torment, make to suffer 730; grieve 4196; harass, attack 1655, 4889; afflict, cause to suffer, harass 1053; anger, trouble 4416, 4613; afflict, grieve 5795; v.intrans. become angry, be angry 719; v.reflex. become angry with one another, quarrel 3021; infin.used as sm. anger, torment 3736 corozant pres.part.used as adj. grieving 1122 corre 1771, 1846, 2170, 2182, 5125; cosre 2100, 2774, 5430, 5683, 7881; pres.indic.3 cort 1720, 1829, 2692, 2701, 4436, 5504, 5561, 7131, 7781, 7982, 7986; pres.indic.6 corrent 3138, 7993; past.def.3 corut 3677, 3748, 3754, 3828, 3832, 3955; past def.6 corerent 1369, corrurent 7695; past part. coru 4007, 4466, 5764, 5859; pres.part. corant 5161, 5821; v.intrans. run; gallop (of horse) (cf.also corant adj.) cors 130, 133, 144, +; sm. body 130, 133, 144, 325, 359, 391, 397, 465, 604, 607, 750, 810, 858, 996, 1166, 1246, 1273, 1278, 1386, 1528, 1634, 1635, 1792, 2024, 2033, 2039, 2060, 2117,

500

Glossary

2137, 2155, 2175, 2204, 2238, 2256, 2284, 2365, 2383, 2514, 2528, 2630, 2645, 2705, 2733, 2757, 2764, 2783, 2790, 2814, 2822, 3030, 3094, 3116, 3144, 3276, 3301, 3307, 3355, 3393, 3426, 3449, 3479, 3483, 3505, 3642, 3693, 3702, 3885, 3960, 4027, 4044, 4137, 4192, 4339, 4347, 4458, 4661, 4789, 5047, 5090, 5128, 5132, 5136, 5197, 5319, 5439, 5447, 5514, 5516, 5983, 5992, 6014, 6028, 6184, 6194, 6420, 6541, 6576, 6634, 6709, 6812, 6817, 6839, 6993, 7014. 7028, 7038, 7042, 7063, 7075, 7147, 7201, 7241, 7246, 7254, 7262, 6264, 7285, 7417, 7456, 7579, 7727, 7781, 7830, 7928, 7941, 7949, 7990, 8041, 8077, 8085, 8099, 8100, 8111, 8126, 8145; person, self (used periphrastically for stress) 479, 1016, 1690, 1956, 2017, 2102, 2129, 2693, 2738, 3179, 3697, 4097, 5178, 5780, 5930, 6333, 6393, 7475, 7530; lo cors pener toil, struggle 810; le cors de Rollant the right to be first to engage in combat with Roland 1246; mon cors guier betake myself 1278; de vostre cors vengiez avenged upon your body 4192, de son cors vengier avenge upon his body 8085 cors sm. speed 2817, 7544 corseinz subj.sing. sm. the Host, the Sacramental bread in the Mass or Communion service; any holy relics; ja n’est ce mie corseinz a deviser this is no Eucharistic bread for dividing/no dead body of a saint whose relics must be shared out, i.e. this poses no problem, there is no difficulty in deciding what to do 8135 * corsier adj. swift 1274 cort sf. court 273, 535, 6324, 7614, 7616, 7683, 7748; a la cort d’eus 7436 on their (judicial) court, on their judgement cose 2597, 7888; couse 172, 7231; chose 767a, 4699, 6348, 6351 sf. thing 172, 767a, 2597, 4699, 6348, 6351, 7888; person, creature 7231 cosir cf. coisir costal sm. small hill, hillside 5507, 6363 coste sf. rib 5966 costé 1677, 2260, 2346, 3570; costez 1784,1788, 2366, +; sm. side (in general) 1677, 2346, 3570; side (of body) 2260, 1784, 1788, 2366, 3447, 3672, 6357; de costé d’un poisson of something like whale-bone (?) or shark-skin(?), or in overlapping scales for strength(?) 7769 [costier] v.trans. go alongside, skirt 2374, 4568 costume sf. manner, way, custom 169 coudre sm. hazel-tree 6123 cous subj.sing. 3137, obj.pl. 3197; queus subj.sing. 3136 sm. cook cous 1164, 1428, + cf. cop couse 172, 7231 cf. cose coute sf. elbow 4458 coutel sm. edge, blade 7907 [craventer]; [creventer] v.trans. break through, smash 3336; destroy 3505, 5047, 5132, 5319; v.intrans. collapse 2438 creance 1742, 4943, 5102, 5165, 5781, 6793; creanze 4527; sf. (Christian) faith, belief, religion 1742, 4943, 5781, 6793; (Muslim) faith, religion 4527, 5102, 5165 creanter 83, 2539 vb.trans. assure, promise, guarantee 83, 3325; approve, agree to 2539 creant 4158, 6676; creanz 3160, 3431 adj. (thing) worthy of belief 6676; (person) faithful, to be trusted 3431; estre c. be certain, truly believe 3160, 4158 creant adj.used as sm. one who believes, believer 4389, 5981, 7734 creim cf. creindre [creindre] variant [creimbre] pres.indic.1 creim 3540, cren 6667, 6784; pres.indic.3 crient 1359, crent 5346, 5737; pres.indic.6 crement 5641; fut.5 crembrez 5246; past part.f. cremue 1624; v.trans.& (ll.6667, 6784) reflex. fear

Glossary

501

[creire], croire 118, 586, 2090; pres.indic.1 croi 888, 1355, 1382, 1487, 2048, 3323, 3331, 3790, 4172, 5205, 6891, 6911, 7156, 7165, 7658, 8061; 3 croit 3499, 5865, 6671; 6 croient 2342, 5057, 5065b, 5315; pres.subj.3 crot 1912; fut.3 cresra 401; fut.5 cresrez 232, 237, 261; past part. creüz 283; v.trans. believe 118, 232, 237, 261, +; ne douce France crot om a reprover nor people think to reproach France 1912 cremetos adj. fearful 6399 cremuz past part.creimbre used as adj. (feared), redoubtable 634, 1826 crenu 3999, 4828, 7582; crenuz (V7’s reading) 634 adj. long-haired (usually used of horse, i.e. of mane and tail) crestre 6534 cf. crostre creü cf. crostre creventer cf. craventer criée sf. loud lamentation 6133, 6150, 7250 crine sf. hair 7058 [croisir], [croissir] v.trans. break, destroy 2115, 2155, 5603; crush 4478; v.intrans. grate, produce a grating sound, crunch 2155(?), 4061, 4078; break, be broken 2155(?), 4409 croistre cf. crostre croler v.trans. shake 1047, 7631 crope sf. croup, rump 2596, 4939, 6357; crupper, harness over rump (abridgement of cropiere) 1577 cropiere sf. crupper (the band, usually of leather, attached to the saddle, with a ring at the end through which the horse’s tail is threaded) 1858, 4993 crostre 6080; crostr(a)e 1358; croistre 6516, crestre 6534; pres.indic.3 crost 3149, 6548; fut.1 crostrai 6253; fut.3 crostra 7648; past def.3 crut 108, 2432, 6242, 6730; past def.6 crurent 2885; pres.part. croissant 6123; v.trans. increase (person’s importance), enhance (person’s status) 6253, 6516, 6534; v.intrans. increase, grow 108, 2432, 2885, 3149, 6548, 6730, 7648; accumulate 1358; grow, flourish 6080, 6123; creü 3487 past.part.used as adj. increased in size, swollen croute sf. crypt 7219 cuer obj.sing. 579, 597, 957, 1123, 1124, 1135, 1196, 1479, 1569, +; subj.sing. 2188; cuers obj.sing. 1970, subj.sing. 2111, 2225, 2367, 2732, +; cor obj.sing. 3512, 4202, 5931, 6803, 6945, 6980, 7874, 8086, 8106; cors subj.sing. 862; sm. heart 597, 827, 862, 1123, 1124, 1135, 1479, 1617, 1650, 2111, 2167, 2181, 2225, 2367, 2732, 2821, 3104, 3392, 3732, 3877, 3907, 4149, 4342, 4380, 4737, 4851, 4918, 5050, 5121, 5366, 5509, 5538, 5586, 5728, 5734, 5828, 5961, 6001, 6029, 6105, 6132, 6192a, 6194, 6227, 6332, 6451, 6472, 6500, 6519, 6633, 6675, 6823, 6847, 6942, 7034, 7106, 7153, 7253, 7306, 7370, 7534, 7535, 7597, 7874, 8076, 8106; courage, spirit 1196, 3487, 6980, 7095, 7098, 7753, 7847; life, soul 2162, 2188, 2225, 7042, 7245; dire son cuer say what one really believes, speak one’s mind 579, 957, 7622; segur de cuer staunch of heart 1135; de cuer from the heart, earnestly 1569 cui cf. qi cuide s.f. prick 1719; a cuide d’esperon spurring on, with a prick of the spur [cuider], [qider] pres.ind.1 cuide 3331, qide 5349, qit 4875, 4898, 6452, 6711, 7486, 7892, 8084; 2 cuides 5747; 3 cuide 2150, qit 2451, 4101, qide 4444, 4786, 6534, 7248, 7968, quide 6260; 6 cuident 2047, qident 5316; imperf.1 cuidoie 3864; past def.1 cuidai 6180, 3 qida 2712, 7980; v.trans. think, expect, intend 2047, 2150, 2451, 2712, 3331, 3864, 4101, 4444, 4786, 4875, 4898, 5316, 5349, 5747, 6180, 6260, 6452, 6711, 6534, 7248, 7486, 7892, 7968, 7980, 8084 cuir sm. leather 5719; skin 8102, 8126

502

Glossary

cure sf. care; n’aveir c.de not to care 268, 6248; scorn, be unwilling (to do something) 2071, 2161; ja de bataille n’averas jameis cure you will nevermore care about a battle OR you would never have cared for a battle* 2276 cuvert 4205, 5549, 8007, 8175; cuivert 6429; adj. ignoble, wretched cuvert obj.sing. 359, 464, 2159; cuverz subj.sing. 7991, obj.pl. 2049; qivert obj.sing. 4037, 5567, 5945; subj.sing. 2241, 2251, 7557; obj.pl. 5800; qiverz obj.pl. 4193 adj. treated as sm. serf; ignoble wretch 359, 464, 2049, 2159, 2241, 2251, 4037, 5567, 5945, 7557, 7991

d dahé 569; daait 1419; dahait 3320, 7893; dahaiz 4471; dehait 4706, 6399; dehaiz 5412 sm. God’s hate, curse; mal dahé evil curse, affliction 569; d. ait accursed be 1419, 3320, 4471, 4706, 5412, 6399, 7893 dalez 1722, 1746, 2205, 2374, 3698, 4414; dales 6746, 7049, 7269; delez 64, 2346; dalie (?) 5862*; prep. beside, alongside, 1733, 1746, 2205, 2374, 3898, 6746, 7269; par dales from alongside 7049; adv. near by, close by 64, 4414, 5862 Damedé 3885, 4392. Damedeo 4136 [= Dominus Deus] sm. Lord God dan obj.sing. 1665, 3127, 3185, 3413, 5585, 6880, 7340, 7636, 7640, 7709, 7835, 7836, 8035; subj.sing. 4973, 7297, 7470, 7571; danz subj.sing. 6264, 6408, 7450, 7856, 8072; dant subj.sing. 6462 sm. Sir, Lord, Milord (a title of respect when speaking of a third person) 1665, 3127, 3185, 3413, 4973, + dangner sm. danger 56 [dangner], [degner], [digner] v.trans. condescend to accept, accept 8; deign 1473, 1946, 2072 [danpner] v.trans. condemn, damn 523 daumage 372, 409, 555, 1474, 1930, +; damage 5930, 6010, 6216, 6625, 6952, +; domage 2603, 6510, 8064; doumage 2279, 3362, 3411; sm. hurt, injury, harm 372, 409, 555, 1930, 5930, 6010, 6626, 7389; loss (and associated pain) 1474, 2964, 2973, 4216, 6216, 6625, 6952, 7211, 7339, 7344; pain, grief arising from loss 2603, 8064 [daumager], [daumajer] v.trans. injure, hurt 4194, 5797; inflict harm/loss on 4406; sunt si daumajez have had such losses inflicted on them 4802 de cf. deo dé 4083 cf. doit [debriser] v.trans. break, smash 1220, 2810 deça 7942 cf. deza [decipliner] v.trans. punish 8173 deci adv.phr. from here 3654 de ci q’en 599; deci q’en 642, 2326, 3680, 5573; deci q’a 1250, 1902, 3520; de ci en 1333; de ci qe 1064, 2272, deci qe 2644, 4338, 4933, 5071, 5660; des ci q’a 3520; prep. from here to, all the way to 599, 1333, 1250, 1902; as far as, right down to 642, 2326, 2272, 2644, 3680, 4933, 5071, 5573, 5660; (+ time) until 3520; des ci q’a from now until 3520; conj. until 1064, 4338 declin sm. end; ne torgent a declin do not collapse, are not defeated 6077; li jor vait a d. the day draws to its end 8012 decliner v.intrans. draw to a close 186; set (of sun) 6297 dedesoz prep. beneath 3480, 6733, 6763

Glossary

503

dedesus adv. above 7133 dedevant prep. in front of, before 4289, 5327, 7782 dedroi (graphy of detrés, detrois) 7554; adv. behind; per dedroi advl.phr. from behind 7554 defendiment sm. defence, protection 3723 defendre 1365, 1936, 7402, 7423, 7459, 7654; [deffendre] v.trans. defend, protect 21, 1936, 3720, 5107, 7423, 7459, 7654, 7703, 7879; forbid, prohibit 282, 2396, 7678; v.reflex. defend oneself 2010, 2290, 2610, 6281, 7402; avoid 1365; l’autre est defendant the other is actively prohibiting this, i.e. does not allow his opponent any rest 2396; va toz jor deffendant is always active in defence, in fighting any attack* 5023 defeniment sm. the end of the world, Judgement Day 2442; prendre defeniment die 3518 defenie 3011 past part.f., defenir, used as adj. finished, over defension sf. resistance 3263 deffaé 564; desfaé 1632, 1759, 2318, 4037, 4771, 5008; adj. terrible 564; infidel, pagan 1632, 1759, 2318, 4037, 4771; hideous, fearsome 5008 defrenée past part.used as adj.f. unbridled 5718 degner cf. dangner [degoler] v.trans. cut heads off 69 deguerpie past part.used as adj.f. abandoned, desolate 3803 dehait, dehaiz cf. dahé dejoste prep. beside, close by, alongside 131, 547, 3780, 6746, 6749 [delaier] v.trans. keep waiting; aler delaiant keep (me) waiting 6938 delez 2346 prep. cf. dalez delez 64 adv. cf. dalez delivrement adv. immediately, promptly 3485, 3820, 3832, 4776 delivrer v.trans. clear of (other) people, leave empty save for oneself 7101, 7109 deloëz past part.used as adj. (deprived of praise), of evil repute 407 demandament sm. request 7145 demanois adv. straightway, immediately 204, 4746, 5493, 6263 demeine 16; demene 1959, 7434 sm. vassal, liegeman demeine adj. own, very 1463; al roi demene to the very king, to the king himself 7621 demeintenant adv. straight-way, immediately 4390 [demener] v.trans. make (noise) 1676, 2194, 2748; show, exhibit (a strong emotion) 4201, 5038, 6973, 6983, 7789 dementer 4259 v.reflex. go mad with grief, moan, lament 4259, 5941; v.intrans. become mad with pain 2504; dementanz pres.part. lamenting 5626 demise adj.f. melted 5694 (see also peiz) demorament 4860, 7414; demorement 5322 sm. delay demorance sf. delay 1741 demorée sf. delay 1578, 4770, 7236 demorer 2085, 4257, 4263, 4577, 5179, 6552, 7287, 8147; fut.3 demora 7215 v.trans. put off, postpone, delay 2513; v.intrans. delay 972, 2085, 3187, 3937, 4518, 4541, 4577, 5179, 5672, 7287; take one’s time, act slowly 1781; stay, continue 4257; wait, be patient 4263, 7873; stay, be forced to wait, be delayed 4533, 6552, 7215; remain (in existence) 5860; aller… demorant (actively) delay, waste time 501, 4923; v.reflex. delay 7424; infin.used as sm. delay 8147 demostrance sf. proof, sign 5779 demostrement sm. sign 7157

504

Glossary

[demostrer] v.trans. show, display 2513, 4440 dener, denier cf. diner deo obj.sing. 8, 94, 236, 1417, 1806, 2048, 2078, 2830,2870, 3769, 8202; dé obj.sing. 1127, 1508, 3380, 3693, 4527, 6012, 6537, 6906, 6997, 7037, 7387, 8069; Deu obj.sing. 177, 182, 359, 486, 625, 649, 899, 1020, 1435, 1442, 1446, 1461, 1503, 1512, 1530, 1539, 1731, 1740, 1758, 1790, 1848, 1879, 1898, 1909, 1921, 1988, 1991, 2122, 2165, 2291, 2342, 2499, 2557, 2562, 2577, 2589, 2845, 2550, 2875, 2986, 3003, 3020, 3067, 3389, 3455, 3505, 3514, 3615, 3759, 3817, 3826, 3882, 3915, 3987, 4041, 4381, 4389, 4440, 5047, 5059, 5065b, 5106, 5123, 5132, 5234, 5268, 5315, 5319, 5550, 5675, 5679, 5865, 5874, 6035, 6048, 6090, 6102, 6134, 6668, 6671, 6744, 6793, 6818, 6821, 6856, 6938, 7013, 7100, 7187, 7193, 7209, 7234, 7283, 7305, 7734, 7961, 8058, 8066; subj.pl. 4706; Deus obj.sing. 1933, 2950; subj.sing. 108, 142, 472, 489, 496, 521, 569, 865, 884, 919, 992, 994, 1038, 1194, 1195, 1222, 1383, 1421, 1598, 1634, 1654, 1835, 1860, 1861, 1954, 1999, 2195, 2251, 2261, 2275, 2312, 2348, 2454, 2482, 2502, 2553, 2633, 2639, 2684, 2694, 2709, 2723, 2739, 2768, 2770, 2771, 1794, 2842, 2909, 2924, 2937, 2962, 3088, 3127, 3220, 3231, 3282, 3401, 3481, 3497, 3510, 3536, 3573, 3600, 3711, 3734, 3742, 3785, 3794, 3825, 3854, 3911, 3925, 3953, 3968, 4069, 4092, 4100, 4112, 4117, 4181, 4185, 4206, 4208, 4216, 4226, 4255, 4263, 4280, 4301, 4394, 4469, 4556, 4742, 4874, 4893, 4895, 4955, 5058, 5082, 5158, 5198, 5335, 5359, 5371, 5421, 5479, 5638, 5769, 5775, 5778, 5863, 5984, 6005, 6088, 6097, 6122, 6232, 6250, 6296, 6307, 6313, 6358, 6520, 6785, 6863, 7154, 7157, 7165, 7181, 7198, 7534, 7704, 7787, 7800, 7851, 7962, 7981, 8021, 8121; obj.plur. 1268, 4658, 4708, 5332; Deus subj.sing. 356, 366, 434, 1586, 2123, 2402, 2574, 3357, 4471, 6535; Dieus subj.sing. 1631, 1656, 3148; sm. (pagan) god 94, 4471, 4527, 4706, 5234; (Christian) God 8, 142, 235, 1127, 1508, + departie sf. separation 3807; font la d. make them separate, force them apart 375 departir 1863 v.trans. divide, tear apart 6617; v.intrans. depart, set off 1863; leave, get away 1185, 7378; separate, go separate ways 3275, 4998, 5358; depart, die 3932 depicer 114, despecier 3751, depecier 7332, depecer 7781 v.trans. break, smash to pieces 114, 7781; tear to pieces 2520, 7332; tear into strips 3751; tear to shreds 4198 deplentin cf. doplentin deporter 15, 988 v.intrans. amuse oneself, take relaxation depris (graphy of despris) sm. contempt; avoir depris hold in contempt 1386 derasnier 330 v.trans. defend by speech, justify, argue in support of, argue 303, 330, 1083, 7706; v.reflex. justify oneself 7331, 7716 derée sf. fault 4334 des prep. from (local) 881, 1250; from (temporal) 4102, 4117, 5029; des or 5413, des ore 7878 from now on; desq’ from 4913; desc’a to, as far as 2437; des qe, des q’ conj. until 3520, 4192; since, because (OR from the time when, ever since) 6787, 8005 desafrez 2364, desafrée f. 5705 past part.used as adj. with the ‘safre’ torn away, i.e. with the hauberc stripped of its damascened finish, or alternatively of its burnish of yellow (acc.to Foulet) or cobalt blue (acc.to Jenkins) enamel 2364; with the byrnie or cuirasse similarly stripped of its finish 5705 desarci cf. desertie descauz adj. shoe-less 5163 desclavelée past part. used as adj.(f) torn, rent 5721 descolorez 680, descoloré 3354 adj. coloured, discoloured descolorie adj.f. without colour, pale, white of face 7082 desconfie sf. rout, destruction 879 [desconfire] v.trans. destroy, rout, vanquish 4681, 4748, 6493

Glossary

505

desenor sm.&f. dishonour, disgrace 1401, 2502, 4206, 4703 deserité past part.used as adj. bereft, dispossessed 3239 deserte adj.f. ravaged, ruined, deserted 1218, 3228 desertez 2515, desertée(f) 5823 past part.used as adj. damaged, destroyed 2515; made desolate, laid waste 5823 desertie sf. desert; torné a desertie laid waste 5857 desertie 2116; desarsis 2595; desarcies 2795, desarcie 3691, 5446, desarci 5604; past part. of desertir/desarcir/desarsir used as adj. broken, shattered, cut to pieces (used specifically of haubercs and cuirasses, cf. desafrez above) 2116, 2595, 2795, 3691, 5446, 5604 [deservir] v.trans. deserve 757, 3005 desevrer 2208, 4268b v.trans. sever 1635, 3104, 6812; separate, break off (battle) 2208; separate, part 3382, 3853, 5686, 6181; v.reflex. separate, part 4268b desfaé cf. deffaé desfaire 61; desfere 1312 v.trans. undo, bring low 1312; v.intrans. disband, disperse 61 [desfermer] v.trans. open, unlock 7203 desfier v.trans. defy (ie declare private war upon, either an overt announcement or by some act of hostility that puts the two parties on a footing of enmity (Whitehead)) 342, 6411, 6416, 7878 desfigurez past part.used as adj. disfigured (by torture or wounds), made unrecognizable 452, 3946 [desfluber] v.reflex. remove, take off (article of clothing) 390 desfroi sm. trouble, expense; a desfroi with great expenditure of effort (?) 7553 desirance sf. desire 5099 desirer infin.used as sm. wish, desire 4303 [desirer] pres.indic.1 desir 3594, 3603, 3933, desire 5982 v.trans. desire, long for 4773, 5982, 7225, 7237, 8160; greatly love (used as intensifier to stress affection in standard phrase aim e desir) 3594, 3603, 3933; estre desirant, estre desiranz yearn for, long to see again 2453, 4897; be eager, impatient 3630, 5081, 5149, 5540; have a desire, have a longing 3871, 4916, 4980 deslacer 3749, [deslacier] v.trans. unlace, unfasten 3749, 7666 desliez 4411, desliée f. 5475; past part.used as adj. unleashed; free, rapid in movement 4411; faithless, pagan, infidel (free of moral ties) 5475 [desloger] v.intrans. dismantle one’s camp preparatory to moving 1023 desloial 5730, deslial 7878 adj. infidel, respecting neither law nor religion 5730; disloyal, traitorous 7878 desloial sm. traitor 6372, 8107 desloiée past part.used as adj.,f. respecting neither law nor religion 4542 desmailier 2882, desmaillier 3844, [desmailler] v.trans. break through mail (of hauberk), cut through 2174, 2516, 2882, 2939, 3425, 3478, 3670, 3844, 5763 desmesiurément adv. beyond measure 2434 desmesurance sf. excess (of pride) 8004 desmesure sf. excess; a desmesure showing lack of moderation, arrogantly 2274 desmesurez 402, desmesuré 4031 past part.used as adj. excessive, immoderate 402; trop es d. you go too far! 4031 desolée adj.f. afflicted 7927 desor 889, 900, 921, 1084, +; prep. upon 889, 900, 921, 1084, 1138, 1402, 2287, 2375, 2547, 2594, 2642, 2715, 2856, 3031, 3210, 3675, 3701, 3799, 3956, 4139, 4310, 4327, 4358, 4689,

506

Glossary

4764, 4812, 5445, 5761, 5762, 6030, 6183, 6184, 6288, 6392, 7373, 7588, 7625, 7922, 7954, 7976, 7983; above 3391, 5955; over 2921, 7066, 7083 desore adv. ira desore will have the advantage, will be the winner 1305 despecier 3751 cf. depicer xxx desplier 7710, [desploier] past part. desplis* 1387, desplois* 4953 v.trans. unleash, carry out, make 1387; unfold (when the gage symbolised by the folded glove is accepted) 7710; desplois past p.used as adj. unfurled 4953 desramer 6160 v.trans. tear into pieces, rend desrengier 7711 v.trans. fall into confusion [desrober] v.trans. rob 6346 desrochier 2262 v.trans. disengage, withdraw desroi sm. disorder, din; el desroi in the noisy impact 7565; a desroi in haste, hastily; sa parole a desroi speaking rapidly, hurriedly (impatiently?) 8067 [desroier] v.reflex. make an impetuous rush (leaving the ranks), gallop off (from one’s troops) 5507 desronpre 3844, 6160; pres.indic.3 desront 2237, 5915, 6488; past part. desrompue f 2478, desronpuz 2731, 3345; desroz 2795, desrot 5604 v.trans. cut to pieces 2237, 2478, 2730, 2795, 3345, 3844, 5604; tear to pieces 5604, 6488; tear to shreds 6160 dessachier 7998 v.trans. pull off, remove dessirer 8145 v.trans. tear to pieces 4029, 6488, 8145 dester 837 cf. destre destor sm. detour 2343; plur. detailed geography of the country, short cuts 6602 [destorber] v.trans. hinder, prevent 4795 destorbier 979, destorber 2635 sm. trouble, disaster, catastrophe 979, 2635, 4313 [destorner] v.intrans. prevent 684 destre 59, 454, 493, 716, +; dester 837; adj. right 59, 454, 493, 616, 738, 892, 1061, 3278, 3288, 3642, 4104, 4457, 4609, 4626, 4664, 4718, 4789, 4808, 5256, 6611, 6665, 6709, 7984; used as sm. right (hand side): sor destre to the right 1392, 4114; used as sm & f. right-hand one 837, 6692; lo dester de ses ganz the right-hand one of his (pair of) gloves, i.e. his right-hand man 837 destrier obj.sing. 516, 939, 985, 1176, 1671, 1682, 1696,1720, 17338, 1770, 1815, 1827, 1845, 2023, 2100, 2134, 2151, 2200, 2210, 2220, 2232, 2257, 2328, 2466, 2478, 2625, 2645, 2699, 2727, 2741, 2752, 2774, 2847, 2877, 2948, 3740, 4317, 4857, 4939, 5152, 5985, 5994, 6239, 6257, 6259, 6285, 6374, 6392, 6434, 6741, 6871, 6876, 7508, 7605, 7693, 7805, 7916; subj.sing. 3828, 6770; obj.pl. 40, 1396, 1525, 3122; destrer obj.sing. 1976, 2631, 2640, 2691, 4828, 5195, 6405, 6431, 7526; obj.pl. 2400; dester obj.sing. 2008, 6247, 6264; destrers obj.sing. 1788; destriez subj.sing. 3480; destriers obj.sing. 6396; subj.sing. 6356, 6370; obj.pl. 973, 1986, 2084, 2591, 2596, 3182; sm. warhorse destrois 107, 855; destroit 8129 past part.destreindre used as adj. distressed 107, 855; in torment, distraught 8129 destroit obj.sing. 3066, 5153; destroiz obj.pl. 1073, 1702, 2347, 2663, 2669, 5169; sm. defile, narrow valley destroite 2116 cf. destreindre [destreindre] pres.indic.6 destreignent 2905; pres.part. destrengnant 3873; past part.f. destroite 2116 v.trans. harass 2905; torment 3873; hack at, destroy 2116 desver 7300 v.intrans. go mad desverie sf. madness, folly 250, 801, 6013

Glossary

507

desvez 396, 679, 4195; desvé 464, 1753, 4019, 6714 past part.used as adj. out of one’s senses, mad 396, 679, 1753, 4019; enraged, maddened 464; wicked, criminal* 6714 detordre 6159; past part.(subj. form) detors 533, detorz 3038; pres.part. detordant 5936; v.trans. wring (hands) 533, 3038, 5936, 6159 detraire 7296; past part.(subj. form) detraiz 2414; v.trans. tear out (hair) 7296; tear apart 2414 detranzant 4972 cf. detrenchier detrenchier 55, 4311; detranchier 117, 4314; past part. detrencié 1184; detrenciez 3980, detrenchiez 4186, 4398; detrencé 5280; detrenchez 5826, 7392; detranchiez 4624; detranchez 4801, 5796; pres.part. detranzant 4972 v.trans. cut to pieces 55, 117, 1184, 3980, 4186, 4314, 4398, 4624, 4801, 4972, 5280, 5796, 5826, 7392 detrianz pres.part.detrier delaying, putting off (departure) 3100 deu [Deu] sm. 177, 182, 359, + cf. deo deus1 [Deus] 108, 142, 356, + cf. deo deus2 247, 856, 944, 1031, 2253, 2301, 2538, 2570, 2717, 2732, 3990, 3991, 4099, 4427, 4618, 4675, 5338, 5498, 5582, 5631, 5659, 5731, 5905, 5916, 5920, 5921, 6327, 6346, 6362, 6436, 6511, 6833, 7201, 7202, 7263, 7311, 7921, 7924, 8097; deu 1178, 5907, 5974; doi subject. 6747, 7262; dui subject 2486, 3518, 4671, 4672, 4775, 5738, 6333, 6423, 6428, 7267; num.adj. two deus3 2602, 5887, + cf. duel [deveir] pres.indic.1 doi 1021, 1086, 1175, 1924, 3232, 3857, 4195, 6328, 6489, 7714, 7718, 8051; 3 doit 47, 87, 111, 414, 703, 806, 948, 974, +; dot 7630; condit.1 devroie 6097, 7300; 3 devroit 6111; 5 devroiz 2078; 6 devreient 3254, devroient 3852; past def.1 dui 6753; 6 durent 2088, 7357; pres.subj.1 doie 1911, 3 doie 7305, 7832; imperf.subj.3 deüst 703, 813, 2537, 2845, 7697; v.intrans. have to, ought to 47, 87, 111, 414, 703, 806, 948, 974, 1021, 1086, 1175,1384, 1386, 1390, 1490, 1911, 1924, 1968, 2144, 3021, 3232, 3256, 3683, 3857, 4195, 4600, 4945, 5666, 6091, 6162, 6163, 6319, 6328, 6489, 6751, 7455, 7457, 7681, 7714, 7718, 8051, 8103, 8146, +; be about to, be due to 457, 495, 984, 6157, 6888; have to (deserve to) 2413, 3795, 8068 devers prep. towards 554, 1475, 2347, 2663, 2669, 5996; from the direction of 1056, 1395, 3656, 4428, 6700; devers Franzois on the French side 2885; por devers nos on our side 2988; devers al 3442, devers aus 5064 on their side, amongst them devier 3859 v.intrans. die devisament 7123, divisement 7131 sm. arrangement, devising, art 7123; par divisement as (God has) devised; (OR graphy of diviseement, thus ‘quite separately’?) 7131 deviser 5193, 7315, 7405, 8135; v.trans. divide, separate 1559, 7315, 8135; devise, arrange 2320, 5004, 7405, 8137; describe 5193; organize, perform 7866 devison, devision cf. division devore, graphy of devorage or devorement (poetic licence) sm. devouring, tearing to pieces 1298 devorer 8101 v.trans. tear to pieces 8101, 8172 dexor adv. from then on, after that 214 deza 4327, deça 7942 adv. on this side [dïer] v.trans. dedicate, consecrate 4090 digner cf. dangner diner 1089, 1166, 5411, 6344, 7171; dener 2199; denier 6278; sm. the twelfth part of a sol, i.e. a small coin of very low value dis 624, 3402, 3432, 5434; diz 7647 sm or f. day; a toz dis for evermore 5434

508

Glossary

dis subj.sing. 214, obj.plur. 662; diz obj.sing. 1980, subj.plur. 4635 sm. words, sayings, speech 214; mes dis my words, what I say 662; vos diz what you say; mal diz evil report, bad reputation 1980 disme ordinal num. tenth 5088 diva! exclam. Come! 4908 division 1712, 2691, 3135, 6049; devison 6054; devisions 8101 sf. art, skill 1712, 2691; division, separation* 6054, 6049; pieces, morsels 8101; per tel division with specific stipulation 3134 diz cf. dis doi cf. deus doiaire 3553, doaire 5750, 6532 sf. dowry doit 3527, dé 4083, doi 5389 sm. finger 4083, 5389; inch 3527 dol 3927, 7369 cf. duel dolantez sf. sorrow 3948 dolçor 4720, 5998, dolzor 6800 sf. pleasantness, beauty 4720; pleasure, delight 5998, 6800 dolée sf. grief, sad news 6966 dolgee(f) 5481, dolgié 7263 adj. delicate [doloser] v.trans. weep for, mourn the loss of 3395 dolosie sf. sorrow, grief 1347 dolçor 5999, dolzor 6800 cf. delçor [dolir] past part.f. dolie 5862 v.trans. sadden, reduce to an unhappy state doncel subj.pl. 2392, obj.sing. 7914; damisel 7849, 7915; damisals subj.sing. 7870 sm. young man, gently born but not yet of knightly rank; squire doner 85, 151, 329, 949, 992, 1195, 1645, 1957, 2079,2213, 2742, 3854, 3872, 4264, 6180, 7031, 7719; pres.indic.1 doins 794, 931, 6318; imperat. doin 4755; pres.subj.3 doinst 569, 1654, 2275; doint 2771; v.trans. give, grant 42, 65, 66, 316, 371, 431, 443, 489, 503, 566, 569, 594a, 794, 931, 939, 945, 960, 964, 1060, 1099, 1106, 1120, 1122, 1253, 1492, 1575, 1634, 1654, 1836, 1972, 2245, 2275, 2332, 2505, 2545, 2593, 2620, 2690, 2771, 2861, 3273, 3389, 3553, 3688, 3794, 3969, 4038, 4335, 4625, 4755, 4779, 4874, 5053, 5254, 5301, 5403, 4625, 4779, 4874, 5053, 5254, 5301, 5403, 5461, 5536, 5559, 5572, 5982, 6211, 6219, 6238, 6318, 6337, 6385, 6530, 6613, 6908, 7152, 7207, 7441, 7537, 7740, 7741, 7767, 7882, 7890, 7900, 8047, 8195 donoier 1653, 1672 v.intrans. waste time, dally (originally with women) doplentin 1679, 7939; deplentin 4950 adj. of double-thickness chain mail dopler adj. lined, double, of double mail 1661; of double thickness 4992 dotanz adj. afraid 3161, 3619; auqes d. fearing nothing, anticipating no further attack 3623 dotanze 1738; dotance 5780, 5794, 6785 sf. hesitation 1738; fear 5780, 5794; ci a fere d. here’s something with most fearful implications! 6785 doter 1012, 1277, 5199 v.trans. fear 300, 1012, 1131, 1277, 4229, 5199, 5236; v.intrans. doubt (about) 1490; v.reflex. be afraid 7432 dotrinez adj. instructed, taught 3676 dous 72 cf. duel dragon sm. flag, standard 1594, 2838, 5327, 5405; dragon 4412 [dricer], [drezer] v.reflex. stand up 235, 4494, 4817, 4892 droit 952, 2439, 4090, 5160, 6082, 6448, 7320; droiz 236, 261, 313, 377, 430, 1095, 1110, 1173, 2490, 4085, 4432, 7282, 7395, 8123, 8134; adj. just, true, noble (addressing emperor) 236, 261, 313, 377, 430, 1095, 1110, 6448, 7282, 7320, 7395, 8123; upright, standing 952; complete, full 1136, 2439; straight 1415, 5508, 6124; right, lawful, just 1173, 2490, 4085,

Glossary

509

4090, 4432; direct 5160; direct, correct 6082; droite la reine in a direct course 3291; faire droite raison answered his prayer fully 3127 droit 976, 1057, 1620, 3712, 4294, 5105, 5821, 6176, 6199, 6303, 6438, 6835, 7353, 7717 adv. straight 976, 1057, 1620, 3712, 4294, 5821, 6176, 6199, 6303, 6438, 6835, 7353; directly 5105; droit juger judge correctly, judge aright 7717 droit obj.sing. 398, 1759, 1849, 5451, 6444, 7706, 7731, 7755, 7872; droiz subj.sing. 303; sm. right, truth 6444; the justice, rights or truth of (someone’s) case 303, 7706, 7731, 7755; avoir droit to be right 398; have right or truth on one’s side 5451; have right or title to 7872; qe droit leur face de that he should give them the victory over 1759; qe droit li tiegne de that he should have the victory over 1849 droitement adv. uprightly, conscientiously 3967 droiturer adj. true, rightful, just, high-principled 4876; lawful, upright, noble 7721 droiz cf. droit dromon sm. large war galley 910, 4537, 4728 dru 4824, 6475; druz obj.sing. 3350, subj.sing. 1820, obj.pl. 282a, 2914 sm. favourite 4824; friend 282a, 2914; friend, comrade in arms 1820, 3350; beloved 6475 drue sf. favourite (applied to city, thus feminine) 2897 drue adj.f. luxuriant 2479 druerie sf. friendship, affection 2837; love, affection 6613, 6907 druz cf. dru duel obj.sing. 451, 675, 678, 720, 741, 1043, 1307, 2288, 2493, 2712, 2945, 2965, 3129, 3233, 3329, 3352, 3405, 3456, 3481, 3530, 3575, 3729, 3868, 4150, +; subj.sing. 7252, 7648, subj.pl. 7093; dol obj.sing. 3927, 7369; dous obj.pl. 72; deus subj.sing. 2602, 5887, 5919, 6862, 6957, 7250, 7276; sm. (physical) pain, anguish 451, 675, 2288, 3405, 3798, 5015, 5678; (mental) anguish, affliction 678, 720, 741, 3129, 3456, 3853, 4340, 5560, 5562, 5669, 5832; grief, sorrow 72, 1307, 2712, 2945, 2965, 3233, 3329, 3530, 3729, 3868, 4185, 4340, 4376, 4490, 4701, 4821, 4894, 5147, 5173, 5884; calamity, disaster 1043, 3481, 3575, 4185, 5925, 6827, 6840, 6952; expression of grief or anguish, lamentation 2493, 4490, 4498, 5894, 5914; faire duel express grief, make lamentation 3352, 4150, 4260, 5894, 5914 dui cf. deus [duire] v.trans. stroke 257 durer 10, 26, 164, 284, 1019, 2730, 3857, 7829; v.intrans. last 10, 284, 410, 5811; hold out (against), remain (intact) 26, 2267, 2730, 4276; live, survive 164, 1019, 3857, 7829 dus obj.sing. 384; subj.sing. 121, 208, 271, 289, 990, 1107, 1156, 1188, 1214, 3079, 3109, 5016, 5087, 5541, 5689, 5792 obj.pl. 1231; dux subj.sing. 3173, 3267, 6558, 6629, 6899; obj.pl. 8014 sm. duke dusq’ 4913, 7443; dusc’ 5622, 6467 prep. as far as 4913, 5622, 7443; until, up to (temporal) 6467

e eage cf. aage [eferir]; fut.3 efira 378; v.impers. (graphy of aferir) be convenient, suit effrééëment adv. in terror, through fear, precipitously 3713 egaus 5912 cf. igal

510

Glossary

eigue cf. aigue el adv. anything else 979; ne …. el nothing else 165, 6480 elenae cf. alene elme cf. eume els1 sm. 342, 394, 1502 + cf. oil els2 pron. 398, 567, 1379 + cf. aus els3 6058, graphy either of oils/uils sm.pl. oils, ointment OR graphy of euls sm.pl. what one has made, i.e. the ointment she has made [emer] graphy of aimer v.trans. love 2834, 6903 en 3257 cf. hon enarché adj. with arched brow, handsome(?), OR arched, i.e. with legs set wide apart in an upright stance(?)* 7665 enarme sf. strap (of shield) 3065, 7875 enbasmer 7289 v.trans. embalm enbastir 1856 v.trans. put, place enbesognez 2311 adj. busy enbler 6192 hide, conceal enbracer 3754, 3832, 4309, 6885; enbracher 6777, [enbracier], [enbrazer] v.trans. put (shield) on arm 2358, 4309, 6376, 6388, 6405; take into one’s arms, put arms around, embrace 3754, 3832, 6885, 7254; v.intrans. put arms around each other, embrace 6777 enbrasée adj.f. lighted 6149 enbron adj. bowed (of head) 256, 3132; with head bowed, heavy-headed 4494 enbronchier 4891, enbroncher 7705, [enbroncer] v.intrans. lean, bend forward 304; fall forward, sink down 3392; lower one’s head (in dismay or despair) 7705; v.reflex. bow down one’s head 4891; enbronchié past part.used as adj. with head bowed, disconcerted, dismayed * 7679 [enbuscher], [enbuschier] v.reflex. hide, take cover 1177; hide, conceal oneself in a wood 6312 [encanter] v.trans. bewitch, cast a spell upon 4032 encaucer 3739, enchaucer 6402, [encaucher], [enchaucher], [enchocher] v.trans. pursue with the sword 2472, 2810, 3739, 4246, 4284, 4809, 6266; v.reflex. spur forward, press forward* 5463; infin. as sm. armed pursuit 6402 enchantement sm. sorcery, evil arts 7137 enchanteör sm. sorcerer 2388 enchaz 6301, enchals 6302 sm. pursuit, chase enclin adj. bowed (of the head), with head bent 1684, 6066, 6083, 7083, 7935 [encliner], incliner 2014 v.trans. bow (the head) 558, 1580, 1895, 2014, 2025, 2322, 4049; bow down before 1675; bow to (as sign of respect) 2061, 3381; v.intrans. bend over, sag from upright position 7124 enconbré 4051, 7528; engonbrez 933; past part.used as adj. overcome 933; encumbered, weighed down 4051; trapped (by his treachery), tarnished by his guilt, therefore facing execution 7528 enconbrier 2263, 2885, 3028, 3734, 4868, 4873, 5158, +; engonbrier 1654, 2963, 4301, 7720; engonbrer 2960; sm. hurt, damage, destruction 1654, 2885, 2960, 2963, 3028, 3734, 4301, 4868, 4873, 5158, 6730, 6784, 6891, 7330, 7701, 7720; obstruction 2263 enconbrier 7627 v.intrans. place the blame, load the burden (of his guilt) * encontre prep. moving towards, therefore alongside 1367; against, in opposition to 2995, 5839; in competition with 2655; (down) to 3289, 3356; towards 4835, 6963, 6976; in rela-

Glossary

511

tion to, in response to (his appearance) 7468; in relation to 7860; encontre a in opposition to, against 1304; adv. against them, to the attack 4923, 4945; towards them 6992; encontre soi against himself, close to himself, closely 7254 encresmé (adj.used as) sm. evil-doer, vilain* 634 encrisme adj. thorough (used as an intensifier with pejorative words like ‘vilain’) 3199, 7021, 7646 endanpner cf. entanpner endementers qe conj. whilst, as the same time as 4670 endosé past part.used as adj. with (byrnies or hauberks) put on 574, 1033, 1767 endoser 1691, 3117, 4241 v.trans. put on (item of clothing, especially armour) endroit adv. la endroit over there 7561 enfant obj.sing. 2403, 2792, 3085, +; subj.pl. 2608, 4161, 6985, obj.pl. 6859, 6869, 7426; enfes subj.sing. 4355, enfans subj.sing. 6818; enfanz subj.sing. 7416, obj.pl. 1204; sm. child 1204, 2403, 2607, 3085, 4355, 4740, 5027, 6005, 6818, 6859, 6869, 6985, 7416, 7426; young (inexperienced) warrior 2792; in relation to the Biblical story of the Fiery Furnace: young people(?), Jews (as children of Israel?) 4161, 5117, 5971 enficée past part.used as adj.f. fixed 4551 enfoïr 6093 infin.used as sm. burial, interment enforcement sm. violence 7139 enforcié adj. increased, redoubled 7252 engal 1640; igal 8111 adj. equal, of equal status 1640; en igal on equal terms 7877; parti igal a flat area 8111; tot engal completely as an equal 5736; see also igal, egaus sm. flat area 5912, 8153 [enganer] v.trans. irritate, cause pain(?) 7959 engegnement sm. invention, creative power 7132 engeignos adj. wily, treacherous 6395 engigne sf. ruse, deceit, spite 12 [engigner], [engegnier] v.trans. deceive, dupe 1996, 2313 enging 84; engin 6431, 6853; enzing 28; engeing 7584; engiens obj.pl. 280; engeins obj.pl. 3474 sm. (abstract) means, (crafty) device 28, 6853; trickery, fraud, deceit 84, 6431; duplicity 7584; (concrete) (war) machine, projectile 280; arrow, missile 3474 engonbrer, engonbrier cf. enconbrier engrés adj. impetuous, fiery 5290, 5300 [engroisser] v.intrans. become heavy, grow sad 6823 enhaiter 5375 v.trans. revive the morale of, encourage enhaltie past part.used as adj.f. provided with a hilt 1343 enhaudeüre 7990, inoldeüre 2268 sf. hilt (of sword) enlumenez 818, enlumenée f. 3663, 6565 adj. rendered illustrious 818; filled with light, illuminated, brightened 3663, 6565 enmi prep. in the middle of 570, 1640, 2029, 2176, 2779, 3685, 3735, 3747, 3815, 4646, 7355; (cf. also mi) ennublée (f) 6147, ennublés (m.pl) 7035 adj. covered by cloud, shadowed, darkened 6147; clouded (of eyes) 7035 enoier 1668, past part. enuiez 2528 v.trans. irritate, annoy 1668; wear out, exhaust; enuiez too exhausted, too incapacitated 2528 enpaloï past part.used as adj. become pale, pallid 7035 enpanez 683, enpené 4176; enpenee f. 5397, 6687; adj. feathered 683, 5397; covered in feathers; winged(?) 6687; winged 4176

512

Glossary

enparentiez past part. related, connected by kinship; des meus enparentiez those of the noblest family or lineage, the best-connected 508 enparlez adj. fluent in speech, eloquent 654 [enpeindre], [enpendre] v.trans. strike down (especially with the couched lance) 2156; wind (the horn), blow lengthily 3048 enpené cf. enpanez enperaör obj.sing. 847, 1331, 1399, 2849, 4700, 6249, 6379, 6624, 7276, 7596; subj.sing. 2491, 5562; enpereör obj.sing. 2623, 2869; enperere obj.sing. 6161; subj.sing. 23, 111, 196, 233, 236, 256, 261, 291, 313, 324, 357, 377, 430, 460, 468, 493, 505, 550, 570, 596, 868, 948, 974, 1024, 1051, 1071, 1095, 1104, 1110, 1124, 1211, 1464, 1486, 1519, 1965, 1998, 2331, 2656, 2895, 2944, 2969, 3014, 3115, 3178, 3192, 3226, 3241, 3362, 3535, 3548, 3716, 4182, 4201, 4320, 4326, 4353, 4424, 4515, 4896, 4927, 5019, 5103, 5168, 5227, 5487, 5695, 5852, 5914, 5926, 5938, 5993, 6002, 6066, 6117, 6135, 6200, 6439, 6448, 6527, 6756, 6764, 6870, 6914, 6974, 6997, 7037, 7113, 7282, 7311, 7320, 7369, 7375, 7395, 7435, 7450, 7491, 7656, 7689, 7730, 7750, 8033, 8045, 8055, 8074, 8082, 8095, 8123, 8134; enpereres subj.sing. 68, 168, 4239, 6152, 7827, 7839; sm. emperor enpere 4995, enpire 6483 sf. army of vassals 4995; gathering of vassals, court 6483 enperere cf. enperaör enpirer 7776, 7938,7970 v.trans. damage 7970,7938; hurt, wound, injure 4807; v.intrans. grow weak 7014; be damaged 7776 enpires sf. rotting, putrefaction 7286 [enplir] v.trans. (fig.) fulfil, carry out fully 432 enprés prep. after 1647, 1919, 6650, 6768; adv. afterwards 7151 [enprendre] v.trans. follow 1390; undertake 7888 enrajer 3840, 7946, [enrager] v.intrans. go mad 2965, 7946; rage, go mad with grief 3840 enragez 2306, 4195, 4412; enragé 8170; past part.used as adj. mad 2306; enraged, raging 4412; mad with grief 4195; mad with rage, wildly excited 8170 ens cf. enz enscanble 120, 545, 4829, 5420; anscenble 124; enscenble 147, 2411, 1412, 3402; ensanble 3186, 3237, 3271, 4840, +; adv. gathered together 1412; together 3402, 4840, 6233, 6777; all together, as a group 6482, 7662; enscanble o, anscenble o, enscenble o, ensanble o together with, alongside, in the company of 120, 124, 147, 545, 2411, 3186, 3237, 3784, 4829, 4887, 4969, 5420, 5576, 6115, 7043, 7650, 7660; together with, as well as 3271; with 7195 enseëlez past part.as adj. sealed 453 ensegnez, graphy of ensaigniez 2529 past part.as adj. covered in blood 2529 enseigne 1029, 1291, 1622, 1851, + sf. standard, banner 4, 1029, 1622, 2113, 2221, 2345, 2676, 3121, 5365, 5367, 5383, 5389, 5406, 5424, 5692, 5715; lance (with banner fluttering at top) 1291, 2238, 5447; battle-cry 1649, 1851, 2077, 2463, 2867, 3176, 3303, 3348, 4229, 4373, 5186, 5505, 5708 enseigner 136, 8124; ensegner 8136; v.trans. point out, designate 327; instruct, tell (who he is) 136; tell, inform, advise 6310, 8087, 8124, 8136 enseigner infin.as sm. education, instruction 6720 enseignez 2294; ensegnez 2529; ensegnié 7257; adj. well-trained, skilful 2294, 2529; learned, well-informed 7257 ensemant 1442, 5968; ensement 3517, 4696, 4854, 5127, 7416; insement 3531 adv. in like manner, likewise 1442, 4854; also, likewise 3517, 3531, 4696, 5127, 5968, 7416 enserer 1702, 6109 v.trans. drive (into), force in, shut in 1702; fasten, put 6109

Glossary

513

ensoine sf. delay, hindrance 5086 ensor prep. upon, on top of 6353 ensus adv. apart, to a distance (of) 3070 entalenté adj. eager 2923 entaliez 4627, entalié 7668, entallée(f) 4550 adj. engraved, chased 4627, 7668; carved, embossed 4550 [entanpner], [endanpner] v.trans. penetrate, bite through 1055; damage, impair (something perfect) 1141 entencion sf. intention, purpose, design; par bone entencion with good intention, to good purpose, with goodwill 1609, 1730, 2564, 6059; par fiere entencion with fierce intent 2568; por voire entencion in real earnest, to real purpose 3912; sans fausse entencion without false intent, without deceitful design 3791; selonc sa entencion according to his wishes, in accordance with his desires 6048 entendance sf. thing to hear, plan 4944 entente sf. accord, agreement 885; fera male e. will attack with bad intent 2775 enters (subj.sing.) 8102, entere (f) 4991 adj. whole, entire 4991; whole, unbroken 8102 entoilement sm. entanglement, involvement, scheming (deceit) 31 entor prep. around 630, 2017, 4121, 6150, 7773, 7995; adv. environ e entor, entor e environ round about 1236, 1591, 1709, 3775 [entorner] v.reflex. turn aside, go away 8039 [entrabaiser], [entrebaisser] v,reflex. kiss one another, embrace 936, 6977 entrals adv. then, at that moment 7246 entre prep. amongst, according to 106; amongst 301, 2249, 2866, 3978, 4000, 4937, 5501, 5960, 6618; between 916, 1048, 1220, 6647, 7263, 7956; shared equally between, thus together 5068, 5074; within 2996, 2999, 6689, 6696, 6704, 7026, 7032, 7074, 7096, 7113, 7121 (** entre- as a prefix to reflexive verbs indicates reciprocal or simultaneous action; with other verbs it may indicate partly completed action.) [entrebatre] v.reflex. strike one another; jus s’entrebatent they strike one another to the ground 5725 [entreferir] v.reflex. strike one another 5738, 7564 entrelaisier 6578, [entrelaisser] v.trans. hinder, delay 6578; v.reflex. gallop forward, charge towards one another* 7899; s’entrelaissent cosre (both) let gallop, i.e. set (their horses) to gallop towards one another 7880 [entreprendre] v.trans. seize, take 4107; entrepris past part.used as adj. attacked, surrounded, in a difficult position 3977; surprised, disconcerted * 632 [entroblier] v.trans. forget (for a certain time) 252 [entrocir] v.trans. kill 2; v.reflex. kill one another 3091 entrois adv. at that moment * 790 enuiez cf. enoier enuit adv. at night, during the night 6636 envaïr 1063, 3591; past def.3 envaï 2674; past part.m. envaïs obj.pl. 2592, subj.sing.* 3464; v.trans. attack, assail 1063, 2592, 2674, 3591 envaïe obj.sing. 5243, envaïes obj.pl. 2793 sf. attack 2793, 5243 envers adv. down backwards, on his back, face upwards 7912 envers 1187, 1728, 1871, 1950, 2517, 2697, 2891, 6167, 6235, 6681, 6947, 7097, 7112, 8058; enver 7213; prep. towards, in the direction of, against 1187, 1728, 1950, 2517, 2697, 2891, 6167, 6235, 6681, 6947, 7097, 7112, 7213, 8058; to 1871; (in battle) against, from (?) 2849

514

Glossary

environ 16, 573, 1236, 1591, 1596, 1709, 1927, +; prep. around 16, 573, 1927, 7501, 8091, 8111; adv. around, round about 1236, 1591, 1596, 1709, 2362, 2762, 3662, 3775, 3905, 5330, 6202, 7763, 7819; (See also entor) enviz sm. ad enviz against one’s will; ad enviz o de gré whether I would or no, whether it suited me or not 8053 envolse past part.envoldre used as adj.(f) wrapped, enfolded 7268 enz 454, 1054, 3093, 3796, 4278, 4840, 5439, 5256, 5573, 5997, 6271, 6312, 6759, 7138, 7684; ens 4292a adv. in, within 4292a; enz ou in 454, 1054; enz es in 3093; enz el in 5439, 5573, 6271, 7684; enz en in 3796; on 4278, 7138; into 4840; inside, within 5256, 5997, 6312, 6759 enz (graphy of einz) adv. rather than this 1463 enzing cf. enging erance cf. esrance erbor 1392, erbos 6397 adj. grassy erer cf. esrer erité sf. heritage, inheritance, lands 4531 errer cf. esrer error sm. metre en error cause anguish, deceive 4219 ersoir 6412, 6443; ersor 6810 adv. yesterday evening es 2022; es les vos 309, 3382; es vos 2013, 2495, 4461, 4905, 5612, 7547, 7603, 7707; ez vos 905,1269, 2749, 3264a; ez vus 1803; este vos 4878; imperative here is; behold; ez li un angle there appeared to him an angel 4258 esauchier 2561; esaucer 3826, 6102; esaucier 6534; [essaucer], [essaucier] v.trans. exalt, increase the honour of, make more illustrious, give greater power to 2561, 3826, 4957, 5485, 6534, 7261; past part.used as adj. honoured, exalted 2929, 6319, 7683 esbahi 5620, 6423; esbaïs 2600; esbaïe 3063, 7199; past part.used as adj. dismayed 2600, 3063, 5620; discouraged, afraid 6423; cast down, crushed 7199 esbanoier 127, 6869 v.intrans. amuse oneself esbaudir 2722, 5591, 6100 v.trans. encourage, incite, put spirit into 2722, 6100; v.reflex. become excited, rejoice 4610, 5591; esbaudi 6404, esbaudiz 1979, esbaudis 2601 past part.used as adj. bold, audacious 799, 1979; fierce, ardent 2601; delighted 6404; encouraged, full of joy 7195; rejoicing, delighted 8013 escachier sm. chess-board 128 escaigne sm. substitute, replacement, someone to replace him 1222 [escanper] v.intrans. (thing) slip from grasp, escape from grasp 457, 1101; (living creature) escape, get away 1056, 3027, 6024 escarboncle sf. carbuncle (bright red gem-stone) 930, 2858, 7977 escaufé adj. (fig.) boiling 477 eschac sm. booty, spoils, reward 4920, 5237 eschaper 4274 v.trans.&intrans. escape 3135, 4274, 4811, 4883, 5168*, 6243, 63222, 6330, 6458, 6661, 7984 eschargaite sf. guard, sentry 4336 eschelle 5018, 5032, 5417, 5477; eschele 5044, 5075; eschile 1407, 5056; eschille 5026, +; sf. division (of an army) 2540, 2571, 5003, 5018, 5026, 5032, 5043, 5044, 5075, 5229, 5263, 5286, 5292, 5305, 5342, 5352, 5417, 5477 escheoir v.trans. be allotted by fate, be granted, belong 6509 escheri adj. few in number; a masnie escherie, a masnée escherie with a small household of servants, with a small entourage of attendants 244, 1042

Glossary

515

eschevi adj. elegant, slim 2024 [eschevir] v.trans. complete, execute completely, fully carry out, fully administer 7848 eschine 3301, 5270; schine 2328; sf. spine, backbone eschis adj.(subj. form sing. of eschif) hated, cursed 625; cut off, separated 3232 eschorcer 8125 v.trans. skin, flay (alive) esciant 348, 809, 826, 845, 1113, 1487, +; escïent 3317, 5048; escianz 3152, 3450; sm. knowledge; par lo men esciant 348, 1113; par le mien esciant 5534a; per lo mien esciant 2980, 6672; per lo mien escïent 3317; per lo mien escianz 3450; to my knowledge 348, 1113, 2980, 3317, 3450, 5534a, 6672; mon esciant, men esciant, mien escïent from what I know (of the matter) 809, 826, 845, 1487, 5048; son esciant to his (certain) knowledge, as he well knows 1966; a esciant, a escianz, a escïent truly, certainly 3152, 4172, 5110, 6298, 6941, 7732 escïentre 3174 sm. good faith, wisdom; per lo mien escïentre by my faith! 3174* esclache sf. subj.pl. splashes of blood 3356 esclairer 4894; esclarier 7106; v.trans. relieve 4894; v.intrans. become lighter 7106; v.reflex. relieve oneself of, rid oneself of 5800; esclairée past part.used as adj.f. relieved, comforted, in lighter mood 4561 esclari 1377; past part. of esclarir (graphy of esclairir) used as adj. bright, shining 1377 esclartissement sm. brightness, shine (of horns and bugles)* 5131 esclicez adj. cut to pieces, divided, dismembered 2304 esclos subj.pl. 6267, 7552, obj.pl.6339 sm. prints (especially of horse’s hooves), tracks 6267, 7552; toz les esclos by the tracks made by his horse 6339 [esçoir] v.trans.(graphy eschivir) elude, escape 881 escoler adj. well-informed, learned 740 esconser 187, 4250 v.trans. hide 6802; v.intrans.&reflex. sink down, set (of sun) 187, 4250 escordablement adv. fervently 5106 escordement adv. fervently 5774 escot sm. what one owes, reckoning, bill 8081 [escremir] v.intrans. fence 129 escroisir v.intrans. break, be broken 2465 escuer 7621; esqïer 7617; sm. squire 5817, 5950, 7603, 7617, 7621, 7712, 7728 esfor obj.sing.1422; esforz obj.sing. 5226, 5260, 5261, 5281, 5284, 5406, 7422; subj.sing. 3037; sm. armed force, army, 1422, 3037, 5226, 5260, 5284, 5406; strength, power (?) 5261; par esforz in force 7422 esforcement sm. forces, army 3725; strength, power 5303 esforcer 2658, [esforzer] v.trans. force 5871; v.reflex. drive oneself, make an effort, struggle 3874 esfrée adj.f. frightful, hideous 5387 esfroi 7553, 8066; esfrois 786, 848; sm. fear, perturbation 786, 848; noise 7553; ad esfroi noisily, with a thundering of hooves*; en esfroi in fear, very anxious 8066 esfroi 7547 (variant graphy of esfrei) sm. shout, uproar, tumult 7547 esgarez 414; esgarée (f) 1625, 6693 adj. misled, misguided 414; strayed, gone astray, far from 1625; lost, bewildered 6693 esgarder 3863, 4364, 5190, 6185, 7063, 7071, 7632; esgardé poet.lic.for esgarder 3366; esgardier 4357; pres.subjunc.3 esgard 6223; v.trans.& intrans. look, look at 395, 463, 603, 682, 760, 953, 1403, 3487, 3522, 3655, 4028, 4357, 4364, 5190, 5525, 5675, 5717, 5990, 6185, 6223, 6352, 6892, 7063, 7248, 7632, 7751, 8043; see, have power of sight 3366, 3863, 3941, 7071

516

Glossary

esgart sm. counsel, judgement 8121 esgrater 7059 v.trans. scratch with one’s nails [esgratiner] v.trans. scratch 7084 [eshaudier] graphy of esalcier, essaucier v.trans. rejoice 7182 [esillier] v.trans. exile, have (someone) killed 6877 esilz sm. ruin, torment, distress 215 esjoïe past part.used as adj.f. full of joy, delighted 6546 eslais 1830, 4940, 7808; eslois 844 sm. bound forward, spring forward 4940, 7808; a l’eslais with all his force, at full speed 1830; d’eslois (lit.) at a gallop, at full speed, (fig.) in haste, impetuously 844 eslaissier 2017 v.reflex. spring forward to attack, gallop forward, charge 2017, 2199 eslaissiez 2315, 6265; eslaissez 4418, 5798 past part.used as adj. bounding forward, eager, impetuous, 2315, 6265; unleashed, leaping forward at full speed 4418; charging, galloping forward 5798 eslecéer 6877 v.trans. give pleasure to, rejoice esleger 1521 v.trans. pay for esliction sf. choice, decision 6055 esliz 205, 1170; eslis 2604 past part.used as adj. elect, distinguished, excellent, most important eslois cf. eslais [eslonger], [eslongner] v.trans. move away from, leave 4554; v.reflex. move away from 6948 eslongié adj. distant, distanced, away 6311 esmaiance sf. fear 8008 esmaiant adj. in a state of dismay, troubled, afraid 494, 4924 esmaier 37, 2012, 6091, 6721, 7610; v.trans. dismay 37; v.intrans. be dismayed, be troubled, worry 2584, 6721; v.reflex. be dismayed, be troubled, be afraid 1299, 1339, 3604, 6091, 6752, 6806, 7185, 7206, 7610 esmal sm. enamel 5510, 5740, 7883 esmer 991, 1549 v.trans. count, estimate 991; imagine 1549 esmeraus sm. emerald 5901 esmeré adj. pure, refined 959 [esmoveir] pres.indic.6 esmoevent 2696; past part. esmeü 4008 v.trans. set in motion, commence [esmoldre] past def.6 esmolurent 7965; past part. esmoluz 2905; v.trans. sharpen, whet; past part.used as adj. sharpened, sharp 2905 [esorber] v.trans. blind 3653 [espandre] pres.indic.3 espant 2299, 6643; v.intrans. spread forth 2299; spread, cover the ground 6643 [espavanter], [espouënter], [espoänter] v.reflex. be filled with fear 1436, 2441, 5233, 5469; espoëntée past part.used as adj.f. filled with fear, terrified 2334 (cf espoënti 5606) [especier] v.trans. break into pieces, cut to pieces 3476 esperiz 1985, 1989; espiriz 6063 sm. (immortal) spirit, soul 1985, 1989; Sainz Espiriz the Holy Spirit 6063 [esperjurer] v.reflex. perjure oneself 7817 esperonée sf. distance covered by a horse when spurred 1771 espés 1662, 5656; espois 5494 adj. thick 1662; el plus espés at the thickest point, in the thick of the struggle 5656 espessier v.intrans. become clouded, become turbid, grow confused 3745

Glossary

517

espié 971, 1176, 1463, 1525, 1847, 2151, 2155, 2228, 2243, +; espi 1247, 2098, 2117, 2171, 2233, 2426, 3328, 4880, 5579; espiez obj.pl. 1415, 1538, 2000, 2365, 2383, 2465, 2522, 2904, +; sm. spear, lance 971, 1176, 1247, 1327, 1415, 1463, 1525, 1538, 1825, 1847, 2000, 2002, 2098, 2117, 2151, 2171, 2228, 2233, 2243, 2365, 2383, 2426, 2465, 2522, 2591, 2630, 2729, 2881, 2904, 3091, 3180, 3191, 3328, 3473, 3483, 3509, 3672, 3957, 4409, 4880, 5094, 5192, 5401, 5427, 5439, 5579, 5602, 5805, 6410 espirver 6779; esprever 6817 sm. sparrow-hawk esploiter 3769, 6528, 6586 v.trans.&intrans. carry out, accomplish, achieve 591, 3769; hasten 4566, 6495; hasten to accomplish 6528; fare, succeed 6304; pensez de l’esploiter focus on accomplishing this 3769; look to accomplish this speedily 6528, 6586 espoëntée 2334 cf. espavanter espoänter cf. espavanter espoënti 5606 adj. terrified [espoërir] v.trans. frighten, terrify 6919 espoir sm. judgement; mal espoir terrible (final) judgement 5346 espois (graphy of espoirs, poetic licence for espoir) 6260 sm. a espois with hope of success, with hope of winning (through?) espoisse sf. thicket, wood 6763 [esposer] v.trans. marry 6570, 6968 espris adj. lighted 6149 [esqarteler], [esquarteler] v.trans. quarter, cut to pieces 2223, 2924, 2937 esqïer cf. escuer esrager 5143; [esraiger] v.trans. pull out 5143; tear out (of socket), pull off (or possible variant graphy of esrachier = tear off) 6710 esramant 1245; esrament 2526, 4118, 4672, 4677, 4842, 5104, 5305, 5373 adv. forthwith, straight-way esrance 4938, 5174; eranche 6798 sf. distress, perturbation 4938, 6798; turmoil, terror, confusion 5174 esrant 1107, 1434, 2451; ?iranz 311?*; adv. immediately, at once esrant 5230, 6271, 6274, 6286, 6290; iranz 311*; adj. travelling 5230, 6271, 6286, 6290; moving quickly, running 311; garzons esrant travelling servant, spy 5230 esrer 20, 80, 199, 4244, 6197; erer 80; errer 1009; v.intrans. act, proceed 20, 199; travel 623; depart, sail away 1009; del erer! infin.used as sm. To horse! Forward! Proceed! 80; penser del esrer think of departure, of getting away 4244; n’i ot qe del esrer there remained only to depart 6197; en esrant pres.part. whilst wandering, whilst swimming about 4164; venir esrant come galloping, come at high speed 6274, 6662 essaucié cf. esauchier essiliez 4634; esilliez 4821 past part.used as adj. exiled, put to death essir cf. issir establer 188 v.trans. stable (horses) 188; establish, set up in camp 1031 [establir] v.trans. establish, draw up, set up, decide 802, 5263, 5286; a bataille establie in fixed battle-order, in pitched battle 876 estajes obj.plur.(graphy of ostages, found 69, 175,+ ) sm. hostages 51 estaje sm. en son estaje on his feet, standing 7488 estajier sm. temporary structure, trestle (?); person occupying (the shroud)(?)* 7054 estal sm. remaneir a estal remain in position, hold firm 1480, 1958; prendre estal remain in fixed position, take up a stance (ready to fight) 5508; stop, break journey 6215; aveir

518

Glossary

estal have resistance, withstand, resist 5514; en son estal firmly to his feet, solidly erect 5731 estaler 195; v.reflex. take up position for a fight estant 4727; estanz 4282 adj. present, there estant cf. ester estaqier 7996 v.trans. attach, tie estature sf. stature 2269 esté sm. en son esté standing upright, erect 952, 1125 estelles cf. estoile [estendre] v.trans.& intrans. stretch out, thrust forward 4111; v.reflex. reach 6314; stretch out 5370, 7125 estendu 2170, 2176, 3998, 5768, 7980; estenduz subj.sing. 2733, obj.pl. 3346; estendue f. 5846 past part.used as adj. (fully) extended 2170; stretched out (on the ground) 2176, 2733, 3346, 3998, 5768, 7980; levelled to the ground, razed to the ground 5846 estens 5684; adj. upright in stance ester 140, 158, 1005, 4256, 4262, 4744, 6182, +; pres.ind.3 esta 376, 980, 1052; imperat. estez 5, 1418, 1540; past def.1 estui 1009; 3 estut 3535, 6892, 7029, 7488; esta 2616; pres.part. estant 3874, 4386, 4594, +; estanz 3618; past part. esté 2, 315, 4451, 4516; esteü 4017; v.intrans. remain, stay, 2, 315, 158, 1005, 1009, 4017, 4451, 4516, 6182, 7029, 7073; stand, take up one’s position 140, 1418, 1540, 4262; stand still, stop, halt 980, 3535, 4256; abide, live 1052, 2616; stand 7488; v.reflex. take one’s position, stand 376; stand still, halt 6892; en estanz present 3618; en estant present, there 7466, 7725; on his feet, upright 3874, 4386; on their feet, standing 4594; to his feet 6117; laisser ester let be, cease to speak of 4744, 6445, 6924, stop thinking of, turn one’s thoughts from, abandon 7283, 7344 ester 4892, 5126 esters 7784 cf. estrier esteü cf. ester estoile subj.sing. 5851; estoilles subj.pl. 5923; estelles obj.pl. 5969 sf. star estonez adj. stunned, dazed 5534, 5542 estor sm. battle, engagement, assault 4, 1239, 1667, 1911, 2197, 2207, 2307, 2601, 2652, 2726, 2735, 2846, 2871, 2880, 2958, 3423, 4305, 4438, 5024, 5544, 5594, 5733; de ma boche … estor geter via my mouth (ie by sounding the horn to summon assistance) …… throw away the chance of battle (between the French rearguard and the Saracen army) 1911; estor chanpal pitched battle 5733 estorage sm. building up, renewal, replenishment 3361 [estordre] v.intrans. twist away, escape 2501 estorer 6155 v.trans. build 6155; create 7871 estormis subj. 704; estormie f. 747, 5661 past part.used as adj. easily frightened, discomposed 704; stirred up 747; overwhelmed with grief, overcome with fear 5661 estormie sf. loud noise, din, violent tumult 3708 estot adj. bold, brave, fiercely proud 556; fierce, violent 1755 estoutie sf. recklessness, daring, audacity 745, 793, 3006, 6913; dash, reckless courage 1852, 2836; (foolish) lack of care, unconcern 4491, 5655 [estoveir] pres.indic.3 stuet 136, estuet 1189, 1521, 1862, 2740, 3865, 3932, 6722, 7107; estoet 2002; pres.subj.3 estoïsse 6115; fut.3 istovra 380, estovra 1923, estora 6103, 7301; past def.3 estut 6755; v.impers. behove 136, 380, 1189, 1521, 1862, 1923, 2002, 2740, 3865, 3932, 6115, 6722, 6755, 7107, 7301; be incumbent upon 6103 estraignement adv. strangely, in amazing numbers (OR in serried ranks?) 5304

Glossary

519

estraier adj. abandoned, rider-less 2883 estre prep. estre ma volunté against my will, contrary to my wishes 8052 estre 414, 580, 703, 710, 806, 813, 823, 1248, 1268, 1384, 1390, 1437, +; fut.3 sera 66, 90, 97, 228, 296, 303, 308, 355, 421, 444, 449, 452, 668, 724, 836, 838, 878, 1074, 1132, 1141, 1150, 1251, 1263a, 1482, 1576, 1628, 1637, 1876, 1887, 1902, 2405, 2603, 2991, 3719, 4724, 4772, 4787, 4926, 5638, 5824, 5897, 5899, 6025, 6026, 6332, 6540b, 6800, 6932, 7378, 7460, 7658, 7674, 7924, 8011, 8180; ert 161, 183, 227, 230, 281, 287, 334, 356, 406, 447, 467, 539, 621, 819, 828, 845, 855, 869, 870, 877, 935, 950, 961, 970, 1218, 1283, 1293, 1294, 1345, 1347, 1348, 1429, 1450, 1629, 1638, 1748, 1755, 1801, 1998, 2772, 2900, 2901, 2902, 2979, 3012, 3013, 3024, 3038, 3161, 3162, 3489, 3552, 3556, 3579, 3612, 3639, 3665, 3703, 3729, 4036, 4096, 4223, 4605, 4634 (2), 4669, 4698, 4734, 4768, 4814, 5064, 5101, 5485, 5898, 5901, 5949, 6254, 6293, 6366, 6399, 6494, 6821, 6862, 6866, 7042, 7411, 7420, 7817, 7818, 8100, 8111, 8147; 6 seront 581, 904, 1481, 1883, 1960, 4009, 4624, 5007, 5013, 6830, 7537; erent 769, 1251, 5005; imperf.3 ert 14, 123, 282, 557, 628, 629, 631, 1044, 1183, 1310, 1630, 1974, 2406, 2731, 3689, 3700, 4000, 4074, 4077, 4119, 4250, 4290, 4321, 4331, 4332, 4354, 4426, 4453, 4495, 4533, 4559, 4633, 4750, 5085, 5088, 5158, 5343(2), 5345, 5387, 5481, 5717, 5808, 5813, 6145, 6514, 6643, 6644, 6651, 6701, 6705, 6706, 6748 (2), 7081, 7141; 6 erent 1595, 3824, 4166, 4308, 4712, 4903, 6759, 6884, 7260, 8065; past def.2 fuz 2734, 3340, 3420; 3 fust 7566; 5 fustes 538, 3227, 3841, 3866, 4620, 6579, 6968, 7165; imperf.subj.1 fuse 7235; 3 fust 684, 815, 6549; 4 fuisons 3653; 5 fuisez 4227; v.intrans. be estrée sf. highway; en l’estrée en route, on my way, moving forward 5400 estregne sf. gift 1292 estreigne adj. foreign 1457, 4231; strange, extraordinary 2343; exceptionally fierce, closefought 2660 [estreindre], [estrendre], [astrendre], [antraindre] pres.ind.3 estreigne 2351; pres.ind.6 estregnent 1732; astrengent 2180; past def.3 antrai (final t omitted for rhyme) 892*; pres.part. estregnant 5998; past part. estreint 2566; v.trans. grip, clasp 8923; clasp, embrace tightly 5998; hold firmly, restrain 2566; v.reflex. strive one’s hardest, make a great effort, steel oneself 1732, 2351; come to grips with one another, become closely enmeshed, thus become involved in hand-to-hand combat 2180 estrier 517, 5196, 5986, 7362, 7807; estrer 2092, 7509; estré 4832; ester 4892, 5126, 7784; sm. stirrup 103, 517, 1787, 2092, 3407, 4832, 4892, 5126, 5196, 5518, 5986, 6362, 7509, 7784, 7807 estris adj. 1376 cf. estrois estris sm.subj. of estrif, estrit noise, combat, violence 2602 estrivere sf. stirrup-strap 4994 [estroër], [ostroër] v.trans. make holes in, pierce in many places 2246, 2755, 3669, 5668, 5704 estrois 789; estris 1376 adj. close, confidential 789; in close array 1376 estroitement adv. straight-way, immediately 2031; in close array 3302 estrot adj. narrow 2290 estuet cf. estoveir esvel sm. mettre en mal esvel place under stress in a desperate, threatened position, (OR bring to a bad awakening, i.e. to death?) 8185 eume obj.sing. 928, 1140, 1580, 1651, 1665, +; heume obj.sing. 1768, 1811, 1856; elme obj.sing. 5181, 7770; haume subj.pl. 5380 sm. large helmet 928, 1033, 1140, 1372, 1405, 1414, 1537, 1580, 1651, 1665, 1684, 1693, 1712, 1735, 1757, 1768, 1783, 1811, 1823, 1842,

520

Glossary

1856, 1895, 2014, 2025, 2054, 2297, 2325, 2397, 2477, 2620, 2642, 2715, 2787, 2924, 2937, 3118, 3132, 3335, 3369, 3392, 3477, 3691, 3749, 4010, 4020, 4039, 4310, 4362, 4929, 5001, 5181, 5380, 5468, 5519, 5526, 5552, 5581, 5648, 5739, 5761, 5786, 6426, 7505, 7768, 7770, 7798, 7935, 7937, 7954, 7969, 7976 eür sm. good fortune, happiness 5258 eürée adj.f. bone eürée happy, blessed 6136 evage adj. dwelling beside the water; i.e. (of the Saracens) who dwell around the Mediterranean (the Levant, North Africa and Spain: possibly used pejoratively as a reference to Arab pirates) 6512 eve cf. aigue evos adj. of the water, of the sea; porcel evos porpoise 5271 [evertuer] v.reflex. revive oneself 4055 ez cf. es

f faille 1467 cf. falie faillir 2581, falir 3592, fallir 6213 v.trans & intrans. fail 2581, 3592, 3700, 4528, 4705, 5414, 6213, 7045; with cuer (heart) fail, become weak 3877, 4149, 4737 faire 118, 589, 834, 1175, 1298, 1379, 1490, 1736, 2389, 3069, 3508, 3557, 3876, +; fere 4236; pres.indic.1 faz 4755; faiz 918; 6 funt 2075; pres.subj.3 fac’ 1477; 4 fazon 2562, fazons 4514; fut.4 feron 1262, 1930, 2553; ferons 1328, 2019, 2068, 2162, 2671, 3625, 5948, 8133, 8154; condit.1 feroie 522, 5758; 5 feroiz 8115; past def.1 fis 3403, 5610, 6281, 6443, 6585; 2 fesis 3403, 7130; 3 fist 17, 35, 100, 139, 143, 192, 242, 250, 474, 649, 924, 977, 990, 995, +; 4 faïsme 3312; 5 feïstes 1607, 3374, 4170, 5965, 5968, 6054, 6057, 6072, 6383; 6 firent 1007, 2573, 3151, 3891, 4704, 5140, 6094, 6583, 7110, 7136, 7287, 7293, 7329, 7589, 7965, 8027; fisent 1533, 2865, 7289, 7358, 7823; v.trans. cause (s.thing to happen) 17, 35, 100, 118, 139, 192, 530, 834, 847, 977, 990, 1176, 1379, 1668, 1736, 2162, 5610, 5905, 6165, 6486, 6510, 6583, 7289, 7293, 7329, 7358, 7589, 7815, 7823, 7965, +; do, act 242, 250, 522, 846, 924, 1007, 1101, 1129, 1175, 1188, 1442, 1477, 1490, 1562, 1781, 2069, 2553, 2865, 3069, 3152, 3374, 3508, 3557, 3625, 3876, 3891, 4620a, 4704, 5140, 5221, 5440, 5749, 5752, 5868, 5948, 6094, 6383, 6401, 6480, 7110, 7287, 7985, 8027, 8066, 8103, 8133, 8154, +; make 143, 474, 589, 649, 768, 995, 1298, 1328, 1533, 1758, 1828, 1898, 1933, 2019, 2068, 2389, 2671, 4170, 4260, 5046, 5610, 5905, 5965, 5968, 6054, 6072, 6127, 6476, 7136, 8089, +; take part in, give (battle) 1262, 1748; grant, give (absolution, pardon) 1607, 1731, 6057; fere vostre gaaigne profit from it, take advantage 4236 fais obj.sing. 1354; fes obj.pl. 5599 sm. burden, load 1354; pain, trouble, difficulty 5599 faitis adj. beautifully made 712 faiture sf. (handsome) figure, personable quality 8044 falance sf. sanz falence indubitably, certainly 4935, 5170; without hesitation, confidently 6790 falie 795; faille 1467; fallie 5872; sf. fail, lack, failure, doubt; fault; (break?, interruption?) sanz falie 795, sanz fallie 5872 without fail, certainly 795, 5872; sance faille without fault (OR without interruption, constantly?) 1467 falir, fallir cf. faillir falsé cf. fauser

Glossary

521

fardaus sm. burden, load 6345 fauconet sm. falcon 6642, 6809 faudestue sm. chair of state, throne 203, 628, 681, 700, 4592, 4817 faus 429, 4584, 5899; fause f. 3791; adj. false fauser 1663, 1692, 7884; [falser] v.trans., used particularly with hauberk, byrnie or shield pierce, cut through 1663, 1692, 2187, 2224, 2247, 2325, 2731, 2908, 3268, 3679, 7884, 7917, 7940, 7957, 7985 fauz subj.sing. sm. falcon 6686 fazon sf. face 604, 1723, 3913, 7028; appearance 6209, 6463, 7760; armé d’une fazon all armed in the same way 7653 fé cf. foi feindre 3110, 3175; [foindre]; pres.subjunc.3 foingne 3257; v.intrans.&reflex. act in a craven or half-hearted fashion 11, 3110, 3175, 3257 fel 1115, 1226, 1398, 2109, 2169, 2313, 2350, 2502, 3110, 3157, 4188, 4872, 5267, 5929, 6244, 6259, 6312, 6430, 6713, 7483, 7545, 7560, 7592, 8005, 8046; felon obj.sing. 473, 606, 767, 906, 1061, 1288, 1430, 1451, +; subj.sing. 7509; subj.plur. 7135; adj. guilty of a breach of feudal loyalty, villainous, treacherous 906, 1061, 1115, 1226, 1398, 2109, 2169, 2313, 2350, 4188, 4872, 5267, 5297, 5929, 6259, 6430, 6713, 7483, 7545, 7560, 7592, 8005, 8046, 8103; infamous, wicked 1288, 1319, 1430, 1451, 1589, 1612, 1883, 1903, 2108, 2127, 2748, 2750, 2798, 2800, 2903, 3921, 4643, 5297, 6045, 6243, 6456, 7600, 8037, 8103; deadly, deathdealing 1686, 3460, 4438, 5300, 5567; sm. one guilty of a breach of feudal loyalty, villain, traitor 473, 606, 767, 2502, 3110, 3157, 3199, 6244, 6312, 6411, 6460, 6668, 7021, 7135, 7509, 7524, 7646, 8003; deadly or death-dealing person 3283; infamous, wicked person 3794, 4593, 5329 felonie sf. treachery 243, 360, 1838; evil thoughts 477, 3004; fury 4489; breach of feudal duty 6896 felor sm folly, imprudent action, madness 6383 feniment sm. end of the world, Judgement Day 4127 (cf. defeniment) [fenir], [finir] v.trans. finish, complete 233, 255, 796, 882, 3069, 5733, 6027; past part.used as adj. finished, dead 877; over 753 fenir infin.used as sm. end of life 3605 fer sm. iron 1143, 1165, 2019, 2057, 2308, 2743, 3668, 5298, 5640, 7500, 7762, 7794, 7805; (iron) shaft of lance 1045, 1785, 2000, 2175, 2757, 5095, 5194, 5472, 5515, 7901 fer adj. cf. fier feranz 321, 825, 3084; ferant 1108 adj. grey, turning grey (of hair) 321, 825, 1108; greyhaired 3084 feranz cf. ferir ferée sf. highroad 5379, 5817 ferement cf. fierement fereor sm. fighter, combatant 4214 ferez 3649, 4728; fesré 4537; feri (poetic lic.) 6409 past part.used as adj. (of ships) strengthened with iron, ironclad 4537, 4728; (of roads) paved 3649; (of spurs) strengthened with iron, made of iron 6409 ferirs 1509; [ferir] pres.indic.3 fiert 2114, fert 4609; imperat.sing. fer 1491, fier 1971; imperat.pl. ferez 1645, 2121, 2142, 2191, 2659, 2708, 2723, 3306, 5354, 5363, 5442, 5450, 5484, 5490, 5505, 5588, 5697; past def.3 feri 2418, 2678, 3369, 7140, 7987; past part. feru 2484, 2505, 3378, 3479, 3483, 3491, 5703; feruz 1822, 2355, 3333, 3342, 3672, 3957; v.trans. strike 1491, 1509, 1645, 1822, 1971, 2114, 2121, 2142, 2191, 2355, 2418, 2484, 2505,

522

Glossary

2659, 2678, 2708, 2723, 3306, 3333, 3342, 3369, 3378, 3479, 3483, 3491, 3672, 3957, 4609, 5354, 5363, 5442, 5450, 5484, 5490, 5505, 5588, 5697, 5703, 7140, 7987; strike pres.part.used as adv. feranz hotly 4286 fermance sf. guarantee; prendre male fermance be guaranteed a bad outcome, an evil fate 5100 fermer 1694, 3118, 3121, 4572, 7995; v.trans. fix, fasten, place in position 514, 1527, 1694, 2058, 3121, 4572, 5034, 5714, 7803, 7995; close 1750, 1783, 3118, 7119; reinforce 5719 [fermier] (graphy of formier) v.intrans. be agitated, boil, froth 5746 feror sf. haughty pride, savage ill-temper 6246 fertez sf. de male fertez full of ill-boding martial spirit, dangerously fierce and proud 3664 feru, feruz cf. ferir fes cf. fais fevre subj.sing. 2933; sm. blacksmith, craftsman fi cf. fil fiance sf. confidence 1735, 4945, 5093, 5785; a fiance in truth, certainly 5172, 8002; confidently, without fear 6797 fiancher 4888 v.trans. swear, promise [ficher] v.trans. place firmly, put 738, 4626; v.reflex. place oneself firmly 2092 fié obj.sing. 6308, 6320, 6497; fiez subj.sing. 724; obj.pl. 85, 1203, 4187, 4628; fez obj.pl. 4508; sm. fief 85, 724, 1203, 4187, 4508, 4628, 6308, 6320, 6497 fier 5019, 5027, 6186; pres.ind.1 je me fi 4975, 7894; v.intrans.&reflex. have faith in, trust in fier 38, 121, 325, 332, 544, 551, 980, 1118, 1275, 1665, 1917, 2089, +; fer 2418, 2782, 2810, 3092, 3500, 3507, 4419, 6840, 7054, 7139, 7493, 7496, 8133; adj. haughty, showing concern for one’s dignity, one’s status 1255, 2130, 7789; proud, noble 38, 121, 135, 1275, 1665, 2782, 3092, 3737, 3758, 3827, 4864, 4872, 7613, 7641, 7686, 7753; fierce, ferocious, ruthless 544, 1917, 2134, 2418, 2549, 2568, 2700, 2753, 2810, 3048, 3130, 3467, 3500, 3507, 4302, 4419, 4436, 4882, 5033, 5525, 5545, 6811, 6840, 7139, 7493, 7653, 7972; noble, princely 325, 332, 551, 980, 1118, 1917, 5128, 5137, 5325, 6227, 6518, 6527, 6532, 6546, 6580, 6584, 6756, 6764, 7054, 7098, 7334, 7381, 7438, 7779, 8072; great, excellent 6546, 7154; terrible, appalling 6222, 6464, 6626, 6721, 6752, 7276, 7496, 8082, 8124, 8133 fierement 644, 1483, 2026, 2107, 2525, 5351, 5375, 5391, 5674, 5937, 6427, 7076, 7255, 7370, 7968, ferement 1063, 6822, 7931 adv. proudly, haughtily 644, 2026, 5391, 7370; fiercely 1483, 2107, 5937, 6427, 7968; in haste, with a sense of urgency 2525, 5351, 5375, 5674, 6822; passionately 7076; intensely 7255 fiez cf. fié fil obj.sing. 53, 177, 236, 383, 528, 2122. 2830, 3003, 3067, 3696, +; subj.sing. 6070; subj.pl. 3724, 4620; fi obj.sing. 5615; fis subj.sing. 764, 5627; fius subj.sing. 2725, 2842, 3280, 3841, 4750, 4791, +; sm. son 53, 177, 236, 383, 528, 764, 2122, 2725, 2750, 2725, 2830, 2842, 3003, 3067, 3280, 3696, 3724, 3841, 3890, 4511, 4620, 4750, 4791, 5065b, 5211, 5222, 5233, 5240, 5261, 5338, 5465, 5565, 5615, 5625, 5627, 5865, 6012, 6070, 6601, 7094, 7165, 7757, 7797, 7974, 8028 finer 74 v.trans. conclude 74, 88; stop 599, 4602, 7407; conclude, cease 7200, 7228; past part.used as adj. at end of life, dead 98, 1792 firmament sm. sky, heavens 3514, 3973, 5052, 7129 fis adj. sure, assured 620, 1501 flor sf. flower 134, 7277; (fig.) flower, best examples, cream 1230, 4266, 6864, 8038; pent a f. painted with flowers 1814; peinz a f. painted with flowers 3190; a flor with flowers 2856; painted with flowers 5552

Glossary

523

florir 2578, 6089, 6112; v.trans. cover in flowers 202, 369, 928, 3701, 3799, 3990, 4453; adorn with painted flowers 1844, 2114, 2796, 5445; v.intrans. flourish 2578, 6089; become hoary 6112; past part.used as adj.to describe beard or hair hoary 357, 743, 1041, 1346, 4089, 5091, 5249, 5617, 6030, 6921, 8015 foi 39,774, 3325, 6020, 6614, 6895, 6903, 7040, +; fi 1590, 3128, 7541; fé 4090; sf. faith 774, 3325, 4090, 6895, 7471, 7549, 7618, 7623, 7672, 8065, 8074, 8086, 8092, 8095, 8132; loyalty, faithfulness 39, 6020, 6614, 6903, 7040, 7854, 8058; de fi truly, assuredly 1590, 3128, 7541; de bone foi of good faith, loyal 7549; por foi 774, per foi 3325, 6895, 7618, 8132 in faith, truly; per ma foi 7471, 7623, 7672, 8074, 8092; par ma foi 8086, 8095 by my faith, in faith foiée 1891, 4546, 4549 sf. occasion, time foille sf. leaf 5553 foillu adj. leafy, covered in foliage 3996 foingne cf. feindre foire sm. sheath, scabbard 687 folage sm. folly, foolish behavious 6511 fontenil sm. spring 4115 foön sm. fawn 7023 forbi adj. furbished, polished, gleaming (used especially of swords and lances) 371, 1843, 3306, 4075, 5094, 5602, 7570 forches sf.pl. gibbet, gallows 7999 forcheüre sf. fork of legs (where legs join trunk) 2271, 8041 [forfaire] pres.indic.3 forfait 4113; past def.3 forfeïst 1711 v.trans. do wrong to, damage, harm forgison sm. blacksmith 7762 fornase sf. furnace 4161 fornir 6101 v.trans. accomplish 364, 421; provide, constitute 5018; bataille fornie pitched battle 370, 5070; mes batailles f. fight my battles, lead my forces 6101; fornie past part.used as adj.f. massive, strongly constituted, hard, resistant 4077 fors adj. (subj.sing. 2813, obj.pl. 3180 of fort) strong 2813, 3180 fors prep. except 7, 12, 31, 5623, 734, 755, 1185, 23511, 3315, 3413, 3595, 6734; adv. out (of it), from there 3076, 3090, 3811; outside 7520, 7523, 8115, 8153; fors ad prep.phrase out of 1881; fors de prep.phrase out of 2298, 3355, 3988, 3995, 4822, 7984, 8040, 8164; except for 3927, 6351; outside of 5392, 7525, 8157; forz qe except for 8110 [forsener] v.intrans. become mad, go wild with rage 458, 677, 3530, 6954 fou sm. fire 450 fraindre 3843; pres.subj.3 fragne 6; past part. fraiz 1025, 2363; v.trans. break down 6; breach 1025; break 1049, 2115, 2173; conquer (?) 4763 franc obj.sing. 6218; subj.sing. 4180, subj.pl. 1798, 2005, 3577, 6093, 6214, 7478, 7633, 7714, 8148; franche f. 4597, 6569, 7009; frans subj.sing.1615, 3109, 3579, 3681, 7088, 7278, 7341; subj.pl. 7381; adj. noble francal adj. French 1645 francor adj. (of the) French 1813, 2489, 2867 frarine adj. wretched, unhappy 7090 frasne 1716; fresne 4991 sf. ash(wood) fremelon adj. shining, gleaming (used especially of hauberk) 1707 fremellon sm. a shining thing(?); freshly made chain mail(?)* 7503; OR blanc come fremellon (as) white as shining, i.e. gleaming white (?)* 7503

524

Glossary

fren sm. bridle 2170, 2806 freor sf. din, uproar 762, 1233, 1235; fear, state of uncertainty 4215, 5541, 6245, 6372; mettre en frëor inspire fear in, undermine confidence of 4215 fresceller 8129 v.intrans. quiver, shudder fresches 6124 cf. frois [freschier] v.trans. cool, refresh 4459 fresée cf. friser friçon sm. fear, dread 3131 fris sm. cold 1385 friser 5191, [freser] v.trans. decorate, embellish 5191; embroider, decorate 5715 frois (graphy of freis, fres, poetic licence) 4954, 6267; froi (final s dropped for rhyme) 7552; fresches f.pl. 6124; adj. fresh 6124, 6267, 7552; newly made, new* 4954 froisser 3746, froser 977, [froiser], [froisier], [froissier] v.trans. break into pieces, shatter 471, 977, 2262, 4039, 4506, 5740, 5860; v.intrans. crumple, collapse, lose strength 3746, 5602 fuirs sm. flight 3503 fuisons, fuisez cf. estre fuisor (graphy of foison adapted to suit rhyme) sm. quantity, mass, abundance 1231 fust obj.sing. 2216, 3205,7901; fuz subj.sing. 2730, obj.pl. 4505 sm. shaft (of lance) 2216, 2730,7901; cudgel 3205, 4505 fust, fustes, fuz cf. estre

g gaagne 9; gaeng 115; gaaigne 2664, 2670, 4236; guaainz obj.pl. 2068; sf. gain, profit, reward [gaaigner], [gaagner] v.trans. gain 5687, 5806 [gaber] v.intrans. boast, brag (especially to utter the conventional exaggerated vaunts usual with warriors on the eve of battle) 3096, 4736, 7371 [gaimenter] v.trans. & intrans. lament, moan, bewail one’s lot 4462 gaires adv. ne …. gaires scarcely, hardly, not very much 171, 721, 753, 3272, 3538, 6502 galart adj. strong, vigorous 1273 galie sf. galley 911, 4537, 4727 galoz sm.pl. les grant galoz at a great gallop 7519 garance sf. madder (a red flower and a plant used to produce red dye) 5163 garant 353, 356, 902, 1186, 1248, 1326, +; garent 2811, 2719, 4124; sm. protection 353, 902, 1326, 1443, 2241, 2788, 2811, 4911, 5588; protector, defence 356, 1186, 1248, 2062, 3719, 4292a, 4382, 4724, 5335, 5464, 6007; guarantee 3635, 4657; guarantor 4124, 6668; tenir a g. call upon as protector 1186, 3451; traire a g. rely on as protector 6007, call on as witness, cite as guarantor 4124, 6668 garde sf. protection, guarantor 3920 garder 142, 189, 1190, 1475, 1907, 2545, 4368, 7464, 7491, 7743; impera.2 1475; fut.5 garrez 601; v.trans. guard, keep 189, 1804, 2344, 2347, 2540, 3145, 3199, 3208, 3793, 4368, 5878, 6024, 6439, 6457, 6466, 7346, 7464, 7531, 7743, 7811, 7838; watch over, look after 142, 386, 529, 1152, 1190, 1907, 2545, 4394, 7246a, 7491; hold back, restrain 601, 799; v.intrans. make sure, take care (to do something) 364, 664, 1387, 1983, 3461, 5086, 5672,

Glossary

525

5776, 7657; v.trans.& intrans. look, look at 1392, 1475, 2157, 3829, 3881, 4114, 4121; v.reflex. restrain oneself (from), protect oneself from, avoid 4479 garentir 2154, 3230; garantir 2215, 2723, 2744; v.trans. guarantee, protect 542, 2154, 2215, 2723, 2744, 3230; guarantee innocence, swear to be free of, defend 7475 [garer] v.reflex. garez en vos! (formalized version of normal exclamation Gare!) interjection, imperative 5 Protect yourself! Take shelter! 5565 garillant 4165, 4913 s.m. and f. marshy ground, here specifically the (extensive) marshland in the southern region of modern Iraq, near the mouths of the rivers Tigris and Euphrates, i.e. just to the north of the Persian Gulf; thus south of ancient Niniveh, but just to the north of the port of Ahwaz (Orcaise in C), 4165, and the eastern boundary of a huge tract of Arab territory stretching from Mecca, beside the Red Sea, in the west, 4913 garir 1860, 2574, 3585, 3935, 4445, 6403; garis (poetic lic.) 1504, 3463; [gasrir]; fut.3 garra 4740; gasra 1729, 7495; v.trans. protect, save 363, 660, 1504, 1600, 1606, 1729, 2037, 2038, 2044, 2120, 2261, 2611, 3585, 3595, 3935, 3971, 4173, 4445, 4740, 5117, 5971, 7578, 7585, 7963, 8021; v.intrans. escape death 2574; go scot free, escape alive, get away 3463, 6403, 7495; recover 3410, 7198, 7495; seek provisions 5834 garison sf. safety 485, 4444, 7512; protection 1302, 2136, 2706, 2763, 3908, 4339, 4507, 5332, 5566; redemption 1605, 1726; a garison for saving (souls), for redemption 1605; faire sa vie garison seek redemption, save one’s soul by confession of sins 1726; faire garison offer protection, protect 2136, 2704, 2763 garni past part.used as adj. fortified, strong, also rich, powerful 241, 4450, 5061 garniment 116, 1378, 1594, 2425, 4948; garnimant 5932; sm. armour 116, 1594; equipment, weaponry 1378, 2425; clothing 4948; protection, armour 5932 [garnir] v.trans. garrison 5876; decorate 6557; equip, provide with guards, protect 7842 garzons subj.sing. 5230; garzon subj.pl. 7516 sm. servant 7516; garzons esrant travelling servant, spy 5230 gaste adj. waste, empty 979 gaster 25 v.trans. lay waste 25, 1024, 2646, 3240, 4762; past part.used as adj. ruined 2646; bereft, emptied 3240 gastine sf. waste-land 6175, 6235 gaudine sf. little wood 6268 [gehir] v.trans. tell, avow, declare 563, 767a, 7180; admit, confess (sins) 1983 generaus adj.m.pl. high-born, of noble birth 5902 genoil obj.sing. 4609; genos obj.pl. 2819 sm. knee genollons 4500; genuillon 7027; adv. a genollons, a genuillon adv.phrase on one’s knees 4500, 7027 gent adj. handsome, comely 384, 391, 858, 2814, 3512, 5128, 5197, 6541; fine, goodly 1256; kind, generous 3788 gent 263, 433, 460, 591, 1457, 1470, 1560, 1623, +; jent 179, 381, 840, 877, 879, 1039, 1393, 1836, 2061, 2264, 2417, 2474, 2894; jant 1322, 5229, 6003; janz subj.pl. 313, obj.pl. 1071, gens obj.pl. 4397, 4410; genz obj.sing.2668, obj.pl. 2351, 4977, 5043, 5171, 5476, 6521; subj.sing. 5029; subj.pl. 6279; sf. people, servants, adherents 263, 313, 381, 1071, 2061, 4133, 4397, 4410, 4455, 4534, 4596, 4685, 4691, 4850, 5115, 5853, 6003, 6678, 6699, 6703, 7204, 7410, 7793; men, forces, thus army 179, 313, 460, 840, 877, 879, 1039, 1322, 1393, 1560, 1699, 1753, 1763, 1808, 1834, 1894, 1896, 2264, 2287, 2289, 2318, 2339, 2417, 2470, 2474, 2487, 2531, 2583, 2658, 2673, 2683, 2697, 2722, 2802, 2815, 2824, 2838, 2890, 2894, 3053, 3063, 3068, 3307, 3539, 3576, 3733, 4240, 4358, 4761, 4764, 4826, 4857, 5134,

526

Glossary

5229, 5356, 5374, 5412, 5460, 5816, 7364, 7371; race, people 433, 591, 1457, 1470, 1623, 1632, 1742, 1759, 1774, 1836, 1849, 1928, 2144, 2503, 3299, 3313, 3551, 3711, 3715, 4121, 4124, 4207, 4374, 4463, 4469, 4472, 4487, 4542, 4552, 4771, 4956, 4986, 5008, 5057, 5065a, 5279, 5291, 5296, 5306, 5350, 5359, 5475, 5484, 5489, 5701, 5946, 6007, 6021, 6064, 6074, 6135, 6145, 6166, 6590, 6618, 6898, 1928, 7091, 7128, 7279, 7390, 7393, 7412, 7426, 7800, 8019, 8023; people in general, individual(s) 4121, 5317, 7272, 7494, 7549; group of people 6271 gentil 769, 3718, 3836, 4134; gentiz 206; gentis 178, 811, 1204, 3219, 3419; gentius 827, 1543, 3761, 4836, 5777, 5790, 6441, 7425, 7821, 8019; adj. high-born, gentle, noble 178, 206, 769, 811, 827, 1204, 1543, 3219, 3419, 3718, 3761, 3836, 4134, 4836, 5777, 5790, 6441, 7425, 7821, 8019 geranz, geron cf. giron gesir 1350, 2128, 2157, 3398, 3910, 5480, 5522, 5913, 6092; jesis (poet.lic.) 3218; pres.indic.3 jist 3210; gist 3779, 4376, 5825, 5892, 7994; git 3480; 6 gisent 1356, 2375, 2401, 2467, 2479, 2884, 3950, 5491, 5502, 6033; fut.5 gisrez 2999; imperf.3 gisoit 3956; past def.3 jut 1064, 3834, 3888, 3958, 3998, 4018, 4025, 4139, 6066, 7083, 8029; past part. geü 4004; v.intrans.& v reflex. lie, lie down; se jut encline bent over 7083; gesir a mort lie fatally wounded, lie dying 3480 geste sf. written account of the notable deeds of a people or family 2489, 2956, 3150, 3494, 5028, 5323 gez sf.pl. jesses, the straps attached to the falcon’s legs 6645 gibier sm. hunting of birds 6526 girfalt sm. gerfalcon, falcon 5406 giron 1708, 5571; geron 3268; gerons obj.pl. 4495; geranz obj.pl. 3446; sm. flap of hauberk, stretching from waist to knee 1708, 3446; part of body stretching from waist to knee 5571; a geron with flaps 3268; painte a gerons decorated with panels of material like that used for making tents, made colourful (hung?) with tapestries (?)* 4495; au giron at my side 5571 [giter] v.trans. throw, throw down 2921, 2926, 3811; shoot 130 [glatir] v.intrans. howl 5654 glavies sf.pl. lances 7933 gloton 2114, 2135, 2157, 2646, 2716, 3142, 5947, 6464, 7589, 8090; gloz subj.sing. 561, 924, 2079, 7856; glotons subj.sing. 8098; sm. miscreant, villain 561, 924, 1595, 2079, 2114, 2135, 2157, 2646, 2716, 3142, 4646, 4900, 4919, 5047, 5947, 6464, 7589, 7856, 8090, 8098 gole 1297, 5826; goule 7994 sf. mouth; gole baee, goule baee with mouth gaping, with mouth wide open 5826, 7994 gonelle sf. long tunic (worn over armour) 2224 gracïer 1087, 6520, 6535 v.trans. render thanks to, thank grailles subj.sing. sf. 2434 hail grasle obj.sing. 2264, subj.pl. 3545, 5130, 5987; grailles obj.pl. 252, 1022, 2082, 2643, 3654, 3715, 4239, 5176; grailes subj.pl.1070; graille subj.pl.1379; graile subj.pl.2573; grasles obj.pl. 3115, 4243, 5374, 6198, 7358; sf. bugle 252, 1022, 1070, 1379, 2082, 2264, 2573, 2643, 3115, 3545, 3654, 3715, 4239, 4243, 5130, 5176, 5374, 5987, 6198, 7358 grasles adj.subj.sing. slender 8042 grasler sm. bugle 6738 grandir comparative adj.(variant graphy of grandur = graignor) greater; a grandir 2152 adv.phrase more nobly, more impressively grans 1078 cf. greins

Glossary

527

grans 560, grant 1297, 1590, 5286, 7785 adv. full, fully graslet sm. small bugle (?)* 6736 gré sm. estre a gré suit, please, be what one wishes 478; venir a gré suit, meet one’s wishes 960, behave as society expects, meet the ethical standard 8044; de son gré as it suits him 582; de mot bon gré very willingly 1129; de gré on purpose, deliberately 3374, willingly 8053; en gré willingly, with pleasure 4779; (See also malgré) grefes obj.pl.(graphy of grafes) sm or f. (small) daggers 3154 gregnor adj. greater 1354, 1800, 5864 greins 640; grans 1078; greim 7935; adj. distressed, bitter, furious grement adv. grievously 4230 grenon sm. moustache 257, 909, 1105, 3141, 3203, 4015, 4021, 4029, 5682 grent graphy of grant, poetic licence 5307; adj. large, great gresius sf.pl.of gresille hail 4403 [grever] v.trans. cause trouble to, injure 584, 5677; hurt, torment 4194; lay heavy on, trouble 6668; oppress, trouble 7948; past part.used as adj. tormented, overwhelmed 3573, 3645; oppressed, troubled, unquiet of mind 634; badly injured 5544; heavy, troubled 7534 grezois adj. Greek 103; feu grezois Greek fire 3156, 6707; cri grezois incomprehensible cry (?) (or scribal error for grevois, of grief, of pain?) * 4651 grezois sm. Greek language sanz grezois plainly 4758 grief 4094; griez 299; grois 774; adj. difficult, hasty 299; severe, uncompromising, harsh in the terms it set 774; heavy, sad, painful 4094 grifaigne 1, grifeigne 2662 adj. forbidding, imposing 1; fierce, cruel, proud 2662 grois cf. grief grondir 5271 v.intrans. grunt gué 3564; guez obj.pl. 6340 sm. ford guenchir 7980 v.trans. incline, bow, bend down 7077; v.intrans. be treacherous, be unfaithful 6407; turn aside, escape 7980 guerers cf. guesrer gueresdon 355, 6038; guiredon 2685; gueredon 3789, 6061; guisredon 6212; sm. reward, recompense gueroier 983 v.intrans. make war 838, 983 guerpir 2714, 6094, 6754 v.trans. remove 715; leave, abandon 2714, 3284, 4527, 5064, 5866, 6094, 6649, 6686, 6690, 6754, 6764, 6901, 7004, 7022, 7546, 7889, 7890; discard, throw away* 6025 guerroi cf. guesrer guesredoner 1555 v.trans. repay 1555; pay for, recompense 1628 guesrer obj.sing. 6724, 6880, 7636, 7640, 7709; subj.sing. 6577, 6749, 7320; subj.pl. 2005, 5154; guerers subj.sing. 1167; guerroi subj.pl. 7563; sm. warrior guier 1278 v.trans. lead 1278, 1290, 2677, 4977, 5340; control, guide 2112; plunge 2117, 5447 guige 5191; guinche 7958 sf. strap by which the shield is suspended round the neck guiredon cf. gueresdon guise sf. appearance 7412 guivres sf.pl. vipers 4412

528

Glossary

h haé cf. aëz hahir 2720; [haïr]; [aïr]; pres.indic.6 héent 5809; v.trans. hate 1836, 1849, 2720, 4487, 5809, 6898, 8023; jent aïe 1836, gent haïe 1849, 4487, 6898, 8023 hated race, i.e. the Saracens; (vide ahir gerund) haïe, graphy of aïve (q.v.); en bone haïe as a good help, to render assistance 6902 haïr sm. cf. ahir haitiez 4423, 4623; haitié 6503; past part.used as adj. in good condition, sound, free of wounds 4423; eager, determined 4623; in good health 6503 hals m.pl. 4401, halte f. 6765 (variant graphy of haut) adj. high hanste 1143, 2057, 2183, +; anste 2267, 2796, 5193; sf. shaft, lance 1143, 1415, 1716, 1847, 2057, 2183, 2201, 2211, 2233, 2243, 2267, 2796, 3424, 3843, 4407, 5193, 5401, 7565 hardiment 2419, 2810, 4122, 7120; hardimen 3259; ardiment 2586, 3507; sm. audacity, boldness; prendre un itel hardimen take such a daring action, make such a bold move 3259 hardiment 6408; adv. boldly hart sf. rope 7995 hasqée 4555; haschée 5478; sf. trouble 4555; suffering, pains 5478 haster 1699; hastier 4249; asteier 195, [aster] v.trans. pursue, attack 5546; v.intrans. hasten 1699, 1778, 6338; be in a hurry 1368; v.reflex. hasten (implies to attack ?) 195; make haste, speed on 351, 1821, 4249, 5042, 5188; aster la voie ride quickly 1619 hastie (graphy of hastive) adj.used as adv. speedily, in haste 880 hauberc cf. auberc haubergié sm. man (soldier) wearing a hauberk 4293 haume 5380 cf. eume hee 1583, 5382; hie 4481, 5653; sf. onset 1583, 5382; a une hie in one fell swoop 4481; en masse, in unison 5653 héent cf. hahir heume 1768, 1811, 1856 cf. eume henor 6544 (graphy of honor, onor) cf. onor herbergement sm. lodging, accommodation 7135 herbergerie sf. place of encampment, camp 869; hostelry, place of overnight accommodation 5856 herberges sf.pl. encampments 985 herbergier 4320, 4867, 6522; herberger 8078; v.trans. shelter overnight, provide with accommodation, lodge 731, 6498; house, domicile, settle 4565; v.intrans. camp 1030, 4320, 4812, 4867, 6145, 6240; find lodging, sojourn 6522, 8078; v.reflex. camp, settle for night 4327 hermin sm. ermine cloak, fur cloak 6084; used as adj. (made) of ermine, fur 7504 heut sm. pommel, hilt 916 hie cf. hee home 11, 327, 561, 706, 1137, 1180, 1302, +; homes 89, 1178, 1291, 1563, 2341, 3053, +; ome 5727, 5966, 7876; hom 49, 87, 270, 414, 822, 1167, 1810, 3262, 6134; hon 264, 289, 292, 588, 651, 690, 816, 848, 915, 1310, +; on 23, 270, 802, 1034, 1090, 1490, 1671, 1682, +; om 2464, 5687, 7527; sm. man in the sense of ‘liegeman’ 49, 89, 264, 289, 519, 1167a, 2557, 2577, 3451, 4617, 4749, 4810, 5285, 5640, 5750, 5801, 6075, 6369, 7878, 8088; (plur.) men, forces, thus army 89, 1178, 1291, 3053, 3159, 3547, 4186, 4745, 5677, 5681; man, per-

Glossary

529

son 292, 327, 414, 561, 588, 651, 706, 816, 822, 1137, 1180, 1302, 1310, 1436, 1535, 1563, 1615, 1630, 1733, 1853, 1855, 1935, 1966, 2125, 2152, 2341, 2441, 2451, 2587, 2703, 3072, 3109, 3112, 3228, 3250, 3367, 3391, 3396, 3419, 3422, 3460, 3556, 3579, 3650, 3665, 3776, 3808, 3947, 4071, 4091, 4117, 4762, 4836, 4987, 5194, 5204, 5209, 5246, 5324, 5598, 5622, 5727, 5966, 6067, 6134, 6246, 6277, 6293, 6319, 6366, 6376, 6378, 6388, 6391, 6570, 6637, 6751, 6841, 6909, 6912, 6998, 7010, 7030, 7312, 7674, 7682, 7876, 8043, 8112, 8137; nus hon de car/char 1535, 5204, 6751, 7030, 8137; nul home charnal/carnal 6366, 8112 no man of flesh and blood, i.e. no mortal man; pron. a man, one, they, (with negative) anyone, no-one 11, 23, 87, 270, 690, 802, 848, 915, 1034, 1090, 1384, 1490, 1549, 1590, 1671, 1682, 1781(2), 1800, 1810, 1815, 1912, 1919, 1967, 1968, 2083, 2097, 2397, 2449, 2476, 2615, 2841, 2880, 3128, 3262, 3589, 3707, 3712, 3858, 4229, 4243, 4309, 4356, 4573, 4945, 4996, 5177, 5228, 5397, 5429, 5597, 5687, 5894, 6157, 6459, 6514, 6866, 6978, 7304, 7457, 7458, 7518, 7697, 7812, 8103, 8116, 8146 honi cf. huni honor cf. onor hontage 2288, 2976, 7318; ontage 1924; sm. shame 1924, 2288, 2976; affront, shameful action 7318 hors 6659, 6735 cf. ors hu 4005; huis subj.pl. 3465l sm. shout, shouting, warcry 3465; sound, cry 4005 hucier 1698, 2012; hucher 6765, 7316 v.intrans. cry out, shout 1698, 2012, 6765; announce, proclaim 7316 huee sf. shouting; les cris e la huee the yells and shouting 5707 hui 481, 796, 1162, 1315, 1799, 2037, 2039, 2121, 2149, 2500, 2501, 2542, 2557, 2688, 2899, 2926, 2943, 3022, 3537, 3580, 3594, 3638, 3898, 3938, 4067, 4233, 4470, 4472, 4957, 5051, 5107, 5118, 5334, 5611, 5614, 5698, 6489, 7601, 7872, 7963; uis 1499; huy 2578, 3228 adv. today huis cf. hu huis sm.obj.pl. doors 7119, 7203 humai 790, humais 5038, huimais 7675, oimais 8122 adv. now, from this point on 790, 5038, 8122; huimais ne puis now or hereafter 7675 humels adj. humble 2064 humilier 2013, [humelier], [umelier] v.trans. humble (?), cause to bow down humbly (?)* 785; v.intrans. bow down 2013; v.reflex. bow down humbly 8018 huni 3305, 3318, 5614; honi 3463, 5355, 5364, +; onie f. 4472; adj. shamed, dishonoured 3305, 3318, 3463, 4092, 4472, 4707, 4709, 5364, 5614, 7818, 7857 [hurter] v.trans. press, urge on, spur 1828; strike 3205

i ice cf. cil icest, iceste cf. cist icestui, iceste cf. cist icil cf. cil ideles sm or f.pl. idols 4529 idonc adv. then 3136 ier adv. yesterday (see also ersoir, ersor) 6539 igal adj. cf. engal

530

Glossary

igal obj.sing. 8153, egaus obj.pl. 5912 adj.used as sm. flat area, level area; en igal on level terms, on equal terms, equals 7877 iloc 195, 3259, 4533, 4715, 4775, 4855, 5258, 5806, 5845, +; iloques 637, 6354; iloqes 2931, 8165 adv. there 195, 637, 2931, 4533, 4715, 4775, 4855, 6133, 6155, 6354, 6523, 6695; then 3259, 5258, 5806, 5845, 6003, 6059, 6815, 8165 incliner cf. encliner inoldeüre 2268 (graphy of enhaudeüre) sf. hilt (of sword) insement cf. ensemant intoschement sm. poison 4116 irance sf. rage and sorrow; faire irance cause rage and sorrow, cause anguish 1744 iranz pres.part.irer used as adv. ardently, eagerly, impetuously 311 irascuz 279, 640, 1821, 2357, 2913, 3335; irascu 2167, 4096; irascue f. 2470 adj. aggrieved, resentful 279, 2913, 4096; angry, enraged 640, 1821, 2167, 2357, 2470; adj.used as adv. angrily, furiously 3335 ire sf. wrath 747, 5718; grief 5927, 6484, 6491, 7016 iré 1124; irié 1732, 2567, 7253, 7680; irez 388, 502, 509, +; iriez 4396; past part.irer, irier (q.v.) used as adj. resentful 388, 502, 2370; angry, furious 509, 721, 1732, 2567; distressed, aggrieved, grieved 1122, 1124, 2370, 4199, 4396, 4805, 7253, 7680; furious, fierce 1732, 2567, 5287; se faire irié be distressed, grieve 7680; adj.used as adv. angrily 374; eagerly, with urgency 4183 iréëment adv. furiously, in great anger 5353 irer 2653; vide iranz & iré above; v.intrans. become angry or animated, full of enthusiasm iror sf. sorrow 4207, 4220, 6237; wrath 6242; par iror in anger, furiously 2496, 3192, 3209; per grant iror furiously 4240; aveir iror vers feel wrath, hatred towards 2855, suffer sorrow 4220; porter iror be angry with 3004 is adv.(poetic licence for i) there 3234*, 3405* iscu, graphy of escu sm. shield 3268 isdeus adj. hideous, terrifying 5345 isi 451 adv. in this wise, thus isi cf. issi isnel adj. quick, agile, prompt 2378; swift 7911; adj.used as adv. quickly, promptly 7913 isnelement 608, 635, 892, 2808, 3484, 3542, 4855, 5564, 6434, 7605, 7749; isnellement 2648b; isnelament 3955, 7790; (isnelemant 5985, isnelamant 8088 with poetic licence) adv. swiftly issi 591, 2068, 3489, 3573, 3968, 4171, 4920, 5237, 5726, 5980, 7390; isi 1101, 3565, 4920 adv.(graphy of ici 59, 446, 600, +) here issir 2580 v.intrans. graphy of essir/asseir sit, be accommodated *2580 issir 2712, 2718, 3586; essir 2150; pres.indic.3 ist 1572, 2129, 3076, 3993, 6324, 6607, 7116; eïs 5317; pres.indic.6 issent 1032, 6336; imperative4 issons 8153; imper.5 issiez 4819; fut.3 istra 6818, 8110; past def.3 issi 4588, 5146, 6779, 6787, 7693; past def.6 issirent 3090, 4162, 6130, 6875; past part. issuz 3423, issu 4822, issi 7032; v.intrans. come out, leave 1572, 2129, 3423, 6324, 6607, 6875, 8110; come out, go forth 1032, 2718, 3586, 5146, 6130, 6336; come forth, disembark 4588, 4819, 4822; come out, issue forth, (of liquid) gush forth 3076, 3993, 6779; go 5317; come (as origin) 6787; come out (of), result from(?) *6818; des sens essir, des sens issir lose one’s senses, lose consciousness 2150, 2712; s’en issir v.reflex. issue forth, go forth 3090, 4162, 7116, 7519, 7693, 8153; come out 5113; issi past part.used as adj. senseless, unconscious 7032 istovra cf. estoveir

Glossary

531

isuée (graphy of eissue, modified by poetic licence) sf. exit, outlet; (sooty) chimney (?)* 6701 itant cf. tant itel cf. tel

j ja adv. in truth, certainly, indeed 178, 237, 346, 387, 541, 656, 684, 1875, 2145, 2775, 3160, 3660, 4305, 4731, 4741, 5534a, 5547, 6162, 6667, 6722, 6891, 7069, 7165, 7215, 7455, 8135, 8147; just 374, 3759, 6007; already 672, 1595, 1596, 1764, 1776, 3094, 3604, 3647, 3661, 3760, 4903, 5227, 5235, 6010, 6283, 6295, 6397, 6716, 7566, 7932; formerly 908, 1494, 3089; soon 3547, 3657, 5007, 5577, 6108, 6332, 6394, 7537, 7857; ever 1436, 1462, 1884, 1900, 3216, 3757, 4736, 6399, 7771, 7840; now 7924; (+ neg.) never 382, 467, 733, 828, 894, 1132, 1141, 1222, 1347, 1348, 1320, 1438, 1443, 1468, 1490, 1542, 1879, 1922, 1933, 1949, 1953, 1964, 2120, 2241, 2276, 2501, 2559, 2562, 2839, 2902, 2963, 2983, 2994, 3027, 3135, 3262, 3287, 3396, 3455, 3489, 3513, 3519, 3556, 3612, 3613, 3628, 3635, 3639, 3665, 3730, 3857, 4036, 4213, 4342, 4511, 4787, 5005, 5021, 5024, 5064, 5279, 5299, 5442, 5598, 5638, 5679, 5727, 5824, 6193, 6458, 6494, 6675, 6806, 6943, 7134, 7194, 7206, 7209, 7305, 7417, 7460, 7495, 7511, 7723, 7786, 8112, 8118, 8131; (+ neg.+ fut.) certainly not, nevermore 347, 500, 539, 601, 970, 3361, 3887, 6494, 7208; ja n’i ait si hardi never so bold (a knight) there might be, thus however bold he might be 7840 jacerant obj.sing. 2237, 2399, 5581, subj.sing. 5467; jaceranz subj.sing. 2053; adj.(used especially of hauberks) of oriental mail, of mail worked in an Eastern fashion jahanz 3341, jaiant 5307 sm.subj.pl. giants 3341; also found as a Proper Name Jahanz 3439, Jaianz 5643 and possibly 5307 jalées 4403 (graphy of gelée 5393 in obj.form pl.) sf. frosts jamay 261 (graphy of jamais 13) adv. never jane 7012 (graphy of jaune 2856) adj. yellow jant, jent, janz v.gent jarrois 5694 (graphy of jarrie? OR scribal error for larrois, graphy of laris, larris q.v. below? OR Proper Name cited by Godefroy as Jarrige ? cf. footnote) heath, waste-land jaus sm.subj.sing. cock 5922 jeüner 4334, 8097 v.intrans. go without food, fast jesis cf. gesir [joer], joier 4354 v.intrans. play (at), play (with) 127, 129, 4354 jogleres subj.sing. 3628, joclere subj.sing. 5031 sm. jongleur, minstrel joindre 5421; past part. jont 97, 162, 184, 228, 264, 441, 666, 2048; v.trans. join, link together 2048; clasp together (hands) 97, 162, 184, 228, 264, 441, 666, 4175; v.intrans. join battle 5421; v.reflex. join, attach oneself (to) 2360 joïr 3598; past def.3 joï 5259 v.intrans. rejoice, be glad, enjoy joïse cf. juïse jone adj. young 6067, 7752 jonture sf. joint, bone 2272 jor sm. day 13, 146, 186, 369, 400, 534, 659, 950, 961, 998, 1199, 1294, 1347, 1436, 1818, 1953, 2035, 2090, 2149, 2280, 2345, 2582, 2770, 2832, 2902, 2963, 3360, 3540, 3574, 3638, 3857, 3938, 4103, 4216, 4364, 4519, 4540, 4704, 4935, 5054, 5111, 5551, 5599, 5703, 6072, 6382, 6930, 7167, 7313, 7930, 8098, 8109, 8113, 8201; date, appointed time 65; day, daylight

532

Glossary

1028, 1067, 1377, 1565, 4257, 7275, 8012, 8032; hui cest jor today, this very day 1162, 2501, 2943, 3022, 3537, 5051, 5614, 7601; cest jor 4190, icest jor 3620, 4756 this day, today; le jor that day 1232, 1400, 2355, 2402, 2473, 2888, 2910, 2937, 4556, 4989, 5479, 5603, 5831, 6240, 7267, 7902; li jor that day 4714; onqes jor on any day, ever 3540; nul jorn 838, nul jor 3519, 4191, 4213, 4605, 4769, 5259, 7209, 7444 on no day, never; tot jor every day, always 199, 1801, 3224, 4393; toz jor every day, always 3224, 4393, 4598, 4901, 5023; toz jorz always 4926; a toz jors for ever 6125; tote jor f.in this phrase the whole day, all day long 3097, 6630; ne …. mais jor not … for a day longer, not … another day 2902, 3730; ja n’ert mais jor there will never be a day hereafter (when) 6494 jornal sm. day’s work 1643; faire son jornal do that day’s work, do the task set him, do his duty 5743 joste prep. alongside 1154, 6657, 6776, 7183, 7269 joste sf. fight, single combat, encounter between two knights 5717; quesre joste seek a joust, a fight, someone to fight 2102 joster 1921, 2103, 2598; v.trans. bring alongside 5417; bring together, draw up (ready for battle) 4999, 5476; v.intrans. give battle, fight, attack 844,1921, 2087, 2211, 2541, 5443; join battle, have a fight (with) 7558; v.reflex. draw close to, approach 1059; ont bataille jostée have joined battle 3054; li cors a la terre li joste falls to the ground 3393; josté past part.used as adj. positioned alongside, drawn near 562; jostée adj.f. positioned nearby, gathered together 4765; joster infin.used as sm. joining of battle 2103, 2598 jovenaus subj. form sing. of jovenal adj. young, youthful 5896 jovent sm. youth 2424, 2812, 3964 jovente sf. youth 2773* jugement 429, 452, 676, 3559, 5133, 7644; jugiment 7422, 8082; sm. opinion, recommendation 429; punishment 452, 676, 3559, 8149; punishment, (death) sentence 5133; judgement, sentence 7646, 8082; judgement, trial 7422, 7655; responsibility for judging (for trying) 7644; torner faus jugement give a false opinion, recommend or counsel with evil intent 429 jugier 1084; juger 1554, 7602, 7717; jujer 7643; v.trans. adjudge, recommend, appoint 308, 337, 405, 420, 435, 487, 500, 1074, 8070; assign 1084; doom 1264, 1431; judge 1554, 6465, 7600, 7602, 7657, 7717; punish 7379; try in court of law, pass sentence, condemn 7616, 7674, 8068; il nos juja he assigned us (the rearguard) and thus doomed us 1399; la lois fu jugée the Law was appointed (ie the Ten Commandments were given by God to Moses) 4544 juïse 7406, 8087, 8096, 8154; joïse 8124; sm. judgement, punishment 8087, 8096, 8124, 8154; prendre son juïse obtain judgement (against), wreak vengeance (on) (?) 7406; faire son juïse mortal carry out the death sentence on him 8154 jus 681, 2156, 3279, 3835, 4856, 5527, 5534a, 5563, 5725, 5729, 5825; juz 3336; adv. down, down to the ground 681, 2156, 3278, 3835, 5534a, 5725, 5825; down 3336, 5527; mettre jus lay (person or thing) down; jus de prep. down from 4856, 5563, 5729 jusc’a prep. jusc’a un an within a year 1349 justise sf. punishment; faire justise de do execution on, punish 4351, 4739, 5948, 7411 justisier 111, 948, 974,8131; v.trans. govern, rule justly 111, 948, 974; bring to court, bring to justice, indict 7673; punish 8131

Glossary

533

l laër, laïer cf. laissier lai cf. laissier laidenger 7698 v.trans. insult, ill-treat laidir 6390 v.trans. insult, ill-treat, torment 1988, 6390, 6412, 7579 laïenz adv. therein 6982; there 2341 (?) [laïer] (cf. laissier to which this verb is assimilated, especially in the future & conditional) lainer sm. coward 2198 lairon obj.sing. 1611, lasron obj.sing. 6461; subj.sing. lere 450, 7573, lerre 7732, leres 3083; obj.pl. lasrons 6346 sm. thief, scoundrel laissier 3029, 6754, 6867; laisser 3765, 6816, 8081; laiser 3821, 6533, 6739; laisier 6766; lasser 7327; [laïer]; pres.ind.3 lasse 1498, 2170; laisse 1771,1846, 2182, 2547, 3877, 5877; lait 1703, 2100, 6218a, 7338, 7520, 7523, 7621, laïet 4794; 6 laissent 1385, 4306; imperat.2 laisse 1614, lai 4958; pres.subjun.3 laist 270, 2045; fut.1 lairai 3624, 6910; 3 laira 3463, lassera 4615; fut.6 laeront 1239; condit.1 lairoie 475, 706, 1160; past def.1 laisai 6745; 3 laissa 144, 1965, 3113, 3330, 4147, 4277, 4329, 4341, 4583, 6975; laisa 996, lassa 1486, 2626, 2948, 4379; p.def.5 laissastes 6503, 6 laisserent 4485, 4656, 4983; past subj.1 laissasse 7627; v.trans. leave, abandon 270, 295, 693, 1239, 1965, 2547*, 3113, 3330, 3624, 3877, 4147, 4329, 4341, 5877, 6975; let, allow 144, 475, 996, 1614, 3463, 4277, 4583; let go, loosen, release 2626, 2948, 4485; (release the reins, halt 2547*?); reject the idea, renounce, put to one side* 6218a; abandon all title to something, cede, give up 6910, 7327; fut. or condit. + ne + subord.clause not fail to, not refrain from 706, 1161, 4615; laissez ester! 4744, 6445, 7283, laisez ester! 6924 Let be! Cease to speak (of it)! laissor sf. loosening (of rein to allow horse to gallop), letting go 2854; cf. laisse corre 1771, 1846, 2182 laiz adj.(m.obj.pl.of lait) hideous, deadly 2341 landon sm. wooden bar or block fitted to dog to prevent its hunting game, clog (to tether animal), muzzle 4434 langues subj.pl. 2059 sf. streamers (of gonfanon) laor sf. width 6241 largitez sf.obj.pl. (instances of) generosity, liberality in giving 821 lariz 1179, 1496, 1976; laris obj.sing. 3989, obj.pl. 3217; larris obj.pl. 623; lairis obj.pl. 5430 sm. high moor, fell, hillside, waste-land 623, 1172, 1179, 1456, 1496, 1927, 1976, 3217, 3989, 5430 las interjec. Alas! 6479, 6617 las 1251, 2661, 4102, 4322, 4383, 4494; lasse f. 4500, 5453, 6693; adj. weary 1251, 2661, 4322, 4494; wretched, unhappy 4102, 4500, 5453, 6693 lasse 6711, 6999, 7156 adj.f. used as sf. unhappy wretch lassier 6889; [lasser] v.intrans. grow weary 2396, 6889; lassez 313, 400, 7588; lasé 4312, lassé 7544, lassée f. 6145 past part.used as adj. wearied, weary lastez obj.pl. form of lastet sf. pains, fatigues 72 latin sm. (colloquial) language, dialect, country wisdom (?) 7944 latiner subj.sing. sm. interpreter 5157 laver cf. lever laz obj.pl. sm. laces, cords 5528, 5787 lé cf. lez lechiere subj.sing. sm. villain 7738

534

Glossary

leçon 3787, 4428; leccion obj.pl. 6052; sf. reading in church, lesson in the liturgy 3787; written account (implying historical validity of document) 4428; prayer 6052 leger obj.sing. 3750, legier obj.sing. 2256, 5136, subj.pl. 129; legiers subj.sing. 6865; legere subj.sing.f. 5206; adj. light-hearted 129; nimble, lithe 2256, 5136; light (in weight) 3750; easy 5206, 6865 legerie 3007, 6912; lezerie 424, 792; sf. recklessness, piece of recklessness 424, 792, 3007; folly, fantasy, fairytale 6912 leissir cf. loisir lere, lerre cf. lairon lés cf. lez letree adj.f. (describing sword) covered with inscriptions 688 letre sf. written document, letter 742; inscription 3627; document, religious text (authoritative because it is written, not oral) 4694, 5111 leu obj.sing. 3396, 4966, 6215; lius obj.sing. 6096, subj.sing. 7408, obj.pl. 2679, 2770, 5596, 5668, 5878, 7060; sm. place 2679, 2770, 3396, 4966, 5596, 5668, 6096, 6215, 7060, 7408; surrounding country 5878 leupart obj.sing.1062; lipart subj.sing.1056, liupart subj.sing. 6041, 6427, 7024 sm. leopard lever 161, 191, 1234, 6166, 7074; laver 1008; v.trans. raise from baptismal font, baptise 96, 161, 227, 665, 723, 1008, 6166; erect (tent) 191; lift up, raise 1234, 1760, 2338, 3093, 3142, 3913, 4270, 4556, 5383, 5401, 5682, 5804, 5808, 5827, 6060, 6128, 6139, 6741, 7074, 7078, 7531, 7547, 7999, 8158; put up (prey when hunting) 6760; stir up 7532; v.intrans. stand up, stand erect 30, 639, 7468, 7707, 7740; get up 1068, 8033; rise (of sun), break (of dawn) 1775, 4543; rise (from tomb) 5054; v.reflex. stand up, get up, rise 1510, 1990, 7858, 7868, 8033; Levez vos en! Stand up! 7195; lo cri l. raise the hue and cry 7531, 7547 levrer 5161, lievrer 6761, lievrier 6768 sm. greyhound lez 112; liez 302, 317, 1123, 2300, 4191, 4354, 4605, 4617, 4816, 5772, 5777; lié 6500, 7681 adj. happy, joyful lez 392; lé 4538 adj. wide; e de lonc e de lé both far and wide 576 lez sm. side 515, 1886, 2362 lez 374, 1677, 1772, 1774, 1795, 1819, 2758, 4477, 4716, 6657; lés 1685, 3980, 4148 prep. beside, alongside 374, 1677, 1685, 1772, 1774, 1795, 1819, 2758, 4148, 4477, 6657; in the vicinity of 3980; lez lor amongst them 4716 lez def.art.(graphy of les) the 7169 lezerie cf. legerie liche sf. blade of sword; faire l. nu unsheath one’s weapon, draw one’s sword; (alternatively liche = lists, field for tournament, thus clear the field for battle?) * 2180 liément adv. gaily 5318, 5361 liemer sm. large hound for hunting deer or boar 8076 lignage sm. lineage, family, kin 1935, 2991, 3889, 4648, 5577, 6079, 6515, 7494 ligners adj.(subj.sing.) of noble race, of noble ancestry 1160 linir sm. linen (?)* 1858 lipart, liupart cf. leupart liranz adj. (lit.) playing the lyre, (fig.adj.used adverbially) lightheartedly 317 lisant adj. wise, learned 4179 lister 17 v.trans. border, edge 17, 1144; decorate with bands, paint border on 1769 lius cf. leu liues obj.pl. sf. league 1584, 3050, 4271

Glossary

locu loee loee loër

535

adj. shaggy, hairy and dishevelled 5011 (graphy of lieuee) sf. a demie l. half a league away 1764 adj.f. cf. loër 139, 990, 1191, 1918, 1922, 2209, 2534, 2738, 3851, 4353, 7299, 7822 v.trans. praise 139, 535, 990, 1276, 1918, 1991, 2209, 2534, 2738, 3851, 4353, 7299, 7822; advise, recommend 267, 3158; s’en l. v.reflex. congratulate oneself/one another over something 1922; France la loée France the renowned 690, 1563, 1892, 2331, 3051, 5390 löer 46, 157 v.trans. pay, reward, hire loier sm. pay, reward 1489, 1518, 2925 loi 8, 149, 182, 646, 740, 889, 921, 1085, +; loy 265, 2010; sf. Christian religion, Christian faith 8, 149, 182, 265, 2561, 3826, 3967, 4957, 5302, 5873, 6102 (sometimes specified, e.g. la l. Deo 8, la l. Deu 3826); (Moslem) religion, faith 646, 740, 889, 921, 3332, 4470, 5209, 5485, 5814, 5866; creed, faith (in general) 898, 5411, 5701, 7557; law, custom, learning(?) 104, 106, 771; Ten Commandments, Jewish Law 3969, 4544; rule, dominion 4754, 4957, 5485; a loi de, a loy de in the fashion of, like, after the custom of 1085, 1267, 1683, 1697, 2010, 5647, 7281, 7876; ou corent mes lois where my rule holds sway 4754; de puite loi morally corrupt 7557; Judeus de la Loi Jewish High Priest and Council, Sanhedrin 8062 lois (Franco-Italian graphy of los) cf. los loisir 169, 3929; leissir 1859 sf. time to spare 1859; par l. 169, a l. 3929 unhurriedly lonc 767b, 1796, 4852; lonz m.pl. 3150; longe f. 1303, 6616, 7029, 7245; longes f.pl.8075 adj. long 767b, 1303, 1796, 3150, 4852, 6616, 7029, 7245; far-reaching, extensive in length 8075 lonc 689; lonz 1472, 5036, 8188; lonq 5650; adv. for a long time 689; far, to a great distance 5650; far away 5036, 8188; e de lonc e de lé both far and wide 576; a lonz far away, at a great distance 1472 lonc prep. along, alongside 2260 longes adv. for a long time 7829, 8083 lor pron.m.pl. them 2866, 4716 lor, lore cf. lors lorer sm. laurel tree 134, 3834, 4591, 6763 lors 61, 501, 940, 7988; lor 1426, 2566, 3345, 4455, 4802; lores 4249, 6755, 7073, 7184; lore 69; adv. then 61, 69, 501, 940, 1426, 2566, 3345, 4455, 4802, 6755, 7073, 7184, 7988; when 4249, 6499 los 775, 846, 2132; lois (Franco-Italian graphy) 1448; sm. advice 775, 846; fame, reputation 1448, 2132 lucent cf. luisir luisanz cf. luisir [luisir] pres.ind.3 luist 986, 1357, 5850; 6 luisent 1405, 5380, 5739; v.intrans. shine pres.part.used as adj. lucent (?)* 1033; luisant 1527, 2397, 2450, 2787, 4375, 5468; luisanz 4585, 5415, 5427 shining [luitier] v.intrans. struggle 4422 lumere sf. sight 3385 lumiere sf. mouth-piece 3046

536

Glossary

m mahomerie sf. Moslem temple 4466, 5855; used as Proper Name Moslem world, Moslem lands 5071, 5660 maint 3973, 4381 cf. menoir maintenant 4153, 6647, 6665, 6674; maintenanz 3625; meintenanz 4296; meintenant 4385, 5536; maintenent 7795; adv. forthwith, at once 3625, 4153, 4296, 4385, 5536, 6647, 6665, 6674, 7795 maire adj. major, greater 3552 maire (graphy of mere, 404, 822, 1137, 1448, 2772, 3947, 6684, 7239) sf. mother 3556, 5756 mais 27, 158, 231, 259, 301, 368, 386, 857, 1066, 1121, +; mas 284, 331, 1416, 3366, 3427, 3437, 4376, 4384, 5620, 6491, 7017, 7704, 8186; meis 856, 1252, 1347, 1363, 1762, 2780, 3114, 6238, 6389, 7718; mes 2458, 3596; adv. I greatly 3426; more, any more 1762, 6238, 6489, 6940; meis de more than 1363; meis qe more than 6389; ne m. qe no more than 2253; ne m. except for, only 3596; ne m. qe except, apart from 259, 2962, 6661, 7841; II longer (time); ever 7258; ne … m. no longer, no more 158, 856, 1762, 2652, 3366, 3620, 3739, 4109, 4222, 4232, 4384, 7017, 7704; ne … m. never 1252, 1347, 2458, 2580, 3435, 3808, 3868, 5686, 6011, 6138, 6293, 6298, 6366, 6637, 6684, 7266, 7817, 8006, 8186; ja ne …. m. jor never again (on any day) 2902, 3730, 6494; ja … m. ne never again 1347, 2120, 2902, 2983, 4511, 4213; ainc m. ne never before 2780, 5544; m. ne never again 386, 2430, 3114; onqes m. ne never 1535, never again 3293, 3868; unqes m. ne never again 5396, 7120; III (with conjunctional force) but 27, 231, 301, 331, 368, 857, 1066, 1121, 1554, 1726, 1861, 2002, 2370, 2559, 2580, 2583, 2597, 2612, 2827, 2889, 2996, 3025, 3305, 3318, 3340, 3371, 3437, 3453, 3507, 3605, 3700, 3837, 3944, 3976, 3992, 4276, 4376, 4422, 4439, 4445, 4479, 4898, 5341, 5538, 5620, 5642, 5769, 5870, 5957, 6112, 6181, 6348, 6491, 6619, 6721, 6769, 6778, 6796, 6804, 6840, 6922, 6971, 6977, 6994, 7370, 7388, 7456, 7511, 7550, 7555, 7561, 7718, 7723, 7753, 7867, 7913, 7927, 7938, 8017, 8054; m. q’il seüst provided that he knew how to 4355; je n’en poi m. I could not do otherwise 6281 maistre 1824, 2359, 3062, 3104, 3137, 5838, 5845; mestre 3198, 4426, 4443; mastre 4437; adj. chief, most important, principal 2359, 3062, 3104, 3137, 3198, 4426, 4437, 4443, 5838, 5845; por le maistre above all, as the most important thing 1824 maistre 6635; mestre 5831; sm. master (as title 6635) maistrie sf. mastery 739 maistrise sf. art, craftsmanship 1708, 1786 maitin sm. morning 6168, 7945, 8032 maitinée sf. morning 1565 maitines sf. Matins, the morning service in church 197, 987 maitinet cf. matinet mal adj. evil, bad 355, 476, 569, 4706, 4873, 6399, 7527, 7539, 7945, 8150, 8185; m. diz evil report 1980; or en ait il m.gré now let him be discontented (or not), now let him put up with that 3581; lor en aiez m. gré may you have dissatisfaction over it 7747; m.espoir terrible (final) judgement 5346; non aiez mal pensé Don’t worry! 7539 (See also dahé, dehait) mal sm. evil 1647, 2833, 4160, 6218, 8179; pain, suffering 992, 1195, 1969, 3273, 4026; damage 2187; a hurt 3403; misfortune 5506, 7389; disaster 5742; evil or suffering, i.e. the Devil or Hell 5516; a mal to a bad end, to misfortune 5506; por m. with evil intent, wrongly 1362; m. faire do harm 3331, 3510; mal soit de may it go ill with, woe betide 29; m. soit cursed be 1479; m. ait cursed be 1956; m. aies tu may you be cursed (suffer

Glossary

537

for it)! 2177, 3338; asez i avra mal he will be in considerable pain; …m. te die! …curse you! 3693; ne sente se m. non feels nothing but pain 3901; m.doner send misfortune, curse 3273, 3854 mal adv. unluckily, for one’s misfortune, in an evil hour 418, 1430, 2302, 3226; in an evil way, to evil purpose 701, 4350; badly 2824, 3901, 4788, 4793, 6036, 6690, 7368; (See also bailier and bailir) [maldire], [maudire] v.trans. curse 359, 1585, 1837, 2108, 2251, 3298, 3711, 4464, 4658, 5301, 5456 maleëzon 486, maliëzon 3793, maleïçon 4663; sf. malediction, curse, damnation malefoi sf. bad faith, treachery 8177 maleïr v.trans. curse, damn 3600 malement adv. badly 584, 3521, 3718, 3722 malentalenté 462 adj. angry, of evil intent maleüré adj. unhappy, of misfortune 4790 malgré sm. grief, trouble, discontent; aveir m. be dissatisfied with, be grieved by, be upset; or en ait il m. so let him put up with it 3581; lor en aiez m.! Then may you be thwarted in this! 7747 [malmetre], [maumetre] past part.malmis 2594, 5438, 7391, 7897, 7901; maumis 622, 934, 7410; v.trans. bring down, abase 622, 934; damage, break, tear 2594, 4079, 5438, 7901; bring to destruction 7391, 7410, 7446, 7897 maltalant 350, 365, 682, 2786, 4899, 4907, 6670, 6926, 7748.; mautalant 1766, 2323, 4388, 5525; maltalent 437, 7910; mautalent 1866, 2418, 4841; mautallant 3146; sm. anger 350, 365, 682, 1766, 2323, 2418, 2786, 3146, 4388, 4841, 4899, 5525, 7748; evil intent 437, 1866, 6670, 6926, 7910 (Mautalant,7551, has been read as the horse’s name (Maltalant 7542, 7582) but it could equally well refer to Gondeboeuf’s emotion.) maltalentis 697, 3456; maltalentos 6402 adj. angry, enraged manant sm. possessor 595 manant pres.part used as adj. remaining, residing 4926 manantie sf. fee, payment, reward 6559; reward (implying a place in Heaven) 7193; possession, wealth 7350 mananz subj.sing. 4729; manent subj.sing 7792; adj. (especially when coupled with ‘rich’) powerful mandement sm. rallying point for troops, fortress 4674 mander 205, 6172, 7834; v.trans. ask for, request 38, 1002; send message, inform via an envoy 148, 172, 219, 281, 709, 4520, 4783, 4811; command 664, 722, 42263, 6527; summon, send for 198, 205, 1230, 2995, 4534, 4781, 6172, 6536, 6594, 6810, 7431, 7834, 8192; Saluz vos mande Sends you greetings (via an envoy) 4800, 6518 mangier 3033, 4306, 6735; mengier 4600, 6663; menger 8084; pres.subjunc.3 menjust 8112; 6 menjucent 3041, 6041; past def.3 menja 4332, 5889; v.trans. eat; lo menger comunal infin. used as sm. the food prepared for everyone to share, i.e. the dishes prepared for the whole court 8112 mangon sm. gold coin 917 manoirs 4751; manois (poet.lic.) 6261 sm.pl. domains, feudal demesnes 4751, 6261 manois adv. immediately 4753, 4947, 5499, 5697, 7742 mante (graphy of mainte) cf. meint mantel 5954 sm. cloak mantenir 2144 v.trans. protect, guard 2144, 4187; maintain, continue 2906

538

Glossary

mar adv. in an evil hour 232, 237, 308, 381, 401, 538, 774, 1137, 1451, 1632, 1903, 1997, 2554, 2734, 3330, 3358, 3359, 3638, 3660, 3866, 3964, 3997, 4238, 4298, 4731, 4919, 5236, 5246, 5550, 5928, 5999, 7091, 7738, 7888, 8044 marchëant cf. merchäant marchiez 2001, marché 8166 sm. bargain, transaction 8166; faire m. de buy 2001 marchis sm. warden of a march; great captain 932; marquis, high noble 7434; Marquis (title) 7765. (When applied to Roland, 932, this may denote possession of high rank or of outstanding prowess, or both.) mariëment sm. marriage 3513 martre sm. marten 389, 794 martrines adj.f.pl. of marten 5917 masel sm. butchery, massacre 2376 masiz adj. solid, massive 203 masnie 244; masnee 1042, 3200, 4558, 4997, 5476, 5483, 7177; sf. household of servants, entourage (?); a m. escherie with a small household of servants 244, 1042 masrie past part.masrir used as adj.f. troubled, plunged into grief 7177 mastins sm.obj.pl. mastiffs, guard-dogs 2047 mastre cf. maistre mater 2538; pres.subjunc.1 mat 1271; v.trans. vanquish, defeat, conquer 811, 1271, 2538, 3578, 4023, 5816, 7756 matinet 984, 7356; maitinet 4863 sm. day-break, dawn maufez subj.sing. 397, maufé subj.pl. 4043 sm. evil spirit, devil, demon mauvais 2408, mauvas 6754; malvaise f. 2844; adj. bad, evil 2408, 6754; shameful 2844; mauvaisse bonté evil return, evil recompense 8049 mauvastiez sf. evil deed, cowardice 5664 maus sm.obj.pl. masts 4572 maus adj. grievous 5895, 5919 me obj.sing.possess.adj.m.(used by Saracen speaker on both occasions) my 59, 942 [meheigner], [mesheigner] v.trans. wound, distress 4230; maim, mutilate 4808 meins adv. less 2991, 3557, 5085; ne …. meins not (emphatic) * 3161 meint 4, 40, 116, 360, 532, 597, 769, 810, 1025, 1589, +; maint 689, 1576, 2019, 2505, 2888, 3491a, 3550, 4087, 4241, 4330, 4365, +; f.sing. mante 4517; m.obj.pl. meins 7644; adj. many, many a 4, 40, 116, 360, 597, 689, 769, 810, 1025, 1576, 1589, 1872, 2019, 2425, 2428, 2473, 2525, 2530, 2566, 2679, 2888, 3038, 3402, 3491a, 3549, 3550, 3718, 3777, 3985, 4087, 4088, 4141, 4241, 4330, 4365, 4519, +; pron. many 3493, 4332 meintenant cf. maintenant [menoir], [manoir] pres.indic.3 meint 3314, maint 3973, 4381 v.intrans. dwell meis cf. mais meissons obj.pl. sf. houses; m. moniaus religious houses, monasteries 5905 mel sm. honey 8127 meller sm. thousand 2957, 6794 mellor obj.sing. 5470, 6369, 7543, 7786; subj.pl. 6465; mellors obj.pl. 5341, 6834; meldre subj.sing. 3966, meltre subj.sing.7239; meudre subj.sing. 273, 830, 2457; meudres subj.sing. 2737, 3422; mels obj.pl. 95; meus 2688, 3202, 6969, 7394; comparat.adj. better 273, 2457, 3966, 5470, 6369, 7543; best 3422, 6465, 6549, 6834, 7239; compar.pron. better 830, 2737, 7786; best 95, 2688, 3202,5341, 6969, 7394 mels cf. mellor (adj.& pron.) or mieus (adv.) memele obj.sing. 6691, mamelles obj.plur. 6812; sf. breast

Glossary

539

menbrance subj.sing. 5784 sf. remembrance, memory menbrer 7388 v. trans. remind 5802; v.intrans. recall, remember 7308; v.reflex. recall, remember 8196; past part. used as adj. memorable, important, wise 692, 1751, 4760; (long-) remembered, famous 1889, 3043, 7986; true, veritable 1514 mendie adj.f.sing. poor, beggarly 7186 menee sf. blast, peal (of horn or trumpet) 1444, 1890, 3047, 3244 menger, mengier cf. mangier menor adj. used as sm. smaller; li grant e li m. big and small 4715; en la m. (referring to echelle) in the smallest mentir 2159 v.intrans. lie 2159, 3231, 7856; v.reflex. lie 7635; m. a sa foi, m. sa foi break one’s word 6020, 7854 menu 7978; menue f.sing. 2898, menuz m.pl.1828, 2359, 2727, 3337; adj. small; denz m. front teeth 2898; li salz m., les sauz m. small leaps, bounds 1828, 2359; in small leaps, bounding 2727; herbe menue short grass 2898; adverbial use: maillez menu closely meshed (chain mail) 7978; e m. e sovent at frequent intervals 2435; over and over again 3726, 4135, 5321 menuëment adv. sparsely; m. ramé(e) covered with a mass of small branches 571, 1772 menzoigne sf. lie, untruth 3057, 7634; sanz m., sanze m. without a lie, i.e. truly 5075, 6295 mer adj. pure 45, 947, 1670, 2200, 2257, 2640, 3017, 3830 meral obj.sing. 8118, meraus obj.pl. 5903 sm. disk, jeton or token distributed to priests and monks for attendance at services, presumably for trading in against basic necessities, such as food; thus basic necessities (for sharing between the monks) * 5903; plein m. a full jeton’s worth, an exact portion 8118 merchäant obj.sing. 6300; mercäant subj.sing. 2416; marchëant subj.pl. 6272, 6289; merchëanz obj.pl. 6308; sm. merchant 6272, 6289, 6300, 6308; trafficker 2416 merci sm. mercy 1609, 4631; clamer m. pray for mercy 1503; aveir m. de have mercy on 3220, 3401, 4041, 5618; m. de m’arme! Have (understood) mercy on my soul! 1792; m. crier beg for mercy 4050, 4253, 6432, 6951; per ta m. in your mercy 5119; les vos m. of your mercy, please, kindly 6282; grant m. vos en rent I render you great thanks 7115; m. cria cried out his thanks 7144 mercier 1020, 1286, 4248a, 4318 v.trans. thank, 1020, 1286, 1351, 4041, 4268a, 4318, 5257, 6538 merir 6116 v.trans. requite 3012, 6026, 6116 mervelle 2330, mervelles 139, 7227; merveilles 3044, 6646; adv. wonderfully, amazingly 139, 1667, 6646, 7227; very much, intensely 2330, 7227 merveller 6728, 6890 v.reflex. marvel (at something), wonder, be surprised 6728, 6890, 6895 merveillos 1070, 1073, 2343, 2692, 2713, 2985, 3073, 3155, 3506, 3587, 3631, 3975, 5533, 6118, 6646, 6663, 6760, 7346, 7900, 8093; mervellos 939, 1667, 2410, 7882; fem. mervellose 1225, merveillose 2265, merveilose 4076, merveleouse 2394, merveilleuse 2660, mervelleose 4290, merveilouse 5473 adj. marvellous, wonderful 939, 1070, 3073, 3631, 6118; amazing, impressive, awe-inspiring 1073, 2343, 2692, 2713, 3587, 5533; terrible, fearsome 1225, 1667, 2265, 2394, 2410, 2660, 3155, 3506, 3975, 4076, 5473, 6663, 7882, 7900; singular, strange 2985; par m. sanblanz in marvellous fashion, wonderfully 1070; per m. samblanz, por m. sanblant in exceptional fashion, exceptionally well 3073, 7346; in terrible fashion, awefully 8093; adj.used adverbially amazingly, terribly 1667, 6646 mes adv. cf. mais mes sm.pl. messengers 546*, 6500, 6786 mesaisez adj. unhappy, troubled, hard-pressed 4410

540

Glossary

[mesaler] v.intrans. fail 3528; loi mesalée mistaken, erroneous religion (?) * 5814 [mesaler] (variant graphy of meseler) v.intrans. be leprous; loi mesalee corrupt, stinking, rotten religion (OR mistaken religion ?, cf. previous entry) * 5814 meschever 7830 v.intrans. have a mishap 4045, 7830 meschie (graphy of mischief) sm. misadventure, misfortune 3376 meschin subj.sing. 1685, 6067; mesquin subj.pl. 7931 sm. young nobleman, squire meschine sf. young (noble) lady, damsel 7087, 7093 mescosir 2716 v.trans. fail to pay adequate attention, misjudge 2716; fail to recognize 3690, 3812; n’i serez mescosi you shall not go unrecognized in this, you shall not be overlooked in this 1171 mescreance sf. doubt, uncertainty, worry 6786 [mescreire] v.trans. disbelieve, doubt 3660 mescreüz 2353, mescreü 4015 past part.used as adj. who refuses to believe (in Christ), infidel, faithless 2353, 4829, 4834, 7589; used as sm. unbeliever, infidel 4015 mesdisanz sm. slanderer 3608 mesire sm. my Lord, my Liege (942?), 6503, 6540 meslée sf. quarrel 693; fight 5809 mesléement adv. confusedly, all involved together indiscriminately 2395 [mesler] v.reflex. join battle with 5676; meslée past part.used as adj.f. joined in battle 3053; mixed in colour, partly grey-haired, ‘pepper and salt’ 6694, 6708a; meslez past part.used as adj. partly grey (haired) 60, 307 mesprison sf. evil, thing worthy of contempt, crime, outrage 491, 7397 mester 946, 1471, 2007, 3023, 4323, 4997, 5716, 6726, 6958, 6967, 7325, 7328, 7376, 7692, 7704, 7723; mestier 329, 1168; sm. need 329, 7692; workmanship 946; work, office 1168; trade 6726; aveir m. de have need of, need 1471, 4323, 7325, 7328, 7376, 7692; aveir m. a be of use, serve any purpose 2007, 3023, 4997, 5716, 6958, 6967, 7723; n’i a m. priere it is too late for praying 4997; or n’ai je mas mestier now I am no longer of any use, I no longer provide any service (to the cause of Justice and Righteousness, in the person of God, who is invoked here) 7704 mestre cf. maistre (sm. or adj.) mesure sf. measure, moderation 3006 meü, meüsent cf. moveir meudre cf. mellor meus cf. mieus mi possess.adj.m.pl. my 927, 965, 1438, 1959, 3792, 4749, 5361, 5491, 5609, 6050, 6465, 7022, 7260, 7599, 8152 mi sm. middle; per mi through the middle (of) 598, 2247, 2365; par mi through the middle (of) 1392, 2471, 2794; tres par mi right through the middle of 2384 miches sf.pl. small loaves of bread 5903 mie sf. crumb; ne … mie not one bit, not at all, (emphatic) not 22, 196, 252, 269, 351, 363, 417, 475, 594, 752, 772, 791, 798, 817, 867, 881, 1050, 1066, 1339, 1358, 1360, 1850, 1867, 1931, 2075, 2104, 2178, 2266, 2830, 3339, 3371, 3488, 3502, 3690, 3702, 3710, 3798, 3812, 4079, 4080, 4085, 4457, 4464, 4474, 4479, 4485, 4667, 4740, 4898, 5065, 5205, 5217, 5226, 5450, 5492, 5665, 5853, 5859, 5860, 5870, 5871, 6017, 6024, 6378, 6422, 6547, 6552, 6605, 6619, 6894, 6906, 6911, 6914, 6922, 7185, 7415, 7906, 7050, 8022, 8135; not…. anything whatsoever 6619 mien 177, 1929, 2980, 3174, 3317, 3450, 5048, 5464, 5534a, 6173, 6634, 6672, 7106, 7718; miens subj.sing. 4190, 7306; moie f.sing. 440, 443, 526, 530, 943, 945, 966, 1096, 1851,

Glossary

541

4100, 4622; possess.adj.(stressed form) my; de moie part on my behalf 440, 443, 526, 943, 945, 966, 4622; moie cope (mea culpa) forgive me, redeem me 4100; par lo mien esciant from what I know of the matter, to my knowledge 2980, 5534a, 6672 (see also under esciant); per lo mien escïentre by my faith! 3174 miens m.subj.sing. 356, moie f. 1364, 2894, 7072 possess.pron. mine mieus 52, 3652; mielz 2869; meus 55, 70, 217, 523, 823, 947, 1013, 1130, 1133, +; adv. better 52, 70, 523, 2869, 3006, 3652, 4742, 5141, 8137, 8142; better off, in a better situation 5897; rather, sooner, in preference 55, 823, 1133, 1452, 1888, 1904, 1936, 1942, 2898, 2976, 4961, 7529, 7946; best 508, 1130, 2688, 3202, 6969, 7394; au meus qe as best, as well as 217; meus de better than, more than 947 miller 125; meillier 2196, 2486; meller 2957; miler 3286; meiller 5140; sm. thousand mire sm. physician 4507 mirer 1547, 1945 v.trans. look on, see, evaluate 1547, 1945, 2028 mise obj.sing. 197, misses obj.pl. 530, messe obj.sing. 987, 1557, 5537, 7360; messes obj.pl. 5904, 6485; mese obj.sing. 1568, 2737 sf. Mass, the Eucharist misoldor 6239, 6257, 6374 adj. magnificent, superb (of horse) missoudor 5542 sm. prize courser, magnificent war-horse moënez (graphy of monoiez) adj.pl. coined, in coin; diners m. coins, money 6344 mogne sm. monk 5085 moie cf. mien moillier sf. wife 53, 526, 5965, 6530 molé obj.sing. 2024; molez subj.sing. 391; adj. well-formed, shapely moluz 1415, 1538, 2729, 2904, 3334; moliz 2243; adj. newly-sharpened, sharp, trenchant (used especially of lances) moncel sm. mound, hillock, (small) hill 7909 moniaus adj.pl. monastic 5905 mont sm. world 363, 4229, 7131, 7224, 7871 monte sf. value, price 7771 monument sm. tomb 5054 [mordrir] v.trans. kill, assassinate 3154, 4649 more sf. point (of sword) 5402 morel obj.sing. 2377, morois obj.sing. 6259, obj.pl. 4952; adj. black or brown (horse) (Also used as name of horse 6335, 6360, 6403, 6407, 7373, 7905) morir 823, 1133, 1210, 1420, 1452, 1468, 1499, 1542, 1862, 1888, 1904, 1942, 2149, 2451, 2577, +; moris (poet.lic.) 3226; pres.indic.1 mur 922, 1493, moert 1793; muert 1973; 3 muert 2900; 5 morez 1505; 6 morent 2423, muerent 2453, 2807, murent 2288, 6655; pres.subj.1 muire 523, 4035, moire 691; fut.5 morerez 675; past def.6 morurent 3155; v.trans. kill 1062, 2288?, 2300, 2356, 2423?. 2485, 2957, 3110, 3648?, 4311, 4762, 4837, 5223, 5454, 5547, 5630, 5812, 6666; v.intrans. die 523, 675, 691, 823, 922, 1133, 1210, 1493, 1505+; (both meanings are possible in ll.2288, 2423, 3648) mort sf. death; a mort fatally wounded, dying 3480 morois cf. morel mortel 745, 2555, 4446, 4868, 5109, 6418; mortal 1644, 6216, 6218a, 6219, 6364, 7879, 8119, 8154; morter (poetic licence for rhyme) 7056 adj. mortal, deadly 745, 1644, 6418, 7056. 8119; cruel, fatal 4468, 5109; deadly, death-dealing, involving many deaths 4446, 6216, 6217, 6364, 7879; per rien m. for any mortal thing 6218a; juïse m. death sentence 8154 mortuaus adj.f.pl. of the dead, funeral 5904

542

Glossary

mostier obj.sing. 1572, 6774, 7048, 7203; moster obj.sing. 2645, 3032, 6155, 6743, 7101, 7109, 7118, 7158, 7293; mostrier obj.sing. 2955; mostiers obj.sing. 7162, obj.pl. 5898; mostriers obj.pl. 6987; sm. monastery, religious house mostrer 19, 79, 1000, 1016, 7070, 7318, 7731, 7825; v.trans. explain, expound 19, 79, 887, 3236, 5399, 7318; show 793, 1000, 1016, 6394, 7070, 7038, 7330, 7755, 7825; show, demonstrate 2280, 5097, 7731; point out, indicate 5389; give a sign (?) 5778 mot 9, 68, 85, 135, 141, 151, 152, 207, 212, 242, 248, 250, 279, 292, 299, 325, 391, 404, 421, 477, +; mult 178a, 385, 399, 456, 462, 476, 494, 534, 651, 697, 717, 831, 1053, 1143, 1145, 1191, 1266, 1386, 1423, 1466, 1469, 1524, 1735, 1763, 1782, 1805, 1821, 1822, 1854, 1918, 1998, 2013, 2052, 2054, 2065, 2084, 2091, 2142, 2148, 2193, 2209, 2210, 2231, 2250, 2265, 2277, 2370, 2552, 2703, 2886, 3247, 4030, 4604, 4841, 5129, 5241, 5620, 7607, 8043; mout 388, 669, 771, 852, 953, 954, 993, 1124, 1224, 1227, 1235, 1242, 1255, 1256, 1282, 1286, 1303, 1526, 1533, 1553, 1555, 1582, 1695, 2147, 3048, 3087, 3383; molt 1797; mut 1852; adv. much, very much, greatly, a great deal 9, 68, 85, 135, 151, 152, 207, 212, 279, 583, +; very 141, 242, 248, 250, 292, 299, 325, 391, 404, 421, 477, +; m. per, m. par a very great deal, very much 399, 651, 5607, 5646, 5777, 6500, 7098, 7686, 7887, 7987, 8013, 8046; mot per a very long time 4004; pron. many, a great many 5304, 5797, 7050 mot obj.sing. 29, 1366, 1497, 1557, 2040, 2074, 2878, 3261, 3286, 3381, 3428, 3475, 4770, 6226, 6849, 6855, 7092, 7242, 7707, 7991, 8181; moz obj.sing. 3928, obj.pl. 275, 2065, 5282; sm. word 29, 275, 2065, 5282; statement 1497, 3928, a icest m. thereupon 1366, 3261, 3286, 3381, 3428, 3475, 7092, 7242; aprés cest m. after saying this 1557, 2040, 2878, 4770, 7707; il ne set mot he does not know anything about it 2074; mot ne lor respont he does not say a word to them in reply 6226; n’en a il m. soné he did not say a word 6849; ne dit mot he did not say a word 7991; teus n’en sot mot such a one did not know anything about it 8181 [moveir] pres.indic.3 muit 4471, 6 muevent 7706; past def.3 mut 6244, 7592, 7720; imperf.subj.6 meüsent 4059; past part. meü 2169, 6477; mu 6479; v.trans. move, displace 4059; move, stir, take action 4471; move (feelings), upset 6479; set in motion, cause 2169, 6244, 6477, 7592, 7720; v.reflex. move, stir 7706 mucer v.intrans. hide 4307 muer 426, 1217, 1223, 2015, 3399, 3744, 4380; v.trans. change 767, 2497, 3524, 3745, 4917, 5175; prevent, restrain oneself (from doing something) 426, 1217, 1223, 3399, 4380 muez adj. (of hawks) moulted, in full feather 153, 223 mult cf. mot musart sm. scatter-brain, idiot 1267 mut adv. cf. mot mut cf. moveir

n [naistre] past def.2 nasquis 5972, 3 nasqui 4157; past part. né 474, 1137; nez 62, 822, 2270, 2997, 3637, 3665, 3947, 4102, 4879, 5666, 5670; nasi 3556; v.intrans. be born nas cf. neis nasel 930, 4010, 7977; nasé 3370; nasal 5519, 5741 sm. nose-piece of helmet naturé (graphy of naturel, poetic licence for rhyme) adj. natural 8162 navie sf. navy, fleet 4536, 4539, 4545; ship 4579; a navie by ship 3712

Glossary

543

[navier] v.trans. navigate 4567 navrer 4367 v.trans. wound, injure 1054, 2514, 2522, 3364, 3477, 3572, 4367, 4782, 4843, 5543, 7947; se sofri a navrer resigned himself/submitted himself for wounding, i.e. allowed himself to be wounded 4367 nazaire (graphy of nasart, or possibly nasel,modified to suit rhyme) sm. nosepiece of helmet, here used simply for helmet 5746 nef 825 cf. neveu nef 4273, nés obj.pl. 4548, 4819, 4822 sf. ship negiez adj. full of snow 4402 negun 1934, 2649, 3367; neguns subj.sing. 6807; adj. (+ ne …) not any, no neiez cf. noier neis 605, nas 7979 sm. nose neïs 1388, 4550 adv. not at all 1388; not even 4550 neiz (graphy of naïz, subj.form sing. of naïf) adj. used as sm. native (to), thus inhabitant (of) 5366 neli (graphjy of nului, nelui) pron. no-one (especially indirect object to no-one) 165 nenel 4471 neg.art. and adj. no nenil adv. no, not at all, not one (used as single, strong negative response to question) 6350 neporquant adv. nevertheless, even so 3025 neqident 3944 adv. nonetheless nes 860 graphy of neïs adv. not even 860; (not at all 3029* ?) neu sm. knot 4022 neveu obj.sing. 527, 2128, 3113, 4340, 4343, 4602, 5149, 5454, 5521, 5699, 5927, 5959, 5960, 5983, 6478; nevou obj.sing. 755, 4210, 6180; nef obj.sing. 852; nevor obj.sing. 1207; neveo obj.sing. 5120; nevo obj.sing. 7322, 7472; niés subj.sing. 871, 1127, 1221, 1240, 1254, 1261, 1593, 1597, 2100, 2104, 3294, 3537, 4211, 4224, 4654, 5895, 5928, 5999, 6016, 6474, 7007, 7300, 8064; sm. nephew nïent 428, 520, 2421, 2429, 2444, 2801, 3503, 3508, 4113, 5316, 5367; noient 594, 815, 951,5057, 6824; noianz 3435; noiant 5236; nianz 3455; niant 3876, 5957, 6009, 6671; sm. (without ne) anything 5236; (+ ne …) nothing 2429, 2444, 3476, 3503, 3508, 3876, 5057, 5957, 6009; used adverbially not at all 2421, 2801, 3435, 3455, 4113, 5316, 5367, 6671, 6824; de n. not at all 428; not anything, nothing 594; por n. in vain, to no purpose 520, 815; without cause, needlessly (ie there is no need to ask, that goes without saying) 951 nigromance sf. magic, sorcery 6789 nivole sf. cloud, mist 5808 no (graphy of non) adv. not (emphatic) 235 noals subj.sing. 2988 sm. disadvantage, worst of it noçoier 6531, 6888 v.trans. marry noef num.adj. nine 6879 noël sm. niello; a noël (of lance) nielloed (ie with an incised pattern filled up with black enamel) 2383; del noël with the lance (abbreviation for del espi a noël) 7918 noëlé past part. used as adj. (of lance) nielloed, inlaid with black enamel 970, 2426; anel a cristal n. ring inlaid with crystal 955 noer 4275; pres.part. noiant 5974 v.intrans. swim; pres.part.used as adj. which swim, swimming 5974 noer 7996, 8144 v.trans. knot, tie in a knot noiant 5236, noianz 3435 cf. nïent; 5974 cf. noer

544

Glossary

noient cf. nïent noier 4299, [neier] v.intrans. drown 4299, 4715, 4810 noier 4322 v.trans. deny nois subj.sing. 777, 6643; nos subj.sing. 5393 (graphies of noi(f)s) sf. snow noisse sf. noise, clangor, clash of weapons, uproar 1380, 2099, 2103, 2598, 3706, 4005, 5131, 5228, 5605, 5700, 6151 noit obj.sing. 8080, subj.sing. 8012; noiz subj.sing. 2449 sf. night nomer 32, 4423 v.trans. name 518; cite by name 1873, 4423, 5783; designate 32, 5007 non sm. name 487, 1598, 2244, 2699, 2829, 3184, 3276, 3406, 3911, 4825, 4879, 5183, 5336, 6053, 7270, 7712, 7727, 8020; nom 1182, 1318, 2270, 5431; nun 7270; jugirent por n. adjudged by name (ie nominated) 487 noncier 982 v.trans. announce, proclaim 982, 3340, 3694, 4772 none sf. the ninth hour of the day, i.e. 3 pm 6799 [norcir] (graphy of nercir, noircir) v.trans. blacken, turn black 2835, 7028, 7034 [norir], [norrir], [nosrir] v.trans. have in household, have in one’s service, thus provide for, support 1978, 4455, 4598; bring up, educate 4146, 5625, 5782, 6575, 6621, 6819, 7240; past part.used as adj. well educated 207; past part.used as sm.(subj.sing.) person one has brought up and educated, protégé 212 norois adj. Norwegian, proud 853, 6264 (or Norois, the name of Oliver’s horse 853 ??) nosches sf.pl. necklaces 944 novele sf. riposte, answer 2229 nueme ord.num. ninth 5075, 5277, 5294, 5313 nun cf. non nus, graphy of nos, possess.adjective our 25 nus, graphy of nuls (nul) adj. no 414, 822, 848, 1535, 1549, 1710, 1800, 1069, 2488, 2841, 3947, 4179, 4219, 4921, 5204, 6134, 6637, 6751, 6978, 7030, 8137; pron. no-one 115, 408, 423, 631, 991, 1407, 2207, 3544, 3595, 4736, 5209, 6091, 6162, 7840, 8199; not one 306, 308, 2852, 5414, 6258, 7124 nus, graphy of nos, personal pron. we (subject.) 268, 922, 2638, 3275, 3444, 4513, 4667, 5948, 7284; us (direct &indirect obj.) 65, 66, 480, 583, 539, 701, 1421, 1464, 1553, 1702, 1779, 1838, 1877, 1965, 2313, 2503, 3440, 4125, 4415 nus plur. of nu adj. bare 732

o o1 neut.pron. that 1066 o2 52, 176, 380, 652, 1337, 2812, 3257, 3741, 4608, 5078, 6096, 6261, 6365, 6616, 7024, 7275, 7459, 8053, 8076; ou 52, 64, 176, 983, 1291, 3010, 3212, 3653, 4275, 4344, 4551, 5498, 5636, 5687, 6148, 6149, 6261, 6676; conj. or, either …. or; ou velle o non whether (one) likes it or not 380, 1337, 2812; o veille o non whether he likes it or not 2369, 3417, 3865; o velle o non whether he likes it or not 3741 o3 adv. yes; ni o ni non neither yes nor no 258, 618, 7025 o4 2113, 2574, 2627, 2997, 3164, +; ou 733, 1052, 1066, 1640, 1842, 2908, 4603, 5093, 5406, 5785 adv. anywhere that 733; where 455, 1052, 1640, 1693, 2113, 2523, 2627, 2997, 3124, 3164, 3340, 3415, 3420, 3772, 3773, 4007, 4161, 4166, 4178, 4273, 4565, 4645, 4754, 4765, 4978, 5494, 5773, 5885, 5886, 6677, 7155, 7188, 7194, 7219, 7507, 7977; how 1066; wherein

Glossary

545

1735, 3687, 4017, 5093, 5785, 6612; in whom 4603; whereon 1842, 2908, 3627, 5406; on whom, to whom 4862, 5362; to which 7694; ou tu fuz where you were, i.e. from which you come 3340; interrog. Where? 2309, 2372, 2574, 3802, 3856, 4781, 6442, 6541, 6887, 6915, 7634 o5 82, 120, 124, 147, 226, 234, 484, 545, 671, 975, 1004, 1014, 1044, 1131, 1413, 1948, 2353, 2411, 2549, 2571, 2610, 2662, 2826, 2863, 3000, 3043, 3052, 3072, 3154, 3186, 3205, 3221, 3222, 3237, 3271, 3292, 3415, 3426, 3566, 3582, 3704, 3725, 3784, 3806, 4184, 4463, 4499, 4509, 4524, 4558, 4717, 4733, 4829, 4852, 4887, 4926, 4969, 4985, 4987, 5076, 5077, 5087, 5134, 5245, 5247, 5304, 5319, 5352, 5420, 5545, 5546, 5576, 5645, 5672, 5814, 5843, 5879, 5884 (2), 6115, 6141, 6209, 6215, 6326, 6463, 6483, 6607, 6694, 6710, 6745, 6829, 6838, 6839, 6872, 6915, 6921, 6993, 7043, 7044, 7059, 7088, 7187, 7349, 7607, 7650, 7660, 7703, 7986; ou 95, 1291, 1370, 1415, 3205, 3213, 4726; prep. with; (See also enscenble o) oberc cf. auberc oblier 196, 2077, 3348, 4372, 6017 v.trans. be forgetful, be remiss (in terms of religious obligations) 196, 3644, 6314, 7181; forget 527, 2077, 4372, 7005, 7369; overlook, leave out, omit 968, 3348, 5003; fail to recognize 3376, 7068; v.reflex. be forgetful, be remiss in one’s duties 6832 occire 5970, ocire 7280, oucire 54; oncire 3506, 4916; v.trans. kill 54, 524, 567, 621, 638, 710, 769, 914, 922, 933, 971, 1184, 1247, 1340, 2252, 2301, 2356, 2386, 2612, 2645, 2647, 2740, 2759, 2766, 2772, 2790, 2911, 2961, 3175, 3229, 3233, 3249, 3270, 3475, 3506, 3541, 3760, 3797, 3980, 3985, 4186, 4193, 4266, 4267, 4287, 4398, 4654, 4668, 4713, 4750, 4752, 4916, 5432, 5584, 5618, 5678, 5699, 5886, 5970, 6006, 6050, 6280, 6512, 6599, 7024, 7391, 7529, 7896, 7903; v.reflex. kill oneself 7280 oïanz 3614, oïance 4941 sf. hearing; en o. in the hearing of others, in the presence of witnesses 3614, 4941 oïe sf. ear 374, 4477, 7183; hearing, sound, distance over which something can be heard 3706, 5650 oil obj.sing. 3778; obj.pl. 743, 1212; subj.pl. 3365, 3384, 3571, 4009, 4040; els obj.pl. 342, 394, 1502, 6006, 6058 (?)*; subj.sing.1336; subj.plur. 3863; sm. (plur. often treated as collective noun with singular verb) eye, eyes; vos els véant in your sight, in your presence 342; nos els voiant before our eyes 6006; a vos els before your eyes 1502; mist ses els a bandon poured tears from her eyes in self-surrender?, wept in self-abasement? 6058* oimais cf. humai oir sm. heir 5338 oïr 656, 2145, 2158, 2573, 3602, 3654, 5597; pres.indic.1 oi 3068, 3538, 3549, 3601, 3726, 5228, 8056; 3 ot 602, 2133, 4109; out 166, 1897, 1907; 5 oiez 19, 1515a,1994, 4760, 5167; oëz 336, 745, 1114, 3715, 7619; 6 oient 4244, oent 5493; fut.2 osras 4698, 4909; 3 osra 67, 1459, 1892, 2978, 3036; ora 1425, orra 1433, 1445, 1919; 4 osrons 7453; 5 orez 32, 1648, 1914; osrez 662; orrez 524, 1305, 5017, 5297; past def.1 oï 1011, 3377, 4271, 4888, 7454; 3 oï 3171, 4005, 4917, 5184, 5574, 6637, 6738; oït (but monosyllabic) 34, 677, 1083, 3001, 3044, 5820; ot (graphy of oït) 373, 3016, 3707, 4532, 4798, 5619, 5772, 5827, 6536, 6823, 7111, 7631, 7952; past subjun.5 oïsiez 1200; oïssiez 5605, 6484; past part. oï 3496, 4261; v.trans. hear 19, 32, 34, 373, 524, 602, 656, 677, +; recognize 3377 oïr sm. hearing 3385 oisel obj.sing. 2291, subj.pl. 4586; oseus subj.pl. 1068; ousel subj.pl. 1567 sm. bird oitisme 5060; uitime 5277; uime 5294, 5313; ord.num.adj. eighth [oloir] pres.indic.6 olent 3962; v.intrans. smell, be scented 3962 olie sf. olive-wood 4476

546

Glossary

olifant obj.sing. 1432, 1473, 1874, 1906, 1946, 2072, 3046, 3061, +; olifanz subj.sing. 3081 sm. ivory horn 1432, 1473, 1874, 1906, 1946, 2072, 3046, 3061, 3081, 3170, 3243, 3534, 3549, 3869, 3984, 4038, 4047, 4967, 5251, 5988; ivory 4592 olive sf. olive-tree 15, 776, 896, 4453 oltré past part.used as adj. exhausted, dead 4797 once sf. ‘ounce’, one sixteenth of the mediæval ‘livre’ or pound (ie about 30 g.) 1570 oncire cf. occire onie cf. huni onor 529, 820, 1203, 1301, 1427, 2122, 2864, 4208, 4849, 6253, 6509, 6516, 6905; honor 1804, 2340, 2865, 3250, 4348, 4706, 4937, 6380, 6820, 7207, 7872; henor 6544 sf. honour shown to a person 4348, 4937, 5883, 6820, 7207; honour, (good) reputation 529, 820, 1427, 1804, 2864, 3250, 4706, 6380, (ironic) 6509; honourable distinction or renown enjoyed by a person 2122, 2340, 2864, 3250, 4208, 4706, 6253; fief, feudal possession, dominion 1203, 4849, 5680, 5878, 6516, 6905; honours (of battle), victory 1301, 7872; distinction, nobility 4937; plur. titles, status, authority 4187; faire h. give privilege to, thus protect 2865; show honour and respect to, treat with honour 5883; do honour to, give high privileges or offices to 6820 onqes 408, 445, 1412, 1416, 1535, 1859, 1873, 2090, 2488, 2715, +; onques 3231; unqes 1289, 5396, 7120, 7335; adv. ever 3540, 6549, 7634; o. jor (emphatic) ever at any time 3540; (+ ne) never 408, 445, 1289, 1412, 1416, 1859, 1873, 2090, 2488, 2715, 2737, 2844, 3420, 3490, 3496, 3608, 4014, 4329, 4332, 4359, 4363, 4921, 5238, 5259, 5514, 5557, 5559, 5618, 5664, 5772, 5894, 6610, 6769, 6778, 6987, 7312, 7544, 7776, 7867, 8058, 8059; o. mas ne 1416, o. mais ne 1535, 3293, 3868, u. mais ne 5396, 7120 never (emphatic); onqes puis never afterwards, never since 2864 [onser] (graphy of oser) cf. ouser ontage cf. hontage or cf. ore ordenéement adv. in good order, in orderly fashion 5352 ordener 1551 v.trans. order, arrange, dispose forces 1551, 4977, 5006, 5043; clerc ordené clerk in holy orders, ordained cleric (OR simply taken as verb arranged, disposed?) 7819 ore (graphy of or, conj.) 396, 1074, 1298, 1299, 3096, 5230, 7090, 7091, 7198, 7314, 7878; adv. now; d’ores en altres from time to time 5461 ore 3520, 4102, 4343, 4544, 5250, 6920, 7357, 7418; or 6542; oure 7379 sf. hour oré 4369; ori 716, 4083 adj. gilded, golden orendroit adv. now, at once, straightway 4153, 4498, 4746 [orer] past def.3 ora 6070 v.intrans. pray orfenis (subj.sing. of orfenin) sm. orphan 2607 orfenté sf. deprivation and poverty, especially being deprived of one’s children, loss of those one holds dear, thus distress, affliction 3575, 4097, 7022 orfrois sm. ‘orphrey’, i.e. an embroidered fabric with gold braid or strands of gold wire worked into the cloth, brocade 102; (of saddles) inlaid with gold, tooled with gold 4949 orguel sm. arrogant man 269, 851; arrogance 340, 381, 1312, 4008, 7664, 7825; pride 834, 2542, 3086, 3322, 5183, 5249, 5390, 7756, 7789; what I took most pride in 8183; l’orguel q’il selt mener the pride that he was wont to show 2542 orieflor sf. oriflamme 1237 orine sf. descent, family, race 7008, 7088 orison sf. prayer 7125, 7200

Glossary

547

orles subj.sing. sm. (in heraldry) a band or border which does not touch the edge of the shield, but follows its shape 795 orlenois adj. of Orleans 4962 ors 1053, 1059, 1061, 8076.; hors 6659, 6735 sm. bear 221, 616, 1053, 1059, 1061, 1229, 4411, 4429, 4436, 6042, 6659, 6735, 6739, 8076, 8097 ors (graphy of or, 17, 45, 86, 102, 103, 116 +) obj.sing. 1161, 3189, 4478; subj.sing. 861, 1693, 1823, 1842, 2908, 5444, 5899; sm. gold ortel sm. toe 5997 osberc cf. auberc oseus cf. oisel ost obj.sing. 61, 253, 874, 1022, 1027, 1031, 1044, 1263, 1559, 1589, 1631, 1908, 2334, 2614, 2826, 3052, 3543, 4406, 5157, 5176, 5244, 5348, 5597, 6086, 6151, 6228, 6303, 6336, 6438, 6598, 6857, 6867, 6875, 6918, 7316, 6364, 7481; subj.sing. 238, 5837, 5841; oz subj.pl. 1457, 1928, 2049, 2070, 2088, 4542, 4997, 5322, 5350, 5416, 5475, 5686; obj.pl. 1547, 2080, 6100, 7315; sf. army, host; ost banie feudal levies 2826, 5244, 6918 ostager 8079 v.trans. give as payment for a night’s lodging ostal 5517, 8109, 8122; ostel 6056, 8079 sm. lodging, accommodation 5517, 6056, 8079, 8122; household, people to look after him 8109 oste sm. host 7499, 7791 osteier 828; ostoier 6725 v.intrans. make war, campaign 828, 6725, 6928, 6931 osteler 190, 2218, 2747, 4358; v.intrans. be lodged, be housed, sojourn, stay 190, 2218, 2747; camp, stay overnight 4358; del osteler infin.used as sm. finding shelter, lodgings 6153 ostelerie sf. inn 4488 ostoier cf. osteier ostors obj.pl. sm. goshawks 153, 223 ostroër cf. estroër otrier cf. otroier otroiement sm. permission, authorization 7111 otroier 49, 2873, 7337; otrier 57, 670, 1545, 4325, 7107, 8132; pres.indic.1 otroi 5243, 7855; pres.subj.3 otroit 1741, 3818, 3925, 4206, 4208, 5082; v.trans. permit 235, 301, 797, 4325; agree to, accept 49, 57, 235, 420, 670, 800, 1545, 7337, 8132; grant 1741, 3818, 3925, 4206, 4208, 5082, 5243, 5253; formally authorise, make (gift) 5243; affirm, accept as true 2873, 7855 ou (graphy of al, au, el = en le) prep.+art. on the 126, 936, 1145, 1308, 2128, 3741; in the 454, 997, 1054, 1350, 1479, 1713, 5530 ou adv. cf. o ou conj. cf. o ou prep. cf. o oucire cf. occire oure cf. ore ousel cf. oisel [ouser], [onser] (graphies of oser) pres.subj.3 oust 328, 1088; fut.6 ouserunt 1949; imperf.subj.3 onsast 29; v.intrans. dare 29, 328, 1088, 1949, 6113 ousez adj. bold, arrogantly foolhardy 540, 2994 outre prep. beyond 3689, 5009 ovre sf. work 796; workmanship 5079, 5092; pl. works, deeds 7645 ovreigne 2341, ovregne 4238 sf. action, deed ovrer 7368 v.trans. work 1786; v.intrans. act, behave 5848, 7368

548 oz oz oz oz

Glossary

adj. bold, rash 5267 cf. ost obj.pl. sm. bones 7288 (variant els) pron.m.pl. them 5265

p paënie 740, 4463; paganie 1834; paienie 3068, 3303; adj.f. (found alongside paiene 105, 1470, 2417, 2470, 2474, 2890, 3063, 5350, 6166, f. of paien) pagan paestez cf. poestez paienie sf. pagan lands, thus Muslim countries 5254; the Muslim race, the Muslim faith 5451 paienor adj. pagan, of the pagans; jent p., gent p. pagan people 1393, 1808, 4207 paile adj. pale 3354 paille 777, 5191, 6473, 7263, 7268; paile 1228, 6609, 7078, 7202; palies obj.pl. 126; sm. piece of rich cloth of gold, brocaded or striped silk (from Alexandria or Egypt) 126, 615, 1228, 5191, 6609; awning 777; shroud 6473, 6775, 7078, 7202, 7263, 7268 [paindre], [peindre] past part. pent 1814, peint 5406, 8030; peinz 3190; paint 7507, 4495, 5035; v.trans. paint 1814, 3190, 5035, 7507, 8030; depict 5406; painte a gerons decorated with panels of fabric(?), painted with heraldic devices (?), made colourful (?) or hung(?) with tapestries * 4495 paire sm. father 5747; applied to God 5313 paissanz subj.sing. sm. countryman, peasant 6341 paissons obj.pl. sm. tent-pegs 5143 [paistre] pres.indic.3 past 4387; past def.3 pot 5890; v.trans. feed on (grass), graze palacin 1683; palazin 3898, 6069; adj. (used of count) of the court, (count) forming part of Charlemagne’s entourage 3809, 6069; sm. paladin, noble warrior of Charlemagne’s court 1683 palefreis obj.pl. 1374; palefroiz subj.sing. 5889; sm. palfrey paluz obj.pl. 2912 sm.or f. marsh, fen pan sm. streamer (of gonfalon) 2137, 2764; piece, portion 5254; skirt (of hauberk) 5513; piece, corner (of cloak) 6084 paor 3195; poor 602, 766, 1210, 1221, 1225, 3420, 3738, 4737, 5234, 5543, 5555, 6712, 6722; poür 4564, 5214, 7545; sf. fear paoros 6655; poroz 5279; adj. frightened, full of fear par, variant graphy per superlative adv., used to emphasize meaning (with adj.) especially, very 2240, 2660; (with adv.) right 2827; thoroughly 3475; tant par so very much 201, 407, 1962, 2851, 3617; tant per so very much 3099, 3153, 3876, 4345, 4750; mult par exceedingly 651; mot per for a long time 4004; a very great deal, greatly 5326, 5607, 5646, 5777, 6500, 7098, 7686, 7887, 7987, 8013, 8046; trop per too much 5745; par dalez right alongside 2205; per dales just alongside 7049; per devant right in front of 761, 2827; per dedroi from behind 7554; per desus right over 7909, on top 8141; per desoz immediately under 7916; per son l’aube, par som l’aube at the very crack of dawn 6606, 8032; per som immediately above 4458, right over 7480; par som at the very beginning 6930; per son right up on 7509; per ont oncis they have (finally and completely) killed 3475; de par moi by me (emphatic) 1884 par 28, 84, 107, 169, 239, 281, 294, 348, +; per 52, 59, 60, 94, 110, 126, 148, 219, 345, 591, +; prep. by 28, 59, 60, 84, 94, 126, 294, 345, 348, 368, 392, 398, 406, 449, 472, 496, 532,

Glossary

549

611, 673, 729, 787, 924, 956, 1152, 1219, 1268, 1348, 2589, 2917, 2996, 3302, 3325, 3423, 3703, 3785, 4015, 4042, 4100, 4173, 4194, 4297, 4475, 4586, 4768, 4795, 5113, 5550, 5567, 5747, 5824, 5978, 6053, 6054, 6055, 6060, 6112, 6293, 6358, 6367, 6431, 6853, 7402, 7459, 7471, 7623, 7672, 7674, 7675, 7682, 7731, 8063, 8074, 8080, 8086, 8092, 8095, 8132; throughout 110, 1621, 2278, 2401, 2467, 2469, 2793, 3543, 4229, 4385, 4534, 4601, 4754, 4946, 5015, 5176, 5597, 5652, 5846, 5850, 5890, 6151, 6867, 7131, 7316, 7532; through, because of 239, 332, 405, 812, 911, 912, 1609, 1626, 1838, 1934, 1938, 2388, 2413, 2555, 2606, 2619, 2861, 3007, 3149, 3376, 3515, 3915, 4189, 4472, 4515, 5454, 5665, 5945, 6218a, 6412, 6926, 6980, 7447, 7897, 7943, 7945, 8004, 8050; through 3828, 3993, 4108, 4109, 4181, 4270, 4324, 4965, 5119, 5260, 5391, 5430, 5504, 5873, 6506, 6859, 6896, 6955, 7144, 7679; via, by means of, through the good offices of 148, 219, 591, 623, 689, 1030, 1324, 1392, 2343, 2346, 2390, 2492, 2513, 2538, 2912, 3225, 3940, 4030, 4473, 4507, 4511, 4525, 4767, 4783, 4811, 4856, 4883, 5138, 5160, 5285, 5332, 5504, 5775, 5779, 6099, 6163, 6175, 6235, 6268, 6349, 6527, 6841, 6985, 6986, 7123, 7132, 7137, 7383, 7458, 7746, 7962, 8112, 8184, 8187; as, in terms of 452, 794; with 578, 639, 648, 760, 1070, 1094, 1255, 1714, 1727, 1730, 1797, 1805, 1816, 1819, 1852, 1862, 1877, 1915, 2130, 2134, 2153, 2189, 2355, 2410, 2418, 2494, 2496, 2498, 2508, 2556, 2564, 2568, 2691, 2692, 2700, 2713, 2753, 2786, 2797, 2810, 2848, 2853, 3048, 3073, 3074, 3130, 3192, 3209, 3470, 3486, 3833, 3932, 3963, 4021, 4060, 4122, 4242, 4491, 4598, 4692, 4943, 5069, 5126, 5340, 5525, 5558, 5655, 5998, 6059, 6118, 6255, 6457, 6670, 7664, 7748, 7908, 7910, 7914, 7949, 7989, 8047; in 2832, 3073, 3155, 4082, 4240, 4436, 4489, 4652, 4907, 5095, 5124, 5183, 5333, 5575, 5718, 5930, 6262, 6372, 6377, 7060, 7422, 7559, 8042, 8042, 8093, 8101; on 1788, 3779, 3950, 4833, 5462; (graphy of por) in order to 243, 7654; for 706, 2189, 2967, 3015, 6218a; as 2967, 4013; (in exchange) for 3952; for, on behalf of 2983, 7731; in respect of 246, 8166; par al otherwise, in any other way 1642; par coi wherefore 281; per qe for what reason, wherefore 7627; per mi through the middle of 598; par mi 1392, 2248, 2365, 2471, 2794; tres par mi right in the middle of 2384; per tans, per tens very soon, promptly 4772, 8096; per un petit almost, very nearly 2965; per un petit ne chiet tot estendu he almost fell full length 5768; per un petit q’il ne s’en vont fuianz they are almost on the point of fleeing 3634; per un petit qe li cuers ne me fant my heart is almost breaking 6001; per un petit q’il n’a le cuer fendu his heart is almost broken 6472; per un petit ne l’a tot trebucié he almost stumbled and fell with her, almost dropped her 7256; per un petit qe nel fiert par dedroi he came close to striking him from behind 7554; per un poi ne l’a fait … he almost caused him to … 7973; de per on behalf of 6379; per foi by my faith!, i’faith 3325, 6895, 7618, 8132; par loisir in a leisurely manner, without haste 169; per mot fiere tenzon in a very fierce attack 4436; per nos cors anbesdous in both our persons, i.e. in single combat 6393; per proisere in number 4987; per raison logically 5752; per soi by his own hand 8063; per tans, per tens very soon, promptly 4772, 8096; par tel division with (such) great skill 2691; per tel division with specific stipulation 3134; par divisement as God has decreed (? OR quite separately?) 7131; per tel condicion with a specific proviso (that) 6457; per terre on the ground 5462; per les testes head first 4315; per un e un one by one 8080; per lo mien esciant/escientre/escient/escianz, par o men esciant to my certain knowledge, by my troth 348, 1113, 2980, 3174, 3317, 3450, 5534a, 6672 parage sm. (noble) birth, good family 764, 2286, 2971 [paramer] v.trans. love deeply, love dearly, be very fond of 344, 2454, 4378, 5992 parastre subj.sing. 1086; persastre subj.sing. 1401 sm. step-father [pardoner] pres.indic.1 pardoins 3380 v.trans. pardon 1611, 3378, 3380

550

Glossary

pardurable adj. everlasting, eternal 3818 [pareir] pres.ind.3 part 4735; 6 perent 1237; v.intrans. appear 1237; v.impers. appear* 4735 paringal 1646; perigal 1940; parigal 8149 adj. absolutely equal parlament sm. conversation, discussion 4852 parlance subj.sing. sf. talk 4940 parllant adj. capable of speech, (once more) able to talk 4170 parmi 1567, 2468, 2678, 3327, 7518; permi 1923, 2155, 2175, 2216, 2449, 2596, 2630, 2675, 2805, 3105, 3248, 3426, 3479, 3483, 3492, 3679, 3754, 3959, 4482, 5270, 5447, 5502, 5514, 5554, 5723, 6123, 6266, 6397, 6831, 6857, 6063, 7881; prep. through, into 1923; through 2155, 2175, 2216, 2449, 2468, 2630, 2675, 2678, 3328, 3426, 3479, 3483, 5447, 5514, 6266, 6397, 6831, 6963, 7518; amongst 1567, 2596, 5502, 6123; around 3754; through the middle of, thus between (?) 3959, 4482; in the middle of 5270, in 7881; throughout 6857; adv. through 3248, 3679, 3492, 5554, 5723; trés parmi right through the middle 3327; (cf. also mi, per mi and par mi) [paroler] v.intrans. speak 631, 5950, 7033, 7098, 7184, 7617, 7620, 7671 [paroir] pres.ind.3 part 4735; 6 perent 1237(?); fut.3 parra 6252; v.intrans. appear, be seen 1237; impers. bien part a .. it is evident from 4735; or i parra se … now it will be apparent if …, now we shall see whether …6252 parti sm. part; parti igal a flat area 8111 partir 2162, 2570, 2717, 5742, 6105, 6940, 7315; v.intrans. take leave of 1178, 2682; leave 1642, 2162, 2188, 2196, 2225, 4458, 6014, 6228, 6896, 6926, 7042, 7081, 7245, 7315, 7577, 8093; leave, implying depart from life, die 6105; be split, break, thus come to an end (through death), end 2500, 2943, 5742; divide, split 2570, 2717, 2794, 3062, 4458, 6419, 6940, 7081; v.reflex. leave, separate oneself from, depart from 975, 2118, 3692, 3919, 5377, 5829, 6065, 6122; li cuer … partir life or soul depart, i.e. die 2162, 2188, 2225, 7042, 7245; Ainz m’ert le cuers el cors parti I shall sooner very shortly die 7042 partison sf. part, share 614 parzoner obj.sing. 669, perzoner subj.sing. 726; sm. partner, joint ruler past cf. paistre paterne subj.sing. 5107; sf. God the Father (= imago paterna, image depicting God the Father) paume subj.sing. 5766; sf. palm (of hand) as measure of size, i.e. about 8 cm across paumer obj.sing. 4878; paumiers subj.sing. 4901; sm. palmer, pilgrim pautoner 6740, 8141; paotoner 8167; sm. churl, low fellow 6342, 6740, 8141, 8167, 8170 pautonere adj.f. base, despicable, dastardly peaus obj.pl. 794; paus obj.pl. 5011, 5920, 7244; piaus obj.pl. 5917; sf. (animal) skins, pelts, furs peça 170, pieza 3011 adv. long ago, for some time past pecheors obj.pl. 999, 1605, 3917; pechere subj.sing. 4102; sm. sinner pecherise sf. (female) sinner 7168 pecier 3843, [peçoier], [pezoier]; pres.indic.3 piece 5438; v.trans. cut to pieces, slice through 2938, 3424, 3843, 4806, 5668, 7903 pein sm. bread, food 8112 [peindre] past part. pent 1814, peint 5406, pl. peinz 3190, f. peinte 8030 v.trans. paint peinturées adj.f.pl. painted, decorated 5002 peitevin adj. of Poitou 1682 peiz 5694, poiz 2835 sf. pitch; plus c’une poiz li est la car noircie his skin is blacker than pitch 2835; peiz demise melted pitch 5694

Glossary

551

pelice subj.sing. 6555 sf. fur-lined cloak, pelisse peliçon 6592, pelizon 7504; sm. fur-lined cloak 6592; fur-lined coat 7504 pels subj.sing. 60; sm. hair pendant sm. slope 2182, 2379, 6363; contreval un p. down a slope 6656 pendant pres.part. pendre used in sense of apendant (cf .6270 for context) relating to it, nearby 6287 [pendier] v.trans.& intrans. hang 5649 pendre 8126; pres.indic.3 pent 7124; past.def.3 pendi 8063, 6 pendirent 3921, 5976, 7137; v.trans.& intrans. hang 797, 1681, 1715, 3921, 5976, 7137, 8126; v.intrans. lean over (from vertical) 7124; v.reflex. hang oneself 8063 pene sf. top part of shield 2235 pene 314, 2482, 3149, 3170, 3971, 4555; peine 3171, 3173; sf. trouble, affliction 314, 2482, 3149, 3971, 4383; a pene with an effort, painfully 3170; a grant peine with great concern, in anguish 3171; sanz pene without difficulty 4555 peneant subj.pl. 345, 499, 1112; sm. penitents penel sm. small cushion placed beneath the saddle-straps, thus here saddle 7916 pener 87, 144, 996, 1553, 2546, 3125, 4277, 4583; pres.indic.3 peine 2667; 6 penent 6201; pres.subjun.3 pen 1855; v.trans. cause to suffer, make suffer 399, 1791, 1879, 8200; take pains, make an effort 583; v.intrans. suffer 144, 996, 1553, 4277, 4583; v.reflex. take pains, make a (great) effort 87, 1855, 2546, 2667, 3125, 6201; s’en a lo cors penez he has caused his body to suffer, he has made great physical demands on himself, he has toiled 810 penez adj.m.obj.pl. feathered 3667 penis adj. (subj.sing.) indefatigable, hardened to suffering 1384 pens cf. pont pensage sm. thoughts, thinking 1929; en pensage plunged in thought 549 pensé 476, 563, 579, 957, 1126, 2027, 3577; penser (cf. verb) 4261, 7453, 7622; pensez subj.sing. 231; sm. thoughts, thinking, opinion 231, 476, 563, 579, 957, 1126, 2027, 3577, 4261, 7453, 7527, 7539, 7622; de mal pensé full of evil thoughts 476, 7527; non aiez mal pensé! Don’t worry! 7539 penser 167, 1949; pres.subjunc.2 as imperative penx 1194, pens 1954; v.trans.& intrans. think, consider 167, 1132, 4798, 6218, 6508, 7950, 8179; (+ a) think (of), remember 3682; (+ de) think about, give thought to 1194, 1954, 2076, 3769, 4244, 6093, 6153, 6423, 6438, 6528, 6586, 6879, 7317, 7360; v.reflex. think, bethink oneself, muse 168, 3531; for infin.used as sm., cf. pensé pensis adj.subj.sing.) pensive, thoughtful 5433 pent cf. peindre and pendre penx cf. penser per sm. peer, noble(man) 16, 5201; peer, equal 77, 1940, 1959; specifically the 12 great peers or paladins of France 344, 833, 904, 1192, 1208, 1280, 1326, 1342, 1366, 1689, 1776, 2286, 2421, 2507, 2575, 2957, 4095, 4267, 6716, 7009; sf., because applied to sword and to fem.place-name peer, equal 4363, 4574 per cf. par perde cf. perte perdicion sf. loss, calamity 2687; perdition, torment 3795 pereres obj.pl., graphy of perrieres sf. projectile-launching machines, siege-engines 114 perfondement adv. deeply, profoundly 5931 perfont 4119; perfonde f. 4290; adj. deep

552

Glossary

perigal cf. paringal permi cf. parmi peron obj.sing. 17, 627, 1573, 2681, 4426, 4671, 4831, 7649; perrons obj.pl. 3991; sm. block of stone (often in front of a building) 17, 627, 642, 2681, 3991; mounting-block 627, 642, 1573, 4671, 4831, 7649; flight of steps, staircase 4426, (possibly 627?, 4671?, 7649?, but unlikely) pers adj. livid, dark blue 3354 persis adj. Persian 5437 persone sf. person; a la personne (at him) in person, personally 8174 pertus sm. opening, hole 5511 perte* 497, perde 1315, 1754, 3559, 4691; sf. loss pés sf. peace 5302 pesance sf. grief, distress 1215, 1726, 4931, 5173, 6791, 8007 pesant 4150, 5024, 5473, 5533, 6625, 6636, 6646, 6729, 7344, 8087; adj. gloomy, threatening (?) 1073, 6636, 6919; grievous 1630, 3440, 4150, 5024, 5473, 5533, 6625, 7344, 7972; heavy 4293, 6646; gloomy, downcast, heavy-hearted 6729; weighty, severe 8087 peser 2537, 7630, 7832; pres.indic.3 poise 1822; pres.subjun.3 poist 791, 7513; post 6906; v.trans. weigh heavily upon, be of great concern to, irritate pesmes adj.(subj.sing.) fierce 299; terrible, evil 3638, 3938 peson 1601; poison 3586; poisson 7769; sm. fish 1601, 3586, 4306, 5974, 7769 petit adj. small 711, 3875, 6743; de petit aé of no great age, not many years old 7752; adj.used as sm. the small 4611; adv. a short distance 7573; un petit adv.phr. a little, somewhat 4613, 7244, 7666, 8025; a short time 168, 1516, 4225, 6184, 6945, 7213; a small amount 7988; un seul petit only a short distance 6948; only a short time 7222; per un petit (qe) very nearly, almost 2965, 3634, 5768, 6002, 6472, 7554, 7554 petitet sm. very short time 3937, 7716 peus (obj.pl. of pel) sm. stakes, (wooden spades ?) 6095 pezon sm. part, portion 2132 pezoier cf. pecier piaus cf. peaus pié sm. foot (part of body) 635, 2910, 2940, 3386, 3616, 3686, 4111, 6665, 6836, 7569, 7580, 7587, 7659, 7678, 7842, 7920, 8076; foot, base (of hill) 1659; foot (measurement of length) 1297, 2705, 7785; bote pié thrusts out his foot 4111 pié 6794 cf. piez piece 5438 cf. pecier piece sf. (period of) time; d’une grant p. for a long time 618; a mot grant p. for a very long time 3101; (cf. peça, pieza) piez subj.sing. 4811, pié subj.sing. 6794 (graphy of pied) sm. man, anyone; (used with ne) not a single one 4811, 6794 pieza cf. peça piment subj.sing. sm. spiced wine 3962 pis adv.& sm. worse 3403, 5440; aveir p. worse happen to him 3399; aveir lo p. have the worst of it 7902 piz sm. breast 1581, 2216, 2248, 3328, 3961, 4099 plaidier v.intrans. speak, speak lightly, joke 6722 plaisir cf. plasir plaisser poetic licence for plaissé, plaissié for sake of rhyme past part.used as adj. twisted (?), conquered (?), fortified (?) * 5153

Glossary

553

plait sm. fight 3293; pact, bargain, deal 3599; a tel p. esmeü set in motion such a dispute; malvas p. m’a mis has imposed an evil bargain (OR a disastrous fight) upon me 7447 plastrie adj.f. plastered 4468 plasir 2165; plaisir 3590, 3928; pres.indic. plait 1442, 1921, 3287; plaist 1633, 3514, 3511; plast 3455, 5119, 7209, 7404; plas 6818; pres.subjunc. place 1435, 1461, 1879, 1898, 1909, 1933, 2562, 5679, 7305; past def. plot 1861, 7868; pleüst 2291, 2845, 2870, 7234; v.impers.&intrans. please; Ne place Deu God forbid! 1435, 1461, 1898, 1909; Ja Deu ne place, Ja Deus ne place God forbid .. ever ..! 1879, 1933, 2562, 5679, 7305; Deu plasir (if it) please God 2165; se Deu plast (plas) (if it) please God 3455, 6818, 7209; se toi plast if it please you 5119; venir a plaisir give pleasure to, accord with one’s wishes 3590; (of thought, idea) occur, come to mind 3928 pleger 7614; v.trans. stand as pledge for, stand surety for plegnier cf. plenier plein obj.sing. 5154; plens obj.pl. 5912 sm. plain pleindre 2970; pres.subjunc.1 plange 1217; 3 plagne 5(?); plegne 13, 1294; plange 3212; v.intrans. lament over, grieve for 1217, 2970, 3212; v.reflex. lament, grieve over 13, 1294 plen adj. full 2, 2109, 2166, 2170, 2843, 3527, 3903, 4516, 5242, 5767, 5882, 6491; (of battle) thorough-going 1629; de plene terre right from the ground (ie without using a mounting-block) 1718, 1827, 4952; a plen fren estendu with bridle fully extended 2170; plene la lance with lance outstretched, with the full force of the lance 2631; plen doit a full inch 3527; a plen at full speed 4245, 5080, 5201; grant plene paume a full palm’s width 5766; plenes les hanstes with lances at full stretch 7565 plenes (obj.pl.of plene) sf. plains 2476 plenier 133, 1164, 1666, 1669, 4305; pleinere 2317, plegnier 7326, plegner 7684, 7772; adj. strong, powerful 133, 1164; large 1659, 1666, 1669; (of battle) fierce, violent, intense 2317, 4305; (of palace) rich, great 7326, 7684, 7772 plenis past part.plenir filled (with) 4841 plens cf. plein plente sf. lamentation 7251 pleüst cf. plaire [plevir] v.trans. pledge, promise, swear, assure 672, 784, 935, 1100, 1103, 1345, 1431, 1434, 1876, 1908, 2618, 2979, 6614, 6903, 7040; pleviz past part.used as adj., subj.sing. pledged, promised 672; sm., subj.sing. pledged man 1175 ploier 5144, 7718; [plier] v.trans. fold, especially used of a glove folded as a formal sign of authority 4625 & 5256, or of challenge to combat 7718; bend (?), or possibly challenge (?) and thus resist the enemy (?) 4471; (of tents) fold, dismantle, pack up 5144 plor sm. weeping, tears; torner a p. bring to tears 4221; estre a plor be in tears (?), OR (poetic licence for estre a plorer?) be pitied (?) 4223 plorantment adv. in tears 4130 plorison sf. weeping, abundance of tears 4651 poc adv. few 2147 pöer cf. pöois pöesteïz adj. powerful 1183 pöestez 449, 673, 729, 812; paestez 2508; sf. force; par pöestez, per pöestez by force 449, 673, 729; par ses grant pöestez with his large forces 812; par dure paestez with harsh force 2508 pognal adj. relating to a hand; piere pognal (the largest?) stone one could carry in one’s hand (used for pelting defenders of town or castle) 5512, 5735

554

Glossary

poi adv. little 7190; few 2583, 3321; a poi ne almost, nearly 458, 679, 3530, 3702, 5563, 5619, 5669, 5820, 6954; a poi qe … ne almost, nearly 3578, 4797, 6014; par poi almost 677; en poi de terme in a short time, very soon 4024, 6646, 8173; un poi a little 468, 1789, 3532, 4024, 4561, 4798, 6028, 6143, 6640, 6680, 6681, 6850, 6893; un poi … ne not much 2815, 5497; per un poi ne almost 7973 [poigner] imperf.3 poignot* 4128 v.trans. attack poin cf. pon poindre 5754; pres.indic.3 point 1675, 1817, 2112, 2774, 3017; 6 poignent 2379; (imperf.3 poignot 4128*?); past def.6 poinstrent 6876; pres.part. poignant 5564, 5817; v.trans. spur 1675, 1817, 2112, 2379, 2774, 3017, 3629, 6360; stab, pierce with pain (4128?) 5931; prick 5754; v.intrans. spur forward 5564, 5575, 5817, 6876 poing obj.sing. 59, 454, 1637, 1902, 3204, 3279, 3288, 6809; obj.pl. 393, 533; pon obj.sing. 1254; puing obj.sing. 2910, 4626, 4846; puig 2940, 4930, 7966; pug obj.sing. 7958, 7966; poig obj.sing. 7984, subj.sing. 715; poinz subj.sing. 3038, obj.pl.1048, 1220, 2059, 1450, 2059, 2296, 3140, 5936, 8143; sm. fist, hand 59, 393, 454, 533, 715, 1048, 1637, 1680, 1220, 1450, 1902, 2059, 2296, 3038, 3140, 3204, 3279, 3288, 5936, 7958, 7966, 7984, 8143; (of bird) talons 6809; les langues d’or en sunt as poinz batanz the golden streamers (of the gonfanon) are flapping (beat) against his hands 2059 poinz subj.sing. sm. point; li poinz del siegle the beginning of life on earth, the Creation 4688 pois sm. pea 5497 pois sm. weight; nel faz pas sor mon pois I do not do it against my will, thus I do it willingly 4755; sor mon pois against my wishes 5699 pois adv.& prep. cf. puis poise 1822, poist 791 cf. peser poitral subj.sing. 5518 sm. breast-plate, breast-strap (of horse’s harness) poiz cf. peiz pomaus (subj.sing.of pomel, pomal) sm. ball-shaped ornament placed on high-point of roof of building 5906 pon 1254 cf. poing pon obj.sing. 916, 7668; subj.pl. 3119; poing obj.sing. 1637, 1902; pont obj.sing. 4083, 4369, 4370; poin obj.sing. 4110; pens subj.sing. 716; poins subj .sing. 1142, 5182; poinz subj.sing. 1680; pons subj.sing. 2922, 2933, 7936; s.m. pommel (of sword) [pooir] pres.indic.1 puis 353, 842, 1000, 1217, 1281, 2535, +; pois 6890; 3 puet 23, 284, 1223, 1279, 1358, 1407, 1549, 1627, 1657, 1663, 1692, 2214, 3366, 3399, 3595, 3861, 3976, 4152, 4374, 4380, +; poit 10, 1409, 3352, 3699, 6182, 7717; pois 7071; pot 2154, 2743, 4042, 4179, 5829, 6727, 6866, 7053, 7286, 7634; poet 6552; 4 pöons 1948; pöon 7017, 7284; 5 pöez 1476, 1945, 2147, 2148, 2538, 3602, 3654, 7635; 6 puent 2452, 2954, 2958, 4246,k 4273, 4307; poent 4312, 6080, 7321; pres.subjunc.1 poise 194; peüse 3597; puisse 4874, 5120, 6190, 6520, 6535, 7461; 3 peüst 26, 3347, 3862; puis 142; puist 329, 992, 994, 1195, 1935, 2659, 2723, 2739, 3600, 3854; puisse 1010, 3585, 4386, 6110, 6459; peüse 1860; poisse 1090; poist 1743, 2083, 3139, 7513; 5 poissiez 733; 6 puissent 4274; poissent 5205; fut.3 porra 3025, 6863, 6944; posra 7536, 8140; 4 porrons 2972; 5 porrez 86, posrez 2017, 2145, 6447; 6 porront 2580, 3033; imperf.1 pöoie 6192; 3 pöoit 5430; condit.3 porroit 7826; 6 porroient 3757; past def.1 poi 6281, 8006; 3 pot 991, 2207, 2261, 2262, 2368, 2730, 2737, 3739, 3838, 3849, 3876, 3880, 3902, 3927, 4055, 4112, 4254, 5440, 5518, 5868, 5934, 6087, 6171, 6773, 6780, 6802, 6893, 6949, 7309, 7312, 7776, 7884, 7970; poït 3988; poïst 4354, 5204, 5597, 5598, 7308; v.intrans. be able 10, 86, 142, 329, 353, 733,

Glossary

555

842, 991, +; have cause (to do something), have good reason to 23, 284, 1948, 2535, 4254, 6171, 6520, 6535; ne puis en avant I can go no further (in objecting), there is nothing else I can say 352; ne puis aler I can go no further, I cannot reach it 3896; vos puis mot amer I find you worthy of affection 2535; je puis bien dire I can affirm, I have good reason to say 3965; puis moi en vos fier? Can I rely on you? 6186; n’en puis el faire I can do nothing about it 6480; toi puis se gracïer I have thus good reason to thank you 6520, 6535; infin.used as sm. cf. pöois pöois 780, 4955; pois 852, 4755, 5699; poër 943; pöoir 5339, 5341, 5341; sm. power, right, authority 780, 852, 943, 4955; strength, forces 5341, 5342; a p. quickly, immediately 5339 poor cf. paor pople obj.sing. 3935; poeples subj.sing. 6153 sm. people, servants, followers 3935; people (in general) 6153 por prep. by, by means of 12, 54, 266, 487, 521, 536, 575, 593, 652, 676, 709, 819, 981, 1302, 2314, 2485, 2903, 3665, 3723, 4010, 4036, 4437, 7261, 7940; by (exclamatory, in oath) 236, 307, 321, 361, 433, 474, 488, 499, 521, 564, 705, 774, 1097, 1112, 1548, 1554, 1598, 2032, 2036, 2499, 3003, 3067, 3615, 4510, 5234, 5354, 5363, 5675, 6012, 6035, 6350, 6856, 6906, 6938, 6997, 7037, 7100, 7319, 7387, 8069; for, on behalf of 39, 147, 417, 530, 583, 765, 999, 1384, 1488, 1605, 1806, 1967, 2946, 3497, 3721, 3917, 5058, 5066, 5908, 6156, 7661, 7759; for 43, 149, 435, 485, 520, 548, 720, 815, 918, 951, 1060, 1464, 1481, 1499, 1584, 1636, 1811, 2198, 2875, 2902, 2986, 3020, 3040, 3127, 3759, 3810, 3848, 3849, 4199, 4209, 4256, 4279, 6209, 6744, 7103, 7283, 7336, 7441, 7612, 7691, 8075; for the sake of 6796, 6804; because of, on account of 395, 413, 473, 480, 551, 572, 585, 695, 748, 878, 883, 931, 979, 1007, 1141, 1180, 1218, 1335, 1436, 1437, 1448, 1462, 1900, 1910, 1979, 2332, 2458, 2587, 2657, 3229, 3523, 3525, 3859, 3873, 4071, 4205, 4210, 4370, 4553, 4691, 4787, 5184, 5730, 5956, 6322, 6330, 6404, 6511, 6543, 6613, 6683, 6712, 6817, 6901, 6936, 7064, 7065, 7153, 7199, 7266, 7544, 7929, 8130, 8179; through 810, 1016, 1022, 5466, 5984; in exchange for 175, 522, 707, 849, 1161, 2416, 3502, 5532, 5758, 5759, 7146; for the sake of 1127; in accordance with 199, 382, 1017; with 682, 1047, 1235, 1242, 1708, 1712, 1720, 1743, 1767, 1786, 1807, 1817, 2135, 3382, 3526, 3591; in respect of 1509, 2445, 2456, 3579, 3842, 3843, 3844, 3845, 4051, 5624, 5625, 5626, 5796, 5989, 6492, 6723, 6724, 6807, 6861, 7040, 7177, 7251, 7694, 8081; in 4008; as 6851, 7279, 7736; (+ infin.) for the purpose of, in order to 15, 24, 36, 41, 42, 79, 127, 130, 153, 154, 177, 189, 198, 988, 1013, 1015, 1190, 1379, 1504, 1547, 1558, 1655, 1690, 1882, 1958, 1989, 2102, 2264, 2389, 2492, 2540, 2545, 2561, 2572, 2652, 2658, 2722, 2931, 3116, 3756, 3935, 4245, 4278, 4584, 5000, 5178, 5317, 5374, 5804, 6444, 6450, 7057, 7063, 7406, 7706, 7781, 7815, 8077, 8142; por bien et por amor in goodly fashion and lovingly 8031; por ce for this reason, because of this 4132, 4374, 5967, 7719; por ce q’il … because they … 6877; por coi wherefore, for what reason 263, 549, 8059; por qoi why, wherefore 540, 2986; why? 2309; because of which 7390; por cui by means of which, whereby 2973; por conoisance according to their armorial badges, under their colours 5171; por dereres adv.phr. behind 1190; por desus over the top 3753; por devers in the direction of, towards 2988, 3656; por la fé in truth, i’faith 4090; por lor amor for love of them 3238; por votre amor for love of you 3618; por soie amor for love of him 6004; por le maistre above all, as the most important thing 1824; por mal wrongly 1362; por merveillos sanblant exceptionally well (ie strictly) 7346; por bon talant generously, graciously 4146; por mautalent angrily, with intent to kill 1866, 2323; por morir to the point of being slain, even if they were to lose their lives 1468, 5021; even if we were to lose our lives 1542, 3284; por quatre fois four times

556

Glossary

over 3242; por tant con puisse, por tant com puisse for as much as I can, insofar as I can (ie I swear by the simplest things I can do) 6459, 7461; por voire entencion in real earnest 3912; por voir in truth 804, 3561, 4126, 5038, 5202, 5349, 5357, 6808; por veritez in truth 1518; (see also par as graphy of por) porcel evos sm. porpoise (literally water pig) 5271 porchacier 3766 v.intrans. search, seek [porfendre] v.trans. cut (right) through 2271, 2326, 2644, 7979 porie cf. porrie poroz cf. paoros [porpanser], [porpenser] v.trans. carefully think out, plan, plot 6715, 7381, 7864; v.reflex. bethink oneself 6351, 6853, 7512 porprandre 554, [porprendre] v.trans. occupy 1172, 1591, 1596; p. sa marge occupy his frontier, implying seize extra land, extend his frontier 554 porqesre 3766; v.intrans. seek, search porrie 7167; porie 7192 past part.used as adj., f. decayed, putrified 7167; rotten 7192 pors subj.sing. 3183, obj.pl. 860, 869; por obj.pl. 1430; porz subj.sing. 3049, obj.pl. 970, 1036, 1044, 1075, 1136, 1187, 1207, 1250, 1321, 1433, 1475, 1522, 1950, 2050, 2347, 2404, 2431, 2437, 2663, 2669, 2978, 3036, 3049, 3078, 3123, 4145, 5152, 5263, 6143, 6157, 7480; sm. (mountain) pass porz subj.sing. 3251, subj.pl. 3041, obj.pl. 6655; sm. pig post cf. peser pouc (Franco-Italian graphy, poco for poi) adj. few 1423 poucelle 1899, 3512; pucele 6934, 6946, 7181; sf. virgin 1899; maiden 3512, 6569, 6934, 6946, 7181 poudrere sf. dust 4270, 4996, 5808 pouldre sf. powder, ashes 6025 poverte sf. poverty 3575 pré 570, 1603, 2029, 2273, 3093, 3568, 3879, 3888, 4025, 6657, 7810, 7828; praël 2384, 7912; sm. meadow, field pree sf. meadow 1621, 2329, 2467, préerie sf. meadow 7355 preïssant 4969 pres.part.of pre-issir going ahead as the forward troops prendre 457, 495, 2047, 3133, 3196, 3462, 4786, 5508, 5870, 7406, 7536, 7601, 7630, 7698; pres.indic.1 pren 362; 3 prent 1123, 2282, 3243, 3963, 4387, 4736, 7047, 7054, 7113, 7121, 7243; 6 prenent 4421, 4640, 4771, 5840, 8156; imperat.2 pren 5755, 7891; 5 prenez 963, 3952,4845, 6400, 7403; pres.subjun.1 prenge 492; fut.3 prenra 372, prendra 5873, 6815, 6933; imperf.6 prenoient 6736; condit.3 prendroit 610, 7041; past def.1 pris 749; 2 presis 1599; 3 prist 141, 166, 186, 187, 484, 787, 912, 1011, +, pris 501, 4010; 4 preïmes 7481; 6 prisent 1943, 3735, 3746; pristrent 3139, 4249; prissent 5591; past part. pris 240, 449, 729, +, pres 3869; f. prise 113, prisse 3512; v.trans. take, capture 113, 240, 912, 963, 977, 2047, 3010, 3089, 3462, 4513, 4517, 4786, 5840, 6449, 7536, 7630, 7690, 7698; afflict, strike 372, 1011; take 457, 484, 492, 501, 787, 1123, 1227, 1518, 1739, 1997, 2341, 2546, 2668, 2673, 3259, 3520, 3731, 3952, 3984, 4230, 4387, 4771, 4774, 4778, 4845, 4846, 5127, 5367, 5996, 6400, 6464, 6673, 6780, 6790, 7047, 7113, 7218, 7384, 7403, 7406, 7418, 7582, 7590, 7601, 7778, 7865, 8060; seize 372, 449, 729, 1669, 3106, 3139, 3243, 3534, 4010, 4015, 4038, 4122, 4476, 5192, 5649, 6305, 7396, 7581, 7592, 7673, 8156; accept 1227, 5755, 5870, 5873, 6460; seize, occupy 1584; hold in custody, arrest 3133, 3196, 7396, 7536, 7581, 7592, 7630, 7673, 7690, 7698; take as husband or wife, marry 3512, 6815,

Glossary

557

6933, 6934, 7041, 7891; capture 6305, 6449, 7536; take away, remove 7054; lift up, raise 7243; choose, opt for 8094, 8150; se je te pren a rien de … if I catch you involved in any … 362; prendre anoncion take flesh, take human form (via the Annunciation) 1599; ne li prendra envie de the desire will not seize for., he will not be seized with the desire for 372; A Damedé en prist …. God was seized with …4392; la teste prendre, les chief prendre cut off (heads), execute 251, 749, 2282, 3279a; prendre nos cuident come mastins soteins they think to surprise (and kill?) us like useless (inefficient) guard-dogs 2047; n’en devons prendre pris we should not take any prisoners (ie we should kill them all) 3462; prendre defeniment lose one’s life, die 3518; prendre estal take up a fixed position, take up a stance 5508; prendre retor make one’s return 6258; prendre son juïse implement the law (ie bring Ganelon to judgement), take his revenge 7406; v.intrans.& reflex (+ a) begin 141, 166, 186, 187, 976, 1187, 1277a, 2004, 3018, 3400, 3745, 3749, 3751, 3753, 3837, 3850, 3963, 4057, 4247, 4249, 4253, 4319, 4580, 4891, 6185, 6432, 7061, 7066, 7095, 7097, 7313, 7609, 7631, 7687, 7784; v.reflex.(+ a) cling on to 5535; a braz se prendre come to grips with one another, struggle together 4421; nus ne s’i prent nobody takes them (the French) on, does battle with them 4736 prés a prés 5359 adv.phrase very nearly in contact, closely approaching one another (prés, ‘nearly’; a prés, ‘close’; rather than the phrase prés a prés, ‘in close succession’) [preseigner] v.trans. make the sign of the cross over, bless 7262 present sm.& adj. en present adv.phrase here, before you 430; here, at hand 2443, 4697, 5118; before him 3958; at hand, ready 7423 presiez cf. prisez presis cf. prendre prestre subj.sing. sm. priest 7031 preu obj.sing. 4959, 5223, 7728; subj.pl.1469, 2652, 4976, 5232, 5496; prou obj.sing. 5942, 7493, 7708, subj.pl.4735; prouz subj.sing. 1939, 5418; preuz subj.sing. 1987; proz subj.sing. 58, 310, 343, 385, 722, 758, 817, 831, 1174, 1465, 1870, 1975, 2119, 2181, 2231, 2256, 2277, 2506, 2590, 2984, 3606, 3842, 5241, 5503, 5639, 5646, 5662, 7556, subj.pl.1943, 7885, obj.pl. 2491; pros subj.sing. 206, 1160, subj.pl.593; pruz subj.sing. 5633; adj. valiant, worthy, noble, honourable 58, 206, 310, 343, 385, 593, 722, 817, 831, 1160, 1174, 1465, 1469, 1870, 1975, 1987, 1939, 1943, 1987, 2119, 2181, 2231, 2256, 2277, 2491, 2506, 2590, 2984, 3606, 3842, 4735, 4959, 4976, 5223, 5232, 5241, 5418, 5496, 5503, 5633, 5639, 5646, 5662, 5942, 7493, 7556, 7728, 7708, 7885; fit 2652; well qualified, worthy 758; [proz read as graphy of prez = ready* 758, 7462, 7833] preu adv. profitably 3499; advantageously, to your advantage 4264 preu sm. profit, advantage 262; service 784 preudon cf. prodome primes 474, 3318; prime 3305; adv. first primereins adj. first, first of all 1259, 2044, 2105, 5463, 6273; a p. firstly, amongst the first, in the first rank 2385 princer obj.sing. 124, subj.plur. 5146 sm. prince, high-ranking nobleman pris sm. worth, prestige, status 4190; de pris of value, valuable 152, 222, 714, 931, 1015, 6593, 7932 pris cf. prendre pris 1166, 2199, 2695, +. cf. priser priser 115, 944, 2648, 2874, 3770, 4870, 6719, 7607, 7688, 7722; proisier 35; proiser 5181; [proisser]; pres.indic.1 pris 1166, 2199, 2695, 4232, 5221, 5226, 5631; v.trans. value, esteem; proisiez 507; prisez 2530, 2929; proissez 4180; prisiez 4419; presiez 4618;

558

Glossary

prissez 4803; proisié 7160; prisié 7676; past part.used as adj. highly valued, noble, of high quality 507, 2530, 4618; highly esteemed 2929, 4180, 4184, 4419, 4803, 7160 privé obj.sing. 6851, subj.pl. 6829, 6838; privez subj.sing.517, 806, 1150; subj.pl.6327; obj.pl.218; privée f.sing. 1576, 7235, 7240, 7924; adj. particular, intimate, close (friend) 517; private, confidential 806; private, own 7240; familiar (applied to sword, thus familiar through contact with its blade) 1576; intimate (as wife and confidante) 7235; closeset, hand-to-hand (battle) 7924; sm. close friend, intimate, member of private entourage 218, 1150, 6327, 6829, 6838, 6851 prodome 3845; proudome 2777; prodon 2946; proudon subj.sing. 5085, 5215, 7945; preudon subj.sing. 5137; sm. good man 36, 2946, 7945; brave knight 2160, 2402, 2777, 3845, 3852, 5022, 5085, 5137, 5215 proef adv. nearly, almost; a bien proef very nearly 6221 proiere obj.sing. 3884, 5963; priere subj.sing. 4997 sf. prayer proisere sf. appraisal, counting; par proisere in number 4987 proisier, proisser cf. priser pros, proz, prou, prouz, pruz cf. preu proz (graphy of prez) 758, 7462, 7833 adj. fully prepared, ready pugnaor 763, pogneor 3872, 7269, 7281; pugneor 1394, 6252; sm. warrior, combatant 763, 769, 1238, 1394, 3872, 4710, 6252, 7269, 7281, 7593 pui sm. hill 251, 1179, 1197, 1198, 1391, 1402, 1409, 1639, 4098, 4139, 5430, 5504, 6229, 6353, 6386, 6648, 6685 puié adj. exceptionally high, lofty 6309; nobly elevated 7664 puis sm. Pit (of Hell) 4484 puis 205, 486, 531, 788, 864, 923, +; pois 2403, 6692, 7231; pui 3557; adv. then 205, 486, 531, 788, 864, 923, 1235, 1513, 1531, 1693, 1739, 1740, 1758, 1771, 1813, 1848, 1992, 2141, 2268, 2410, 2482, 2567, 2761, 2826, 3155, 3206, 3245, 3249, 3270, 3279a, 3303, 3341, 3488, 3696, 3750, 3783, 3915, 4292a, 4456, 4466, 4618, 4628, 4646, 4818, 4823, 4892, 4894, 5129, 5181, 5317, 5584, 5916, 6023, 6211, 6225, 6502, 6687, 6692, 6977, 7126, 7229, 7231, 7508, 7517, 7606, 7610, 7767, 7799, 7838, 7873, 7937, 8076, 8090, 8144; afterwards 978, 1985, 2403, 2864, 3557 puis 3559, 3620, 4544, 6649; pois 6824; prep. since (temporal), after puis qe 1274, 1467, 1941, 2840, 5039; pois qe 7372; conj. after, now that, once 1274, 1467, 2840, 7372; since the time when, ever since 1467, 1941, 5039 puit 3551, 5744; puite f. 2341, 7557, 8023; pute f. 4487, 4552; adj. vile; de puite ovreigne vile in their deeds 2341; de puit aire of vile birth 3551, 5744; de puite loi of vile religion 7557 pulent adj. stinking, foul 4119

q qan, qant cf. quant qanqe cf. qenqe qar (graphy of car) conj. for, because 2561 qarals cf. quarals qarraus sm. marble 5898 qart num.adj. four 7167 (?quart?)

Glossary

559

qasser 6426, [quasser] v.trans. break, smash 471, 2363, 2629, 4039, 4478 qe obj.sing. 5, 6, 17, 59, 115, 172, 203, 265, 446, 528, 741, 899, 965, 1091, 1157, 1162, 1173, +; obj.pl. 101, 147, 155, 175, 275, 662, 663, 849, 999, 2962, 3821, 3849, 3942, 4341, 4370, 4378, 4379, 4465, 4499, 4654, 4983, 5177, 5319, 5371, 5590, 5782, 5802, 6224, 6353, 6490, 6660, 6834, 7607, 8056, 8070; subj.sing. 361, 746, 816, 825, 834, 916, 1057, 1205, 1364, 1419, 1575, 1907, 2033, 2048, 2297, 2699, 3514, 3586, 3776, 3892, 3915, 3919, 3995, 4061, 4199, 4381, 4456, 4459, 4849, 4914, 5376, 5552, 5624, 5671, 5683, 5699, 5796, 5816, 5881, 6253, 6551, 6679, 6940, 7128, 7224, 7225, 7240, 7368, 7419, 7804, 7873, 8200; subj.pl. 566, 2168, 3284, 3293, 4187, 4219, 5231, 6299, 7464; indir.obj.sing. 992, 1195, 3250, 3854, 5082; qi obj .sing. 345, 499, 1112, 1634, 3817, 3848, 3911, 4471, 6339, 7069, 7227; obj.pl. 152, 156, (899), 1860, 1999, 3031, 4937, 4975, 5225, 5944, 7161, 7329; subj.sing. 8, 12, 26, 29, 30, 66, 111, 123, 132, 136 +; subj.pl. 45, 104, 126, +; indir.obj. 3126, 3139, 3287, 3516, 5412, 5863, 7832 7832; cui obj.sing. 646, 871, 1215, 1837, 2350, 2454, 2665, 2671, 2838, 2917, 2973, 3839, 4032, 4508, 5100, 5132, 5848; obj.pl. 1151, 5014, 5070, 5132, 6099, 6163; indir.obj. 569, 1574, 1654, 2195, 2292, 2340, 2406, 2429, 2502, 3114, 3187, 4206, 4208, 4683, 5122, 6509, 6550, 6968, 7513, 8021; indir.obj.form for possession 255; rel.pron. who, whom, which 5, 17, 59, 101, 155, 172, 203, +; that which, what 19, 237, 336, 586, 886, 1114, 2614, 2934, 4335; whoever, anyone who 267, 539, 854, 834, 857, 1088, 1216, 1419, 1477, 1488, 1725, 1801, 1985, 2081, 2281, 2424, 2493, 2537, 2811, 3120, 3216, 3255, 3345, 3499, 3540, 3542, 3683, 4328, 4331, 4333, 4357, 4368, 4925, 4956, 5448, 5332, 5599, 5630, 5868, 6550, 6942, 7154, 7307, 7474, 7680, 7817, 7829, 8151; any which, anything that 1089, 4387; ne sa ge qe face I do not know what to do 3357; cui q’en poist ne qi non whether anyone is upset over this or not 7513 qe 2553, 2709, 2767 interrog.pron. what? qe adv. how 108, 4668, 5941, 6641 qe conj. that 10, 70, 72, 94, 95, 169, 179, 217, 235, 267, 281, 367, 427, 439, 475, 486, 490, 523, 621, 624, 665, 691, 695, 700, 702, 710, 723, 800, 820, 922, 931, 957, 979, 991, 1044, 1048, +; (+ time) when 13, 962, 1294, 1399, 1792, 2007, 2149, 3520, 3594, 4102, 4103, 4343, 4544, 5842, 6467, 7357, 7378, 7418, 8078, 8190; than (that) 56, 71, 523, 1133, 1452, 1888, 1904, 1937, 2834, 2976, 3006, 3314, 4012, 7530; whether 300; rather than 492, 1884; than 823, 1270, 1300, 1721, 1829, 1942, 2834, 2869, 2934, 3006, 3314, 3588, 3611, 3962, 4012, 4742, 4962, 5296, 5431, 5694, 6427, 6554, 6701, 6998, 7012, 7086, 7277, 7942; with the result that 69, 406, 856, 2367, 3527, 3673, 3919, 4036, 4048; so that, in order that 857, 1418, 1540, 2020, 2261, 2433, 2755, 3284, 3308, 3544, 3690, 3918, 4001, 4048, 4446, 5206, 5440, 5448, 5583, 5789, 6024, 6041, 6064, 6065, 6077, 6110, 6425, 6773, 6841, 7102, 7159, 7412, 7544, 7724, 7739, 7788, 7883, 7901, 7918, 7963, 7977; as 115, 404, 1409, 1590, 4090, 6323, 6895; for, because 306, 386, 894, 1821, 2033, 2263, 2491, 2611, 2643, 3095, 3177, 3434, 3561, 3645, 3702, 3838, 3855, 3930, 3976, 4112, 4151, 4492, 4528, 4741, 5142, 5522, 6058, 6094, 6120, 6729, 6780, 6893, 7033, 7225; until 632, 3069, 3208, 4346, 4616, 5860; only, no more than 846; qe ne … without (that) 971, 1091, 1162, 4153, 7496; bien va qe ne s’escrie he came near to shouting out 8025 qel qe obj.sing. 735, 881; qeus qe subj.sing. 1305, 2776 adj. whatever 735; pron. whichever 881; la qeus ira desore (of swords, f.) whichever one will be the winner 1305; li qeus qe vence whichever (side) may be the winner 2776 qenqe 86, 522, 3543, 4055; qen q’ 5221; qanqe 4477, 8182 adv.phr. as much as 86, 4055, 4477, 8182; as many as 3543; however much, whatever 522, 5221; (cf. quan, quant) qerre cf. quesre

560

Glossary

qevage sm. tribute, poll-tax, i.e. payment made by vassal to lord or by conquered to conqueror 1937 qi cf. qe qi q’ pron. whosoever, no matter who; qi q’en poist ne qi non whether anyone is upset over this or not 3139; cui q’en poist ne qi non no matter who is troubled by this or who is not 7513; qi q’en plor ne qi chant whosoever weeps or sings (rejoices) over it 5025; qi q’en plor ne qi rie no matter who weeps or laughs about it 5448; qi q’en ait joie whosoever may be pleased about this 6942; qi q’en doie peser no matter whom it may trouble 7832 qiarelon (graphy of quaregnon?) sm. sheet of paper folded into four, letter 484 qief cf. chief qier 475 cf. chier qier 37 cf. quesre qierement adv. dearly 1628 qiet cf. chaïr qisanz (graphy of cuisant in subj.form sing.) adj. biting, sharp 5640 qite adj. relieved of, absolved from (sins) 1511; clamer q. renounce all rights over 442; doner q. give/hand over, whilst renouncing all rights 4755 qite, qitier sf. cf. quitiez qivert cf. cuvert quant 34, 225, 354, 375, 427, 466, 495, 506, 525, 619, 641, 793, 920, 984, 1039, 1102, 1201, 1425, 1439, +; qant 263, 281; qan 2406; conj. when quant 318, 2449, 7124; quan 3699; pron.n. amount, degree; au quant de mes talents to the best of my ability 318; au quan q’il poit insofar as he could, as best he could 3699; ne … tant ne quant not .. at all, not … the slightest sign of 2449; ne tant ne quant to the slightest degree 7124 quant adv. how many 4590 quarals subj.sing.1663, obj.pl.3474; qarals subj.sing.1692, 1829; carrals obj.pl.3668; s.m. quarrel, bolt of cross-bow quarez past part.used as adj., plur. squared, shaped (used of wood); thus bien quarez well shaped 393 quarte ordin.num.pron.(f) fourth 5008, 5273, 5290 quarter 2202, 3752, 7971 sm. quarter (fraction) 3752; (heraldic) quarter or quartering of shield 7971; escu de quarter quartered shield (heraldry) 2202 que interrog.pron.neut. what? 3295 quens cf. cons quesre 2102; qerre 3821; pres.indic.1 qier 37, 1007, 1246, 1552, 1555, 1700, 4322, 7786, cher 4577; 3 qiert 982, 2829; 6 qerent 345, 499,1112; fut.3 quesra 4345, 6 quesrunt 621; pres.part. qerant 3822, 4612, 5466, 5960, 6272, 6291; querant 3883, 4730; plur. qeranz 3622; past def.3 qist 7864, 6 qisent 7149; v.trans. seek, ask (for), request queus 3136 cf. cous quietez 229, qitier 1284, qité 4794; sf. peace, tranquility; full, undisturbed, peaceful possession quinte num.adj.& pron.(ordinal) fifth 5018, 5274, 5291

r rabinos (graphy of ravinos) adj. swift, impetuous, fiery 6396 [raconzer] (graphy of racoucher) v.trans. lay flat again, return to supine position 7201

Glossary

561

raençon 914, 3917; raiençon 3135, raënzon 4656, 6458; sf. redemption, ransom raier 5156 v.intrans. flow, spread 3701, 4022; shine, sparkle 5156 raison 79, 233, 295, 609, 644, 655, 692, 1255, 1751, 1756, 2133, 2556, 3044, 3127, 3236, 3902, 3919, 4430, 4670, 4760, 5216, 5325, 5399, 5752, 6206, 6959, 7030, 7164, 7196, 7228, 7230, 7402, 7458, 7511, 7645, 7861; raisson 1095, 1096, raiezon 2138; sf. speech 233, 295, 644, 655, 2133, 3044, 7164; ability to distinguish what is true, thus true meaning 246; views, opinion 609, 1095, 1096; what was written there, contents of a letter 772; speaking, discussion 772, 1756, 2138, 7228, 7230; matter, amount of information, description 5216; instructions 6206; permission 7511; reasons, justification 7645; justice 8104; raison menbrée wise remark, memorable phrase 692, 1751, 4760; por sa raison mostrer to act as his spokesman 79; mostrer une raison set out a (logical) argument 3236, 5399; faire droite raison grant a prayer, fully satisfy a request 3127; dire raison speak plainly, intelligibly 3902, 3919, 4430; discuss the matter 4670, 7196; conter sa raison set out one’s opinion, speak out 6959; avoir raison do anything about, prevent 7030; par raison in a sensible manner, with a good argument, logically 2556, 5752, 7402; par grant raison by means of firm evidence 7458; senz autre raiezon without further argument 2138 raler 7301 v.intrans.& reflex.(compound of s’en aler) return home ralier 2264 v.trans. rally 2264, 2838, 3698, 5652 [raloer] v.trans. replace 7202 ramananz sm. 2532 cf. remenanz ramananz adj. 1074, 3158 cf. remanant randon sm. speed, impetuosity 1720, 2692; por tel r. so impetuously, with such speed 1720; mettre en grant r. put in confusion, throw into disorder 6450 randonée sf. impetuous, rapid gallop 2336, 3244 rant cf. rendre ranpoigne 2139, 2274 sf. taunt [ranpoiner] v.trans. mock, rail at, quarrel with 2986 [rasacier] p.def.5 rasaciastes 5975; v.trans. satisfy (hunger of), feed 5975 raser 156 v.trans. fill to brim 156, 224 ratrazon sf. reproach, censure 2139 recelée sf. concealment 1762; a r. in concealment, stealthily 1032 recevre 2079 v.trans. receive 182, 265, 2079, 2688, 3304, 8155 reclamer 1914, 2081 v.trans. call upon (person to do something), invoke (God) 1790, 2031, 3606, 3882, 4156, 4292, 5487, 5642, 5774, 7013, 7223, 7871, 7975; shout out, raise (battlecry) 1914, 2081; claim as one’s due, lay claim to 4768; sa coupe reclamer recite one’s mea culpa aloud, make open confession of one’s sins 3387 [recollir] v.trans. receive 5258, 5867, 6018, 6905, 7846; take, accept (as wife) 6899; hardiment r. gather one’s courage [or welcome the challenge, or recoil, draw back, preparatory to springing boldly forward?] 6408 reconoisance 1734; reconuissance 8004 sf. means of recognition, armorial bearings, thus here tunic with coat of arms 1734; avor r. obtain expiation, obtain punishment 8006 (hapax legomenon)* OR recognize as one’s vassal (?) recovrance sf. recovery 5789 recovrée 6706; recovreigne 2666, recovregne 2672 (graphy modified to suit rhyme); sf. recovery, remedy 6706; sanz nulle r. without any possible remedy or escape; inevitably 2666, 2672 recovreigne cf. recovrée

562

Glossary

recovrer 5206, 6431, 7284; v.trans. make good (loss) 5206; recover, get back 3639, 5824, 6431, 7284, 7925, 7974; son cor r. regain one’s spirits or courage 6980, 7950 recovrer 54, 2649, 3022, 3771, 5148; recovrier 2956; infin.used as sm. recovery, remedy, means of escape, way of avoiding something recreant 589, 1251, 2413, 4608, 7486, 7739; adj. ready to accept defeat 589, 1251, 1331, 3611, 4608; disloyal, foresworn, guilty of perjury 2413, 7739; guilty of breaking oath 3464; estre recreanz de be weary of something, be ready to give up 828, 1283, 3166, e morz e recreanz 5624, mort e recreant 7486 dead and defeated, dead and admitting his crimes (thus dead and damned?) recreü past part. recroire used as adj. vanquished, forced to admit defeat, discouraged 2890, 3489 recreüe sf. retreat 4067 redossez past part.used as adj. worn out, knackered (of horse) 734 redotanz adj. inspiring fear, redoubtable, formidable 1077, 2063 redricer 4304 v.trans. raise up again, lift again to upright position 7249; v.reflex. stand again, stand and fight 4304; stand up again 4319; sm. rising to one’s feet again 6422 refreschier 4324 v.intrans. refresh oneself, take food, revive one’s strength, become refreshed refroidier 3756, refroider 6882; v.intrans. cool down; rest, become refreshed refuser 1552, 8146 v.trans. reject, refuse 267, 1562; avoid 1552; avoid, refuse to do battle with 1057; reject 8146 regeïssant adj. making confession 4155 regreter 3400, 3850, 3963, 7298; v.trans. mourn the death of, lament over 3400, 3806, 3850, 3963, 4210, 6236, 7298, 7951 reigne 62, 2452, 2563, 2608, 2997, 3225, 3340, 3694, 4124, 6260, 6897, 6905, 7385, 7435; regne 6556; sm. kingdom reigné 4534; reignier 1285; reignez obj.pl. 71, 1463; sm. kingdom reisne 1619; resne 2466, 2626, 2948, 7886; renes obj.pl. 2478; sf. rein 102, 2400, 2466, 2626, 2806, 2948, 5422, 6622, 7886; torner sa reisne, torner sa resne turn (the horse’s head by means of) the rein 1619, 7886 rejoster 2492 v.intrans. do battle again relever 6423; v.trans.(used with sus) raise up again 6696, 6971; v.intrans. stand upright again, rise again (to feet) 3675, 7913; sm. rising to one’s feet again, regaining an upright position 6423 remanance sf. staying behind, remaining (in position) 8009 remanant 904; ramananz subj.sing. 1074, 3158; pres.part.used as adj. remaining, left alive 904; staying behind, left behind (in rearguard) 1074, 3158 remenanz 2066, ramananz 2532; sm. the rest, the remainder 2532; en remenanz for the rest, moreover, furthermore 2066 [remaneir and variants remaindre, remeindre] pres.indic.5 remanez 416; pres.subjunc.3 remeigne 7840; fut.3 remanra 859, 1284, 8102; remandra 5323, 5376; 4 remanrons 1480, 1958; 5 remandrez 7186; 6 remanront 2540; past def.1 remes 6693; 3 remest 714, 1191, 2296, 3741, 4423, 5719, 7118, 8198; 6 remestrent 5659; past part. remés 1208, 1342, 2253, 3296, 4189, 4594, 5570; remest 5765; v.intrans. remain, stay 416, 714, 2296, 3741, 8102; be left (as possession) 1284; remain, stay in position 1480, 1958; remain, stay behind 904, 1074, 1134, 1191, 1208, 1342, 2540, 3296, 3741, 7118; remain, be left 1342, 2253, 4189, 4423, 5323, 5376, 5659, 5719, 5765, 6693, 7186, 8198; come to an end, cease, be over 859, 5570

Glossary

563

remenbrée sf. remembrance, memory 6712 remenbrer 2078 v.intrans. remember, recall, think of 2078, 3529, 5781, 5959, 8026; remind 1203 [remener] pres.indic.4 remenon 4432; v.trans. take back with one, escort back to one’s home remire sm. remedy, relief, comfort 7015 remuer 4365 v.intrans. move back, retreat 4071; move, fluctuate, shimmer 4365; v.reflex. move, give way, become weakened or damaged 4061 ren sf. thing 7070; something, anything whatsoever 362; sor tote ren above anything at all, above everything 7070; ne …. ren nothing 4260, 4329, 7389 renc obj.sing. 2359; rens obj.plur. 126, 1324, 1944, 2168, 5668, 7448, 7679; sm. rank 1324, 1852, 1944, 2021, 2168, 2180, 2359, 5668, 7448, 7679 rendre 1361, 7486; [randre]; pres.indic.1 rant 3885; 3 rent 6365; v.trans. give up, surrender 281, 4013, 4513, 5845, 7596; make (bow) 636; make, cause to be 1331, 1361, 7486; give back 6253, 6467, 7540, 8048; give 6620; v.reflex. se r. al cheval regain the saddle, mount one’s horse again * 6365 rengier 1551, 1857; v.trans. place in ranks, place in position, draw up (troops), marshal 1551, 1857, 3767, 7662 renoi adj. renegade, false 7555 [renoier] v.trans. deny, renounce, abjure (faith) 4632; renoié 568; renoiez 2301, 4188, 4193, 5800; past part.used as adj. (568, 4188, 4193, 5800) & sm. (2301, 4615, 6307a) renegade [reocire], [reoncire] v.trans. kill (with implication in prefix ‘next’, ‘and then killed’) 2387, 2647 repaire subj.sing. sm. withdrawal, retreat, return home 4814 repairer 47, 2045, 3735, 7694; repairier 976, 4321, 4874, 6521; [s’en reparer] v.intrans. come back, return 382, 409, 434, 490, 976, 2045, 3414, 3548, 3716, 6521, 7694; turn back 1062, 2316; go back, return 3735, 4321, 4874, 6303, 7694; v.reflex. betake oneself, go back 47, 1211; sm. return 3024 reparlance sf. talk again and again, frequent mention 8011 repentiment sm. retraction, reservation; sanz nul r. unreservedly 7156 replenie 4490, 5854; replanie 7169; adj.f. filled, full [repondre] past participle repu 4000 v.reflex. hide reposement 7148; repousement 7417; sm. rest, repose 7417; place of rest, place where body lay (?) 7148 [repovoir] past def.5 repeüstes 6442; past part. repu 4000; v.intrans. be able again (with idea of repetition transferred from dependent verb, to meet requirements of metre)* 6442 reprocie 1438; reprocine 3308; sf, reproach, dishonour, blame reproçon sm. reproach, blame 3262 reprovage sm. reproach, dishonour, blame 1934 reprover 1912, 1935; v.trans.&intrans. reprove, reproach 1100, 1912, 1935; sm. reproach 2020, 2982 repu cf. repovoir reqigner 8130 (graphy of rechignier) v.trans. show (teeth); clench (one’s teeth) [requerre]; pres.indic.3 requert 7968; 6 requirent (for requierent) 2680, 5655; past def.6 (or historic present 6?) requirent 6418, 7570, 7931; past part. requis 7887; v.trans. seek (make pilgrimage to) 2680; attack 5655, 7887, 7968; v.reflex. attack one another 6418, 7570, 7931; sm. return, returning; del r. of the return, about returning 7317 (and possibly 1947, 2073: for the return ?)

564

Glossary

rer prep. behind 3330 resaler v.intrans. leap again, spring again 2210 resbaudir 5596 v.trans. encourage, animate, give fresh heart to 1851; re-animate, give fresh strength to, ardently re-engage, (OR resound?) * 2103; hearten, encourage, (OR resound?) * 5596 reseigne sf. fear 1289 resoigner 135 v.trans. fear, redoubt, hold in awe 135, 2308, 4616 respitier 6114; respiter 7485, 8083; v.trans. give respite, delay execution, allow to live 6114, 7485, 7658, 8083, 8105 resplendis adj. shining, gleaming bright, resplendant 1378 resplendor sf. splendour, brightness 2859 [restreindre] pres.indic.6 restregnent 2192; v.reflex. close ranks or embrace one another several times 2192 [restre]; pres.indic.6 resunt 2507; past def.3 refu 6133; v.intrans. still be, continue to be 2507; once again be 6133 [resvigorer] v.trans. reinvigorate, give fresh strength to 4027 reteniment sm. holding back, slackening of pace 2809 [retenir] v.trans. hold, keep 667, 2902; hold back 5341; take prisoner, hold prisoner 5776, 7581 [retentier] v.intrans. reverberate, (of heart) throb, thump 2111 [reter] v.trans. blame, reproach, accuse 5664, 7397, 7472 retorner 1908, 1947, 2073, 6438, 7061, 7629; pres.subjunc.3 retort 2229; v.trans. cause to return, take back 1560, 1947, 2073, 3052, 6438, 6699, 7243, 7317; v.intrans. return, come back 408, 1425, 1908, 2229, 2519, 2979, 3287, 6658, 7629; turn back 7061; v.reflex. s’en retorner turn back from something 6301, 63022, 7343, 7591 retraire 3550 v.intrans. resound, draw out/prolong their peal (soner e r. blare out and fade, ululate?)* 3550; retraire a resemble, belong to 3157 revel sm. rebellion, revolt, pride 2388 riche adj. rich 595, 2425, 4989, 4993, 5603, 6209, 6452, 6607, 7354, 7408, 7409, 7792; powerful 63, 536, 651, 1310, 2125, 2392, 2703, 2849, 2993, 4729, 6015, 6113, 6210, 6327, 6539, 6909, 7276, 7421, 7596, 7639, 7761 [‘indique plutôt la puissance fondée sur la richesse que la richesse elle-même; c’est un synonyme plus colorié de ‘puissant’ - Foulet] richement adv. powerfully, strongly 765; richly 2023; in favoured position (?) 2579 rimor sf. noise 1200 rivoier 41, 153 v.intrans. hunt (especially waterfowl) roée adj.f. (used of shield) adorned with patterns of wheels, circles or roses 1034, 1581 roge adj. red 6673 roide f.sing. 7804, roides f.plur. 1538, 4953, 5034; adj. (applied to lance) firm, rigid [rojir] v.intrans. become red, redden 3709 roiemant pres.part. raëmbre with variant graphy roiëmbre, used as adj.&sm. redeeming, redeemer 3148 romein sm. chronicle in the vulgar tongue, account in the vernacular 8199 romesin (graphy of romoisin) sm. kind of small Roman coin of very low value 7938 roncin sm. pack-horse 1556, 7996 ros adj. (applied to horse) chestnut, bay 5503; sm. bay (horse), chestnut 6400, 6403 rosere (graphy of rosaire) sm. garland of roses 4990 rote sf. rout, defeat 6626; company 6833 rotes past part.of rompre used as adj.f.plur. applied to reins broken, in pieces 2806

Glossary

565

rovent adj. bright red 4115 rover 165, 1905, 4255; pres.indic.3 rove 419, rueve 895, 5853; past.def.3 rova 3110, rouva 3175 v.trans. ask 165, 1905, 4255; command, order 419, 895, 5853; wish 3110, 3175 rusez past part.of ruser (graphy of reüser/refuser) m.plur. repulsed, put to flight 5667 ruste adj. forceful, robust 1805, 1957, 2485, 2863; violent, fierce 2278, 2903, 4030; vigorous 2505 rustie sf. boldness 760

s sab sm. pine, fir-tree 1369 sachanz 5633, sazanz 5644; adj. knowledgeable, experienced 5633; aware 5644 sachier 5138, [sacher]; v.trans. tear out 2298; pull, drag 4016, 4422; pull out 5138, 7992 safrez 1139, 1406; safré 2399, 2464, 5001, 7987, 7957; adj. damascened or burnished (applied to hauberk 1139, 1406, 2399, 7957 and to the metal plates on the byrnie 2464, 5001, 7987) sage obj.sing. 333, 2967, 5622; subj.sing. 1465, 2289 (at end of line); subj.plur. 546, 7497; sages obj.plur. 33, 270; subj.sing. 414, 5209, 5241; subj.plur. 104; saje obj.sing. 2281; sajes subj.sing. 771; saives subj.sing. 292, 6787, 6801; saive subj.plur. 7348; adj. clever 33; learned (in) 771; wise, knowlegeable 104, 270, 292, 333, 414, 546, 1465, 2281, 2289, 2967, 5209, 5241, 5622, 6787, 6801, 7348, 7497 sage (graphy of sache) 6514 cf. savoir sagete subj.sing. sf. arrow 7521 [sailir], [saillir], [salir], [sallir]; pres.indic.3 sail 260, saut 374, 681, 940, 1770, 6372, 6375, salt 1816, 3448, 5195, 5731; 6 saillent 375, 4292a, 4952, sallent 6256, 6263; past def.3 salli 7922; 6 sallirent 6262; past part. sali 782a, sailli 4030, saillie f. 7197; sailliz subj.sing. 1718, sailuz subj.sing. 1827, salli 7726, sallie f. 4482; v.intrans. leap 260, 374, 681, 1718, 1770, 1816, 1817, 1827, 4292a, 4952, 5195, 5731, 6375, 7922; leap forward 782a; (of liquid) spout forth 3448; protrude 4482; leap forth 7726; enz sailir sally forth 4292a, 6256, 6262, 6263; sus sailir leap up 374, 6372, 7197 saintie adj.f. sacred 5873 (cf. santisme 1598) sainz obj.plur. 6485, 6987; (graphy of sein) sm. bell; sainz soner toll the bells (in mourning) sairement cf. serament saives (graphy of sage) cf. sage sal sm. salt 2185, 8117 salt obj.sing. 5408; salz obj.plur. 1828; sauz obj.plur. 2359, 2727; sm. jump, leap salu sm. greeting, salutation 39, 636, 2360, 4800, 6518; account, report 3498 salu sf. escape from eternal damnation or death, salvation, safety 4106; a salu in safety, i.e. you may rely on me 4827 [salver] v.trans. save, protect from death 4226 salveté sf. salvation 7530 samblant obj.sing. 6292; samblanz obj.sing. 3073; sanblant obj.sing. 4735, 6640, 6649, 6663, 7346, 7482, 8093; sanblanz obj.sing. 1070; senblant 7070; sanblans subj.sing. 135; sm. appearance, mien 135; deeds, behaviour * 4735; views, opinion, 6640; appearance, expression 7070; what he thought, his impression/interpretation (of this) 7482; par merveillos s. to singular effect, with arrogant flourish 1070; with intense feeling

566

Glossary

3073; in terrible fashion 8093; por merveillos sanblant. in singular fashion, i.e. with exceptional security 7346; faire le samblant give the appearance, seem 6292; faire merveillos sanblant give a fearsome impression of …, look terribly as if …6663; ainc puis n’en vi sanblant I never since saw anything like it 6649 samin sm. fine silk material 7940 sanblance 5167, senblance 5778, 6799; sf. opinion, views 5167; visible appearance, sign of (his) presence 5778; appearance, image, sight 6799 sanbue sf. saddle 6680 santisme adj.(superlative of sainti) most sacred 1598 sarcie (graphy of sartie, past part.f. sartir) cf. sartir sarqeus obj.plur. sm. coffins 7290 sartir 2715; past part.used as adj.f. sarcie 4081; v.trans. decorate, damascene, reinforce sauf subj.plur. 4162; saus subj.sing. 645, 1985, obj.plur. 2045, 5909; adj. kept safe, protected 645; saved, redeemed 1985; saus et seins 2045, sain e sauf 4162 safe and sound saüst cf. savoir sauters obj.plur. sm. psalters 6486 sauvament *4682, sauvement 5112; adv. safely sauvation sf. salvation 490 sauz cf. salt savoir 3533; pres.indic.1 sai 219, 231, 439, 1995, 4172, 4514, 4590, 4668, 4853, 5110, 5167, 5193, 5470, 6071, 6298, 6359, 6941, 7720, 8074, 8107, 8124, 8136; sa 1453, 3316, 3357, 8061; 3 set 1195, 1590, 2074, 3128, 4152, 6602, 6619; 6 sevent 1039, 1066, 3561, 4275, 5834, 6841, 7275, 7493; pres.subjunc.3 sage 6514; 5(imperative) sachez 696, 1518, 4804; sachiez 727, 1128, 1779, 3038, 4413, 4526; 6 (jussive) 7377; fut.2 savras 3276; 4 savrons 7405; 5 savrez 652; past def.1 soi 8059; 3 soit 29, 5773; sot 739, 3814, 6726, 6789, 6950, 7368, 7820, 8181; imperf.3 savoit 7697; 6 savoient 2444; imperf.subjunc. 3 saüst 2973; v.trans. know (fact), be aware (of) 170, 219, 231, 333, 439, 652, 696, 727, 1039, 1066, 1128, 1195, 1453, 1518, 1590, 1779, 1995, 2074, 2444, 2973, 3038, 3128, 3316, 3357, 3561, 4152, 4172, 4413, 4514, 4526, 4590, 4804, 4853, 5110, 5167, 5470, 5773, 6071, 6298, 6359, 6514, 6619, 6841, 6941, 6950, 7275, 7377, 7493, 7697, 7720, 8061, 8107, 8181; be acquainted with, know (person) 528, 5663; be familiar with (thing) 3276, 5834, 6602; know, be learned in 739, 3814, 6726, 6789; know how to 4275, 4668, 5193, 7368, 7405, 7820, 8074, 8124, 8136; infin.used as sm. wisdom 276, 5337, 5340 schernir (graphy of eschernir) v.trans. mock, jeer at, insult 702 schine cf. eschine [scoisir] (graphy of eschoisir) v.trans. take out (ie es-choisir) or look at, possibly scan, read 742; see, discern 4121 se (graphy of si) adv. so, such a 5209; thus, as follows 7622, 7877; se com just as 32; conj. if 83, 84, 194, 262, 281, 297, 323, 329, 334, 356, 362, 370, 398, 415, 434, 436, 478, 489, 524, 662, 670, 720, 721, 727, 730, 750, 752, 753, 844, 890, 901, 911, 919, 922, 982, 983, 1222, 1248, 1271, 1277, 1279, 1313, 1362, 1364, 1442, 1493, 1505, 1545, 1633, 1793, 1874, 1907, 1921, 1973, + 92; whether 5636, 6349; so, therefore, thus 276, 457, 790, 865, 975, 1006, 1128, 1490, 1506, 2304, 2780, 2978, 3002, 3143, 3197, 3344, 3351, 3373, 3748, 3869, 3989, 4625, 5028, 5771, 6178, 6674, 6685, 7348, 7397, 7622, 7877, 7914, 8051, 8081, 8116, 8154; cum se as if, as though 6689 secorre 1168, 5804; past part. secoruz 1152; v.trans. help, aid, bring succour 1152, 1168, 1487, 1877, 1893, 2373, 2980, 3037, 3112, 3632, 3645, 4433, 4525, 4797, 5681, 5804, 6659; e car me secorez! Now help me! (with car intensifying the imperative) 2373; secorant

Glossary

567

pres.part. used as adj. providing help 6659; secoranz pres.part.used as adj. capable of being helped, able to be helped 4297 seëler 4369; v.trans. seal sege 90, 254, 448; seje 81; sm. siege segner 6523, 6737; seiner 3743; [seinner], [seinier]; v.intrans. bleed 357, 3365, 3743, 6523, 6737 segnorie 247, 418, 2119, 2828, 3688, 4088, 6548; seignorie 5062, 5255; sf. rank, nobility, power 247, 2119, 2828, 3688, 5062, 5255, 6548; lordship, seigneurial (or parental?) authority 418; dominion 4088 segnoril 2609; segnori 7841; segnorie f. 8019; adj. (of things) seigneurial, noble, great 2609; (of people) princely, rich, noble, powerful 7841, 8019 segur (graphy of seür) cf. seür seigner 1674, 3836; v.trans. make the sign of the cross over, give benediction to 1530, 1674, 1988, 1991, 3319, 3619, 3817, 3836, 5059, 5123; v.reflex. cross oneself 7228 seignorage sf. nobility, power 557 [seinner], [seiner] cf. segner selonc prep. according to 6048, 7220, 7872 seme ordinal adj. seventh 5044, 5060, 5276 semoignant cf. semondre [semondre] pres.indic.1 semon 293, 1643, 6207; 3 semont 6224; pres.part. semoignant 1324, semonant 2232, 4982; v.trans. assure, swear to 293; invite, call to take part in 1643; urge on, encourage 1324, 2232, 4982; call upon, summon to assemble 6207, 6224 sen obj.sing. 464, 6805, 6828; siens obj.pl. 2632, 2712; sm. sense 464; meaning, significance 6805, 6828; (plur.) senses, consciousness 2150, 2632, 2712; views, judgement, wisdom 5634 senblance cf. sanblance senblant cf. samblant [senefier] v.trans. signify, mean 83, 92, 106, 193, 245, 497, 1043, 5403 senestre adj. left 515, 1784, 1843, 4846, 6691 senez adj.(subj.sing.) wise, learned 58, 722, 817, 1870, 2506, 5662, 5671 sengler sm. boar 6427 senpres 303, 2979, 3427, 3450, 6433; adv. straightway 303, 2979, 6433; (+ fut.) be about to, be on the point of 3427, 3450 sens (obj.plur.) cf. sen seoir 293, 322, 776; pres.ind.3 siet 3255, 5467; past def.3 sist 15, 629, 988, 1713, 1812, 2377, 2625, 2699, 2727, 2752, 2847, 3324, 3589, 5212, 5429, 5459, 5503, 6264, 6608, 6627, 6679; 6 sistrent 126; v.intrans. sit down 293, 322, 776; (of person) sit, be seated 126, 629, 5212; (of object) be set 1713, 1812; sit, be mounted (on horse) 2377, 2625, 2699, 2727, 2752, 2847, 3255, 3324, 3589, 5212, 5429, 5459, 5503, 6264, 6608, 6627, 6679; suit 5467; v.reflex. sit, be seated 15, 988 [sepelir] v.trans. shroud, put in sepulchre 7175 sepoucre 145, sepulchre 1880, 2034, 5977; sm. sepulchre serament 7848; sairement 7859, 7860, 7929; sm. oath, taking God as witness to what one affirms serer 1944, 8143; serrer 8075a; [sesrer] v.trans. hold close, clasp 1581; draw close together 1944; keep closed 7119; bind tightly, fasten, lock 8075a, 8143, 8169; serré past part.used as adj. close-meshed, closely woven 1662; close-structured, firmly sealed 7970

568

Glossary

serie adj.(fem.of seri) serene, peaceful, tranquil 1040, 8012 (cf. Valie Serie 6900); soft, gentle 3806 serin adj. (of wind) very light, becalmed 7930 serin sm. evening, nightfall 6072 serjans obj.plur. 189; serjant subj. plur. 4922, 5950, 6126, 7516, 8091; sm. serving-men 189; men-at-arms 4922, 5950, 6126, 7516, 8091 sermon sm. speech, discourse 608, 5757; sermon, edifying discourse 1724, 1977, 4909, 5040; speechifying 2689; (after negative) a word 3776; speech, words 4640; speech, language 5331; sermon traire make a speech 608, 5757 seror cf. suer serrer cf. serer ses 1245, 1865, 2085, 5427; sis 4973, 5921; abbrev.of si les so …… them ses poss.adj. cf. son sesrer cf. serer seul obj.sing. 3940, 3943, 5411, 6087, 6096, 6178, 6661, 6948, 7222; subj. sing. 6661; seus subj. sing. 691, 694, 3774; sol obj.sing. 4216; adj. alone 691, 694, 3774; single 3940, 4216, 5411, 6096, 6278; adj.used as sm. single one 3943, 6087, 6661; un seul petit only a little 6948, only for a short time 7222 seür 620, 2321, 3064, 4564, 4956; segur 1135; adj. sure, reliable; segur de cuer loyal, whose courage may be relied upon 1135 seürance 4934 sf. protection, security; faire seürance offer protection* seüremant adv. securely, in security 5933 [seürer] v.trans. assure 1512 seürté sf. safety 1136; formal promise, pledge (of loyalty) 4774, 4778 seus cf. seul seüs adv. assuredly, confidently 2196 sevrer 1003a; [severer]; fut.4 sevesron* 6043; v.trans. separate, part 823; separate, cut off (lit.) 2909, 3642, 4789, 5582, 6709, 6710; separate, cut off (fig.) 6079; draw away, draw back 5398; separate, set apart, thus distinguish, select 6043, 6833; v.intrans. depart from, take leave of 1003a sevrance sf. separation; faites sevrance separate (them) out, make a (visible) distinction between 6064, 6076 si poss.adj. cf. son si graphy of se reflex.pron.(reflex.vb. needed) 2092, 3305, 4917 si adv. so, thus 255, 319, 432, 540, 846, 863, 986, 1005, 1059, 1083, 1084, 1258, 1322, 1433, 1445, + 57; also, likewise 213, 356, 391, 394, 412, 661, 1002, 1108, 1155, 1158, 1503, 2358, 2428, 3049, 3389, 3965, 4517, 5220, 6581, 7891; so, to such a degree, such 541, 557, 704, 816, 834, 915, 929, 1140, 1630, 2069, 2152, 2168, 2228, 2432, 2476, 2482, 2603, 2637, 2802, 2814, 2840, 2852, 2880, 2994, 3087, 3224, 3226, 3274, 3400, 3536, 3637, 3638, 3641, 3656, 3706, 3761, 3786 + 65; so much, to such a degree 3195, 3250, 4362; like 6368; even so, nevertheless 7658; si fait such 566, 2332, 6484; conj. so then, and then 18, 167, 186, 257, 260, 483, 504, 617, 783, 953, 973, 1028, 1040, 1068, 1087, 1176, 1215, 1376, 1564, 1573, 1745, 1768, 1976, 1990, 2151, 2184, 2234, 2324, 2711, 2829, 2907, 3026, 3030, + 58; so, consequently, therefore 161, 183, 300, 463, 694, 864, 932, 1078, 1131, 1181, 1380, 1394, 1505, 1898, 1929, 2125, 2156, 2380, 2428, 2498, 2499, 2573, 2645, 2790, 2831, + 50; so that 713, 1817, 1828, 1944, 1992, 2646, 3112, 3385, 3477, 3906, 5017, 6688, 8060; if, whether 409, 652, 1292, 2008, 2873, 3135, 4814, 6297, 6729, 7851; si come, si com, si cum as, just as

Glossary

569

708, 902, 1011, 1389, 2021, 2036, 2088, 2361, 2682, 3251, 3331, 3790, 4271, 5228, 6063, 6157, 6762, 6831, 6988, 7023, 7380; si qe so that, such that 7412, 7518 siens 2632, 2712 cf. sen sifaitement adv. in such a way, thus 7333 sigant, siguant cf. sivre siste ord.num. sixth 5032, 5292 sit 2; set 22, 238, 2651, 4589, 4644, 7841, 8128; card.num. seven [sivre]; pres.indic.3 suit 4886, 5261, 7561; 6 suigent 5794; fut.3 sivra 226, 6 sivront 3965; pres.part. siguant 1529, sigant 4589, sivant 6299, (plur.) sivanz 4283; v.trans. follow, pursue [smarir] graphy of esmarir * v.trans. trouble, distress 7177 soavet adv. gently 7609 soduiant 1111, 1115, 1319, 2407; sosduiant 4593, 4900, 5945, 6276, 6669, 7738, 8090; sosduianz plur. 2049; adj. false, deceitful, knavish 1115, 1319, 2407, 4593, 4900, 5945, 6669, 7738, 8090; sm. knave, deceiver, traitor 1111, 6276 soe cf. son soé 3373; söef 3755, 3962, 4146, 4692, 5962, 5995, 6621; adv. softly, gently 3373, 3755, 4146, 4692, 5962, 5995, 6621; sweetly 3962 soferte 1316; sofraite 3241; sf. need, want, plus the idea of distress, suffering due to loss sofraitos adj. in need, in want, deprived 3686, 6479; (with the idea of suffering loss 6479) sofranz (de) adj. patient (over, in respect of), indulgent 3088 [sofreindre] pres.subjunc. 3. sofreigne 2340; v.intrans. be lacking soif obj.sing. 3873, 3904; soi obj.sing. 8119; sois subj.sing. 3060; sf. thirst soion sm. bond 3140 sois sm. sorrel* 5691; (?) sois (graphy of sié with s added for rhyme) sm. seat (on horse)* 5691 (?) sol adv. merely 136; only 3532, 6192, 8110; sol le fer the iron (spearhead) on its own, solely the iron 5194 solauz cf. solel soldoier sm.plur. mercenaries 46, 157 [soleir]; pres.indic.3 selt 2542, 4258, 7303; 6 suelent 6165; imperf.3 soleit 4620a, 5 soliez 7069, 6 soloient 6100; v.intrans. be wont, be accustomed solel obj.sing. 2448, 3189, 3632, 4256, 4954, 5969; subj.sing. 2859, 7485; solauz obj.sing. 8109; subj.sing. 187, 1357, 1377, 1566, 1775, 2450, 4250, 4282, 4319, 4585, 5314; soleuz subj.sing. 200; solaiz subj.sing. 986; solaz subj.sing. 5415; solouz subj.sing. 6297; sm. sun solement 3315; soulement 4936; adv. only 3315; on their own, without support * 4936 solemnité sf. solemn ceremonies, rites 4529 som cf. son somellant adj. dozing, nodding, half-asleep 5952 somier 734, 3207; somer 3142; sm. packhorse 734, 1023, 3031, 3142, 3207, 6129, 6139, 6674 son 916, 1588, 1694, 1713, 1812, +; som 4458, 6930, 7480, 8032; sm. top; en son in the top 916, 1694, 1713; at the head 1588; on top 3207; at the top of 4099, 4138, 4572, 5906, 6154, 6386; to the top of 6648, 6685, 6742; el son auzor right at the top 1812; (spatial) per som le coute above the elbow 4458; par som les porz passant passing over the passes 7480; per son l’estrer, par son l’estrier up on the stirrup, up by means of the stirrup 7509, 7807; (temporal) par son l’aube esclarie at the very first light of dawn

570

Glossary

6606; par som avril issant immediately after the end of April 6930; par som l’aube at break of dawn 8032 son m.sing. 30, 36, 74, 144, 151, 167, 191, 220, 257, +; ses obj.sing.(m) 1824, 3476; (f.) 516, 812, 3684; subj.sing.(m) 49, 135, 208, 210, 212, 354, 517, 518, 622, 680, + 42; (f.) 808; obj.plur.(m) 88, 89, 198, 205, 218, 226, 243, 278, 280, 319, + 76; (f.) 114, 162, 184, 228, 264, 279, 441, 483, 648, 821, + 26; subj.pl.(m) 63, 507, 666, 1069, 2993, 8008; (f.) 3743, 7467; sa fem.sing. 79, 169, 233, 257, 312, 395, 460, + 210; si plur. 16, 112, 519, 942, 975, 1529, 2548, 3138, 3613, 4316, 4617, 4810, 5146, 5201, 5285, 5640, 5814, 6462, 6469, 6829, 6838, 7468, 7736; sis subj.sing. 212a(V7), obj.pl. 3983; (before vowel) s’ f.sing. 4, 515, 529, 715, 759, 1029, 1573, 1597, 1616, 1897, 1789, 3246, 3687, 3805, 4057, 4106, 4454, 4461, 5186, 5447, 5505, 5692, 5708, 6253, 6432, 6905, 6916, 7200, 7427, 7925, 7992; poss.adj. his, her (cf. stressed form of poss.adj. and pronoun, suen, soe, soie) soner 29, 1022, 1233, 1906, 1931, +; v.trans.& intrans. sound 252, 1022, 1070, 1379, 1424, 1432, 1444, 1459, 1874, 1884, 1890, 1906, 1931, 1946, 2072, 2082, 2264, 2552, 2573, 2825, 2977, 2992, 3035, 3047, 3074, 3081, 3082, 3106, 3109, 3115, 3170, 3178, 3243, 3525, 3534, 3543, 3550, 3705, 4239, 4243, 5130, 5176, 5373, 5650, 5987, 6198, 6738, 7358; ring out, resound 1233; (of bell) ring 3636, 3858, 3887; sainz soner toll the (mourning) bells 6485, 6987; (with negative) mot soner (not) say a word 29, 6849 sor 17, 418, 429, 466, 516, 584, 853, 893, 896, 897, 969, 985, +; sore 4007, 4436, 7781, 7982, 7986; prep. on, upon 17, 466, 516, 853, 893, 896, 897, 985, 1116, 1148, 1241, 1274, 1391, 1496, 1523, 1577, 1787, 1990, 1993, 2023, 2051, 2114, 2172, 2202, 2213, 2222, 2235, 2245, 2259, 2593, 2702, 2787, 2898, 2968, 3120, 3142, 3147, 3162, 3207, 3335, 3346, 3363, 3406, 3486, 3493, 3569, 3610, 3616, 3617, 3755, 3830, 3835, 3879, 3901, 3961, 3998, 4055, 4130, 4174, 4200, 4242, 4406, 4426, 4456, 4459, 4481, 4494, 4609, 4661, 4784, 4817, 4828, 4831, 4894, 4939, 4952, 5083, 5104, 5125, 5152, 5212, 5393, 5459, 5462, 5500, 5503, 5526, 5552, 5595, 5741, 5786, 5935, 6080, 6139, 6142, 6182, 6187, 6241, 6259, 6264, 6285, 6355, 6400, 6415, 6434, 6435, 6468, 6522, 6608, 6627, 6638, 6644, 6680, 6688, 6834, 6871, 6876, 6964, 7085, 7451, 7461, 7525, 7604, 7644, 7784, 7862, 7869, 7873, 7874, 7904, 7926, 7939, 7969, 7987, 8052, 8159, 8167; above (fig.), more than 584, 1206, 1624, 1833, 2357, 2726, 3096, 3822, 7235, 7238, 7291; above (spatial) 4327; above (sound), louder than 1338, 1777, 5988, 6070; against 429, 1119, 2086, 2417, 2796, 4616, 4986, 5057, 5065a, 5082, 5097, 5484, 6101, 6511; over 418, 1734, 4955, 5180, 5892, 6058, 7062; in respect of 969; sor destre to his right 1392, 4114; sor lui against his will, in spite of his resistance 1046, 1088; sor mon pois against my will 4755, against my wishes 5699; sor vos against your wishes 5168; sor mer overseas (especially over the Mediterranean Sea) 5180; sor tote ren over anything at all, in respect of every little thing 7070 sor adj. red-brown, sorrel (describing horse) 1816; adj.used as sm. sorrel (horse) 7590; soirs (?)* 5691 sorcis obj.plur. 3993; sm. brows, temples sordois sm. shame, suffering, pain, disaster, downfall 845, 5500 sore prep.used adverbially against, towards; corir sore run at, fall upon, attack 4007, 4436, 7982, 7986 soreins adj.plur. quiet, untroubled * 2047 [sorhaucher] v.trans. raise on high, elevate, highly exalt 4633 sorqidez adj. arrogant, bold, presumptuous 2303, 4820 sosduiant cf. soduiant [soslever] v.intrans. be raised, rise up 3527 sospeçon sm. doubt, uncertainty, suspense 4441

Glossary

571

sospois sm. fear, worry 4745 [sosprendre] v.trans. surprise, take by surprise 3976, 4123 sospiranz 2064, 3080; sospirant 5926, 6631; adj. sighing, soft-spoken 2064; sighing, full of sighs 3080, 5926, 6631 soudées obj.plur. 7146; sf. payment, recompense soul cf. seul soulement cf. solement sovin adv. on one’s back, supine 3888 soz prep. under, beneath 15, 251, 590, 749, 776, 896, 1177, 1213, 1580, 1677, 1757, 1818, 1895, 2025, 2152, 2851, 2935, 3834, 4166, 4453, 4538, 4591, 6165, 6473, 6563, 6772, 6775, 7010, 7263, 7770 soz subj.sing. (of sot) 5268, 5283; sm. someone lacking in judgment, a fool 5268; sm. come soz like a fool 5283 spee cf. espee speroner (graphy of esperonner) v.trans.& intrans. spur 4247 spines obj.plur. 6088; sf. thorn bushes spiritaus adj.m.obj.plur. of heaven, heavenly 5911 stuet cf. estoveir suel cf. soleir suen m.obj.sing. 94, 382, 649, 650, 7949; m.obj.pl. 821; suens m.subj.sing. 231; soie f.obj.sing. 1060, 5165, 6004, 6551; soies f.obj.pl. 1790; soe f.obj.sing. 1743, 2116, f.obj.pl. 4181; stressed form of poss.adj. his, her 94, 231, 382, 649, 650, 821, 1060, 1743, 1790, 2116, 4181, 6004, 6551, 7949; por soie amor for love of him 1060, 6004; des suens m.obj.pl. 2964; li suen m.subj.pl. 3416; soule f.subj.sing. 1296; la soie f.subj.sing. 5185; poss.pron. his 1296, 5185; his (men) 2964, 3416 suer 383, 6209, 6217, 6231; seror 2998, 6933, 7797, 8047; sf. sister 383, 2998, 6209, 6217, 6231, 6529, 6581, 6933, 6960, 7185, 7797, 8047 suigent cf. sivre sus 2327, 2352, 7695; prep. on 2327; on, over 2352; upon, against 7695 sus 126, 374, 1234, 1546, +; suz 3414; adv. up, on high 126, 374, 1234, 1546, 1783, 3265, 3414, 4030, 4674, 5813, 5851, 6372, 6696, 6971, 7129, 7197, 7247, 7326, 7465, 7595, 7605, 7772, 7999 susjeccion sf. en ta s. in subjection to you 6062

t tables obj.plus. sf. backgammon 127; altar-tables 5899 taborie subj.sing. 3705; sf. blaring noise, tumultuous sound taigne, teigne cf. tenir taisir 2576; [taissir]; v.trans. keep silent about 720; v.reflex. be silent 259, 305, 308, 2576; taisanz pres.part.used as adj.plur. silent 320 talant 594a, 1110, 1118, 1337, 4146, 4598, 4917, 4979, 5017, 5465, 6118, 7010, 7367, 7473, 7626, 7692; talent 805, 3500, 3506, 3816, 3975, 4695, 4698, 5044, 5049, 5126, 5128, 5365, 7159, 7794; sm. inclination, desire, satisfaction (of desire) 594a, 1337, 3506, 3631, 3816, 4979, 6118, 7010, 7626, 7692; opinion, views 735, 1110, 5017, 5634, 7159, 7367, 7473; disposition, character, bearing 299, 3500, 1118, 3087, 4695, 5049, 5128, 5365, 5465; a lor talant

572

Glossary

as much as they desire 594a; a ton talent to your satisfaction, very pleasing to you 4698; a son talent (talant) to his satisfaction, as suits him 5044; to his taste 7794; au quant de mes talanz as well as I can 318; en talent aveir have an inclination, desire 805; (se) mua son talent his spirit became inflamed, he was moved to anger, his desire (to attack) increased 4917; son talent his opinion, whateven he likes 7473; por bon talant, par bon talant with goodwill, graciously, generously 4146, 4598; per bon talent courteously 4692, 5126; par merveillos talant with terrible longing 6118 tanir cf. tenir tanser cf. tenser tant 6, 23, 33, 344, 533, 589, 591, 623, 699, 757, +; itant 4898, 5944, 6130, 6282; adj. so many 533, 589, 590, 811, 827, 1034, 1035, 1396, 2094, 2097, 2098, 2397, 2398, 2399, 2400, 2402, 2426, 2447, 2464, 2465, 2466, 2477, 2478, 2605, 2606, 2607, 2735, 2770, 2773, 2794, 2803, 2881, 2882, 2883, 2909, 2910, 2911, 2924, 2925, 2926, 2937, 2938, 2939, 2940, 2968, 3117, 3121, 3228, 3447, 4064, 4069, 4082, 4142, 4309, 4310, 4311, 4556, 4651, 5014, 5143, 5144, 5145, 5423, 5424, 5479, 5625, 5704, 6220, 6844, 7296, 7435, 7593; pron. so many 991, 1408, 4604, 5202, 5245, 5452, 5480, 5812, 6044, 6706, 6874; adv. so, to such a degree 6, 33, 1077, 1157, 1167, 1917, 5238, 7255, 7460, 7676, 7779, 7789; so much 23, 344, 591, 699, 757, 819, 855, 858, 1636, 1999, 2454, 2628, 2738, 3365, 3375, 3462, 3507, 3554, 3571, 3585, 3598, 3603, 3640, 3742, 3842, 3848, 3851, 3931, 3933, 4019, 4254, 4279, 4345, 4378, 4505, 4597, 4603, 4616, 4722, 4773, 4803, 4870, 5710, 5807, 5910, 5928, 5941, 5944, 5982, 5992, 5999, 6013, 6171, 6178, 6367, 6596, 6771, 6903, 7047, 7237, 7260, 7274, 7299, 7303, 7722, 7841, 7891, 7970; in such great numbers 1404, 4903; so long 1243, 4017, 4451; so far 4731; tant arere et avant both back and forth 6658; tant de so much 1196, 3497; tante de so much 3497; tant mar in such as evil hour 1451, 2554, 3359, 3866, 3997; tant par so very much 201, 407, 1962, 2851, 3617; tant per so very much 3099, 3153, 3876, 4750; ne … tant ne quant not anything at all 2449, not to the slightest degree 7124; tant soit hardi however brave he may be 3737; tant fust … however much he were … 4179; tant soit however much it be 5719, 6147; tant esfréez however much afraid 6361; tant a esté such a long time he has been (on the journey), i.e. after a long journey 4451; conj. tant qe until 98, 6190, 6315, 6497, 7180; tant … qe so far … that (fig.) 367, (lit.) 623, 4731; so long that, i.e. until 814, 1009, 3145, 3208, 3597, 3874, 4346, 4641, 4830; tant cum for as long as 164, 230; tant com as much as 176; as long as 185, 829, 1019, 2267, 3183, 3624, 8011, 8084, 8186; tant … com as much … as 960; tant ai alé qe … I have gone so far, been so presumptuous that … 367; por tant c’on puisse … as surely as one can … 6459, por tant com puisse .. as surely as I can … 7461; a itant forthwith, immediately 4898, 6130; en itant then, straightway 6282 tapace obj.sing. 1694, subj.sing. 1713; tapaze subj.sing. 1812; sf. topaz [tapir] v.reflex. conceal, hide oneself 4000 targe 1034, 1581, 1769, +; tarze 2114, 2477, 2856; sf. targe, shield 1034, 1581, 1769, 1844, 2114, 2222, 2477, 2796, 2856, 4992, 5445, 5704, 5719, 7923 targer 2003, 4877, 5135, 5142, 8071; targier 4257, 6875; tarzer 2876, 6727, 7110; [tarjer] v.intrans.& reflex. tarry, ride slowly 547; tarry, delay (in doing something), be slow (to do something) 504, 1120, 2003, 2285, 2421, 2801, 2876, 3544, 3819, 4284, 4818, 5142, 5367, 6502, 6824, 6875, 7110, 8017; s’en tarzer prevent oneself from doing something 3838, 6727; (cf also atargier) tarjance 5096 sf. delay tasiaus obj.plur. sm. metal frogs which serve to fasten the two edges of a mantle 5915 teinz adj. dark (of complexion), dark with anger, purple with anger 680; purple 3354

Glossary

573

teinz obj.sing. 2246, 5740; subj.sing. 2203, 2214, 2236, 5036; tent obj.sing. 2755; sm. paint tel 115, 371, 459, 678, 726, 834, 840, 1047, 1133, 1217, +; tels sing. 435; tes plur. 214, 8010, 8103; teus sing. 340, 347, 500, 3556, 3559, 7574, 8180; plur. 214, 2772, 2871, 4074, 6098, 8022; itel sing. 2393, 2416, 2786, 3259, 8186; ites plur. 2657; iteus plur. 3085; adj. such, so great 115, 340, 459, 678, 1047, 1133, 1217, 1233, 1298, 1334, 1395, 1478, 1888, 2135, 2153, 2187, 2586, 2598, 2691, 2712, 2786, 2945, 2964, 2965, 3113, 3134, 3259, 3283, 3310, 3382, 3526, 3530, 3559, 3662, 3683, 3868, 3902, 3927, 4021, 4066, 4737, 5101, 5258, 5396, 5605, 5669, 5914, 5934, 5938, 6098, 6383, 6484, 7120, 7273, 7888, 8007, 8010, 8186; such, of such a sort 214, 834, 840, 1720, 1817, 2158, 2229, 2250, 2393, 2416, 2657, 2841, 3085, 3254, 3556, 3599, 3612, 3808, 4008, 4351, 4698, 4706, 4708, 4933, 4945, 4956, 6457, 6799, 7123, 7136, 7574, 7688, 7893, 8103; such, like, similar (equal) 1416, 3490, 3496, 4074, 5029, 5701, 6637, 6912; (indefinite) some, certain 8022; pron. such a one, such a man 347, 435, 500, 2740, 4074, 8181; such men 2772, 2871, 4074; such a thing (= blow) 371; auqes de teus a few more such men 2871; adv. so 4014; onqes tes ber ne fu never was anyone so valiant (as he) 4014; (indefinite quantity, + numeral) some, approximately 2927, 3286 [tendre] pres.indic.3 tent 455, 470, 3914; tient 6544, 7735; v.trans. hold out, offer tenebror adj. deep in shade, dark, gloomy, sombre 1197 tenebror sf. darkness 7273 tenir 2148, 2571, 2721, 4852, 5518, 6102, 6104, 6109, 6320; [tanir]; pres.indic.1 ting 2967, 3451; pres.subjunc. 3 taigne 10; teigne 270; tiegne 1849; fut 3 tenra 163, 229, 435, 725, 856, 1018, 4531; tanra 4343; 5 tenrez 50; past def.2 tenis 7151; 3 tint 256, 258, 517, 578, 683, 953, 1186, 1254, 2126, 2278, 2641, 2762, 2769, 2922, 2932, 2936, 3334, 3687, 4020, 4075, 4110, 4665, 4673, 4930, 5126, 5388, 5548, 5986, 6083, 6373, 6622, 6795, 6851, 7027, 7032, 7096, 7362, 7451, 7875, 7908, 7920, 7936; 6 tindrent 5196, 5986; past subjunc. 3 tenist 1818, 5161; v.trans. take prisoner, capture 10, 730; hold 256, 517, 578, 683, 1254, 2278, 2641, 2769, 2922, 2932, 2936, 3334, 3687, 4020, 4075, 4110, 4673, 4930, 5126, 5986, 6083, 6373, 6622, 7027, 7032, 7096, 7362, 7875, 7908, 7920, 7936; hold (as a possession) 50, 63, 229, 552, 725, 1018, 2126, 2762, 4508, 4531, 4665, 5388, 5548, 5680, 6320, 7451; hold, think, consider 435, 2148, 2721, 2967, 6851, 6913, 7279; hold (conversation) 4852; bear 6104; maintain 7151; maintain (siege) 254; keep (peace), adhere to 5302, 6102; take, accept 915, 928, 6444, 7484, 7730; take hold of, grasp 953; hold back, stay (in position) 2571; hold back, restrain 5518; hold to, rely on (as guarantor or protector) 1186, 3451; qe droit li tiegne that the verdict should be in his favour, that he should have the victory 1849; tenez me convenant keep your promise to me 903; tenir lonc parlament hold a long conversation, spend a long time talking 4852; v.intrans. hold out (against enemy), maintain the fight 856; tenir a be a characteristic of, derive from 6795; nel tint pas a enfance it did not derive from childishness, was not the result of childishness 6795; v.reflex. keep (oneself), remain 258; hold to, adhere to 270; endure (the pace), keep up with 1818, 5161, 6362; keep on, continue (on one’s way) 6292; hold back, delay 6476 tenple sf. temples, forehead 3103, 3169 tenpler sm.&f. temples, forehead 2644 tenprez past part.used as adj. tempered (of steel) 1140 tens 178a, 3583, 4117, 4792, 5029, 6585, 7555, 8096; tans 2408, 4772; tenps 3924; sm. time 178a, 2408, 4117, 4320, 5029, 5474, 6585; time, life-span 3583, 4792; tot tenps always 3924; toz tens always 4073, 5030; per tans, per tens very soon, 4772, 8096; tot tens all the time 7555

574

Glossary

tenser 2215, 2744; [tanser]; v.trans. protect 542, 2215, 2744, 4042 tent cf. teinz tentir 6371; v.intrans. resound, echo, reverberate 3049, 6371 tenzaille sf. quarrel, dispute 1471 tenzon sf. fight, battle 2696, 5570; quarrel, argument, dispute 4650; par mot fiere tenzon in very fierce attack 4436; del Sarazin est remés la t. for the Saracen, the fight is over 5570 terce cf. tierç [terdre] past def.3 terst 7143; v.trans. wipe teriene adj.f. earthly 7190 terme obj.sing. 4024, 6646, 8173; termes subj.sing. 66, 3594; sm. term, length of time 66; appointed time 3594; en poi de terme in a short time, very shortly, very soon 4024, 6646, 8173 terne sm. hill 5855 terral obj.sing. 8110; terraus obj.plur. 5890; sm. domaine, embankment, slope 5890; mud, drain 8110 terrer sm. earth rampart, mound 6731 terst cf. terdre tertresel sm. small hill 2374 tes, teus cf. tel tes cf. ton tierç 146; terce 3106, 4976, 5272, 5289, 5308; num.adj.& pron. third tinbres obj.plur. 7359; sm. drums tirant adj. obstinate, stubborn * 4159; (OR sm. tyrant?) tirer 6159 v.trans. pull out, tear 533, 6159; pull, pull at, tug 1212, 4021, 4197, 5937, 6487; pull away (horse), steer (horse) off in certain direction 2040; pres.part.used as adj. (capable of) pulling, which can pull strongly 8151 tochier 2261; [tocer]; past def.3 toça 7867; v.trans. tap, strike 1241; touch 2261, 7867 toie cf. tuen [toldre]; pres.indic.2 tols 403, 3274, 3802; pres.3 tol 3802; 5 (imperat.) tolez 8055; 6 tolent 4288; pres.subjunc. 3 tolle 6278; fut.3 toldra 582; past part. toluz 278, tolu 4009, 4504, 5452, 6220, 6969, 7394, 8038, 8183; tolue f. 4072, 4634; v.trans. take away (from) 278, 582, 3274, 3802, 4009, 4288, 4504, 5452, 6220, 6278, 8038, 8055, 8183; tu li tols moi … assez you take enough away from me …, you rob me … of enough in this matter (?) OR you disparage me …. enough to him, you carry enough tales of me .. to him ? 403; tolent lor voies … capture their roads, cut off their (escape) routes 4288 tolir 5600; past def.3. toli 2649; v.trans. take off 2649; v.reflex. lose 3802, 5600 ton sm. tone, pitch, voice; a un ton in a loud voice 2131; a aut tons in a loud voice 4502; a aut ton loudly 4653 [toner] v.intrans. thunder 6134 tonoire sf. thunder 2433; lightning 4402 tor (graphy of torn) sm. turn; fai le tour franzoi make the ‘French turn’! (a manoeuvre in fighting when one pretends to flee in order to return to the attack with greater force) 7559; reprendre son tour advance again at speed 7886 tormente sf. severe punishment 2771 [tordre]; past part. tors 863, 2820; v.trans. twist; twist round, bend over 863; past part. used as adj. twisted 2820

Glossary

575

torner 192; tornier 999; turner 5205; pres.subjun.3 tort 5506, 6 torgent 6077; v.trans. turn 192; fall 215; return 379, 5205; turn back 999; brandish 1526; torner faus jugement return a false verdict, sentence falsely 429; torner a mal turn out badly for 5506; torner a declin fall into decline, be routed, die 6077 tors cf. tordre tort obj.sing. 719, 1389, 1470; torz subj.sing. 282, 2312; sm. wrong tost 421, 694, 751, 1409, 1721, 1782, 1806, +; adv. quickly, speedily 421, 694, 751, 1409, 1721, 1782, 1806, 1826, 1829, 1854, 2084, 2339, 2615, 2701, 2741, 2808, 3122, 3182, 4454, 4536, 4859, 4866, 4922, 5045, 5681, 6325, 6856, 7109, 7310, 7373, 7600, 7790, 7823, 7928, 8088; early, soon 5498, 7577, 7583, 7826, 7953 tot adj. m.obj.sing. 110, 199, 479, 484, 596, 707, 1161, 1252, 1801, 2182, 2784, 2834, 3924, 4229, 4534, 5159, 5226, 5652, 5758, 5788, 5949, 6286, 6952, 7013, 7224, 7555, 7591; subj.sing. 363, 373, 3663, 6708, 8180; obj.plur. 975, 1268, 1280, 3332, 3647, 4198, 4509; subj.plur. 112, 565, 942, 1130, 1192, 1371, 1646, 2427, 2798, 2817, 2961, 3211, 3648, 4429, 7429, 7662, 7683, 8149; f.obj.sing. 2336, 3704; toz m.obj.sing. 6710; subj.sing. 313, 737, 3583, 4352, 5898, 6223, 7162; obj.plur. 27, 117, 213, 278, 480, 487, 580, 584, 1184, 1206, 1375, 1777, 1833, 1996, 1257, 2362, 2518, 2524, 2529, 2726, 3094, 3096, 3391, 3412, 3591, 3813, 3822, 4073, 4324, 4385, 4411, 4501, 4751, 4901, 4946, 5023, 5187, 5610, 5678, 5979, 5988, 6070, 6108, 6125, 6188, 6339, 7203, 7291, 7332, 7345, 7403, 7546, 7598, 7632, 7711, 8130, 8192, 8194; subj.plur. 473, 676, 1431, 1981, 2666, 3441, 3647, 3852, 6660, 7682; tuit m.subj.plur. 137, 337, 581, 1264, 1545, 2672, 3091, 3746, 3762, 3765, 3824, 4162, 4235, 4594, 4668, 4877, 4970, 5011, 5095, 5231, 5232, 5903, 5939, 6465, 6469, 6587, 7117, 7516, 7580; subj.sing. 7737; tote f.sing. 241, 254, 613, 906, 1016, 1022, 1218, 1846, + 46; tuta f.sing. 433; totes f.plur. 1611, 1624, 2333, 3138, 3221, 3448, 3466, 3743, 4280, 4503, 4529, 7235, 7238, 7467, 7586, 7619; all; toz dis always, ever 3432; a toz dis for evermore 5434; tot jor always 199; toz jor always 3224, 4393, 4598, 4901, 5023; toz jorz always 4926; a toz jors for ever, still 6125; tote jor throughout the day, day-long 3097, 6630; toz les plusors most of them, many of them 2529; tot tenps 3924, toz tens 4073, 5030, tot tens 7555 always, all the time tot sm. the whole, all; del tot completely 6480, 7005 tot 361, 368, 737, 962, +; toz 117, 213, 924, 1410, 1431, +; tuit 7737; pron. everything 361, 368, 737, 962, 1548, 1554, 1656, 1740, 1839, 1848, 1861, 1954, 2275, 2585, 3020, 3596, 4755, 5589, 7635; everyone, each one 3583, 3946, 7737; all 117, 213, 924, 1015, 1098, 1113, 1410, 1431, 1981, 2356, 2586, 2707, 2818, 3195, 3591, 3813, 3942, 4248, 4314, 4429, 4624, 4626, 4713, 4955, 4972, 6187, 6207, 7345, 7848, 8174 tot 138, 459, 681, 698, 1259, 1323, 1920, 2022, 2058, 2105, 2175, 2176, 2180, 2237, 2246, 2271, 2284, 2326, 2366, 2644, 2791, 2812, 2883, 3089, 3198, 3344, 3355, 3463, 3692, 3712, 3774, 3875, 3968, 4171, 4183, 4294, 4418, 4437, 4516, 4777, 4861, 5105, 5108, 5112, 5153, 5319, 5369, 5736, 5741, 5768, 5857, 5873, 5051, 5980, 6199, 6284, 6303, 6371, 6416, 6438, 6835, 6878, 6883, 7045, 7256, 7337, 7353, 7609, 7856, 7868, 7880, 7912, 7959, 7979, 8111, 8114, 8145, 8170, 8173; toz 2, 258, 272, 311, 415; adv. fully, completely, quite; sit ans toz plenz a full seven years 2; tot al in quite a different fashion (before long) 7880; tot primier 138, tot primer 6883, tot primereins 1259, 2105, toz primereins 5463, 6273 first of all; tot le chief primier leading the way, right in the lead 6878 traïment sm. betrayal 7136 traïn sm. pursuit, killing 2823; massacre 3310, 3891 trainer 2543; trainier 1556; v.trans. drag 1556, 2400*, 2543

576

Glossary

[traïr] (graphy of trahir), [treïr] v.trans. betray 12, 213, 891, 1113, 2110, 2179, 2618, 3521, 6968, 7019, 7852, 8005, 8059 traire 5753, 5757, 8115; pres.indic.1 trai 885, 4124, 6668; 3 trait 743, 2335, 2714, 4499, 5397, 5732, 6373, 6387; pres.subjunc.1 traie 7946; past def. 5 traissistes 5979; past part. trait 608, 1181, 1815, 3750, 5932, 7932; traiz (subj.sing.) 1826, 2358, 6257; v.trans. tear (hair or beard) 743, 4499; draw (sword) 1181, 2309, 2358, 2714, 2742, 5732, 6373, 6387; draw forward, bring 1815, 1826; draw forth, release 1882, 5979; pull off, take off 3750; shoot 5397, 5490; drag along (by galloping horses), tear apart 5753; take away 5932; + avant urge forward 6257; traire (un) sermon make a speech 608, 5757; traire au qief accomplish, bring to a successful conclusion 885; an traire draw forth (hand from glove), or raise (hand) as swears an oath (?) 892; en trai vos a garant I call upon as witness to you, cite to vouch for this to you 4124; te traient a garant they rely on you, look to you to protect them 6007; Deu en trai a garant as God is my witness 6668; traire a fin bring to end (of life), put to death, kill 7932, 7946; fors traire drag out, haul out 8115; traiez vos za come hither! 1485, 2942, 2949, 5017; v.reflex. draw oneself (direction indicated with adverb), thus draw near 1689; arier se traire draw back 2335; ensus se traire draw apart 3070; relate to, rely on, trust in 6808; traiez vos en ça! Trust in that! 6808 traïsor (graphy of traïson modified to suit rhyme) 1226, 1398; sf. treachery trait sm. length (of shot); lo trait a un archier bow-shot 2095, 2101 traïtor obj.sing. 2511, 4205, 6219, 6250, 6364, 6442, 6456, 7368, 7600, 7879, 8007, 8168; subj.sing.(voc., interjection) 5549; obj.plur. 8103; subj.plur. 1319, 1612, 3921, 7134; traïtres obj.sing. 4350; subj.sing. 7421, 7555, 7573; traïtes subj.sing. 568, 6359, 6377, 6429, 7430, 7495, 7525; traïte subj.sing. 8049; traïtre subj.sing. 2407, 4188, 6477, 7732, 8175; (v.Introduction page 92–93) sm. traitor trametre cf. tremetre [transglotir] v.trans. swallow 4164, 5110 transiz obj.plur. past part. transir used as sm. those who have died, dead (men) 7285, 7288 [travailier], [travaillier], [travailler], [travaler] v.trans.& intrans. work hard, toil, make every effort 399, 583, 6496, 7661; plough (ground) 2805; harass, weary, hurt 4391; disturb, torment 4400; v.reflex. make a great effort, fatigue oneself, labour, strain 1470, 1488 traval sm. torment 8114; a traval with sustained effort, in anguish 3170 travers sm. crossing 4567; breadth, side 5507; en travers crosswise, sideways 7953 trebal sm. tribulation, suffering, torment (cf. traval) OR uproar, agitation (graphy of trepeil) 2189 trebucher 4315, 7973; [trebucer], [trebucier]; v.trans. knock down, strike down 2138, 2217, 2239, 2745, 6421; throw down, cause to fall 4315; let fall, drop 7256; v.intrans. fall, stumble, fall down 6770, 7919, 7973, 7983 trecherie sf. treachery 362 [trechier] v.trans. cheat 7259 [trecier], [triser] v.trans. plait, braid 1577, 4197, 7667 tref 511, 512, 1620; trez 191, 1042; trief 988; tré 5156, 8161; sm. tent tremente sf. (tormente?) [tremetre], [trametre]; past def.3 tramist 101, 1862, 6078, 6382, 7161; tremist 1344; 5 tramesistes 248; past part. tramis 218; v.trans. send 40, 53, 101, 218, 248, 282a, 287, 1344, 2862, 6078, 6382, 7161

Glossary

577

trepaser 1953, 6862; trespasser 4276; come to an end, expire 66; go beyond, cross 559, 3823, 3989, 4555; omit to do, fail to do 664; go beyond, live beyond, survive 1953; cover (distance) 4276; go through 4685; (fig.) go beyond, exceed 6564; (fig.) undergo, suffer, bear 6863 tres num.adj. three 2367, 8097 tres adv. right; tres …. jusc’al right from …. until 2448; tres parmi right in the middle of 2384; prep. tressc’ up to, as much as, as long as 194; tresqe right up/down to, as far as 2271, 2745, 2789, 5764; until, right up to the time that 4608; tresc’a right up to 511, 534, 909, 1704, 2809, 2823, 4103, 4343, 5811, 5842, 6325; as prefix with perfective force, cf. tresaler, tresalir, trestot tresaler v.reflex. faint, lose consciousness 4026 tresalir cf. tressaillir treschier 6868, [trescher] v.intrans. dance 6868, 6984 treslice adj.f. cf. tresliz tresliz obj.sing.m. 2247; treslis obj.plur.m. 1371; treslice f.sing. 4994; adj. plaited (literally with three threads) trespaser cf. trepaser [tressaillir]; [tresalir]; pres.indic.3 tressaut 679, tresal 7953; v.intrans. shudder, quiver 679; jump right over, jump clear again 7953 tressc’, tresc’, tresqe cf. tres [tressuer], [tresuer] v.intrans. be covered in heavy sweat, sweat copiously 678, 1766, 6634, 8057 [trestorner] v.trans. overturn, change, countermand 467; turn (right) over 2118, 2205, 2249, 2384, 2706, 2765, 2866, 5462; turn aside, deflect (from purpose) 4787, 5005; change the meaning completely 6805; turn (horse) right round, turn (horse) straight back 7909; v.intrans. (person) return, come back 970; v.reflex. turn back, retrace one’s steps 7914; ja n’en ert trestorné nothing can prevent it 467, 970 trestot (stressed form of tot) adj. absolutely all, the whole of 50, 266, 341, 410, 929, 961, 1456, 3950, 4267, 4607, 4669, 6011, 6127, 6167, 8091; trestot le monde absolutely all the world, (absolutely everybody?*) 6167; adv. completely 259, 978, 2186, 2273, 3562, 3657, 7117, 7379, 7742, 8125, 8172; (with negative) at all, in the slightest degree 2085; trestot manois absolutely immediately 7742 treton (graphy of troton) sm. fast trot; le treton at a fast trot 4429, 4642 treü sm. tribute 4094, 4013; par ton treü as your tribute, as what you owe 4013 treüsage 6507; treüage 6513, 7496; sm. tribute 6507, 6513; what is due to him, recompense 7496 trez cf. tref trisée past part.f. of triser cf. trecier tristor sf. sorrow, distress 4203, 4211, 4701, 5884 tronzon obj.sing. 4664; tronçons subj.sing. 2304; tronchons obj.plur. 5498; sm. stump 4664, 5498; stump (of broken lance) 2304 tros obj.plur.of trot 2818; sm. trot; as tros at a trot, trotting 2818 trosier 1022; trosser 5145; troser 6436; v.trans. load 1022, 3207, 5145, 6139, 6436 trover 81, 4273, 4766, 5183, 6435, 6442, 7634, 8140; pres.indic.1 truis 370, 890, 970, 1271, 1279, 1292, 1313, 1362, 4630; 2 truis 4428; pres.subjunc.5 trussez 5214; v.trans. find 5, 81, 90, 370, 559, 888, 890, 901, 920, 970, 1088, 1271, 1279, 1292, 1313, 1362, 2412, 2455, 2527, 2529, 3028, 3648, 3652, 3777, 3781, 3782, 3783, 3830, 4249, 4273, 4346, 4428, 4467, 4630, 4732, 4766, 4767, 4868, 4978, 5028, 5040, 5183, 5214, 5862, 5961, 6332, 6435, 6442, 6625,

578

Glossary

6695, 6697, 6811, 6834, 6983, 6994, 7205, 7249, 7428, 8094, 8133, 8140; find, meet with, encounter 1969, 3437, 4675, 6826, 7634; think up, imagine 3056; find out, discover 4780; trover plus espois find the thickest of the fighting 5494 tuen m.sing., toie f.sing. 2894, 4100, 5118, 7174; toies f.plur. 6358; poss.adj.(stressed form) thy, your tuit cf. tot tumulte sf. noisy disturbance, uproar, alarm 7532 turner cf. torner turqoi 3324; turqée f. 4552; adj. Turkish tuta cf. tot

u uis (graphy of hui with -s added for rhyme) adv. today 1499 uitime, uime cf. oitisme umelier cf. humilier unguement sm. ointment 7150 usé past part.used as adj. worn out, exhausted 6843 uxor obj.plur. 1204; sf. wives

v vair 2060, 5083, 7805; ver 3084; adj.used in heraldry for pieces of fur in two colours (of eyes) bright, shining, clear (ie with white and iris clearly distinguishable) 394, 2060, 2783, 7438; partly grey-haired, “pepper-and-salt”, implying becoming old 3084; (of horse) piebald 7805; used as sm. piebald horse 5083 valance sf. worth, valour 8000 valant 2231; valanz 310; vallan 3419; vallanz 3433; vallant 4738, 5587, 7478; valent 5366; adj. valiant, worthy 310, 2231, 3419, 3433, 4738, 5366, 5587, 7478 valcel sm. vale, small valley 2379 [valeir] pres.indic.3 vaut 269, 795, 917, 1300, 2429; valt 2185, 2203, 2236, 2756, 2934, 3006, 4742, 5411, 5497, 5553, 5957, 7191; val 2869; 6 valent 947; imperf. 3 valoit 5085; past def. 3 valut 5529; past subjun.3 valsist 3503; pres.part. vallant 6278; v.trans. be worth 269, 795, 917, 947, 1300, 2185, 2203, 2236, 2756, 2934, 3006, 4742, 5411, 5497, 5553, 7191; (with negative) be of any help to, avail 2429, 3503, 5957; pres.part.used as adj. worth 6278 valet sm. vale, valley 6881 valez subj.sing. 2303; veslez subj.sing. 5896; vaslez subj.sing. 7711, 7715, 7726, 7982; sm. young warrior, squire, man at the beginning of his knightly career valoi sm. valley 7562 vasal 5417, 7876, 8106; vasaus 2600, 3432, 5497, 5666; adj. valiant vasal obj.sing. 1082, 1494, 1651, 5470, 5505, 5663, 5734, 6367, 6370, 6995, 8121; subj.sing. 1109, 1493, 1869, 1974, 2231, 2457, 2734, 3173, 6411, 6543, 7877, 7972; obj.plur. 2968, 3347, 6158, 6220; subj.plur. 1135, 2193, 2401, 2467, 2479, 2491, 2884, 5022, 7003, 7548, 7885, 8152; vasaus subj.sing. 273, 2915, 3219, 3489, 3607, 4073, 5891; obj.plur. 1316, 2911, 3240, 3804, 5666; subj.plur. 3651; vassaus obj.plur. 6342; sm. vassal (ie someone holding a

Glossary

579

fief of an overlord by military tenure); one possessed of qualities (especially bravery) appropriate to a vassal; a vasal adj.phr. knightly, brave 6367; de vasal adj.phr. noble, valiant 1651, 5734, 8121 vasalage sm. the qualities which distinguish a vassal, especially valour 35, 325, 332, 551, 556, 1276, 1466, 2042, 2280, 2836, 2929, 3359, 3423, 6506, 7949 vavasor subj.sing. 7272, vavasors subj.plur. 5879; sm. (in general sense) lord (usually title next in dignity to a baron) 5879; (more precisely) subsidiary vassal of baron or lord (ie lower rank of fief-holder) 7272 vegiles sf. vigils of the dead, religious offices for the dead 6486 veïment sm. sight 7141 veingier 1614; v.intrans. take revenge (on) veir cf. voir veïr 2147, 3863, 7071; pres.indic.1 voi 477, 1395, 1701, 1753, 2042, 2653, 2968, 3068, 3229, 3360, 3555, 3895, 4186, 4214, 4274, 5228, 5489, 6005, 6092, 6098, 6112, 6890, 6917, 7705, 8057, 8150; 2 voiz 6007; 3 voit 358, 427, 848, 1336, 1482, 2305, 2497, 2711, 2014, 2964, 3313, 3333, 3342, 3395, 3839, 4007, 4079, 4093, 4107, 4113, 4299, 4403, 4560, 5522, 5586, 5667, 5816, 6163, 6430, 6958, 7780; veit 1393, 1961; 4 veions 3185; 5 veez 59, 446, 1303, 1343, 1502, 2987, 2996, 5410, 5676, 6010, 6353, 7279; 6 voient 2588, 2681, 3321, 5555, 6081, 7961; imperative 2 vei 5390, 5 430; fut.1 vesrai 369, 3887, 4960, 6194, 8186; 3 verra 386; vesra 3114, 3310, 3513, 3727, 6978; 4 viesrons 6604; vesrons 7755; 5 vesrez 346, 1164, 1920, 2016, 2087, 2543, 2587, 3435, 3604, 6531, 6799, 7857, 8129, 8145; verrez 2998, 3930, 5727; 6 vesront 3039, 7412; imperf. 3 véoit 4411, 7812; condit.2 vesrois 840; vesroiz 8101; past def.1 vi 1003a, 1004, 1009, 4782, 6649, 6707, 6709, 6735, 6740, 6772, 6775, 6777, 7091, 7544; 3 veïst 1034, 2097, 2397, 2464, 2476, 2537, 2880, 2951, 3120, 3345, 4309, 4365, 6880, 7307; vit 772, 915, 1404, 1412, 1535, 1582, 1589, 1621, 1800, 1810 + 55; 4 veïmes 2554; 5 veïstes 3359, 4298, 7258; 6 virent 1202, 1093, 2403, 5371; past subjun.5 veïsiez 2802; veïssiez 3117, 6146, 6158, 6290, 7295; veïssez 4241, 5143, 5155, 5383, 5423, 5594; pres.part. véant 342; voiant 4452, 4460, 6006; veiant 4501, 4626; past part. veü 1454, 1925, 3496, 4001; veüz subj.sing. 273, obj.plur. 1411, 1534; veüe f.sing. 4, 4058; v.trans. see 4, 273, 342, 386, 427, 477, 840, 1003a, 1004, 1009, +; be witness to, live to see 3887; (with period of time) be present for, be alive for, see out 346, 4960; vos els véant before your eyes, in your sight 342; voiant sa baronie in the sight of his barons 4452, 4460; voiant toz … in the sight of all …4501, 4626; nos els voiant before our eyes 6006 velle adj. cf. viels velle v.intrans. cf. voleir veltre obj.sing. 4434; veutres obj.plur. 154, 221; sm. hound vencre 5485; pres.indic.3 veint 4439; pres.subjunc.3 vence 2776; fut.4 vencron 2708; past part. vencu 3714, 7739; vencuz 811, 1245, 1540, 3166, 3556, 3665; v.trans. conquer, defeat 811, 1245, 1540, 2708, 2894, 3166, 3556, 3666, 3714, 5816, 7739; bataille vencre win a battle 3165, 4064; v.intrans. win 2776, 4439, 5485, 5636 vene sf. vein 3062, 3104 vener 154; v.intrans. hunt vengement sm. vengeance 3520, 3731, 7418 vengeson 492; vengison 610; vengneson 7403; sf. (used with prendre) (take) revenge, (wreak) vengeance venir 1393, 1395, 2537, 2580, 2997, 3584, 3597, 3917, 4429, 4656, 4896, 6107, 6370, 6740, 7157; imperative 2 ven 5752; pres.subjun.3 veigne 3341; viegne 3418, 4628, 4753, 6177; vigne

580

Glossary

6218a; fut.3 vendra 6462; venra 671, 960, 1014, 3309, 4785, 5024, 6807, 8078, 8113; vegnera 3533; 4 venrons 6603; vendrons 7347; 6 vienront 7401; past def.2 venis 1605; 3 vint 206, 260, 574, 643, 873, 985, 992, 1057, 1107, 1155 + 45; 4 venimes 7479; 6 vindrent 4285, 4714, 5154, 5835, 6144, 6240, 6992, 7174; past part. venuz subj.sing. 140, 271, 311, 633, 641, 1155, 1158, 1159, 1830, 1832, 2352, 2724, 2726, 2918, 4623, 4915, 5164, 5348, 5736; venu subj.sing. 1240, 6470; subj.plur. 624, 2168, 4105, 4642, 4840, 4903, 5202, 5939, 6482, 7583, 7586, 7652, 8034; v.intrans. come; venir a plaisir 3590, venir a plasir 3928 give pleasure to, please, accord with one’s wishes; venir encontre come to meet 6992; infin.used as sm. what is to come, future; au venir in the future 2584 venjance 6464, 7319, 8162; venjanche 7601; sf. vengeance ventaille obj.sing. 1750, 5138; ventalles obj.plur. 3118, 5392; sf. part of hauberk covering cheeks and chin [ventlier] v.intrans. flutter 1353 ventoser 6523; v.trans. treat with a cupping-glass to draw off blood from infected part of body venture sf. fate 2275 ver cf. vair veraiement 3316; verasement 3791; adv. truly verais subj.sing.m. 7181; veraie f. 5041; adj. true [verdier] (graphy of verdoyer, verdeier) v.intrans. grow green verge sf. rod, staff 4550, 5920 verger 2642; [vergier]; v.trans. chase, ornament with vertical stripes, inlay 2297, 2642, 4407 verger 119; vergier 191, 1666, 7811; sm. garden vergoigne 5082; vergogne 8057; sf. shame, dishonour vergognos adj. disrespectful, wanting in honour 6398 vergonder 2739; v.trans. cover in shame, dishonour [vermeillier] v.trans. make red 4408 vermeil m.sing. 1716, 1825, 2819, 4990, 6034; vermelles f.plur. 1328, 2781; vermeis m.obj.plur. 1374; vermeus m.subj.sing. 2941, 5163, and obj.plur. 5002; adj. vermillion, bright red 1361, 1328, 1374, 1716, 1825, 2221, 2781, 2819, 2941, 4990, 5002, 5163, 6034 verser 5595; v.trans. bring down, throw to ground 3674, 3949, 5706; v.intrans. be brought to the ground, fall down 2447, 5595; versez past part.used as sm., obj.plur. those brought down, the fallen 3949 vertuz 639, 1417, 1539, 2354, 4704; vertu 1047, 1836, 2135, 2166, 3470, 3497, 3526, 4021, +; vertut 7908; sf. strength, power, force 1047, 1417, 1539, 1836, 2135, 2153, 2903, 3470, 3526, 4021, 4838, 5760, 5767, 6471, 7908, 7928, 7967, 7974, 7989; courage, valour 2166, 4826, 4835; manifestation of divine power, miracle 3497, 5282, 6122, 7981; miracle (used sarcastically) * 4704; (divine) power 4100, 5775; par vertuz 639, per vertut 7908, par vertu 7989 with force, forcefully, vigorously (?); per vertu through physical strength OR valiantly (perhaps both) 7908, 7914; per grant vertu in great force 3470, 6262; per tel vertu with such force 3526, 4021; de Deu aiez vertuz may you have strength from God, may God give you strength 1417, 1539; a force e a vertuz (vertu) in force and with valour, energetically and valiantly 2354, 4826, 6481; de grant vertu very strong, of great strength 5760 vertuos adj. strong, powerful, courageous 669 vespre obj.sing. 6258, 6412; subj.sing. 7485; vespres subj.sing. 3188; sm. evening vespree subj.sing. 1028, 4766, 5702, 5835, 6144; sf. evening veüe 3528, 7929; sf. (faculty of) sight

Glossary

581

veutres cf. veltre veve adj. bereaved, widowed 2606; bereft 4233 veveneis adj. of or pertaining to Vienne in France, famed for its steel, or possibly pertaining to Viana in Galicia (cf. Whitehead, Index of Proper Names: vianeis) 1373 viaire sm. face, visage, countenance 680, 3555, 5197 viels subj.sing. 808, 816, 3659, 4521; viel subj.plur. 6067; vels obj.plur. 3422; velle f.sing. 3150; adj. old vienois obj.plur. 4951; vienoi obj.plur. 7564; adj. pertaining to Vienne in France* or Viana in Spain, cf veveneis above (although Boissonnade p.359 sees this as a reference to Poitevin steel) vif 4043, 4344, 8125; vis subj.sing. 706, 1279, 2861, 3404, 5442; adj. alive 706, 1279, 3404, 3461, 4344, 5442, 8125; (of devil) incarnate 2861, 4043 vigorance sf. strength 1743 vileins subj.sing. 7944; sm. villein, peasant vilté sf. degradation 481, 567, 2039, 5729; degrading or dishonourable action 1133 vis sm. opinion; ce m’est vis it is my opinion, it seems to me 7392, 7436, 7444 vis sm. face, countenance 133, 373, 698, 936, 980, 1273, 1528, 1665, 3398, 3758, 3827, 2860, 3906, 3913, 3986, 4131, 4140, 4864, 4872, 5105, 5128, 5530, 5788, 6178, 6518, 6527, 6562, 6584, 6596, 6750, 6756, 6764, 7047, 7292, 7334, 7381, 7438, 7449, 7613, 7751, 7891, 7923, 8072 vis adj. cf. vif visanz sm. mink (fur)* 6555 vitalle sf. victuals, food 4333 [vivre]; fut.3 vivera 185; vivra 829, 3538, 8084; past subjunc.1 vesquisse 3597; 3 vesquist 3511; vesqist 4046; pres.part. vivant 341, 1252, 4072, 4605, 4607, 5470, 5949, 6011, 6277, 6293, 6637, 6939; vivanz 3453, 3620, 3624; vivent (with ending modified for rhyme) 3730; v.intrans. live, be alive 185, 385, 829, 1941, 3511, 3538, 3597, 4046, 6039, 8084, 8186; tant com je vive as long as I live 8186; pres.part. used as adj. living, alive 3620, 3624, 5470, 6277, 6293, 6637, 6939; pres.part.used as sm. life, lifetime 341, 1252, 3453, 3730, 4072, 4605, 4607, 5949, 6011; (person) living person, person alive 6661 [voer] v.trans. dedicate, consecrate; voée en son non dedicated to him, consecrated to him, undertaken in his name 5336 voiant cf. veïr voider 2651; [voidier]; v.trans. empty 2651, 3804, 5924 voir 727, 804, 2444, 3561, 3660, 4126, 5038, 5202, 5349, 5357, 6046, 6808, 6920, 7127, 7144, 7151; voirs subj.sing. 105, 850, 2036, 4747, 5981; veir 1995; voire f.sing. 3912, 4943, 5073, 5107, 6049; (adv. 352) adj. true 105, 850, 2036, 3912, 4747, 4943, 5073, 5981, 7127, 7144, 7151; adv. truly, in truth, indeed 352, 4155; de voir in truth, truly 727; por voir in truth, truly 804, 3561, 4126, 5038, 5202, 5349, 5357, 6808; dire veir 1995, dire voir 2444, 3660, 6046, 6920 speak truthfully, tell the truth; il n’i savoient ni dient voir nïent they knew nothing about it and therefore did not speak an atom of truth about it 2444 voirement 1262, 1613, 3716, 3720, 3786, 3968, 4132, 4158, 5108, 7176; voiremant (graphy to suit visual rhyme) 1330, 4171, 5966, 5980; adv. truly, certainly, most surely volage adj. winged, flying 2291 [voleir]; [voldre]; pres.indic.1 vei *19; voi *4885; veil 1096; viel 1133, 2976, 3765, 7601; vuel 1452, 1905, 7337, 7719; vel 3766; 3 velt 165; vuel 1340; 6 volent 83, 2003, 4890, 7998; pres.subjunc. 3 velle 380, 586, 1337, 2812, 3741; fut.1 volrai 6116, 6191, 7365, 7367,

582

Glossary

7628, 7946; voldrai 6861; 3 voldra 3026, 6816; volra 3125; 6 volront 3123; imperf. 1 voloie 7629; condit.6 voldroient 4981; past def.3 vol 31; volst 147, 999, 2090, 2716, 3095, 3542, 3592, 3819, 3821, 3872, 5183, 5769, 5870, 6510, 7146, 7867, 7915; volt 2643, 2714, 2953, 3348, 4361, 5142, 5179, 7620; vout 4360; 6 voldrent 1005; volsent 4570, 6094, 6735, 6769, 6778, 6875, 6877, 7110, 7287; past subjunc.3 vousist 5532, volsist 7700, 8054; v.intrans. wish volsure sf. rolled band of material used to ornament robes, a kind of piping 6557 volunté sf. what one wishes, wish, will 586, 4796, 6540; s’il est qe croire velle ma volunté if there is here what my strong intention (will) wants (drives me) to believe (ie if there is the opportunity for revenge through treason) 586; par bones voluntez 1871, par bone volunté 8047 through goodwill, generously; estre ma volunté against my wishes 8052 vos sf. voice 5271 voutiz 1177; voltiz 2245; voltis 2593; adj. (of shield) curved vuel sm. wish, will 382

z za cf. ça zant num.adj. one hundred 1440 zo cf. ça

IV Proper Names A Abel 2385: a reference to the son of Adam and Eve, killed by his brother, Cain (cf. Genesis iv.1-8); the name may be chosen as implying a Middle Eastern origin Abilent 5313: a town and district in the Middle East; probably the town of Abila (now Nebi-Abil) at the foot of the Anti-Lebanon mountains (cf. Romania 9 (1880), 29) Abiron 2126, 4666: Abiram (cf. Numbers xvi.12, 27 and details given under Dathan): the formula ‘Dathan et Abiron’ was much used in mediaeval contracts Abisme 2829: a name signifying ‘abyss’, usually ‘abyss of Hell’; a Saracen noble killed by Turpin Acars (li Mors) 210: French baron, brother of Almiz Adant 5964: Adam, the first man according to Genesis Aiglent 5311: a pagan country in this text, although in other chansons de geste this name was usually given to the fief of Milon, the husband of Charlemagne’s sister Berte (or Gille) Aiguent 7796: an unidentified place or area. cf. Gui d’Aiguent Ais 47, 63, 159, 731, 813, 1052, 4426; Aiz 180, 182; Haiz 4351: Aix-la-Chapelle, the modern Aachen, which was Charlemagne’s imperial capital Albanie 5309: this may possibly be a reference to Scotland, i.e. a reminiscence of the Picts and Scots mentioned in Geoffrey of Monmouth, rather than Albania Alborion 5691 cf. Auberis Alemant 5010, 5141: German Alemeigne 4235; Alemegne 6499: Germany Alexandre 712; Alixandre 4538: Alexandria in Egypt Almarie 6507 cf. Aumarie Almice 3490, 3627: the name of Turpin’s sword Almiz 210: French baron, brother to Acars Alvergne 5046, 6204: Auvergne (cf. Garin d’Alvergne) Amaugin 6635; Amaugis 6588: the well-educated confidential clerk to Girart de Vienne, he is almost certainly identical with the high status retainer/court official named in C 6588; he came originally from Coutances in the Cotentin peninsula in Normandy (Amaugis, strictly the subject form, is used for the direct object in C 6588) Amauriz 1182; Amauri 2345; Amauris 2352: Saracen king Amiens 7451: the cathedral city in Picardy Aminoi 7568: the region around Amiens (cf. Gui d’Aminoi) Amorois 781 cf. Braüz l’Amorois Anage 5277: (unidentified) pagan country Anceume 122: one of Charlemagne’s barons, identified in O 3008 as Antelme de Maience Angevin 5495, 5695, 6068: inhabitant(s) of Anjou; specifically, reference to Tierri d’Anjou as Li Angevin 7943 Anglés 5621: a Saracen, but the choice of name, Anglés, may reflect reaction to the sixth century westward sweep of the Germanic tribe and/or the enmity between French and English during the twelfth century

584

Proper Names

Anjou 110, 123, 209, 4234, 5690, 6171, 6568, 6832, 7352, 7641, 7974; Angou 5671: an area rather larger than the modern Anjou, stretching across what is today northern Poitou (cf. also Jofroi d’Anjou) Anseïs 121, 2209, 2729, 3782: one of the Twelve Peers, killed by the Saracen Malqidanz. According to Bédier (1921, vol. 4) this name was well-known in the twelfth and thirteenth centuries as that of an ancient king of France – and Anseïs1 (not the peer) is listed by Langlois (1904, 34) as King of France in Jean Bodel’s Chanson des Saxons; in this connection, Boissonnade (1923, 356) suggests Ansegisilus, the grandfather of Pépin le Bref and also St. Anségise, Archbishop Anségise of Sens and Bishop Anseïs of Troyes, Chancellor to King Raoul of Burgundy (cf. also Hunez) Antoine d’Avignon 2761: a French baron, killed by Grandoine. He is identified in the text as the seigneur of Valence, the ancient city on the Rhône with the steep cliffs of the Massif Central on the opposite, western bank Aostin (Seint) 3897: St. Augustine of Hippo, 354–430 AD, the famous Christian philosopher Apolin 646, 2697, 3889, 4467, 4658, 4743, 5329, 6076: Apollo, the Ancient Greek sun god, whom the mediæval Christians coupled with Mahomet and Jupiter to form a pagan Trinity supposedly worshipped by the Moslems. Usually coupled with Mahomet in the text, but le lignage Apolin (C 3889) and la gent d’Apolin (C 6076) are used to designate ‘the pagans’ Aqiteigne 2677, Equiteigne 2668, 2673: Aquitaine, seen as being an ally of the Saracens, perhaps because of the (twelfth-century) wars between the French and the Plantagenets (see the name Anglés above, which may represent Anglais; cf. also Roudez d’Aqiteigne) Arabe cf. Erable Arabis object form plur. 630, 2651; Arabiz object form plur. 2249; Arabi subject form plur. 5601: the Arabs (the adjective arabi 6415, 1986, 2596 is applied to horses) Archant (L’) 4568: L’Archanz, named as a Mediterranean port and the scene of the battle in the Chanson de Guillaume, has been identified as the Aliscans of Arles by Bédier (1922, vol. 1, 83). In the context in C, it might alternatively be a corruption of Alicante in Spain, or at least a confusion of the two names. Langlois (1904, 45) simply identifies it as where Vivien was defeated and killed Archedeclin (Seint) 3893: this name is a corruption of Archetriclinus, ‘master of the feast’, the title used in the story of Christ’s first miracle at the marriage in Canaa (John ii.1–11). It was assumed in error to be the bridegroom’s name and he was thought to be a saint Ardene 4428; Ardenois 5690: the Ardennes area in North-Eastern France. cf. also Tieri d’Ardene, Tieris d’Ardenois Argoigne 5087: the Argonne region to the east of Paris, between the R. Aisne and the R. Aire (cf. also Terris d’Argoigne) Argüel 5654: an unidentified pagan country (based on Aroer?, the name of two places in Palestine mentioned in the Old Testament (cf. Numbers xxxii.34; I Samuel xxx.28; II Samuel xxiv.5; I Chronicles v.8; Isaiah xvii.2; Jeremiah xlviii.19). Langlois (1904, 47) links it to Argoilles, a pagan country. Boissonnade (1923, 201s.) suggests Archelaïs > Eraclea/Eregli/Héraclée in Cappadocia, the site of a Crusader victory in 1097 Aspre (les porz d’) 1250, 1475: probably the Col d’Aspe in the Pyrenees (today the Col du Somport above the Vallée d’Aspe), but emendation to porz de Cipre (as per C 869, Cibeing misread as As-) or to porz d’Espagne (as in O 1103) would also be possible; cf. footnote to C 1475. Charlemagne’s army would cross via the port de Cize to reach St Jean-Pied-de-Port

Proper Names

585

Aspremont 6873, 7708: possibly Aspromonte in Calabria, Italy. Otherwise an imaginary place-name (cf. Richer d’Aspremont) Asragons 3438: inhabitants of Aragón in N-E Spain. This was a Moorish kingdom from early in the eighth century until 1131. Saragossa was its most important city Athon 5219 cf. Otes 1 Aton (d’Estampes) 6454: one of Charlemagne’s barons Auberis subj. 5365; Alborion obj. 5691: Saracen baron, appointed standard-bearer to Baligant Aude subject 3517, 4220, 6233, 6463, 6475, 6541, 6548, 6560, 6561, 6572, 6577, 6596, 6608, 6619, 6627, 6631, 6639, 6679, 6719, 6727, 6750, 6821, 6825, 6839, 6883, 6895, 6906, 6910, 6923, 6937, 6976, 6996, 7006, 7011, 7017, 7025, 7029, 7049, 7053, 7061, 7067, 7069, 7076, 7083, 7087, 7092, 7095, 7111, 7118, 7125, 7163, 7185, 7197, 7200, 7206, 7222, 7228, 7230, 7242, 7259, 7265, 7302; obj. 2998, 3512, 3805, 4143, 6178, 6191, 6529, 6796, 6804, 6861, 6885, 6944, 6962, 6975, 6993, 7032, 7047, 7218, 7253, 7264, 7323 and in the envoi. Audein object form 7292, 8027; Auden obj. 7298 (it may be purely coincidental that a different graphy is used for the object form only after Aude’s death): Oliver’s sister, betrothed to Roland; she was the niece of Girart de Vienne, in whose household she was brought up after the death of her father, Renier de Geneve Aufricans 2724: African (perhaps specifically North African, i.e. Arab); the adjective africant, describing a robe or cloak, is also found in C 5954 Aufrique 2724: Africa Aumarie 3694, 6609, 6897, 6929, 7078; Almarie 6507: Saracen town and kingdom: Almeria in Spain (cf. Boissonnade 1923, 73) Autaclere 1971; Auteclere 2309, 2641; Hauteclere 2932, 3334: the name of Oliver’s sword Autehoïe cf. Autevile Autevile 251; Autehoïe 749: the place where Charlemagne’s two earlier envoys to Marsile, Basan and Basile, were put to death. The name derives from the practice of placing gibbets on hills at the borders of a lord’s fief. Boissonnade (1923, 87s.) identifies this as one of a series of ‘puy’ outside Barbastro Avignon 2761: Avignon cf. Antoine d’Avignon

B Babiloine 4520, 6931: possibly Babylon, but this was also the usual name for Old Cairo in the Middle Ages, in C 4520; quite definitely Babylonia, the ancient kingdom around Babylon, in C 6931 Babuer 76: one of Marsile’s barons Babureigne 1287: a Saracen kingdom, possibly ‘Barbare-reigne’, ‘the kingdom of the Berbers’, a term generally used to designate the North African countries to the west of Egypt Baile 5278: a Saracen town. Boissonnade (1923, 217) suggests that this may be Balis, also called Balide (O reads Balide here), an important town on the way from Aleppo to Bagdad, mentioned in pilgrim itineraries because the tomb of St Bacchus was there Baiver 5141, 7637, 8123; Bauver 7695: Bavarian (cf. Naymon lo Baiver, Neme li Baiver) Baivere 4988: Bavaria, part of Charlemagne’s eighth-century empire; a duchy by the twelfth century, possibly earlier

586

Proper Names

Balaguer1 75, 1272; Balaguez 3678: presumably in these 3 instances Balaguer in Catalonia. In the Middle Ages, this fortress town stood at the eastern frontier of the territory ruled by Saragossa and commanded an important position on two great routes (cf. Boissonnade 1923, 91; cf. also Clarin de Balaguer and Rubion) Balaguer2 4568: from the context, this must be on or near the coast, so not Balaguer in Catalonia; but C’s geography is very suspect in C 4568–4569 and place-names may be imaginary. Possibly a corruption (via metathesis) of or scribal error for Badalona in the province of Barcelona or Barcelona (formerly Barcino > Barcinona) itself Balcaire 5748: Langlois (1904, 79) suggests Beaucaire (Gard), an arrondissement of Nîmes, but this does not fit the context. Read as Balc[l]aire it would be a reference to the ‘plains of Beauclere’, identified as a Saracen country on the border of Syria. Otherwise, this may be bel Caire, ‘beautiful Cairo’ or a corruption of el Caire Baldestoz 3441: an unidentified and probably imaginary pagan land inhabited by giants Baligan1 subj. 4532, 4828, 4886; obj. 4014, 4037, 4520, 5164; Baligans subj. 4546, 4563, 4780, 4815, 4841, 4995, 5101, 5215, 5303, 5325, 5339, 5360, 5476, 5589, 5614, 5619, 5693, 5714, 5734, 5744, 5757; Baligant subj. 4613, 4902, 5222, 5351, 5398, 5463; obj. 4587, 4723, 4847, 5047, 5082, 5676, 5712, 5786, 5796, 6933; Baliganz subj. 4637, 5419, 5623; Balligan obj. (= possessive form) 5723; Balligans subj. 4798, 5819, 5825; Balligant obj. 5578; Balliganz subj. 5637: The ruler of the Islamic empire, presumably identified with the Caliph of Baghdad, but Boissonnade (1923, 190–192) suggests the Sultan of Cairo as an alternative; he is killed by Charlemagne Baligant2 599; Balliant 4157; Baliant 5972; Biauliant 6296: Bethlehem in Judea Balne 3267: Beaune in Burgundy Barbarie 2144: Barbary, a province of Africa, the land of the Berbers. Cf. Romania 55 (1929), 136 Barbarins 1266: Berber; or this might simply be the substantive or adjective meaning ‘barbarian’, ‘barbarous’: cf. barbarine, adjective feminine, C 4469 Barbemor 2625: the name of a horse, probably that belonging to Cliboïs (cf. footnote to C 2625) Basie 249, 748; Basille 354: a French baron, sent on an earlier occasion by Charlemagne as ambassador to Marsile and put to death by the latter; companion-at-arms to Basin Basin le Bergoignon 6203, 7638, 7836: one of Charlemagne’s barons, specified as ‘the Burgundian’ to avoid confusion with the Basin mentioned earlier (see below), who was dead before the events of this story began Basins 249; Basin 748; Basant 354: a French baron, companion-at-arms to Basie and sent with him as envoy to Marsile, who had both barons put to death Baudöin 385; Bauduin 528: the son of Ganelon and Berte, therefore half-brother to Roland and nephew to Charlemagne Baudois 782: inhabitant of Balaguer, the fortified town to the east of Saragossa (Clarins li Baudois is also known as Clarin de Balaguer) Baulande 6580, 6598: Langlois (1904, 79) identifies this simply as a town by the sea, perhaps Nice (cf. Hernaut de Baulande) Bedöin 6045: a Saracen race, presumably Bedouins Belcler 4569: a place or country on the route sailed by Baligant from Babylon (or Cairo) to Saragossa. It may well represent a port on the Spanish coast (Barcelona?), but could equally well be imaginary (cf. also Balcaire and Balaguer2 above) Berart cf. Bernart

Proper Names

587

Berengers 1156, 2242; Berenzer 1780, 2760; Berenger 3763, 5219: French duke, one of the Twelve Peers, the companion of Otes. Boissonnade (1923, 38, 41s.) identifies him with Raimond-Bérenger III le Grand, count of Barcelona, one of the bravest adversaries of the Saracens in Spain, but the name Bérenger is very common in chansons de geste Bergognon 5576; Borguegnon 6121; Bergoignon 6203: Burgundian(s) (cf. Basin le Bergoignon) Bernart 6588; Berart 6628: vassal of Girart de Vienne Berte 6209, 6233, 6960, 6966, 6970, 6973, 6976: Sister to Charlemagne, widow of Duke Milon d’Aiglent and mother of Roland; then married to Ganelon, by whom she had another son, Baudöin Besençon 6204; Besenzon 7890: Besançon, the capital of the Franche-Comté Betanie 7166: Bethany, the home Lazarus, Martha and Mary, in the Holy Land (cf. New Testament: John xi.1) Biauliant 6296 cf. Baligant2 Biterne 1183: a Saracen town; identified by Whitehead (1942, 168) as Viterbo in Italy Blanzardin obj. 31, 562; subj. 643; Blanzardins subj. 99, 619, 713, 3065, 3071; Blançardins subj. 34, 58; Blançardin obj. 79, 138; subj. 139, 782a.: Saracen baron, counsellor to Marsile, sent by the latter as envoy to Charlemagne Blaives 6199, 6482, 6483, 6531; Blaive 6522, 6826, 6835, 6982: Blaye on the Gironde Blois 109, 4759, 6170: the city of Blois on the Loire (cf. also Gerart de Blois) Bloz 5272: a pagan people according to Langlois (1904, 101). These were perhaps the Polovzias or Cumans who inhabited the north and west shores of the Black Sea in the twelfth century Boidïent (Castel) 4675: an unidentified (possibly imaginary) place in Turkey Boloigne 5083; Bologne sor mer 7451: Boulogne-sur-mer Bone 8086 cf. Bovon Boneval 7354: the town of Bonneval, on the R. Loir, just south of Chartres, in the département of Eure-et-Loir; present population c.5,000; its Abbey of St Florentin was founded in 841 Bonevent 4844: a Saracen city, which might be Benevent in Italy, but mention is made of a roi de Bonivent in C 5114, where Bonivent is clearly a scribal error for Niniven, ‘Niniveh’. The same error may well have occurred here Bordele 2219; Bordel 2386: Bordeaux (cf. also Engeler de Bordel) Bos de Lions 6589: vassal of Girart de Vienne Boteroz 5266: Butintro in Epirus, since legend had it that this was where Judas Iscariot was brought up (cf. C 5267); alternatively, closer to the Holy Land, the valley of Butentrot in Cappadocia (cf. Bédier 1922, 44; Romania 7 (1878), 435–437; Boissonnade 1923, 199–201) Bovon 3137: Chief cook to Charlemagne. Also called Sanson, C 3198. Since ‘chief cook’ could imply ‘chief steward in charge of provisioning the army’, this was no mean post; it may therefore be the same Bovon who is referred to later as a baron (see below) Bovon 7835; Bone 8086: one of Charlemagne’s barons Braibant 7413: Brabant in Belgium Braimimonde 4461, 4699, 4731, 4835, 5890, 5882; Brasmimonde 4559, 4696, 5813; Braismimonde 5844: the wife of King Marsile (see also Braimonde) Braimonde 940: named in error in C and V7 as the wife of Brans de More (cf. note to C 940); O and V4 agree in having just one Bramimunde, the wife of Marsile Brans de More 937; Brans 942: a vassal of Marsile according to C and V7, probably falsely identified with Braüz l’Amorois; in fact, Brans de More is a corrupt reading of the name Braimimonde (cf. note to C 940)

588

Proper Names

Braüz l’Amorois 781: a Saracen baron, vassal of Marsile; possibly the same person as Brans de More. The sobriquet may be a play on moreis = ‘black, dark brown’ and amoros = ‘pleasant, likeable’, ‘worthy’ and/or amouros = ‘amorous’, ‘compassionate’, ‘faithful’ Brehas 4883: the wife of king Noitier; instrumental in the release of the pilgrim Reiner who warns Charlemagne of Baligant’s approach. This may be a reference to a popular story of the time, now lost; the names Brehas and Noitier are found only in C and V7 (cf. Moisan 1986, 258) Bretaigne 2348; Breteigne 2679, 4234; Bretegne 7847: Brittany (cf. also Seint Mallo de Bretaigne, Salemons de Bretegne) Bretons 5032: Bretons Briance 4932: a Saracen kingdom, according to Langlois (1904, 115), perhaps also alluded to in C 5169, the destroiz de Bruance (q.v.). Possibly Brihuega, a market town with ancient walls, in La Alcarria, part of the province of Guadaljara in Castilla la Nueva, on R. Tajuna which cuts through the plateau north-east of Madrid. However, it is coupled with Turs de Cople (q.v.) and there may be both corruption and confusion here, since an allusion to ancient Byzantium (Bisance > Briance perhaps) would fit better with nearby Syria Brigart 1269: possibly Berbegal in Aragon, near Barbastro (cf. Revue critique 2 (1858), 174; Bédier 1921, vol. 3, 291; Boissonnade 1923, 95s.). Cf. also Malprimes de Brigart and Malpui de Mont Pregal, C 2184: matching these lines, O 889 refers to Malprimis de Brigant and O 1261 to Malprimis de Brigal. The Spanish place-names Briance (above) and Bruance (below) may perhaps also be alternative graphies Bruance (les destroiz de) 5169: possibly Brihuega, since it is situated on a plateau; cf. Briance above. Or, if this is a graphy of Brigart (q.v.), perhaps Berbegal Brum de Val Fondee 5712: a Saracen Brunsoir 5344: a land of pagan giants Buiron 4648: a Saracen, ancestor of Gibon; mentioned only in C and V7, otherwise unknown

C Cafarnaön 4665: Capernaum, the ancient city by the Sea of Galilee where Christ preached Caïn 3794, 6079: Cain, the first murderer, who killed his brother Abel; his crime and punishment are recounted in Genesis iv.1–15 Calabre 552, 2896: Calabria, in southern Italy (the region forming the ‘heel’ of Italy); with Puglia, this fell beneath Lombard rule until Charlemagne defeated the latter Canabes, Canabex, Canabeu cf. Carmilleu de Dorree Capadoce 2751: probably Cappadocia, an ancient country in Asia Minor which lay to the west of modern Armenia, today part of eastern Turkey. The name would be familiar in mediæval times because it is mentioned in the New Testament (Acts and the Epistles) Capamor 5248; Capamorz 5262: a Saracen king, the Emir of Lerie (or Leurie) q.v. Capoë 2750: a pagan, the father of the Saracen Grandoine, king of Cappadocia Carcasone 577: Carcassonne, the famous mediæval walled city in the Aude département in south-western France Carmilleu de Dorree 5386; Canabés 5520; Canabeus 5545, 5616, 5624, 5628; Canabeu 5552: a Saracen, king of Turre (q.v.), brother to Baligant, killed by Charlemagne.

Proper Names

589

Despite the considerable difference between Carmilleu and Canabeu, this must surely be one and the same person; the variation may be indicative of different sources Cartage 3297: a Saracen city, possibly Carthagena on the east coast of Spain, but according to Boissonnade (1923, 157s.), after the town of Carthage was destroyed, this name was used for the whole of the old Roman province in North Africa. Since this is one of the domains of Marsile’s uncle, the Caliph, the latter makes better sense Caudie 3297: another of the domains ruled by the Caliph, Marsile’s uncle; possibly Gandia in Spain (cf. Langlois 1904, 130), but this again seems too small (cf. Cartage above). It seems better to identify this with Chaldæa, the term used in the Old Testament for the whole of Babylonia. Strictly, it is the land between the head of the Persian Gulf, the desert of Arabia and the River Euphrates Cesaire 2849; Jule Cesar 3153: Julius Caesar, the first Roman emperor, certainly in C 3153; in C 2849, Cesaire may simply refer to any Roman emperor Chaëlon 6455: Châlons-en-Champagne (formerly Châlons-sur-Marne) Charlle subj. 1, 437, 472, 1433, 2687, 3312, 3330, 3552, 4255, 4268, 5912, 7047; obj. 660, 1597, 1649, 3294, 3348; Charlles subj. 10, 63, 107, 350, 401, 419, 438, 664, 708, 722, 833, 850, 855, 1020, 1050, 1064, 1089, 1206, 1216, 1251, 1294, 1307, 1425, 1445, 1459, 1498, 1559, 1875, 1892, 1907, 1947, 1978, 2073, 2132, 2178, 2542, 2628, 2993, 3036, 3051, 3078, 3107, 3111, 3130, 3163, 3225, 3339, 3533, 3563, 3580, 3582, 3603, 3629, 3649, 3704, 3929, 4070, 4089, 4133, 4185, 4251, 4368, 4376, 4391, 4396, 4401, 4410, 4421, 4721, 4761, 4805, 4820, 4837, 4853, 4876, 4917, 4938, 4955, 5059, 5083, 5134, 5215, 5225, 5252, 5404, 5408, 5433, 5548, 5586, 5639, 5710, 5716, 5767, 5772, 5838, 5865, 5876, 5919, 5934, 5951, 5990, 5999, 6009, 6012, 6022, 6035, 6046, 6083, 6085, 6090, 6097, 6131, 6155, 6168, 6172, 6198, 6203, 6206, 6224, 6226, 6250, 6304, 6313, 6316, 6321, 6441, 6451, 6533, 6536, 6542, 6842, 6855, 6892, 6921, 6923, 6943, 7002, 7045, 7107, 7215, 7257, 7309, 7334, 7340, 7355, 7363, 7381, 7420, 7534, 7599, 7633, 7636, 7704, 7754, 7813, 7898, 8189; Charllon obj. 38, 482, 647, 1093, 1263, 1593, 2141, 2902, 3127, 3145, 3208, 3390, 3497, 3547, 3725, 3890, 3892, 3922, 3987, 4209, 4435, 4531, 4630, 4654, 4669, 4758, 4809, 4904, 5157, 5184, 5214, 5249, 5320, 5335, 5348, 5362, 5402, 5466, 5611, 5715, 5722, 5736, 5761, 6231, 6448, 6584,7026, 7252, 7648, 7766, 7823, 7863; subj. 3658, 4872, 5562, 5697, 5753, 5799; Karllon obj. 81, 260, 3282; subj. 7463; Charlon obj. 736; subj. 1316 (cf. footnote); Charles subj. 1282; çarlle subj. 1472; Karlle subj. 2160; Carllon obj. 2569, 4464; Karlles subj. 3146, 3171, 4599; Carlle subj. 4999, 5283; Charll’ subj. 8198: Charles the Great, the emperor Charlemagne, often referred to specifically as king of France Charlles li Maines subj. 119, 1223, 1346; Karles li Maines subj. 201; Charlle Meine obj. 746, 825, 859, 2971; subj. 980, 1041, 1459, 3495; Charlle Maine obj. 807; Charlle Mene subj. 836, 4224, 4247, 6885; obj. 4209, 4730; Charlle Meinne subj. 1041; Karlle Meine subj. 2405; Carlle Meine obj. 2431; Charle Meine subj. 3209; Charllo Meine obj. 4145; KarlleMene obj. 4209, 4279, 4512, 5673, 7854; subj. 4371, 6518, 6756, 6820, 6971, 7962, 7981, 8000, 8015, 8019, 8035, 8132; Charllemene subj. 4963, 6694, 6810, 7032, 7204, 7207; Charlles li Maignes subj. 4985, 6002; Carllomene subj. 5538; Charlemene obj. 6624; Charllomene subj. 6981, 7774, 7816: Charlemagne (cf. Charlle above) Chartres 109, 4759, 4879, 7353: the cathedral city of Chartres Chastelle 2220: Castile, Spanish Castilla, the Christian kingdom which emerged in the eleventh century, occuping the central table-land of Spain (the reference is to horses: destrier de Chastelle) Cipre (pors de) 869: the Port de Cize, today the Porto de Ibañeta, a pass in the Pyrenees on the route from Roncevaux to St Jean-Pied-de-Port; or possibly a reference to St Jean-

590

Proper Names

Pied-de-Port itself, for it was the capital of the small region of Cize in southern Navarre. Cf. also Aspre (les porz de) Clarenbalt 5409; Clarenbalt de Mascon 7505: a French baron from Mâcon, provided both references are to the same man. However, the Clarenbalt in C 5409 is coupled with Naymes and described approvingly as ‘valiant’ or ‘noble’, whereas Mâcon was probably in Ganelon’s territory, since his wife Berte was residing there, and it is Clarenbalt de Mâcon whose helmet is put on Ganelon (C 7505) Clariel 4619, 4782; Clariaus 4722; Clarïez 4799: a Saracen nobleman, the son of Matragiez and brother to Effragiez; the brothers were sent by Baligant as messengers to Marsile Clarin de Balaguer 75; Clarins li Baudois 782: Saracen baron (cf. Balaguer) at the court of Marsile Claris 6597: the previous owner of the mule which Aude rides; a Saracen killed by Girart de Vienne at Baulande (q.v.) Clarvent 5055, 7799: this place has not been identified and the name may be imaginary, since in both cases it provides the rhyme needed. However, C 5055 relates to Josue de Clarvent, who is named as one of two leaders of the seventh division, composed of Auvergnats. Clarvent may then be a corruption of Clermont the ancient capital of the Auvergne (later Clermont-Ferrand)? It was there, in 1095, that Pope Urban II convoked the council at which the First Crusade was preached, so the name would be well known. The reference to Conubre de Clarvent, C 7799, offers no help with identification, unless the mention of his helmet indicates a metal industry Cliboïns 782; Cliboïs 925: a Saracen baron at Marsile’s court; probably the same person referred to in C 2613 and 2616 (cf. footnote to C 2613) and certainly the baron killed by Oliver, C 2642–2646 Cologne 5079: Cologne, the great city on the Rhine, noted for its steel in mediæval times Comarcis 6591: the domain of Girart de Vienne, possibly equivalent to that of Bovon, count of Vienne, Arles and Provence, who proclaimed himself ‘king of Burgundy’ in 879 (or de Comarcis may simply be a corruption of del comtat cis, ‘of this county’) Constantinoble 553; Costentinoble 2897: Constantinople, founded by the Roman Emperor, Constantine the Great, on the site of ancient Byzantium as his eastern capital. In both cases in the text, it is coupled with Saxony and Italy to show the extent of Charlemagne’s dominions as Holy Roman Emperor, although its inclusion may well reflect its relevance to the Crusades Conubre de Clarvent 7798: it is not clear whether de Clarvent implies that this is a French baron (cf. Clarvent and Josue de Clarvent for possible Auvergnat connection) or whether Conubre should be equated with Cornuble (q.v.), a Saracen baron killed by Roland. Since there is no mention of the latter’s helmet being sliced through (cf. C 2270–2271), it might later have been taken as booty and here used in the arming of Pinabel. Otherwise, this is an unknown knight or baron Cople (Turs de) 4547, 4932, 5658 (cf. Turs de Cople) Cordes 81, 90, 113, 461, 975, 7782: Boissonnade (1923, 128s.) suggests Cortes, a small town near Saragossa; Whitehead (1942, 169) in his edition of O opts for the more obvious Cordoba. Although the latter might seem to lie much too far south, the poet’s geography is not always reliable Cornuble de Mont Nigre 1352; Cornubles 2254; Cornuble 1360, 2270: a Saracen baron, killed by Roland (cf. also Mont Nigre) Corsablis 1265; Corsabrins (de Barbarie) 2143: a Saracen, king of Barbary, at Marsile’s court; presumably one of his Berber allies from North Africa

Proper Names

591

Cortein 7775; Corteïn 7989: the name of Ogier’s sword Costance 6788: the town of Coutances in the Cotentin peninsula in Normandy

D Daire 3559: Darius, king of Persia; either Darius I, whose army was defeated with heavy losses at the battle of Marathon in 490 BC, or more likely Darius III (ruled 335–330 BC), who was defeated by Alexander the Great; cf. also the Old Testament Book of Daniel, chap.xi for the prophecies of Darius’ downfall Danemarche 2848: Denmark Daniel 1600, 5116: one of the 4 great Old Testament prophets; the story of how God saved him in the lions’ den is found in the Book of Daniel, vi.16–23. Popular mediæval theology transformed him into a Christian saint, as found in C 1600 Datan 2126; Dathan 4666: Dathan, one of the sons of Eliab, who with his brother Abiram rejected Moses’ call to serve God in the tabernacle; they were swallowed up by an earthquake; cf. the Old Testament Book of Numbers, xvi.12–33. Cf. also Abiron. The formula ‘Dathan et Abiron’ was much used in mediæval contracts Denise cf. Donis Denois 5688, 6872, 6946, 6962, 7822; Danois 6132: Danish, applied in this text to Ogier li Denois (q.v.), the term li Denois often being used to designate him without mention of his baptismal name Digon 3267: Dijon, the capital of Burgundy Donis (Seint) 708, 3987, 5441, 6594, 6603, 7385, 7441; Denise (Seint) 1350, 2454, 7407; Denis (Seint) 2608, 2955; Dionie (Saint) 4084: Saint Denis, the apostle of the Gauls and the first bishop of Paris (martyred on the hilltop of Montmartre whose name commemorates this event). Considered (i) simply as a saint, 4084, 7441; (ii) as the patron saint of France (= la reigne Seint Donis, 2608, 7385) and (iii) as the patron saint of Charlemagne or as symbolising France in li rois de Seint Donis, 708, 3987, 6594, 6603. The town of St Denis lies to the north of Paris; it is alluded to in C 1350: ou borc de Seint Denise. Its famous abbey church (eleventh–thirteenth century) was built on the site of a monastery, founded by Dagobert I in 626 and reconstructed by Charlemagne’s father, Pépin le Bref, which was the richest and most highly reputed in France (note the reference to its library of MSS, C 2454–2455 and 2955); it contained the tombs of the kings of France until they were desecrated and destroyed in 1793. Thus: a Seint Denise, C 2454 and 7407; a Seint Denis mostier, C 2955. St Denis was also invoked in Charlemagne’s war-cry: Monjoie St Denis! C 5441 (cf. Montjoie) Dorree 5386: an unknown place, unless it is the ancient port of Durrës/Durazzo in Albania, whose name might be familiar as a port of embarcation for the Crusaders or because of the Norman kingdom set up in that country in the twelfth century. If this identification is correct, the Turre of which Carmilleu de Dorree is king could represent the capital, Tirana Dotaire 5759: no king of this name has been identified; it may represent a corruption of Clotaire, the name of four early Frankish kings, and the allusion to wealth perhaps provides evidence of a folk memory of the tribute amassed at Soissons by Clotaire I Duralt 5407: a pagan land, which has not been identified by Langlois (1904, 183) and may well be imaginary. The -alt ending is needed for the rhyme

592

Proper Names

Durance 5094: an unidentified place; it might again refer to Durrës/Durazzo (cf. Dorree above), which was known for its iron. Alternatively, it could simply be used as a common noun to indicate the ‘enduring quality’ of the metal tips to the lances Durendart 1142, 1304, 1441, 2278, 2714, 2769, 2787, 2922, 2935, 3247, 3626, 3687, 3985, 4052, 4057, 4081, 4110, 4142; Durendal 1364, 1449, 1885, 1901, 1917, 4062, 4068, 4122, 6664; Durendor (ending needed for rhyme) 1428; Durenda (ending needed for rhyme) 1491, 1971: the famous sword given by Charlemagne to Roland Durestant 1250: according to Boissonnade (1923, 76s.), a fortress town marking the southern extremity of the Moorish state in Spain. He suggests either Daroca de la Rioja on the route to Compostella or preferably the great Arab fortress at Daruka (Daroca) in the Jalon basin, near the town of Durera, but there is no definite identification. Whitehead (1942, 870) says simply: «a place in Spain, not identified, but obviously marking the Southern limit either of the whole country or of the Christian kingdoms»

E Effragier 4619: a Saracen baron at Baligant’s court, son of Matragiez and brother to Clariel, sent with the latter as messenger from Baligant to Marsile Egricent (porz d’) 2437: the port of Wissant which lies between Calais and Boulogne Elogne 5074: unidentified place, possibly in France, but perhaps more likely to be in Saxony, since Joseran d’Elogne is mentioned in relation to the 9th division of Charlemagne’s army, composed of 300 knights from Sansoigne. This name might simply be esloigne, ‘distance’, ‘far away’, transformed into a place name (the -ogne ending is needed for the rhyme) Enfrus 5643: an unidentified pagan people (Enfruns in O, and the adjective enfrun means ‘greedy’, ‘avaricious’, ‘gloomy’, possibly even ‘grimacing’, en frume, so this may be a coined name) Engelers 1158, 2219; Engeler 2022, 3783; Enzeler 2386, 2627, 2636: one of the Twelve Peers, often referred to as li Gascon or de Bordel or li Gascons de Bordele, i.e. from Bordeaux Enmauriz 209: a French count Equiteigne cf. Aqiteigne Erabe cf. Erable Erable 1454; Erabe 4822, 5313; Arabe 5407: Arabia; the whole region inhabited by the Arab race, the Moslem Middle East Ermines cf. Hermines Esclavés 5294: probably the same race as the Esclavons (q.v.), since this name is found at the end of a line in laisse rhyming in -és, -ers. Langlois (1904, 152) identifies Clavers as a pagan people Esclavonie 4480, 5072: the name frequently used for Yugoslavia in the twelfth–thirteenth centuries, i.e. the Slavonic-speaking East coast of the Adriatic Esclavons 3439; Esclavon 3780, 4645, 6044; Esclavoz 5273: Yugoslavs: despite their conversion to Christianity, Slavs were often equated to Saracens. Boissonnade (1923, 184) points out that the first Crusaders looked on them as rough, wild and untrustworthy Escler 4581: Slavs confused with Saracens or pagans Escoz 5273: Scots

Proper Names

593

Espaigne 2, 50, 229, 266, 629, 1056, 1207, 1208, 1230, 1403, 1894, 1922, 1949, 2028, 2127, 2347, 2471, 2517, 3248, 3283, 3415, 3458, 3720, 3735, 6506; Espagne 56, 71, 163, 1284; Espeigne 442, 598, 667, 724, 860, 906, 1017, 1024, 1288, 1443, 1925, 2050, 2495, 2526, 2663, 2669, 2680, 3027, 3214, 3291, 3581, 4140, 4227, 4468, 4475, 4601, 4669, 4683, 4720, 4734, 4754, 4848, 4986, 5990, 6513, 6521, 6700, 6725, 6927; Espegne 856, 917, 1249, 1395, 1522, 3552a, 4131, 4516, 5748, 5823, 6504, 6533, 6651, 6714, 7351, 7442, 7904: Spain Espanie 6622: an unidentified place, found only in relation to Gerart d’Espanie (q.v.), who is not mentioned anywhere else in C unless he should be identified with the Gerart de Blois (C 6170) sent by Charlemagne as one of the messengers to Girart de Vienne (Espanie occurs as the last word of the line, so the ending may be suspect) Espiriz (Sainz) 6063: the Holy Spirit or Holy Ghost Estanpes 64; Estampes 6454: Etampes in the Seine-et-Oise; an ancient town, mentioned by Grégoire de Tours in relation to the Merovingian period, visited by Pope Calixte II in 1119; the castle was a royal residence from the tenth to the late twelfth century (cf. Aton d’Estampes) Estomarin 77; Estramant 1318; Estormiz 2244: one of the Twelve Saracen Peers, companion-at-arms to Estorgant (cf. C 1317, 1318). He was killed by Berenger, C 2244–2247, and O 1304 names him in this context as Astramariz, so Estormiz is probably a corruption of Estromiz or Estramariz. There seems to be considerable scribal confusion in C over Saracen barons whose names begin with Esto- or Estra-; they may be four rather than three in number if Estomarin should be differentiated from Estramant Estorgant 1317, 2234: Saracen Peer, companion-at-arms to Estomarin; killed by Otes Estormis de Valterne 1309; Estormi 2222: a different Saracen baron, one of the Twelve Saracen Peers. He is killed by Engelers de Bordel (C 2222–2225) and O 1291 reads Escremiz de Valterne in that context, distinguishing him from the Estormiz of C 2244 (cf. Estomarin above; cf. also Vauterne) Estormiz cf. Estomarin Estoz de Lengres 1159; Estouz de Lengres 1687, 2011, 2166; Estouz 1697, 2766; Estolz de Lengres 1723: a French baron, one of the Twelve Peers, son of Odon de Lengres (q.v.); killed at Roncevaux by Grandoine (cf. also Lengres) Estramant cf. Estomarin Etiopés 3298: apparently Ethiopia, but the name was generally used in the Middle Ages for ‘black’ Africa (cf. C 3299). As Ethiopia itself was Christian from the earliest times, following the rites of the Coptic Church, it would be an error to describe it as a ‘cursed land’ Eve 5965: the first woman, created as companion to Adam; cf. Old Testament, Genesis i.27 and ii.21–23 Evraus 5918: probably Evreux in Normandy, the original Ebrocas becoming Everous, then Evereus (in the Norman graphy) from about 1200

F Fabrin d’Espaigne 3248: a Saracen baron killed by Roland Falsagon cf. Fausiron Fauseron 2297: a Saracen baron killed by Oliver, C 2293-2300; probably identical with Fausiron, but he is killed by Oliver, C 2138. V7 2249 has the same reading, indicating an error in the CV7 shared source; O 1353 reads Malun in this context which may represent the original reading or be O’s correction of an earlier error

594

Proper Names

Fausiron 779; Falsagon 1259; Fauseron 2124, (2297?): a Saracen duke, brother to Marsile, lord of Limois (C 779) and also of Dathan and Abiron (q.v.) (C 2126). The Fauseron killed by Oliver, C 2138, is specifically identified as Marsile’s brother, C 2125; C 2293– 2300 apparently repeat this incident, but without the attendant details provided earlier (cf. note to Fauseron above) Fauvel 6415: adj.used as Proper Name brownish-yellow, russet – thus, as horse’s name: Russet Faviens 518: the uncle of Ganelon Fergalem de Mont Nu 4825: a Saracen baron in Baligant’s army Feron (Seint) 4159: in the Légendes Épiques, Bédier (1921, vol. 4, 290), identifies this saint as Saint Faron, who founded an abbey at Meaux. Faron was a high official at the court of Dagobert I and a personal friend of the king who took religious orders later in life. Nothing is known to justify describing him as ‘the tyrant’ or ‘the obstinate’ (as here in C), but there may be confusion with the title ‘Pharaoh’ (Fr. ‘Pharaon’) or a deliberate play on words (cf. footnote to C 4159) Fesrant 7588; Ferrant 7778; Ferant 7873, 7874; Feranz 7919: the name of Thierry’s horse, identified in C 7778 as belonging originally to Duke Reiner. The name is based on the adjective ‘grey’ and the noun meaning ‘a grey horse’ Flandre 5061; Flandres 7413: Flanders Floire 5255: a Saracen king Floires 6601: a vassal of Girart de Vienne; the son of Paris Florence cf. Sorence and Pinabel de Sorence/Pinabel de Florence Florent 6900, 6935: an unknown (Saracen?) king of Valdormant and/or Valserie Florent 7791: a burgher of Laon Fochiers de Vilance 5791: a French baron Foqe del Mans 6454: a French baron Franc 111, 285, 730, 948, 974, 1209, 1224, 1251, 1419, 1623, 1921, 2087, 2193, 2395, 2469, 2472, 2527, 2581, 2609, 2810, 2887, 2905, 3028, 3306, 3319, 3561, 4246, 4283, 4287, 4325, 4329, 5478, 5685, 5795, 5806, 5812; Frans 1655, 2541, 2674, 3237, 5635, 5657; Franz 3444: Franks, strictly speaking, but used in C as a synonym for the French, showing how the idea of national identity had evolved in the two centuries since O was copied. (Note: Franc de France, 2527, 2609, 2887) Of course, the term ‘Franks’ may be residual evidence of the original composition of Charlemagne’s army, still retained in the later remaniement France 47, 108, 125, 132, 159, 180, 199, 204, 215, 225, 419, 481, 508, 525, 690, 746, 813, 850, 868, 877, 1014, 1016, 1027, 1052, 1193, 1211, 1218, 1237, 1285, 1293, 1349, 1427, 1445, 1452, 1462, 1560, 1564, 1619, 1638, 1737, 1746, 1748, 1892, 1904, 1907, 1912, 1934, 2062, 2120, 2147, 2331, 2432, 2447, 2483, 2527, 2609, 2656, 2703, 2887, 2893, 2901, 2945, 2970, 3009, 3013, 3051, 3218, 3227, 3239, 3308, 3330, 3360, 3390, 3515, 3516, 3549, 3574, 3598, 3602, 3651, 3656, 3695, 3803, 3933, 3939, 3974, 3986, 4074, 4088, 4092, 4096, 4143, 4212, 4223, 4235, 4314, 4317, 4337, 4347, 4351, 4416, 4599, 4606, 4329, 4636, 4710, 4730, 4761, 4784, 4837, 4862, 4875, 4891, 4897, 4090, 4042, 4962, 4988, 5042, 5088, 5122, 5135, 5200, 5283, 5373, 5390, 5434, 5453, 5457, 5496, 5537, 5617, 5657, 5680, 5682, 5711, 5728, 5777, 5798, 5871, 5872, 6140, 6245, 6369, 6407a, 6494, 6699, 6787, 6794, 6800, 6866, 6910, 7116, 7212, 7317, 7325, 7328, 7347, 7424, 7488, 7676, 7691, 7699, 8005, 8010, 8038; Franse 651; Franze 2069; Franche 3211, 5168, 5362, 7612: France François 61, 333, 336, 375, 398, 1030, 1065, 1293, 1410, 1417, 1423, 1442, 1497, 1507, 1510, 1539, 1550, 1621, 1640, 1830, 1965, 1986, 1990, 2003, 2108, 2121, 2130, 2161, 2164, 2191,

Proper Names

595

2461, 2659, 2661, 2797, 3007, 3107, 3116, 3179, 3193, 3229, 3321, 3728, 4199, 4208, 4268b, 4304, 4623, 4937, 4944, 4946, 4979, 5469, 5989; Franzois 309, 326, 496, 564, 593, 702, 783, 842, 873, 1023, 1119, 1151, 1199, 1281, 1306, 1365, 1484, 1977, 1999, 2008, 2064, 2094, 2142, 2197, 2354, 2376, 2422, 2425, 2480, 2484, 2543, 2553, 2560, 2574, 2588, 2633, 2709, 2723, 2734, 2767, 2792, 2801, 2868, 2885, 2960, 3111, 3131, 3412, 3551, 3630, 3890, 4298, 4372, 4414, 4465, 4713, 4735, 4756, 4802, 4806, 4812, 4861, 4959, 5154, 5353, 5371, 5414, 5442, 5487, 5491, 5493, 5576, 5590, 5593, 5655, 5667, 5794, 5833, 5858, 5863, 6236, 6255, 6262, 6265, 6303, 6567, 7487, 7533, 7679, 7695; Francis 1380; Franceis 5601: French; a Frenchman, the French Fremin (Saint) 6074: most probably St Firminus I (variant graphies Firmin and Fermin), bishop of Amiens, who was martyred there c.290 AD. Born in Pampluna, where he and his parents became Christian converts, he travelled across France as an evangelist before settling in Amiens, from where some of his relics were transported to Pampluna in 1186. His statue forms the trumeau of the left portal of Amiens cathedral and his life is recounted on the marble friezes around the choir (further details are given in the footnote to this line) Frise 5061, 7413, 7490, 7550: Friesland (cf. Guandeboes/Gondebuef de Frise) Frison 7522, 7768: inhabitant of Friesland (cf. Guandeboes/Gondeboef le Frison) Frondaus 5892: an unidentified place (the Cons de Frondaus is Oliver’s uncle)

G Gabrials (Saint) 4393; Gabriel (Sains) 5770: the Archangel Gabriel, the Angel of the Annunciation (cf. Luke i.26–38). In O, he has a role in Roland’s death scene (O 2389–2396), but in C he is mentioned solely in relation to Charlemagne, as the emperor’s guardian angel and special messenger from God Gaifer de Lonbardie 7639: one of Charlemagne’s barons, from the Italian side of the Alps Gainelon subj. 12, 1996; Gainellon subj. 259; obj. 3133, 3196, 4205; Guenes subj. 213, 275, 352, 365, 376, 388, 417, 427, 456, 470, 482, 494, 498, 502, 506, 512, 520, 544, 555, 558, 592, 617, 619, 640, 682, 686, 756, 800, 815, 843, 867, 935, 941, 951, 972, 985, 1046, 1081, 1114, 1195, 1226, 1398, 1553, 2109, 2169, 2502, 3083, 3854, 4238, 5456, 6020, 6244, 6246, 6259, 6339, 6387, 6389, 6404, 6713, 6792, 6953, 6968, 7466, 7473, 7483, 7560, 7578, 7584, 7592, 7617, 7620, 8005, 8023; obj. 927; Guene subj. 335, 340, 438, 469, 568, 704, 714, 759, 791, 798, 804, 829, 839, 886, 891, 903, 992, 1076, 1092, 1837, 2313, 2350, 2407, 2414, 2685, 4188, 4872, 5929, 6413, 6430, 6477, 7471, 7545, 7782, 8046; obj. 1083, 6024, 6440, 7330, 8166, 8172; Guenelon subj. 462, 538, 633, 7019, 7489; obj. 2617, 4349, 6211, 6305, 6322, 6330, 7345, 7366, 7376, 7396, 7419, 7447, 7452, 7498, 7581, 7602, 7643, 7647, 7652, 8040, 8156, 8163; Guenelons subj. 476, 625, 6181, 7852, 7863; Guenellons subj. 654, 718, 6421; Guenellon subj. 693, 6268; obj. 332, 358, 603, 787, 938, 953, 981, 1517, 1614, 3139, 3149, 3521, 3599, 5945, 6036, 6449, 6669, 7319, 7410, 8191; Gaynellon obj. 1101; Gaynelon obj. 2554; Guenon obj. 1218, 7389, 7608, 7661: Ganelon, the treacherous baron who is brother-in-law to Charlemagne and step-father to Roland. Whitehead (1942, 170) cites Wanilo, archbishop of Sens in the reign of Charles the Bald (Charles II of France, 823–877), as the generally accepted historical prototype of this character, but Rejhon (2005, 26) draws attention to evidence of another Wanilo, bishop of Laon and contemporaneous with Charlemagne, as cited by Martinet (1978, 41s.; 1994, 148s.)

596

Proper Names

Galafre (l’amanzor) 2862: the emir in question may be the Emir of Aleppo; cf. Boissonnade (1923, 205) Galant 5524: here called le roi Galant, but this seems to be Wayland, the legendary blacksmith (Völand or Wieland in Norse legend). Cf. Rajna (1956, 445) and Romania 29 (1900), 259–262 Galie 5068: Galicia, the region in the extreme north-western corner of Spain, a Christian kingdom by the time of Charlemagne Galilee 7176: Galilee Gandebuef, Gandebues cf. Guandeboes Garamon 2752: the name of Grandoine’s horse Garillant 4165, 4913: ‘the Marshland’, the common noun garillant used as a place-name to designate the extensive marshes in the southern part of (modern) Iraq, around and between the rivers Tigris and Euphrates and to the north of the port of Orcaise (q.v.) near (modern) Basrah on the Persian Gulf, but south of ancient Niniveh, C 4165; forming the eastern boundary of a huge stretch of Arab territory Garin d’Alvergne 6204: one of Charlemagne’s barons, sent as one of the messengers to his sister, Berte Garin de Mont Arsor 6247: another French baron, clearly not the same person as Garin d’Alvergne, since the latter is already on his way to Berte in Mâcon Garins de Mont Agu subj. 7587: a French baron, possibly identical with Garin d’Alvergne if one reads Auvergne as the region and Mont Agu as a particular place, but probably simply another nobleman, his purely imaginary name being coined to provide the rhyme Garnier cf. Guarner Gascogne 211, 6481; Guascogne 122; Gascoigne 1202, 2628, 2637, 2760, 7590: Gascony Gascon 2022, 2636, 3783; Gascons subj. 1158, 2219: Gascon Gautiers subj. 1163, 1174, 1185; Gauter obj. 1171, 2344, 3475; subj. 2348, 2357, 3454; Gauters subj. 2361, 2368, 2374, 3421, 3430, 3435, 3459; Gauter de Luz obj. 3413, 3764; Gauters de Lus subj. 3468: Gautier, Roland’s vassal Gelers cf. Gerers Gelins cf. Gerins Gelmafi 5612: a Saracen noble in Baligant’s army Gerart1 1803, 1820, 1832; Girart da Rosion 3784; Gerart de Rosion 3271; Gerart de Rosellon 7767: one of the Twelve Peers, comrade-at-arms to Sanson; killed by Marsile. Count Gérart de Rousillon was an historical personnage, Count of Paris, Marquis of the duchy of Lyon, regent of the kingdom of Provence at the time of Charles le Chauve (ninth century), founder of the abbey at Vezelay; in fact identical to Girart de Vienne historically (cf. Bédier 1914–1917, vol. 2, 1–92; Boissonnade 1923, 357–359) Gerart2 3517, 6190, 6472, 6505, 6517, 6520, 6535, 6536, 6577, 6623, 6632, 6678, 6750, 6785, 6797, 6829, 6880, 6883, 6947, 6950, 6979, 6990, 6992, 7282, 7320, 7334, 7636, 7709, 7820; Girart de Viene 4144; Girart 6176, 6499, 6572, 6588, 6621, 6628, 6753, 6819, 6822, 6838, 6877, 6883a, 6954, 7015, 7048, 7211, 7236, 7257, 7307; Girarz 6558, 8072; Gerarz 7205; Gerart de Viene 7297: an important French baron, son of Garin de Monglane, uncle of Oliver and Aude. He was an historical personnage, Count of Arles and Provence (cf. Gerart1 above and Bédier 1917, vol. 2, 1–92; 1921, vol. 4, 184). The thirteenthcentury chanson de geste, Girart de Viane, forms part of the Charlemagne cycle, although it is probably a ‘prequel’ to the main elements of this

Proper Names

597

Gerart de Blois 6170: a French baron sent by Charlemagne as a messenger to Gerart de Vienne Gerart de Monflor 8035: a French baron, possibly to be identified with Gerart le Normant Gerart de Seint Omer 7362: one of Charlemagne’s barons Gerart d’Espanie 6622: a French baron, either a vassal of Girart de Vienne or perhaps another name for Gerart de Blois Gerart le Normant 7340: a French baron in Charlemagne’s army, not otherwise identified unless this is Gerart de Monflor; cf. Monflor for discussion of this question Geremie 6552: the Old Testament prophet Jeremiah; according to the legend of the ‘Harrowing of Hell’, he was one of ‘the Just’ freed from Hell by Christ in the interval between the Crucifixion and the Resurrection Gerers subj. 1155, 1765, 1774, 2378; Gelers subj. 2181, 2190; Gerer obj. 2759; subj. 3762: Gerer, one of the Twelve Peers, the companion-at-arms of Gerin Gerins subj. 1155, 1733, 1749, 2377; Gelins subj. 2198; Gerin obj. 2754; subj. 3762: one of the Twelve Peers, the companion of Gerer Geronde 6522: the Gironde estuary Gibon 4647: an unidentified Saracen king whose palace was clearly at Saragossa. He may be purely imaginary, but the statement that he is ‘of the lineage of Buiron’ may indicate a reference to some popular contemporary story Gifeüs 780; Virfalle 4667; Girfalli 4791: the son of Marsile Girart cf. Gerart2 Girart da Rosion v. Gerart1 Girart de Terascoigne 5084: an otherwise unknown baron in Charlemagne’s army; he may be from Tarascon-sur-Ariège near Foix, but Tarascon-sur-Rhône (Bouches du Rhône) is a much more ancient and important city, well known in mediæval times Girart d’Orion 6205: another French baron, sent by Charlemagne to summon his sister, Berte, the mother of Roland Girfalli cf. Gifeus Gloz 5278: normally used as an adjective meaning ‘miscreant’ or ‘villain’ (cf. C 561, 924, 1079, 7856), but here treated as the name of a pagan race or country Gondeboef, Gondebuef, etc. cf. Guandeboes Goz 5274: a pagan country (Saracins de Goz), but the name may arise from confusion with the Goths, who extended their domains from around the mouth of the Visula in Poland right across the south-east of Europe (the Ostrogoths into what is now Bulgaria, the Visigoths into Spain and Italy). Langlois (1904, 296) simply equates Goz to Goths Grandoine 2545, 2548, 2749, 2777: an important Saracen, king of Cappadocia, who leads half of Marsile’s army; the son of Capoé Gualane 76: a Saracen noble, vassal to Marsile and one of the latter’s envoys to Charlemagne Guandeboes subj. 7460; Gandebues subj. 7477; Gandebuef de Frise obj. 7490; Gondeboef le Frison subj. 7522, 7768; Gondeboef de Frise subj. 7550; Gondebuef obj. 7538; subj. 7582; Gondeboef subj. 7553, 7556, 7562, 7576: the king of Friesland, a vassal of Charlemagne Guarner 122; Garnier 7638: a French baron Guenellon, Guenelon, Guenelons, etc. cf. Gainelon Guenes, Guene, Guenon cf. Gainelon Gui d’Aiguent 7796: the nephew of Pinabel de Sorence. The place-name has not been identified: possibly Aignan, the ancient capital of Armagnac. Or this should perhaps be

598

Proper Names

located in Eastern France, perhaps Franche-Comté, to coincide with the region from which Ganelon’s relatives come. It may well be fictional Gui d’Aminoi 7568: a French baron from the region around Amiens Gui de Gascogne 122, 211; Guion obj. 2760: one of the Twelve French Peers, killed by Grandoine Gui de Montamis 6589: a vassal of Girart de Vienne. The place-name has not been identified Gui de Nevers 6205, 6959: a French baron, sent to summon Berthe, Charlemagne’s sister; Nevers was an important city, the capital of the Nivernais, governed by a count in the eleventh–fifteenth centuries, later a dukedom Gui de Seint Omer 6170: a French baron, sent by Charlemagne as messenger to Girart de Vienne. St Omer, founded as an abbey in the seventh century by the saint of that name, was a thriving city in the Pas de Calais by the thirteenth century and an administrative centre Guibelin 5584: a French baron killed by Balligan Guibor 6542, 6553, 6574; Giborc 6540a; Gibor 6545, 7238; Gibors 6565: the wife of Girart de Vienne Guinemanz obj. 4963, 5417; subj. 5443; Guinemant1 obj. 5580: a French count, one of Charlemagne’s barons, killed by Balligan Guinemant2 6628: a French count who accompanies Girart de Vienne and Aude; the name could well arise from confusion with either Gui de Seint Omer, Charlemagne’s messenger to Girart, or Gui de Montamis, Girart’s own vassal (cf. Guinemer2 below) Guinemer1 (el havre) 4578: this place-name is not known, but the harbour is clearly situated close to Saragossa Guinemer2 7835: a French count, from the context clearly of some importance, but not mentioned elsewhere unless the name Guinemer (providing the rhyme at the end of the line) is to be read as a variant of Guinemant2 above (which also ends the line), and also possibly as one of the French barons named Gui (perhaps Gui d’Omer for Gui de Seint Omer – or Gui de Nevers)

H Haiz 4351 cf. Ais Hauteclere cf. Autaclere Henoier 2904: inhabitants of Hainaut (in modern Belgium) Hermines 3439; Ermines 5275: strictly speaking, Armenians, but a term used loosely for Saracens Hernaut de Baulande 6580: a son of Garin de Monglane and brother to Girart de Vienne Hervié, Herver, Hervier cf. Hués de Lion Hongrie cf. Ongrie Hué del Mans 7568, 7572: a French nobleman Hues de Lion 7650; Hervié 7660; Herver 7802, 7825; Hervier 8008: a French noble, lord of Lyon; a relation of Ganelon and cousin to Pinabel. Lyon, initially a tribal capital of the Gauls, then a great Roman city, continued to flourish in the Middle Ages, so the emphasis is on the power of Ganelon’s relations Hugon 3266: one of the Twelve Peers of France, killed at Roncevaux. He is apparently not mentioned earlier in the poem, but Langlois cites Huon and Hugon as scribal errors for

Proper Names

599

Oton cf. Otes1, who is mentioned six times, but whose death is not otherwise recorded. (The Oxford MS, O 1891–1892, names him ‘Bevon’.) He is Duke of Beaune and Dijon, i.e. Duke of Burgundy Hugrent 5308: Hungarians, used as a general term for pagans. Boissonnade (1923, 181s.) draws attention to Magyar attacks on ill-disciplined, pillaging Crusader bands Hunez subj. 1157: one of the Twelve Peers; listed here apparently with Engelers, who is normally cited as the companion-at-arms of Anseïs (q.v.), so this may be a variant graphy Huns 5308: a pagan people; according to Boissonnade (1923, 182) the Huns, whose reputation as pagan barbarians was preserved in legends and in folk memory (cf. Goz preserving the memory of the Goths) Hurepois 110, 5488, 5696: according to the Chanson des Saisnes, as cited by Brasseur (1990, 214s.), a huge area stretching from Mont St Michel to Château-Landon (due south of Paris, half-way between Nemours and Montargis), probably a regional name for the ancient kingdom of Neustria in C 110; the inhabitants of this region in C 5488, 5696

I Ificoz 5276: the city of Jericho in the Holy Land, today in Jordan Inde 601, 4912: India Irois 4861: Irish: probably an allusion to the Danish invaders who established colonies along the Irish coast, seen as pagan enemies of Charlemagne, therefore a Saracen race (cf. Bédier 1921, vol. 4, 44) Ivoire 3270, 3763, 3781, 5220: one of the Twelve Peers, the companion-at-arms of Ivon Ivon obj. 3270, 3781, 5220; Ive subj. 3763: One of the Twelve Peers, the companion-at-arms of Ivoire

J Jafer de Valdormanz 5632: a Saracen baron Jahanz 3439; Jaiant (de Valproissié) 5307; Jaianz 5643: a pagan race of giants (the adjective jaiant being used as a Proper Name) Jarrois 5694: an unidentified locality, possibly simply a graphy of jarrie (sf) ‘heath’, ‘moor’ or ‘wasteland’, modified by poetic licence; carbon de jarrie would then signify ‘charcoal’. Godefroy also cites Jarrige as a Proper Name. In the V7 ms., the upper-case I could be read as lower-case l or even upper-case L with a vestigial foot, giving Larrois or Larris, also meaning ‘heath, waste-land’, and Godefroy cites an rocky area of this type west of Châtillon-sur-Seine, presumably called Larris and indicates that the word larris was retained very late in the vocabulary of Burgundy, Picardy and the Beauvaisis. The phrase would then mean ‘a charcoal-burner from either Jarrige or Larris’ Jaünois (graphy of Johannois, Johenois?) 5502: a Saracen race Jerusalem 912: Jerusalem Jessemanie 7172: Gethsemane, where Christ prayed immediately before he was betrayed by Judas Iscariot and arrested by the Jewish High Priest (Matthew xiv.32s.) Jeus 4584: the Jews Jesus, Jhesu, etc. cf. Yhesu

600

Proper Names

Jocaïnne 5055: one of Charlemagne’s barons Jofroi d’Anjou 123, 5690, 7974; Jofroiz d’Anjou 209, 6568, 6832; Jufroiz l’Amoros 5880; Jofroi d’Angou 5671; Joifroi d’Anjou 6171, 7641; Joifroi 7757: Geoffrey of Anjou, Charlemagne’s standard-bearer; presumably an ancestor of the Plantagenet nobleman of the same name who was the second husband of Matilda, daughter of Henry I of England, and the father of Henry II of England – although folk-memories tend to confuse famous people from different periods and the sobriquet would be apt for the later nobleman (cf. Braüz l’Amorois) Joiose 4363, 4371, 4931, 5551, 5785: the sword of Charlemagne, the name meaning ‘joyful’, ‘happy’ or simply ‘beautiful’; but this may be a corruption of Ysares, the name of the woman smith who made it, according to the story in the ‘Floovant’ fragments Jonas (Seint) 1601, 4163, 5108: the Old Testament prophet whose story is told in the Book of Jonas. The appellation ‘Saint’ is of course erroneous, but typical of mediæval popular theology Joseph (Sainz) 7145: Joseph of Arimathea, who obtained Jesus’ body from the Romans and put it in the tomb he had had constructed for himself (Matthew xxvii.57–60; Mark xv.43–46; Luke xxiii.50–53; John xix.38–42) Joserant 5016, 5986; Joseran d’Elogne 5074; Joseran 5671: one of Charlemagne’s barons Josue de Clarvent 4575: one of Charlemagne’s barons Josuer 4575: the king of Tortelose, perhaps Tortosa at the mouth of the River Ebro, in Spain; if this is a reference to another story familiar to contemporaries, the allusion is no longer recognized today Judas 5267, 7170, 8061, (used as a common noun, meaning ‘traitor’, C 2842): Judas Iscariot, the disciple who betrayed Christ (Matthew xxvi.14–16 and 47–49; Mark xiv.10–11 and 43–45; Luke xxii.3–5 and 47–48; John xviii.2–5; cf. also Boteroz) Judeus de la Loi 8062: the Sanhedrin, the Supreme Council and Court of Justice in Jerusalem Jufroiz l’Amoros cf. Jofroi d’Anjou Jule Cesar cf. Cesaire Jupin, Jupins cf. Jupitel Jupitel subj. 2390; Jupin obj. 4659, 5862; Jupins subj. 5629: Jupiter, chief of the Roman gods, looked upon as one of the gods forming (with the prophet Mahomet and the Greek god Apollo) a Moslem ‘trinity’ to parallel the Christian Trinity (Father, Son and Holy Spirit) Justin de Valfondee 2324: a Saracen baron killed by Oliver

K Kent 5309: the English county seen by the writer as a pagan land at the time of Charlemagne, perhaps due to the influence of Geoffrey of Monmouth and the story of the Saxon conquest of Kent involving the giants Hengist and Horsa

L Lazaron 488; Lazeron 1602; Saint Lazaron 3786, 4169; Lazarons (s added for rhyme) 5039; Lazare 7166: Lazarus, who was raised from the dead by Christ (John xi.1–46). Leg-

Proper Names

601

end has it that he travelled to France with Mary Jacobe, Mary Magdalene, Martha and others, landing at Les Saintes-Maries-de-la-Mer and spreading the Gospel northwards. His (supposed) skeleton may still be seen in the magnificent cathedral of St Lazare in Autun, which was built to house his relics; but these were brought to Marseille by Gérard de Roussillon in 1079 Le Mans 6454, 7568, 7572: the chief town of Maine, now the départements of Sarthe and Mayenne, in southern Normandy (cf. Foqe del Mans, Hué del Mans) Lengres 1159, 1687, 1723, 2011, 2166, 7638: Langres in the Haute-Marne, on the plateau de Langres in southern Champagne (cf. Estoz de Lengres, Odon de Lengres) Lerie 5248; Leurie 5443: since Capamor, the emir of Lerie, is part of Baligan’s army and coupled with the king of Persia, this is unlikely to be Lerida in Spain, which Langlois (1904, 394) suggests. In the same contexts, O 3205 mentions Dapamort, un altre rei leutiz and O 3360 un rei leutice. Whitehead (1942, 171) offers two alternative meanings for this adjective: (a) pertaining to the Wiltzes, a Slavonic people settled in what is now Mecklenburg or (b) Lycian, citing Bédier (1922, 515) and Constans (Roman de Troie, 1904–1912, vol. V, 62). As Whitehead points out, the latter alternative is better suited to the context. Lycia was an small country on the south-west coast of Asia Minor given a significant importance by Homer in the Iliad; today it is part of southern Turkey Liganors 2303: a Saracen nobleman killed by Oliver Limois 779: not Limoges, because Fausiron de Limois is a Saracen; he is also referred to as lord of Dathan et Abiron (q.v.), which implies that his dominions lay in the Holy Land Limojes 6308: Limoges (Haute-Vienne) Lions 6589; Lion 7650, 7765: Lyon, the important city at the confluence of the Rhône and Saône; a tribal capital of the Gauls, a great Roman city situated on a major route from Italy through France to Boulogne-sur-Mer, it continued to flourish throughout the Middle Ages (cf. also Bos de Lions, Hues de Lion) Loënel 78: a Saracen baron, counsellor to Marsile Loereigne 7834: Lorraine Loereng 4329: inhabitants of Lorraine Lonbardie 7639: Lombardy in North Italy (cf. Gaifer de Lonbardie) Longis 3981, 7139: the blind Roman centurion who, according to legend, pierced Jesus’ side as he hung on the Cross and thus obtained pardon and sight. This story embroiders (allegorically) on John xix.34, which reports that, after Jesus had died, ‘one of the (Roman) soldiers with a spear pierced his side’, and on Matthew xxvii.54, Mark xv.29 and Luke xxiii.47, which all record that the centurion recognized Jesus as ‘a good man and a just’, ‘the Son of God’ Luz 3413, 3764; Lus 3468: an unidentified place in France, but there is the small town of Luz (with ruined castle) in the Central Pyrenees, not so very far east of the pass through to Roncevaux, and on the Atlantic coast, close to the Spanish border, there is St-Jean de Luz, so Gauter may well be a French baron from that region (cf. Gauter de Luz)

M Mahomerie 5071, 5660: the Moslem lands, the Islamic empire Mahomet obj. 9, 564, 611, 1234, 2831, 4485, 4510, 4935, 5282, 5354, 5363, 5862; subj. 1248, 2108, 5166, 5253; Mahon obj. 645, 898, 2697, 4042, 4292, 4297, 4467, 4658, 5102, 5328,

602

Proper Names

5577, 5408, 7506; subj. 3281, 3644, 4445, 4743; Mahons subj. 1300, 4504, 5240, 5464, 5627; Mahomez subj. 4702: Mahomet, the Prophet of Islam, considered by popular mediæval theology to be one of the gods of the Moslem ‘trinity’ (cf. Apolin, Jupitel) Mahon v. Mahomet Maiance 5092: Mainz in modern Germany (French: Mayence), on the River Rhine Malduz 2725: a Saracen king, the father of Malquidant; the latter is referred to as ‘an African from Africa’, which might allude to the former Roman province in North Africa (modern Tunisia and northern Libya) but is more likely to be ‘black Africa’ to the south, the usual mediæval significance of this term Maleüz 3421: a Saracen or a Breton vanquished earlier by Gautier de Luz or a place conquered by him Cf. Romania 12 (1883), 114 for a discussion of Breton names based on mael-maglo by Gaston Paris). Boissonnade, (1923, 124s.) prefers to see O’s Maelgut as a corruption of Monteagudo, near Tudela in Spain Mallo (Seint) de Bretaigne 2348: St Malo, the sixth-century Celtic evangelist who established a Christian bishopric in the area around what is now the port of St Malo (Ille-etVilaine) in Brittany as a result of the Celtic invasions Malprimes subj. 5211, 5222, 5236, 5240, 5459, 5503, 5627; Marprimes subj. 5233; Malprime obj. 5615: the son of Baligant, killed by Naymes Malprimes de Brigart 1269: a Saracen baron, named as Malprimis de Brigant in O 889 and as Malprimis de Brigal in O 1261, although C 2184 (echoing O 1261 in an expansion of O’s laisse) refers to Malpui de Mont Pregal Malpui de Mont Pregal 2184: a Saracen baron, probably the same person as Malprimes de Brigart (see note above). The scribe may perhaps have inserted Mont to obtain an extra syllable, after shortening Malprimes to Malpui in error Malqidanz 2725: a Saracen prince, probably from ‘black Africa’ (cf. Malduz) Maltalant 7542, 7582; Mautalant 7551: ‘Fury’, the name of Gondeboeuf de Frise’s horse (the common noun transformed into a name) Mansel 6068: the inhabitants of the ancient French province of Maine, equivalent to the modern départements of Sarthe and Mayenne Marbrie 4574: a place in (or near) Spain, perhaps Manresa in Catalonia; but since Marbrie is closely coupled with Marbroie, perhaps more likely to be either Majorca or Minorca. These islands lie due east from the mouth of the R. Ebro (cf. Boissonnade 1923, 73) Marbroie 4574: possibly an unidentified place in Spain, but more likely to be either Minorca or Majorca (cf. note to Marbrie above and Boissonnade 1923, 73) Marche 8036: an ancient province of France, comprising the modern département of Creuse, plus parts of Haute-Vienne, Indre and Charente; here, clearly a county Margarie de Sebie subj. 1332; Margariz subj. 2254, 2519; Margarit obj. 2255; Marpris subj. 2336; Margari obj. 2511: a Saracen, king of Seville (the origin of his name being very probably margariz, meaning ‘convert from Christianity to paganism’ (cf. Romania 14 (1885), 416–420, and 27 (1898), 8; cf. also Marpris below) Marie obj. 236, 3003, 3067, 3696, 5065b, 6012, 7165; Marie (Seinte/Sainte) subj. 366, 1835, 2122, 2842, 3682, 3801, 4137, 4957, 6314; obj. 2830, 5865, 8020, 8028; Marïen obj. 1447, 1899; la vierge Marie 5067: the Virgin Mary, mother of Christ Marie 6057: Mary Magdalene, variously identified as the woman who poured precious oil over Christ’s head or feet, as the woman taken in adultery and as Mary the sister of Lazarus (cf. Maudelene) Maries (les Trois) 3883, 7149: Mary Magdalene, Mary the mother of James and (Mary) Salome, who are recorded in the Gospels as going to the Tomb on the first Easter Sunday in order to anoint Christ’s body with ointments and finding it empty (Mark xvi.1, 9)

Proper Names

603

Marmoise 5311: a pagan country, perhaps Marasch, a strong-hold of the Crusaders (cf. Boissonnade 1923, 216) Marmorin 6073: probably the Sea of Marmora, between Europe and Asia, connected with the Aegean Sea by the strait of the Dardanelles and with the Black Sea by the Bosphorus. This might well represent (to the mediæval mind, or at least to popular thought in Western Europe) the farthest extent of the known world eastwards Marmoris 2699: the name of Valdebron’s horse, probably meaning ‘Dapple’, rather than ‘Marble’ or ‘marbled’. The Island of Marmora, famous as a source of marble, lies in the Sea of Marmora, cf. Marmorin above Marprimes v. Malprimes Marprinant de Mer 78: a Saracen baron, one of Marsile’s envoys to Charlemagne Marpris 2336: a Saracen; certainly this line is intended to refer to Margariz de Sebie, who has not been heard of since C 2263. This identification is confirmed by C 2511 Marsille subj. 8, 14, 74, 80, 88, 218, 237, 242, 250, 478, 653, 677, 791, 804, 839, 863, 884, 895, 1230, 1253, 1286, 1320, 2001, 2570, 2824, 3010, 3166, 3288, 3295, 3566, 3576, 3641, 3719, 3722, 3945, 4442, 4449, 4493, 4514, 4518, 4558, 4690, 4716, 4744, 4760, 4781, 4807, 4844, 5827, 5861, 6508; obj. 37, 318, 328, 407, 783, 788, 1240, 1254, 1520, 1644, 1763, 2100, 2104, 2125, 2495, 2518, 2524, 2809, 2823, 2837, 2891, 4066, 4561, 4695, 4787, 5822, 6383, 6478, 6714, 7734, 7853, 8053; Marsilion subj. 148, 612, 2547, 2799; obj. 263, 288, 440, 643, 656, 760, 907, 1102, 1260, 1547, 1592, 2683, 3280, 4622, 4662, 4890, 7020; Marsilions subj. 160, 171, 181, 639, 685, 697, 771, 952, 1013, 1654, 1779, 1807, 2340, 2534, 2647, 2662, 2668, 2673, 2994, 4496, 4724; obj. 636, 2337; Marsilles subj. 277, 739, 1002; Marsile obj. 3264a, 6671; subj. 4476; Marsiles subj. 3546; Marsire subj. 6493; indir obj. 6793; ind.obj. of possession 7481: Marsile, the Saracen whom the Chanson de Roland identifies as king of the whole of Spain Marsire v. Marsille Martin (Saint) 6078: St Martin, the fourth-century Roman soldier who became a disciple of St Hilaire of Poitiers and eventually Bishop of Tours. The most famous story relates how he cut his cloak in half in order to share it with a beggar Marzaus (Sainz) 5897: St. Marcel, either the fourth-century Pope Marcel I or, more credibly in the context, the late fourth-/early fifth-century Bishop of Paris; both were canonized. In the text, the reference is to a monastic foundation dedicated to St Marcel Mascon 6208, 7505: the town of Mâcon on the Saône, formerly the capital of the Mâconnais, today the administrative centre of Saône-et-Loire. This area may have been the domaine of Ganelon, since his wife Berte was living in Mâcon (C 6208–6209) and he wears the helmet belonging to Clarenbalt de Mascon (C 7505) q.v. Mater 5192; Maté 5579: the name of Baligan’s lance, meaning ‘Conquer’ or ‘Conqueror’ rather than ‘conquered’, of course Matragiez 4620: a Saracen king, father to Clariel and Effragiez who, like him, were used as reliable envoys by Baligant Maudelene (La) 1607; Maudalene (La) 3788; Marie 6057; (also la pecherise 7168-9): Mary Magdalene, mentioned by name in Luke viii.2 and xxiv.10, but also generally identified as the woman who poured precious oil over Christ’s head (Mark xiv.3) or His feet (Luke vii.37), the woman taken in adultery (John viii.3–11) and (less frequently) as Mary the sister of Lazarus (Luke x.38–42; John xi) Médee 3249: Medea, the land of the Medes, lying to the south of the Caspian Sea Meqe 3733, 4913, 5071, 5660: Mecca, the Holy City of the Moslems, where the Prophet Mohammed was born and to which they make pilgrimage. It lies in Saudi Arabia, fairly close to the port of Jiddah on the Red Sea

604

Proper Names

Mer 78: if this is a location in Spain, it might possibly be Murcia, a region which became first a kingdom after the defeat of the last Visigoth king, then part of Moorish Spain. However, Moisan (1986, tome 1, vol. 2, 237) identifies La Mare as Marrah in Asia Minor (which was captured by the Crusaders and is mentioned in the Chanson d’Antioche, Le Chevalier au Cygne et Godefroi de Bouillon, Les Chétifs, and Récits de la Première Croisade). Alternatively, Mer may simply indicate that Marprinant is an ally of Marsile from overseas (cf. Marprinant de Mer) Merguileis 5196: a Saracen in Baligant’s army Merinde 240: a Saracen town in Spain, probably either Miranda or Merida (cf. Boissonnade 1923, 123). Miranda (on the Ebro) may be the more likely, in that it lies in northern Spain, just south of Bilbao, on the borders of Vieille-Castille and Navarre. Boissonnade and Paris (Romania 11 (1882), 489) reject Merida as lying too far south-west in the Extremadura, on the frontier of Castile and Portugal, but it would probably fit the thirteenthcentury political situation better Micher (Seint) 48; Michel (Seint) 65, 2437, 3897, 3972, 5970: (i) the archangel Michael, cited in Daniel x.13, 21 and xii.1 as a ‘great prince’ and in the Book of Revelation xii.7–9 as the leader of the heavenly host and slayer of the great dragon; (ii) his feast-day, Michaelmas Day, is 29 September and it is to this date that lines 48 and 65 make reference; (iii) in C 2437, the reference is to the famous abbey of Mont St Michel on the Channel coast, the island community near the border between Brittany and Normandy, only accessible at low tide (intriguingly, it is also used as a reference-point in designating a large area of France in the Chanson des Saisnes; cf. Hurepois) Mile de Besençon 6204: One of Charlemagne’s barons, from Besançon in Franche-Comté Miles (li Joïz) 211: one of Charlemagne’s most important barons (barons esliz) on whose counsel he relies; probably identical to Mile de Besençon, mentioned later in the text. ‘The Joyful’ refers to his temperament, whereas Besançon is the capital of his domains Millon 6210: the Duke Milon, usually known as Milon d’Aiglent, the first husband of Berte, Charlemagne’s sister, and the father of Roland. He died before the events of this story took place Moïsent 3969, 4117; Moïsant 5029, 5474; Moïsés 5301: Moses, the famous Old Testament Jewish prophet, leader and law-giver whose story is recounted in the Book of Exodus. It was to Moses that God gave the Ten Commandments on Mt. Sinai (Exodus xx.1–17), as alluded to in C 3969–3970 and 5301–5302. Since this relates to the early history of the Jews, possibly even as early as the thirteenth century BC (although this is disputed), the reference to ‘the time of Moses’ in C 4117, 5029 and 5474 expresses the idea of ‘the far distant past’, ‘the dawn of history’ Monflor 8035: the fief of one of Charlemagne’s barons, Gerart de Monflor (q.v.); the only Montfleur found is a small town to the north-east of Bourg-en-Bresse. However, since there are so many different Gerart in the text, it is tempting to identify this one with Gerart le Normant (C 7340). Monflor might be read as a scribal error for Honflor, but the port of Honfleur in Normandy was probably too small and unimportant in the thirteenth century. On the other hand, one of the great contemporary families was the de Montfort dynasty: Amaury de Montfort was Constable of France in the early thirteenth century and the family lands were in Normandy and Brittany. So this may well be the underlying allusion, although the name was modified because -or and not -ort was needed for the rhyme Monjoie 1914, 2081, 2141, 2333, 2463, 2658, 2722, 2867, 3069, 3242, 3349, 3681, 3699, 4373, 5098, 5426, 5496, 5596, 5710, 5790; Monjoie la Roial 2190; Monjoie la Carllon 2569;

Proper Names

605

Monjoie Saint Donis 5441: ‘Monjoie-Saint-Denis’ was the war-cry of the French army throughout the Middle Ages. It was reduced very often to ‘Monjoie’; other variants are shown here. In one case, C 3069, ‘Monjoie’ is not a battle-cry but symbolises Charlemagne and France itself Monleon 109, 7519; Mont Leon 3498, 7392, 7400, 7433, 7595; Mont Löon 7326, 7411, 7429: Laon, the cathedral city in north-eastern France whose position on a 100 metre-high limestone ridge is evoked by the appellation ‘Mount Laon’. It was the capital of Carolingian France (eighth–tenth centuries) and boasted a large royal palace on the hill-top, just inside the eastern city gate; unfortunately, no trace of it remains above ground. Laon was also a famous religious centre and seat of learning in the Middle Ages: the school established around 1100 by Anselm attracted pupils from far and wide. The Abbey of St Vincent de Laon is said to have had a library of 11,000 volumes, before it was ravaged by fire in 1359. Rejhon (1993) links the abbey of Saint-Vincent de Laon to the story of Roland, but the evidence is fragmentary and the legends she cites could arise from the popularity of the Chanson de Roland. More recently, Rejhon (2005, 26s.) expands on the ‘Laon connection’. In addition, Martinet (1972, 28–34) provides interesting information about the involvement of Laonnais knights in the Spanish Crusades and in the First Crusade Monmartre 7386: the hill just outside Paris in the thirteenth century, the ‘Mont des Martyres’ whose name commemorated the Christians martyred there in pagan times Mont Agu 7587: the locality is not specifically identified in the text, but in Auvergne Montaigu-le-Blin and Montaigut-le-Blanc both had castles in the Middle Ages. The de Montagu were an important noble family (cf. Garins de Mont Agu) Montamis 6589: an unidentified locality, possibly Mont-Aymé(?) (cf. Bédier 1921, vol. 4, 414; cf. also Gui de Montamis) Mont Arsor 6247: an unidentified locality, probably fictional; the name provides the rhyme required (cf. Garin de Mont Arsor) Mont Cenis 7892: Mont Cenis in the Alps, the pass of the Petit Mont Cenis offering a route into Italy in the Middle Ages. The allusion may provide additional evidence that Pinabel came from somewhere to the east of the Saône and/or Rhône Monte Calvaire 5758: Calvary, the hill near Jerusalem upon which Christ was crucified Mont Leon, Mont Löon cf. Monleon Mont Nigre 1352, 5269: Whitehead (1942, 172) and Boissonade (1923, 93–95) identify O’s Munigre (O 975) as Los Monigros to the east of Saragossa or the town of Monegrillo on the Ebro. In C, the reference to camels (C 1356) and to ‘a harsh country’ (C 2270) points rather to the Nigri Montes in Cappadocia, i.e. south-eastern Turkey, which would also justify the reference to ‘Turks from Mont Nigre’, C 5269. Although Cornuble de Mont Nigre (q.v.) is at Marsile’s court, he could well be an Arab ally from a Middle Eastern kingdom rather than a locally-born baron Mont Nu 4825: an unidentified locality which provides the necessary rhyme, therefore possibly fictional (cf. Fergalem de Mont Nu) Mont Pantés (le roi de) 5293: an unidentified locality, again providing the rhyme needed; the king of Mont Pantés leads one division of Baligant’s army but is not otherwise mentioned Mont Pregal cf. Malpui de Mont Pregal, Malprimes de Brigart and Brigart Morant 5584: a knight in Charlemagne’s army, killed by Baligant Morel1: the adjective morel, normally applied to a horse and meaning ‘black’, ‘brown’ or ‘dun-coloured’, is used without an article as a Proper Name: for Otes’ horse 6335, 6360, 6403, 6409; for Naymes’ horse 7373; for Pinabel’s horse 7905

606

Proper Names

Morel2 8187: a city coupled with Rome and Valencia as conquests of Charlemagne; perhaps Morella, south-east of Saragossa and roughly equidistant from there and Valencia Murgalans 3659: a Saracen in Marsile’s army

N Naimeri cf. Naymer de Pavie Naimes, Naymes, Namon cf. Neimes Nairon 87: either a collective noun meaning ‘the infidels’ (variant graphy Noiron) or a scribal error for Paien, ‘the infidels’, as found in V7 87 and O 77 (cf. footnote to text) Naymer (li sires) de Pavie 6023; Naimeri 7572: Aimeri de Narbonne, son of Hernaut de Beaulande, count of Narbonne; V4 inserts his story immediately after Charlemagne’s return to France, but it is not found in C, V7, P, T or L Neimes subj. 271; Neyme subj. 289, 3109; Naymes subj. 990, 5016, 5541, 5566, 5986, 6183, 6867, 7278, 7372, 7821; Neymes subj. 1107; Nayme subj. 1214, 3079, 3173, 5689, 6852; Neymon obj. 4865; Naym subj. 4975 (final syllable of 1st hemistich); Namon obj. 5004, 6032; Nemes subj. 5126, 5534, 5574, 5791, 5880, 6131; Naymon obj. 5045, 5074, 5409, 5525, 5564, 6030, 6468, 6851, 6988, 7108, 7363; Naimes subj. 5509; Neme subj. 6855; Neme li Baiver subj. 8123; Naymon lo Baiver obj. 7637; Duke Naimes, who is Charlemagne’s friend and principal and wisest counsellor; identified as Duke of Bavaria in C 7637 and 8123 (and Bavarians are linked with French in composing Charlemagne’s army, C 4988, 5141, 7695) Nemon de Galie 5068: a nobleman from Galicia, the Christian kingdom in the far northwest of Spain, fighting in Charlemagne’s army Nevelon 6169: a French nobleman Nevers 6205, 6959: the old capital of the Nivernais, situated close to the confluence of the Loire and the Allier; a fortified town in 52 BC, when Julius Caesar used it as a depot. Governed by the Count of Nevers in the eleventh century, this area became a duchy in the sixteenth century (cf. Gui de Nevers) Niniven 4166, 5114: Niniveh, the ancient Assyrian capital, situated on the R. Tigris. It is cited here in relation to the story of the prophet Jonas (cf. footnotes) Niticoral 5515: a graphy of niticorax, ‘night-jar’; possibly the name of Naimes’ lance. Duggan (2005, 505) emends V7 5633 to jusqu’al coral but more drastic emendation is called for: to juridical Nobles 240, 3089: a Spanish town conquered by Roland. O 1775–1779 calls it Noples, which both Whitehead (1942, 172) and Boissonade (1923, 117) identify as the fortress of Napal, near Barbastro Noiron 1603, 4659; Noirons 4483: the basic identification is with the Roman emperor Nero. The pré Noiron (Prata Neronis in Latin) was the land on which the Vatican was later built (cf. footnote to C 1603). Nero was considered as a devil (C 4483) and also as one of the gods worshipped by the Moslems (C 4659) (the word Noiron was also used as a collective noun meaning ‘the infidels’: cf. Nairon) Noitïer 4881: a pagan king, presumably ruling a kingdom in Spain Norman (dan Richer le) subj. 4973, 7470; Normant (dan Richart le) obj. 5585, 7340; Normant subj.pl. 5018, 6068, 6121; Normanz subj.pl. 5488, 5696; Normant subj.sing. 6788: Norman, from Normandy (see comment on ‘Normandy’ below)

Proper Names

607

Normandie 6022, 6908, 7640: Normandy, an anachronistic term in relation to Charlemagne, since the invasion of France by the ‘Northmen’ began only at the end of his reign Norois 853: Oliver’s horse, the name meaning ‘Norwegian’ or ‘proud’ Noz 5277: a pagan race whose members form part of Baligant’s army; not otherwise identified Nubie 3689, 6555: Nubia in Africa, covering modern Sudan and southern Egypt; the Nubian Desert stretches eastward from the Nile to the Red Sea Nubles 5272: a pagan people; the Nubians (cf. Boissonnade 1923, 192)

O Océan, Océanz cf. Olchéan Odes 5037: the Lord of the Bretons and father of Salemons Odon 1722, 2766; Odon de Lengres 7638: the father of Estouz de Lengres (q.v.) and himself an important baron; one of the judges appointed by Charlemagne for the trial of Ganelon Ogiers subj. 206, 5662, 5792, 5880; Ogier subj. 1081, 1188, 4975, 6469, 6884, 7096, 8106; obj. 4865, 5918, 6453, 6872, 6892, 7108, 7637, 7775; Ogier li Denois subj. 5689: the important baron from Denmark who leads Charlemagne’s vanguard on the return from Spain, in place of Roland; usually referred to as a duke (although in C 5662 as a count), his historical prototype may perhaps be Duke Autcharius, who lived in the reign of Charlemagne’s father, Pépin-le-Bref, or possibly Otharius, but there were a number of nobles, heroes of the Crusades with this name (cf. Boissonnade 1923, 350– 354) Olchéan 5295; Océan 5346, 5653; Océanz 5642: a pagan region, perhaps Oxiane, the region through which the River Oxus flows, i.e. the modern River Amou-Daria in Uzbekistan (cf. Boissonnade 1923, 231s.) Oliferne 3297, 5366: the name Oluferne was often used in later poems as a synonym of Haleb = Aleppo: cf. Boissonnade (1923, 206). On the other hand, O 1915 here reads Alfrere (or Alferne) which Boissonnade (1923, 158s.) links to the Beni-Ifren Berbers and a confederation of North African tribes Olivier subj. 120; Oliver subj. 212a, 298, 989, 1341, 1422, 1453, 1465, 1471, 1515a, 1660, 1924, 1939, 1945, 1955, 1994, 2071, 2193, 2285, 2310, 2316, 2393, 2419, 2497, 2670, 2975, 3005, 3019, 3067, 3232, 3760, 3860, 4804, 5895, 6164, 6525, 6652, 6748, 6766, 6887, 6915, 7052, 7068, 7090, 7159, 7232, 7782; obj. 343, 853, 1484, 1685, 1751, 1795, 1819, 1850, 2090, 2259, 2634, 2872, 2951, 2966, 3001, 3235, 3326, 3353, 3529, 3822, 3848, 3927, 3931, 3941, 3982, 4147, 4345, 4377, 4448, 4655, 4869, 4910, 4958, 4966, 4981, 5049, 5121, 5218, 5224, 5455, 5698, 5892, 5943, 6017, 6128, 6529, 6581, 6757, 6776, 6782, 6925, 6994, 7001, 7039, 7061, 7104, 7179, 7183, 7226, 7269, 7292, 7293, 7299, 7322, 7702, 7896, 8027, 8073 (and in final words); Olivers subj. 373, 638, 831, 858, 872, 1154, 1192, 1280, 1314, 1381, 1391, 1400, 1402, 1411, 1534, 1543, 1870, 1889, 2133, 2293, 2322, 2506, 2664, 2689, 2814, 1981, 1990, 3320, 3333, 3342, 3364, 3377, 3383, 6902, 6932, 7008; obj. 1325, 1963; Oliviers subj. 2930: Oliver, the son of Renier de Genes, brother of Aude, nephew of Girart de Vienne; one of the Twelve Peers and companion-at-arms to Roland Ongrie 1344, 1845, 2834; Hongrie 5658: Hungary, but looked upon (like all of South-Eastern Europe) as a pagan land, an alien place. This may also be due to folk memory of the

608

Proper Names

ninth-century arrival of the pagan Magyars and their tenth-century wars with the Saxons Orabloi 4569: a locality on the Spanish coast, between Portpaillart and Belcler, but not otherwise identifiable Orbrise 5294: a Saracen country, possibly Prussia because it lay beyond the eastern frontier of Charlemagne’s empire. Cf. Romania II (1873, 332 and 480) and Gaston Paris (1909, 581) Orcaise 4165: a place-name not mentioned in the Biblical narrative in the Book of Jonah. The town of Ahwâz (pronounced Ahk-waz) on the R. Kârûn (or Dujayl) in Khuzistan was the capital of the province and an important and thriving town with canals to bring fresh water into it and considerable traffic by water (cf. Le Strange 1930, 232–234). It still exists today, just to the north-east of Basrah at the top of the Persian Gulf; and it lies due south of Niniveh and just south of the extensive ‘marshland’ in southern Iraq, near the rivers Tigris and Euphrates (cf. Garillant and Le Strange 1930, 6; Moisan also lists it as Archaise, found in Godefroi de Bouillon 4327, 4431, 4661 (amiral d’A), but identifies it simply as a distant pagan region) Orcois (les vals d’) 6265: as Ganelon is galloping south from St Jean-Pied-de-Port towards Saragossa (C 6169), it is possible that this indicates the Valle de Arce (just south of Roncevaux) in Spain Orebe 77: a Saracen baron, one of several chosen by Marsile to accompany Blancandrin on his mission to Charlemagne, but never mentioned again Orgalifes 3438: a pagan people in Marsile’s army, but not identifiable Orion 6205: an unidentified locality, presumably in France since Girart d’Orion is one of Charlemagne’s messengers to his sister Berthe in Mâcon. Although Garin d’Alvergne is mentioned in the preceding line, it is just possible that the title here might be a corruption of de Riom, which was the ancient capital of the Dukes of Auvergne. In fact, Charlemagne sends barons from Burgundy, Auvergne, Besançon (Franche-Comté) and Nevers, as well as Girart d’Orion: all from roughly contiguous areas, who might be expected to have some knowledge of the routes to Mâcon, so Riom would fit in; it lies due west from Mâcon Orlenois 5489: the old province of the Orléanais, later a duchy; covered the area of the three modern départements of Loiret, Loir-et-Cher and Eure-et-Loir Ormanois 5343: it has been suggested that this pagan people might be the Canaanites, but Boissonnade (1923, 173) identifies them with the Jarmlenses, inhabitants of Ermland or Ormaland, i.e. Letts or Latvians. Or they might represent the ‘Omanitae’ of Roman times, who lived along the coast in southern Arabia, in the area covered by Oman and Aden today. It is doubtful whether the Ormanois should be equated with the Orvalois, C 5290 (q.v.) Orquenois 5287: a pagan race forming part of Baligan’s army Orsaire 5751: identical to the Cesaire (variants Orcanes, Orsaires, etc.) in the Chanson d’Antioche: a person famed for his wealth, he may be brother to Angobier Orvalois 5290: a pagan people, possibly the Canaanites (cf. Ormanois) Otemer cf. Otoier Otes1 subj. 1156, 2231; Ote subj. 2226; Oton obj. 1705, 3763; Athon obj. 5219: one of the Twelve French Peers, usually coupled with Berenger. Apparently C does not recount how he met his death, although he is listed amongst the dead, C 3763; but Langlois (1904) lists Hugon (q.v.) as an error for Oton, and Hugon’s death is recorded in laisse 205, C 3262–3269 (cf. Bédier 1922, 180s.)

Proper Names

609

Otes2 subj. 6005, 6264, 6296, 6306, 6316, 6323, 6328, 6334, 6347, 6358, 6377, 6389, 6395, 6408, 6410, 6421, 6441, 6445, 6448, 6451, 7843; Ote subj. 6363; Oton obj. 6288: a French count who becomes one of the chief barons of Charlemagne after the death of the Twelve Peers, when he brings back the escaping Ganelon Otoier 7450, 8134; Otemer 7836: a French count, lord of Amiens and Boulogne-sur-Mer

P Palatine 7094: a noble lady of the court or palace Palernes 8188: Palermo in Sicily Paris1 64, 674, 2603, 4629, 7386, 7433, 7898, 7900: Paris, the French capital city; li rois de Paris: Charlemagne 7433; escuz de Paris: shields made in Paris, or in the Parisian style 7900 Paris2 6601: Paris, the father of Floires and presumably, like his son, one of Girart de Vienne’s retainers or vassals Pavie 6023, 6558: Pavia, the capital of Lombardy in Italy (cf. Naymer de Pavie) Pepin 3890, 6070: Pépin le Bref, the father of Charlemagne, so that the latter is indicated in both instances as le fil Pepin Persant 899, 4610, 6045, 7481; Persanz 3072, 3438, 5420, 5635: a Persian, (the) Persians Perse 4554, 4568; Persie 5247, 5647, 6021: the ancient kingdom of Persia (modern Iran), possibly extending to cover the lands of the old Persian empire Piere de Rome (Seint) 1300; Piere (Seint) 1603, 4084: St. Peter the Apostle, who established the Christian Church in Rome and was the first Pope Pinabel 527, 2413, 7735, 7747, 7749, 7756, 7788, 7793, 7807, 7826, 7831, 7860, 7862, 7866, 7876, 7880, 7887, 7902, 7905, 7909, 7935, 7942, 7954, 7964, 7976, 7994, 8004; Pinabel de Florence 7613; Pinabeaus 7649, 7856; Pinabaus 7659, 7665, 7684, 7725, 7737, 7741, 7744, 7790, 7858, 7983; Pinabel de Sorence 7925: the nephew of Ganelon and his champion in the judicial combat, killed by Tierri d’Anjou; he is mentioned in O 3783 as Pinabel del Castel de Sorence (and simply Pinabel de Sorence, O 3915), confirming C 7925 (cf. Sorence) Pinaus 2677: a Saracen baron, not otherwise mentioned Poier 5696: the Picards (cf. Langlois 1904, 530) Pol (Seint) 1604, 6074: St Paul Pol li chauf (Seint) 4160: C’s reading was Policarf, but Polycarpe, Bishop of Smyrna, was the disciple of St John and one of the Apostolic Fathers; he was martyred c.155 AD. Details of his life are given by St Irenaeus (Bishop of Lyon, but born in Smyrna c.130 AD and well known to Polycarpe) and in the anonymous Martyrium Polycarp, a contemporary account of his martyrdom, as incorporated in Eusebius’ Ecclesiastical History iv.15 (cf. Bibliotheca Sanctorum 1968, vol. 10, 985–990). There is no reference to ‘many sins’ in Polycarpe’s final speech (in which he says he has served Christ for 80 years) or in any other account. In C the copyist may have read this as a clerical pun, a play on the words poly (‘many’) and carpe (= corpe = colpe = culpae, ‘sins’). V7 4270 reads Pol li chauf, whom Duggan very plausibly identifies as St Paul of Tarsus; the reference to ‘many sins’ then becomes very apt, as instanced by St Paul’s own admissions in his Epistles Poitiers 1161: the capital of Poitou, today the administrative centre of the Vienne

610

Proper Names

Port Palart 4569: a town by the Mediterranean Sea in Spain; originally ‘pagus Palliarensis’, near Ribagorza and Urgel. Langlois (1904, 536) opts for the variant Port Palant Preciose 5185, 5368, 5587, 5709: ‘Precious’, the name given to Baligan’s sword and also his rallying-cry Priamus 76: a Saracen baron: the name derived from Homer’s King Priam of Troy Prince 241: a locality in Spain conquered by Roland, possibly Pina with its famous monastery of San-Juan de la Peña, since O 199 reads: la tere de Pine. V7 reads: prince (no caps.), so this may simply be the substantive (m. and f.) meaning ‘principality’, although the definite article is then misplaced (through poetic licence or scribal error) Provance 4933: Provence Puille 552: the Italian region of Puglia, immediately north of Calabria which forms the ‘heel’ of Italy and is mentioned in the same line; from the seventh century, it was ruled by the Longobards (= Lombards), but they were conquered by Charlemagne in 774 AD

R Rabel 4963 (scribal error for Ranbel?); Ranbalt 5068; Ranbaus 5418; Rembauz 5428; Ranbiaus 5435, 5441: a French count, vassal to Charlemagne, appointed by him to be one of the leaders in battle; in command of the Flemish and Frisian contingent Rainer 3841; Reiner1 6579, 7099, 7323, 7778: Renier de Genes, the son of Garin de Monglane, brother of Girart de Vienne, father of Oliver and Aude Rapadans 2648: a Saracen baron killed at Roncevaux by Oliver Reger cf. Richer d’Aspremont Reiner1 v. Rainer Reiner2 4879: a French pilgrim, originally from Chartres Reins 310, 1730, 1987, 2041, 2387, 2564, 3476, 3482, 3614: Rheims, the cathedral city in the Marne département, an archbishopric since the fourth century (cf. Turpin de Reins) Rembauz cf. Ranbalt Rés 5288: an unidentified pagan people; if Rés is a corruption of Ros, then Boissonnade (1923, 184) suggests that these are Russian merceneraries Richart de Vermendoi 7567; Richart 7571: a baron from the Vermandois, the region around St. Quentin in north-eastern France; the Counts of Vermandois were important mediæval barons, of royal blood in the eleventh/twelfth centuries (Count Raoul I being the grandson of Henry I of France) Richart le Normant, Richart de Normandie, Richart cf. Richer le Norman Richer d’Aspremont 6873, 7708; Reger 6884: a baron who accompanies Charlemagne when he rides to greet Girart de Vienne; perhaps from Aspromonte in Calabria (q.v. re its subjection to Charlemagne following the defeat of the Lombards), although there are two small places in Provence called Aspremont, one on the R. Buëch just north of the Montagne de St Genis, the other on the R. Var just north of Nice Richer le Normant 4973; Richart le Normant 5585, 7340, 7470; Richart de Normandie 6022, 7640; Richart 7837, 7843: Richard, Duke of Normandy; a reference to the historical Duke Richard I, ‘Richard le Vieux’ (943–996) (cf. Bédier 1921, vol. 4, 1–8). Presumably Richard is struck, but not killed, by Baligan, C 5585; otherwise, scribal error has occurred Rohés 5292: a pagan kingdom. This could perhaps be Rhodes (Latin: Rhodus), where the rule of the Byzantine Empire was frequently challenged by Arab raiders

Proper Names

611

Rollant 120, 212, 232, 234, 297, 331, 340, 349, 370, 381, 396, 411, 426, 427, 428, 444, 463, 492, 540, 569, 574, 588, 610, 621, 637, 668, 725, 852, 857, 866, 871, 881, 887, 894, 901, 919, 932, 969, 977, 989, 1029, 1058, 1083, 1093, 1116, 1126, 1132, 1138, 1167, 1170, 1178, 1186, 1191, 1221, 1225, 1246, 1271, 1281, 1292, 1313, 1325, 1340, 1362, 1381, 1383, 1394, 1400, 1421, 1424, 1426, 1432, 1435, 1444, 1446, 1460, 1465, 1478, 1482, 1515a, 1522, 1552, 1558, 1561, 1569, 1586, 1588, 1593, 1616, 1623, 1639, 1650, 1653, 1656, 1668, 1683, 1699, 1760, 1777, 1831, 1853, 1863, 1867, 1878, 1897, 1909, 1932, 1939, 1961, 1994, 2013, 2037, 2089, 2096, 2111, 2227, 2267, 2305, 2312, 2330, 2372, 2392, 2418, 2456, 2475, 2498, 2504, 2536, 2618, 2634, 2653, 2667, 2682, 2684. 2694, 2721, 2769, 2779, 2813, 2872, 2900, 2913, 2920, 2942, 2951, 2964, 2979, 2984, 3000, 3019, 3034, 3035, 3042, 3043, 3058, 3081, 3092, 3102, 3124, 3129, 3158, 3168, 3177, 3185, 3195, 3213, 3235, 3246, 3252, 3260, 3272, 3304, 3313, 3341, 3350, 3363, 3375, 3379, 3391, 3395, 3397, 3409, 3418, 3428, 3467, 3487, 3500, 3522, 3537, 3555, 3568, 3578, 3584, 3593, 3606, 3664, 3669, 3675, 3685, 3690, 3739, 3748, 3758, 3780, 3785, 3819, 3827, 3831, 3839, 3847, 3860, 3867, 3894, 3900, 3926, 3936, 3040, 3954, 3960, 3974, 3977, 3992, 3997, 4006, 4011, 4016, 4018, 4023, 4025, 4045, 4049, 4063, 4075, 4093, 4098, 4107, 4113, 4120, 4139, 4180, 4222, 4224, 4237, 4376, 4447, 4602, 4663, 4717, 4752, 4788, 4803, 4869, 4910, 4959, 4966, 4981, 5048, 5217, 5223, 5431, 5454, 5559, 5927, 5942, 5959, 5983, 5989, 6016, 6128, 6164, 6237, 6473, 6525, 6530, 6550, 6583, 6613, 6615, 6652, 6662, 6664, 6724, 6749, 6758, 6767, 6770, 6775, 6782, 6813, 6815, 6888, 6899, 6916, 6924, 6934, 6940, 6967, 6995, 7000, 7007, 7040, 7052, 7077, 7083, 7087, 7105, 7227, 7233, 7291, 7294, 7338, 7472, 7702, 7713, 7722, 7728, 7733, 7852, 7865, 7895, 8026, 8050, 8054, 8064, 8120, 8184, 8196 + final line of MS; Rollanz subj. 829, 1153, 2050, 2271, 2711, 3074, 3086, 3165, 3217, 3353, 3430, 3454, 3456, 3774, 3798, 4054; obj. (end of line rhyme) 1076, 4298; Roillanz 5891: the nephew of Charlemagne, Count of the Breton Marches, killed at Roncevaux; the Hrodlandus of Eginhard’s chronicle Rome 551, 812, 1300, 2896, 3152, 8187: Rome, the capital of the Roman Empire, but in mediæval times the spiritual capital of Christendom, the Holy City where the Christian Church was founded and the seat of the Pope (cf. Seint Piere de Rome) Romer 5196: a Saracen Roncivaus 920; Ronchivals 1278, 1312; Rencivals 1290, 3164, 3870, 3948, 3978, 4183, 4341, 4680, 4705, 4867, 4984; Roncivals 1321, 1361; Ronchivaus 1340, 4313; Rencevals 3417, 3891, 4281, 4379, 4397, 4790, 4801, 5491, 5803, 5885, 6130, 6510, 6618, 6702, 7019, 7391, 7480; Rencevaus 4094, 4654, 5887, 5925; Rencivaus 4321; Renchivals 5432; Renchevals 6137; Renchevaus 6493; Renceval 8120; Roncisvali on the very last line: Roncevaux, the valley in the Pyrenees, on the Spanish side, where the rearguard of Charlemagne’s army was attacked and wiped out; it lies almost due south of St. Jean-Piedde-Port, over the Col d’Ibañeta Rondel 6771: the name of Oliver’s horse Rosion 3271, 3784; Rosellon 7767: since Girart de Roussillon was a historical figure with fiefs in Burgundy and also Marquis of the Duchy of Lyon, Boissonade (1923, 357–359) identifies this as the ancient stronghold on the Rhône 20 km south of Vienne. However, by the time of Charles le Simple (tenth century), the Counts of Roussillon had dominion over a considerable territory inland from the Western Mediterranean coast, the capital being Perpignan; i.e. the same area as the modern Roussillon. (This may explain why, in C, Girart de Roussillon and Girart de Vienne are seen as two separate people.) In 1278, the last Count of Roussillon, Guinard or Gérard II, left this territory in his will to Alfonso, king of Aragon. The province became French again only in 1659 (cf. also Gerart1)

612

Proper Names

Roudez d’Aquiteigne 2677: a Saracen baron, presumably an ally from north of the Pyrenees, the reference to Aquitaine as ‘enemy territory’ presumably arising from the French wars with Henry II of England (and his wife, Eleanor of Aquitaine) in the twelfth century Rualent 4673: a Saracen Rubion 3678: the king of Balaguer (q.v.) in Catalonia

S Saissoigne cf. Sansoigne Salemon 913: Solomon, King of the Jews on the death of his father, David; famous for his wisdom and for building the Temple in Jerusalem (cf. Old Testament, I Kings ii.12s.) Salemons obj. 5037 (final s needed for rhyme); subj. 7764; Salomon subj. (-on rhyme) 7395; Salemon obj. 7641, 7837; Salemons de Bretegne subj. 7847, 7859, 8095: in C, a Count of some importance at Charlemagne’s court, the son of Odes of Brittany; historically, Salomon was Duke of Brittany 857–874 Samarie 1333: since this is linked with Seville (Margariz de Sebie… de ci en Samarie), one would expect it to be a locality in Spain; but the obvious allusion is to Samaria in the Holy Land, familiar to the copyist and his readers through the New Testament: the woman of Samaria and the Good Samaritan. This would surely be the extreme of poetic hyperbole. However, O 956 has a reading similar to C’s: Cazmarines, which Whitehead (1942, 169) tentatively identifies with Camariñas in Galicia. On the other hand, V7 1391 reads terre… d’Aumarie (and V4 1904 simply a la marine/al Amarine): cf. Aumarie, Almarie = Almeria in Spain, which makes good sense in the context. So Samarie may simply be a scribal error, caused by the copyist’s familiarity with religious texts Samués (la gent) 5291: the Saracens; certainly a pagan people, perhaps the Bulgarians, ruled by Tsar Samuel, who attacked the Byzantine Empire in the twelfth century Sanses1 (li dus) subj. 121, 208, 1820; Sanson obj. 2702, 2711, 3782: a French duke, brother to Terriz (who is probably the Duke Terris d’Argoigne, q.v.); he was one of the Twelve Peers and comrade-at-arms to Anseïs or Gérart; he was killed by Valdebron at Roncevaux Sanses2 subj. 6326; Sanson obj. (3198?), 6453: a rich baron in Charlemagne’s army, the close friend of Otes; he may be the same Sanson (= Bovon, C 3137) who is referred to as the emperor’s ‘master cook’, i.e. the man in charge of victualling the army (C 3198), possibly the quartermaster-general; the allusion to him as Count Sanson (C 6453) may explain the scribal confusion of cous and cons in C 3136, 3137 and 3197 (cf. Sanson3 below) Sansoigne 553, 5076; Saissoigne 2897: Saxony; the Saxons were finally defeated by Charlemagne in 803 and their lands, stretching between the R. Elbe, the Harz Mountains, the R. Rhine and Friesland, became a dukedom after 850 AD Sanson3 3198: the ‘master cook’ to Charlemagne, the same person as Bovon (q.v.), C 3137; if this title implies that he was quartermaster-general to the army, he may possibly be identical to the ‘rich baron’, Count Sanson, of C 6326 and 6453 (cf. Sanses2 above) Saracins obj.plur. 117, 1382, 1404, 2509, 4065; subj.sing. 547; Saracin subj.plur. 545, 1367, 3471, 3577, 3623, 4581, 4903; obj.sing. 1589, 4648, 4905; Sarazins subj.sing. 603, 786, 1310, 2230, 2843, 4019, 5158, 5162, 5238; subj.plur. 1032, 2067; obj.plur. 1454, 1502, 1736,

Proper Names

613

2063, 2376, 3054, 3214, 3249, 3345, 3437, 4389, 4618, 5271, 5381, 5500, 6087; Sarazin subj.plur. 1676, 2748, 3072, 3090, 3560, 4053, 4566, 4643, 5361, 5866, 6512; obj.sing. 2613, 2775, 3777, 3999, 5570, 5730, 6101, 6599, 7057; Sarracin subj.plur. 2926; Saracinois subj.plur. 4749: the Saracens, the Arab invaders who had conquered most of Spain and established the Moslem faith there Saragoze 7, 192, 253, 288, 302, 317, 338, 422, 447, 597, 624, 633, 671, 728, 762, 2613, 2616, 3553, 4286, 4443, 4450, 4503, 4523, 4557, 4573, 4578, 4621, 4642, 4757, 5769, 4774, 4778, 4785, 4830, 4849, 5810, 5835, 5837, 5852, 5876, 6269, 6652, 7733; Saragoce 1000; Saregoze 14, 1233: Saragoza in Spain, the old capital of the kingdom of Aragon; it stands on the R. Ebro Sartegne (peron de) 2681: Sardine-stone, a precious stone with a deep blood-red colour which is a variety of cornelian Sebie 1332: Seville in Spain, a region of Andalucia and also the name of its chief city; held by the Moors from 712 to 1248. Boissonnade (1923, 73) rejects this identification and suggests Sediles or Sevil near Barbastro (1923, 96–99) (cf. Margariz de Sebie) Sebre 3565, 3712, 4289, 4327, 4358, 4570, 4576, 4714, 4726, 4764, 4784, 4812: the R. Ebro in Spain Sein Leri (port) 5622: an unidentified locality, but presumably one recognized at the time for its learning Seint Omer 6170, 7362: the town of St. Omer in the Pas-de-Calais which grew up around a seventh-century monastery (cf. Gui de Seint Omer, Gerart de Seint Omer) Seint Denise (borc de): the town of St. Denis, north of Paris 1350; Seint Denise: the abbey of St Denis, in the town of the same name 2454, 7407; el reigne Seint Denis 2608, el reigne Seint Donis 7385: The kingdom of St Denis, i.e. France (cf. Donis (Seint) for extra details) Seinte Sophie cf. Sophie Seint Jame cf. Seint Jaqe Seint Jaqe en Espeigne 2680; Seint Jaqe 6082; Seint Jame 7443: St James the Greater, the Apostle who suffered martyrdom under Herod Agrippa (cf. Acts xii) and whose dead body was supposedly transported by boat to Spain, where it aided in the conversion of the inhabitants; in about 820 AD, his corpse was miraculously re-discovered uncorrupted in its tomb in the ‘Field of Stars’ (Compostella). The great cathedral built over it became a major place of pilgrimage in the Middle Ages: Santiago de Compostella, in Galicia; indeed, according to Bédier (1921, vol. 3, 300; 1912, vol. 4, 430–433), the Chanson de Roland had its source in the stories recounted to pilgrims on their way there. The reference in each case is to Santiago, the shrine of St James, although there may also be a ‘personal’ element to ‘seeking St James’, C 2680 Seint Johan 6144, 6156: St Jean-Pied-de-Port in the foothills of the Pyrenees, at the foot of the pass from Roncevaux, 6144; this is confirmed by C 6155–6157, in which Charlemagne establishes a monastery dedicated to St John (the saint as a person, 6156) ‘as piez des porz’. The original town was razed to the ground by Richard the Lionheart in 1177 and rebuilt shortly afterwards on a nearby, but more easily defensible, site; its original site is now known as St Jean-le-Vieux Seint Micher 48; Seint Michel 65, 2437, 3897, 3972, 5970: Michaelmas, the feast day of St Michael, 29 September, C 48, 65; the famous Mont-St.Michel with its great abbey church, off the coast of Normandy, linked by a causeway to the mainland at low tide, C 2437; the saint himself, C 3897, 3972, 5970 (cf. Micher, Michel for further details)

614

Proper Names

Sene 6831: Saintes (Charente-Maritime), an important city in Roman Gaul and in the Middle Ages; it lies to the north of Blaye, slightly off the route from Vienne, but was the capital of the Saintonge, which stretches right down to the northern coast of the Gironde, so the allusion may be to the whole area Serence cf. Sorence Sers 5296: a Slavonic race, possibly the Sorbs who lived just north of Dresden Siduel (Val de) 8188: clearly, from the context, one of the boundary points of Charlemagne’s empire, but not firmly identified. If it derives from the Roman Sedunium, it might be Sedrun on the upper Rhine, or Sion (Sitten) on the infant Rhône, both in Switzerland. Or this could be the Vale of Siddim which encloses the southern end of the Dead Sea. Gasca Queirazza, in his edition of V4 (1954, 314), reads V4 6049 as la val de si duel, ‘the valley of his grief’, i.e. Roncevaux, but Beretta (1955, 373, and V4 6011) rejects this, maintaining that this must be a place-name Simon 7499: a burgher of Laon Singlorel 2387: a Saracen sorcerer, killed by Turpin. The mediæval mind equated the great, but often to it abstruse, learning of the Moors with necromancy Solteins 5288: a pagan race (perhaps sol-teins, ‘sun-burnt’, indicating an origin in black Africa). Boissonnade (1923, 176ss.) offers a number of hypotheses Sophie (Seint) 4083: not the famous cathedral of St Sophia (= divine Wisdom) in Byzantium, but the Saint Sophia who was martyred in Rome during the rule of the Emperor Hadrian; her feast day is 30 September Sorence 7925; Serence 7890; Florence 7613: Boissonnade (1923, 135) identifies the castle and fief of Pinabel as possibly the modern Sarrance in the Val d’Aspe in the Pyrenees. He is an important relative of Ganelon, who has domains in the Aspe area (around the River Aspe, cf. C 6260: aspois). However, Ganelon’s main base seems to be Mâcon, and he is linked in C 7650 with Hues de Lion (Lyon). It is therefore possible that Pinabel’s fief lies within the same area. The place-name most nearly approximating is some distance to the north-east: Sourans in the Jura, not far from the R. Doubs (and Pinabel offers Serence and Besançon to Tierri, C 7890, implying that he has lands in FrancheComté). Alternatively, perhaps the scribal error Florence C 7613 points to another Italian city, Sorrento; this would place Pinabel’s lands just on or slightly beyond the border of Charlemagne’s empire, which extended down through Italy to Naples (the graphy Sorence has been given precedence because this is found in O.; cf. also Pinabel de Sorence) Sorges 6240: a place on a river, on the direct route north from St Jean-Pied-de-Port to Blaye, (V7 6360 reads: Sorges, but V4 4484: a uns loges, possibly indicating that the name was not recognized). This is probably Sorde l’Abbaye, which lies on the old pilgrim route from St Jean-Pied-de-Port to Dax and is shown as a halt on the Carte des Chemins de St Jaques de Compostelle of 1648. It lies E-SE of Peyrehorade, not far from the confluence of the Gave d’Oloron and the Gave de Pau. Bédier (1921, vol. 3, 334–340) makes this identification, giving a brief history of the Abbey of St Jean de Sorde from its foundation in the tenth century and particularly its later bid for independence, founded on false documents relating to Charlemagne Surie 6608, 7351: Syria, in both cases referred to as a source of mules Susanë 1606: Susannah in the Apochryphal story, ‘Susannah and the Elders’, the virtuous woman against whom the lascivious elders bore false witness when she refused to be blackmailed into giving them her favours

Proper Names

615

Symon 6056: the Pharisee at whose house Jesus dined, Luke vii.36, 40, 43 Synaï 3970: Mount Sinai, on which God gave Moses the Ten Commandments (Exodus xx)

T Tardis 5429: the name of Ranbalt’s horse Terascoigne 5084: as Girart de Terascoigne (q.v.) is one of Charlemagne’s barons, this is likely to be Tarascon (and the region around it), near Arles in the Bouches-du-Rhône, an important centre ever since the Greeks established Massalia (Marseille); not Tarragone in Spain, because this would have been a Saracen domaine at the time Terre Major 1201, 2490, 2622, 2868, 3099, 4721, 5548, 6235: France Terriz 208; Terris d’Argoigne 5087: the Duke of the Argonne, brother to Duke Sanson (one of the Twelve Peers); there seems no reason to see Terriz and Terris d’Argoigne as two separate people with the same name (cf. Argoigne, Sanses1) Tervagan, Tervagans, Tervigan, Terviganz, Tervigant cf. Trivigant Tieri d’Ardene 5670; Tieris d’Ardenois 5690: an important baron from the Ardennes, mentioned in other chansons de geste as a duke Tieris subj. 7712; Terris subj. 7727, 7760, 7783, 7855, 7858, 7870, 7874, 7893, 7904, 7909, 7920, 7952, 7992, 8016; Terri obj. 7750, 7772, 7915, 7937, 7963; subj. 7849, 7887, 7922, 7942, 7947, 7964, 7969, 8031: Tierri, the son of Jofroi d’Anjou and squire to Roland; Charlemagne’s champion in the judicial combat with Pinabel (in other versions, the brother of Jofroi d’Anjou) Tinodel 2380: a Saracen Tiois 5495; Tioirs 5695: the Germans (Langlois, 1904, 642). Duggan (2005, 832) suggests specifically Teutons, perhaps because O 3795 here reads Alemans e Tiedeis Tolete 6261: Toledo in Spain Torchis de Tortolose 1295; Torgis 2211: a Saracen count, one of the Peers in Marsile’s army, killed by the French Peer, Anseïs, C 2217 (but cf. Torgins below) Toregne 7352: Touraine in France Torgins 2302: a Saracen warrior killed by Oliver, but not mentioned elsewhere; use of his name may arise from the uncritical amalgamation of two distinct versions of the battle, since he could readily be identified as Torchis (Torgis) de Tortolose (q.v.), who has already been killed by Anseïs: this is involuntary repetition of an incident through scribal carelessness Tortolose 1295; Tortelose 4575, 5748: a Spanish town, perhaps Tortosa, near the mouth of the R. Ebro, or, as Boissonnade (1923, 104s.) proposes, Tarazona-Tudela (Latin: Tórtoles) just over a mile from Tarazona, since in the twelfth century this thriving Moorish town dominated the main Madrid-Saragossa road. Its listing alongside Spain and Cairo in C 5748 implies considerable size and importance Torz cf. Turs Trivigant obj. 898; Trivigan subj. 3646; Triviganz obj. (but rhyme needed) 4292; Tervagan obj. 4477, 5328, 5608; subj. 4481; Tervigan obj. 4659; Tervagans subj. 4694, 5629; Terviganz subj. 4703; Tervigant subj. 4742; obj. 5234: a Saracen god, often seen as one of the Moslem Trinity. Boissonnade (1923, 248) links the name with the Scandinavian Thor or Germanic Thingsus/Tuisco Tulles de Persie 5247; Turlex 5262, 5437: the king of Persia

616

Proper Names

Turchez cf. Turs Turleus cf. Tulles Turpin de Reins subj. 310; obj. 3614; Turpis subj. 1495; Turpin obj. 1785, 1678, 1704, 2678, 3814; subj. (for rhyme) 3894; Turpins subj. 1683, 2150, 2863, 3816, 3922; Turpins de Reins subj. 1730, 1987, 2041, 2387, 2564, 3482; obj. 3476: the archbishop of Rheims, apparently one of the Twelve Peers, who was killed at Roncevaux. His historical prototype, the late eighth-century archbishop of Rheims of the same name, was a peaceful administrator, but his eleventh-century successor, Manassès I, was a fiery individual with family links with the Spanish crusades (cf. Boissonnade 1923, 322–327) Turre 5387: an unidentified city or country; possibly Tyre Turs obj.plur. 3439, 4301, 4675, 5343, 5482, 5643; Turchez subj.plur. 3470; Turc subj.plur. 4610, 5972, 5606, 6045; obj.sing. 3985, 5722, 5730; Torz obj.plur. 5269: the Saracens or Turks, the terms being used indiscriminately Turs de cople 4547, 4932, 5658: an error for Turcople, the name given to the sons of a Turkish father and a Christian mother, who formed a class apart in the Syrian population at the time of the First Crusade. Later they seem to have furnished light cavalry troops who fought in the Turkish fashion

U Ugon de Val Color 8036: a French baron

V Valance 4929, 5986; Valence1 8187: since both Charlemagne and Baligan wear an eume de Valance, this must surely be a reference to Valencia in Spain. The city was the capital of the Saracen kingdom of the same name, which was conquered by the Cid in the eleventh century and incorporated with Aragon in 1238. It was noted for its high quality steel, so the reference probably specifies the origin rather than the style of helmet. In C 8187, Valence is listed with Rome and Morel amongst the conquests of Charlemagne (cf. also Boissonnade 1923, 73) (to read valance as the substantive, meaning ‘valour’ or ‘worth’, does not fit the context so well or accord with normal practice in the text, where sources of weapons and other material are usually cited) Val Bruient 5310: an unidentified (probably imaginary) Saracen country Val Color 8036: probably Vaucouleurs on the R. Meuse just over 40 km. to the west of Nancy (cf. Ugon de Val Color) Val Dormanz 5632; Val Dormant 6935 (= Val Serie 6900): the name has been transcribed as written in the ms., but perhaps this should be Val d’Ormanz, linking it with the pagan Ormanois listed in Baligan’s army, thus a valley in Oman or Saudi Arabia. On the other hand, Talos 1992 suggests that Val Dormant simply signifies ‘land of eternal sleep’, thus ‘land of death’; the initial use of Val Serie as the name of this kingdom may have a similar implication: seri, adjective, means ‘peaceful’, ‘tranquil’ but also ‘of the evening’, ‘of shadows’. The very fact that Charlemagne changes the detail of his story (Florent is first king of Val Serie, then of Val Dormant) may be a deliberate poetic ploy to indicate that he has difficulty in telling lies

Proper Names

617

Valebron 781, 905; Valdebron 2698: a Saracen baron who helped in the capture of Jerusalem; killed by Roland. The name probably derives from the Val d’Hebron (Vale of Hebron) 30 km south of Bethlehem where the tombs of the patriarchs, Abraham, Isaac, Jacob, etc. stood (cf. Boissonnade 1923, 208s.) Valence2 2761: the city of Antoine d’Avignon q.v. The latter is a French peer and his domains are described as ‘Valence and the rock round about’, so this is likely to be Valence on the Rhône with the steep hillsides and lofty crags of the Massif Central crowding the west bank of the river and the foothills of the Alps just to the east of the town. (O 1626 reads …Valeri e l’enurs sur le Rosne, but the rhymed versions all read Valence and Whitehead (1942, 48) emends O accordingly; Bédier’s discussion and rejection of Jenkins’ proposed identification of Valeri as Saint-Vallier (a district of Valloire in the Drôme) may be found in Romania 64 (1938), 238s. Valentir, Valentis cf. Velantif Valfondee 2324; Val Fondee 5712: an unidentified (probably imaginary) Saracen city or region (cf. Justin de Valfondee, Brum de Val Fondee (the ms. graphies have been retained) Valfonte 5313a (V4): an imaginary Saracen place or country, probably identical with Valfondee. Mortier (1941, 101) reads Valfonde in V4 3430; Beretta (1955) V4 3430 has Valfonte Valie 7353: there is no town or area of this name (nor does V4’s variant ‘Varie’ exist). It could be a corruption of either Val de Loire or, alluding to a major tributary of the R. Sarthe, Val de Loir: Val de Loir > Val Loir > Valloir > Valie, the ending being modified to suit the rhyme. Duggan (2005, 519) identifies this in the note to Critical Text line 7535 as the battle-cry of the Angevins, citing also evidence of its identification with the valley of Beaufort, on the right bank of the Loire between Angers and Tours. However, the place-names relative to this journey are listed in chronological order and correspond to the ancient ‘route d’Espagne’: from Blaye, Charlemagne’s army travelled first north to Saintes, then to St Jean d’Angely and on to Poitiers in the north-eastern corner of Poitou. Taking the main road (today the N10) to Tours would mean (in the late twelfth century, when this section of the poem was written) passing through a ‘southward bulge’ of Anjou and on through Touraine to Tours, where the road crossed the Loire and continued northward to Chartres (the army stopping overnight at Bonneval, just south of Chartres). Thus C’s and V7’s Valie must lie between Tours and Bonneval. The main place-names today on this route are Château Renault, Vendôme and Châteaudun; but the road just south and north of Châteaudun closely follows the winding River Loir in its valley. Thus Valie might be a corruption of Val Loir, already modified to suite the rhyme Val Mortal 2860: an imaginary Saracen country (the Vale of Death, perhaps) Valproissié 5307: an imaginary Saracen country, peopled with giants Valsegnee 4547: an imaginary Saracen country Val Serie 6900: an imaginary Saracen country, ‘Vale of Shadow’, dreamt up by Charlemagne to give substance to his story to Aude, hiding Roland’s death from her (cf. Val Dormanz) Valsevree 5388: an unidentified location, forming the boundary of the kingdom of the Saracen Carmilleu de Dorree, Baligan’s brother. If Dorree (q.v.) equates with Albania, this could possibly be either the Gulf of Corinth or the Sea of Marmora Val Tenebre 4285: the Val Tenebrus of O 2461, a valley down which the Saracens retreat from Roncevaux towards Saragossa; Whitehead (1942, 175) suggests the valley of the R.

618

Proper Names

Irati (the gorge of Foz), referring to a note in Jenkins’ edition of O, and Boissonnade (1923, 143s.) Valterne cf. Vauterne Val Troblé 4524; Val Troblee 5381: an unidentified ‘troubled valley’ within Baligan’s dominions. In view of all the fighting in the Holy Land prior to and during the Crusades, this might possibly be a reference to the Jordan valley Vauterne 241; Valterne 976, 1309: a Saracen town in Spain, probably Valtierra near Tudela, an historically significant town in the Spanish Crusades which commanded an important cross-roads (cf. Boissonnade 1923, 106s.; cf. also Estormis de Valterne) Velantif obj. 1148, 1523, 1577, 2051, 2713, 3569, 3617; Velantis subj. 1868, 3406; Velantir obj. 3589, 3610; Valentir obj. 3740; Valentis subj. 5431: the name of Roland’s horse Vermendoi 7567: the County of Vermandois in north-east France of which St.Quentin (Pasde-Calais) was the chief town (cf. Richart de Vermendoi) Viene 1666, 4144, 6176, 6490, 6497, 6607, 6889, 7297, 8073; Viane 7709: Vienne in the Rhône Valley, the ancient city just south of Lyon, the whole region surrounding it being the fief of Girart de Viene (q.v.) Vilance 5791: an unidentified locality, presumably in France; perhaps a graphy of Valence on the Rhône, just south of Vienne (cf. Fochiers de Vilance, Valence2) Virfalle v. Gifeüs

Y Yhesu Crist subj. 657, 3973, 7758; Yhesu obj. 1633, 2031, 4178, 5774, 7247, 8028; Jhesu obj. 3684, 3696, 3967; Yhesus subj. 2121; Jhesus subj. 2585, 2659, 3809, 5039; Jhesui indir.obj. 4268a (V7); Jesus subj. 4873, 6038, 6222; Yesu obj. 7126, 7223; Jesu obj. 7975: Jesus Christ with variant graphies based on the early abbreviation of the name Gésu: YHS and IHS were both used in the thirteenth century and both graphies are found in C, written in full and in part, as well as the two forms of the abbreviation Ysaïe 3811: the great Old Testament prophet, Isaiah, whose writings are found in the book of the same name; as one of ‘the Just’ who lived before the birth of Christ, he was released from Hell by Christ, during the three days before the Resurrection, according to the popular belief as exemplified in the legend of the ‘Harrowing of Hell’ Ysorez 6326: a rich baron at Charlemagne’s court and Otes’s close friend and confidant

V Bibliography V. Editions (Roland) Das Altfranzösische Rolandslied, Text von Châteauroux und Venedig VII, ed. Wendelin Foerster, Heilbronn, Verlag Gebr. Henninger, 1883 (Altfranzösische Bibliothek, 6). La Chanson de Roland, ed. Cesare Segre, Milano/Napoli, Riccardo Ricciardi, 1971 (Documenti di filologia, 16). La Chanson de Roland, édition critique, ed. Cesare Segre, Genève, Droz, 1989. La Chanson de Roland nel testo assonanzato franco-italiano, ed. Giuliano Gasca Queirazza, Torino, Rosenberg & Sellier, 1955 (L’orifiamma, 1). La Chanson de Roland, Oxford ms., ed. Frederick Whitehead, Oxford, Blackwell, 1942. The Franco-Italian Roland (V4), ed. Geoffrey Robertson-Mellor, Salford, University of Salford, 1980. The Song of Roland: the French Corpus, Part 3: The Châteauroux-Venice 7 Version, ed. Joseph J. Duggan, Turnhout, Brepols, 2005. The Song of Roland: the French Corpus, Part 4: The Paris Version, ed. Annalee C. Rejhon, Turnhout, Brepols, 2005. The Song of Roland: the French Corpus, Part 5: The Cambridge Verson, ed. Wolfgang van Emden, Turnhout, Brepols, 2005. Il testo assonanzato franco-italiano della Chanson de Roland: cod. Marciano fr. IV (= 225); edizione interpretativa e glossario, ed. Carlo Beretta, Pavia, Università degli studi di Pavia, 1995. Les Textes de la Chanson de Roland, ed. Raoul Mortier, Paris, Éditions de la Geste Francor, 1940–1944.

V. Secondary Literature and Dictionaries Aebischer, Paul, Rolandiana Borealia, La Saga af Runzivals bardaga et ses dérivés scandinaves comparés à la Chanson de Roland, Essai de restauration du manuscrit français utilisé par le traducteur norrois, Lausanne, Publications de la Faculté des Lettres, F. Rouge & Cie S.A., Librairie de l’Université, 1954. Battisti, Carlo/Alessio, Giovanni (edd.), Dizionario Etimologico Italiano, 5 vols., Firenze, Università di Firenze, Istituto di Glottologia, 1950–1957. Bédier, Charles Marie Joseph, Les Légendes Épiques: Recherches sur la formation des Chansons de Geste, 4 vols., Paris, Librairie Ancienne H. Champion, 1914–1921 (La Chanson de Roland in vol. 3, 1921). Bédier, Charles Marie Joseph, La Chanson de Roland, Commentaires, Paris, Édition d’Art H. Piazza, 1927. Beretta, Carlo, Per la localizzazione del testo rolandiano di V4, Medioevo Romanzo 10 (1985), 225–248. Beretta, Carlo, Studio sui rapporti fra i manoscritti rimati della «Chanson de Roland», premessa di Cesare Segre, Potenza, Facoltà di lettere e filosofia, Università degli studi della Basilicata, 2001.

620

Bibliography

Bibliotheca Sanctorum, Istituto Giovanni XXIII della Pontificia Universita Lateranense, vol. 10, Roma, Imprimatur Tusculi, Tipografia «Mariapoli» di Città Nuova Editrice, 1968. Bisson, Sebastiano, Il fondo francese della Biblioteca Marciana di Venezia, Roma, Edizioni di storia e letteratura, 2008. Boissonnade, Prosper, Du nouveau sur la Chanson de Roland. La genèse historique, le cadre géographique, le milieu, les personnages, la date et l’auteur du poème, Paris, Champion, 1923. Bosi, Roberta, Lo Studium a Bologna, in: Massimo Medica/Stefano Tumidei (edd.), Duecento: Forme e colori del Medeoevo a Bologna, Venezia, Marsilio, 2000. Braghirolli, Willelmo/Meyer, Paul/Paris, Gaston, Inventaire des Manuscrits en langue française possédés par Francesco Gonzaga I, Capitaine de Mantoue, mort en 1407, Romania 9 (1880), 497–500. Brasseur, Annette, Etude linguistique et littéraire de la «Chanson des Saisnes» de Jehan Bodel, Genève, Droz, 1990. Brook, Leslie C., The Alexandrine Sequence in the Post-Oxford «Roland» Manuscripts, Olifant 15,1 (1990), 15–32. Brunel-Lobrichon, Geneviève, Mise en page et format des MSS littéraires du XIIIe siècle en occitan conservés à la Bibliothèque Nationale de Paris, Revue des Langues Romanes XCVIII,1 (1994), 115–126. Busby, Keith, Codex and Context: Reading Old French verse narrative in manuscript, Amsterdam, Rodopi, 2002. Colliot, R., Adenet le Roi, vol. I, Berte aus grans piés, Paris, Éditions A & J Picard, 1970. Coniglio, Giuseppe, Mantova: La Storia, vol. I, Dalle origini a Gianfrancesco, Primo Marchese, Mantova, Istituto Carlo d’Arco per la Storia di Mantova, 1958. Coniglio, Giuseppe, Mantova: Le Lettere, vol. I, Mantova, Istituto Carlo d’Arco per la Storia di Mantova, 1959. Coniglio, Giuseppe, I Gonzaga, Milano, dall’Oglio, 1967. D’Arcais, Francesca Flores, Les illustrations des manuscrits français des Gonzague à la Bibliothèque de Saint-Marc, in: Essor e fortune de la chanson de geste dans l’Europe et l’Orient latin, Actes du IXe Congrès international de la Société Rencesvals pour l’étude des épopées romanes, Padoue-Venise, 29 août–4 septembre 1982, vol. 2, Modena, Mucchi, 1984, 585–616. DEAF = Heidelberger Akademie der Wissenschaften (ed.), Dictionnaire étymologique de l’ancien français, founded by Kurt Baldinger, Tübingen/Québec (today Berlin/Boston), De Gruyter, 1974ss. Degenhart, Bernhard/Schmitt, Annegrit, Frühe angiovinische Buchkunst in Neapel: Die Illustrierung französischer Unterhaltungsprosa in neapolitanischen Scriptorien zwischen 1290 und 1320, in: Friedrich Piel/Jörg Traeger (edd.), Festschrift für Wolfgang Braunfels, Tübingen, Wasmuth, 1977, 71–92. Destrez, Jean, La Pecia dans les manuscrits universitaires du 13e et du 14e siècles, Paris, Éditions Jacques Vautrain, 1935. La Entrada en España: poema épico del siglo XIV en franco-italiano: «El cantar de Roldàn», preliminary study plus full-colour facsimile and translation of Marciana ms. Fr.Z.21 (= 257), ed. Carlos Alvar, Valencia, Ediciones Grial, S.L./Venezia, Biblioteca Nazionale Marciana, 2003. Faccioli, Emilio, Mantova: Le Lettere, vol. I, Mantova, Istituto Carlo d’Arco per la Storia di Mantova, 1959. Faral, Edmond, Les Jongleurs en France au Moyen Age, Paris, Champion, 1910.

Secondary Literature and Dictionaries

621

Faral, Edmond, La Chanson de Roland, Étude et analyse, Paris, Mellottée, 1932. Faral, Edmond, Le manuscrit 19152 du fonds français de la Bibliothèque Nationale: proverbes populaires, fabliaux, etc., with photocopy of the original ms., Paris, Droz, 1934. FEW = Wartburg, Walther von, et al., Französisches Etymologisches Wörterbuch. Eine darstellung des galloromanischen sprachschatzes, 25 vols., Bonn et al., Klopp et al., 1922– 2002. La Folie Lancelot, a hitherto unidentified portion of the Suite du Merlin, contained in mss. B.N.fr. 112 and 12599, ed. F. Bogdanow, Tübingen, Niemeyer, 1965 (Beihefte zur Zeitschrift für romanische Philologie, 109). Fouché, Pierre, Le verbe français (Etude morphologique et historique du français), Paris, Éditions Klincksieck, 1967. Foulet, Lucien, Petite Syntaxe de l’Ancien Français, Marseille, Champion, 1982. La Geste Francor: edition of the «Chansons de Geste» of ms. Marciana Fr.XIII (=256) with glossary, introduction and notes, ed. Leslie Zarker Morgan, Tempe, Arizona Center for Medieval and Renaissance Studies, 2009. La Geste Francor di Venezia, Edizione integrale del codice XIII del Fondo francese della Marciana, ed. Aldo Rosellini, Trieste, Editrice la Scuola, 1986. Girart de Roussillon, ed. Winifred Mary Hackett, Paris, Société des Anciens Textes Français, 1953. Girart de Vienne, chanson de geste, edited according to ms. BXIX-Royal of the British Museum, ed. Frederic G. Yeandle, New York, Columbia University Press, 1930. Godefroy, Frédéric (ed.), Dictionnaire de l’ancienne langue française et de tous ses dialectes, du IXe au XVe siècle, composé d’après le dépouillement de tous les plus importants documents manuscrits ou imprimés, 10 vols., Paris (Abbeville), 1880–1902. Guilleminot-Chrétien, Geneviève, La Dispersion des livres de Versailles (collections princières et dépôt littéraire) de 1789 à 1804, in: Antoine Cohen (ed.), Actes du XVIIe Congrès International de Bibliophilie, 1991, Paris, International Association of Bibliophiles, 2003, 91–125. Hasenohr, Geneviève, Les chansons de geste, in: Henri-Jean Martin/Jean Vexin (edd.), Mise en page et mise en texte du livre manuscrit, Paris, Éditions du Cercle de la Librairie-Promodis, 1990, 239–244. Hassell, James Woodrow Jr., Middle French Proverbs, Sentences and Proverbial Phrases, Toronto, Pontifical Institute of Mediæval Studies, 1982. Henry, Albert (ed.), Les œuvres d’Adenet le Roi, Bruxelles, Presses Universitaires de Bruxelles, 1963. Historical French Reader, Medieval Period, edd. Paul Studer/Edwin George Ross Waters, Oxford, Clarendon Press, 1924. Holmes, Urban Tigner, Jr., Adenet le Roi’s «Berte aus Grans Piés», ed. with introduction, variants and glossary by Urban T. Holmes (with facsimiles), Chapel Hill, University of North Carolina, 1946. Holtus, Günter, La versione franco-italiana della «Bataille d’Aliscans»: Codex Marcianus FR.VIII [= 252], Tübingen, Niemeyer, 1985 (Beihefte zur Zeitschrift für romanische Philologie, 205). Holtus, Günter, L’État actuel des recherches sur le franco-italien: corpus de textes et description linguistique, in: François Suard (ed.), La Chanson de Geste: Écriture, Intertextualités, Translations, Paris, Université de Paris X, Nanterre, 1994 (Cahiers du département de français, Centre des Sciences de la Littérature, Littérales, 14), 147–171.

622

Bibliography

Horrent, Jules, La Chanson de Roland dans les littératures française et espagnole au moyen âge, Paris, Gembloux, 1951. Illuminated manuscripts in the Bodleian Library, edd. Otto Pächt/Jonathan James Graham Alexander, Oxford, Clarendon Press, 1970. Intavulare, Tavole di Canzonieri Romanzi (serie co-ordinata da Anna Ferrari), I Canzonieri Provenzali, vol. 5, Oxford, Bodleian Library: S (Douce 269), ed. Luciana Borghi Cedrini, Città del Vaticano, Biblioteca apostolica vaticana, 1998ss. Jacob-Hugon, Christine, L’œuvre jongleresque de Jean Bodel: l’art de séduire un public, Bruxelles, De Boeck Université, c.1998. Jacoby, David, La littérature française dans les états de la Méditerranée orientale à l’époque des croisades: diffusion et création, in: Essor e fortune de la chanson de geste dans l’Europe et l’Orient latin, Actes du IXe Congrès international de la Société Rencesvals pour l’étude des épopées romanes, Padoue-Venise, 29 août–4 septembre 1982, vol. 2, Modena, Mucchi, 1984, 617–646. de Joinville, Jean, Chronicles of the Crusades and of Louis IX, king of France, Bohn’s Antiquarian Library, 1897. Jourdain de Blaye (Jourdains de Blavies): chanson de geste, ed. Peter F. Dembowski, Paris, Champion, 1991 (Les classiques français du Moyen Age, 112). Langlois, Charles Victor, Saint Louis, Philippe le Bel, les derniers Capétiens directs, 1226– 1328, in: Ernest Lavisse et al. (edd.), Histoire de France jusqu’à la Révolution, vol. 3, pt. 2, Paris, Hachette, 1901, 18–102. Langlois, Marie-Louis Ernest, Table des noms propres de toute nature compris dans les chansons de geste imprimées, etc., Paris, Librairie Émile Bouillon, 1904. Le Strange, Guy, The lands of the eastern caliphate: Mesopotamia, Persia and Central Asia from the Moslem conquest to the time of Timur, Cambridge, Cambridge University Press, 1930. Lestocquoy, Jean, Aux origines de la bourgeoisie: Les Villes de Flandre et d’Italie sous le gouvernement des patriciens (11e au 15e siècles), Paris, Presses Universitaires de France, 1952. Lettere bibliografiche Autografe deli abate M L Canonici al abate Carlo Maria Masnago, 1788–1793, British Library ms. Add. 26.059. Letters to John Strange, 1780, 1782, British Library ms. Add. 23730. f.349. Letters to Sir John Strange, British Consul at Venice, 1758–1785, British Library ms. Eg. 2002. f.149. Maillard, Jean, Charles d’Anjou, roi-trouvère du XIIIe siècle, Musica Disciplina 21 (1967), 7– 66. Malacarne, Giancarlo, Araldica Gonzaghesca, la storia attraverso i simboli, Giancarlo Malacarne; presentazione di Carlos Gonzaga di Vescovato; saggio introduttivo di Giuseppe Papagno, Modena, Bulino, 1993. Malacarne, Giancarlo, I Gonzaga nella storia e nell’arte, Modena, Bulino, 1994. Malacarne, Giancarlo, I Gonzaga di Mantova, una stirpe per una capitale europea, Modena, Bulino, 2004. de Mandach, André, Naissance et Développement de la Chanson de Geste en Europe, Genève, Droz, 1961. Manuel de paléographie latine et française du 6e au 17e siècle, suivi d’un dictionnaire des abréviations, accompagnée d’un album de 24 planches, ed. Jean Maurice Prou, Paris, Alphonse Picard, 1924. Marchesan, Angelo, Treviso Medievale, Treviso, Tipografia Funzionali, 1923.

Secondary Literature and Dictionaries

623

Marchesan, Angelo, L’Università di Treviso nei secoli xiii e xiv e cenni di storia civile e letteraria della città in quei tempi, Sala Bolognese, Forni, 1974. Marcon, Susy, Estudio Artístico de las Miniaturas, cf. La Entrada en España 2003. Martinet, Suzanne, Montlaon: reflet fidèle de la montagne de Laon de 1100 à 1300, Laon, Imprimerie du Courrier de l’Aisne, 1972. Martinet, Suzanne, Laon, ancien capitale de la France, Laon, Imprimerie du Courrier de l’Aisne, 1989. The Medieval French Roman d’Alexandre, Version of Alexandre de Paris, 6 vols., ed. Edward C. Armstrong, Princeton, Princeton University Press/Paris, Presses Universitaires de France, 1937–1976. Merolle Tondi, Irma, L’abate Matteo Luigi Canonici e la sua biblioteca: i manoscritti Canonici e Canonici-Soranzo delle biblioteche fiorentine, Roma, Institutum Historicum Soc. Iesu, 1958. Meroni, Ubaldo, [catalogue for] Mantua, Biblioteca comunale, 15 settembre–10 ottobre 1966: Mostra dei Codici Gonzagheschi, 1328–1540, Mantova, Biblioteca comunale, 1966. Millet, Hélène, Les Chanoines du Chapitre cathédral de Laon, 1272–1412, Roma, École française de Rome, 1982. La miniatura a Padova: dal Medioevo al Settecento, progetto e coordinamento scientifico, [catalogue] edd. Giovanna Baldissin Molli/Giordana Canova Mariani/Federica Toniolo, Modena, Franco Cosimo Panini, 1999. Mitchell, J.B., Trevisan and Soranzo: some Canonici manuscripts from two eighteenth-century Venetian collections, Bodleian Library Record 8.3 (1969), 125–135. Moisan, André, Répertoire des noms propres de personnes et de lieux cités dans les chansons de geste françaises et les œuvres étrangères dérivées, Genève, Droz, 1986. Morawski, Józef, Proverbes français antérieurs au quinzième siècle, Paris, Champion, 1925. Moreno, Paola, Sui manoscritti veneziani del «Foucon de Candie». Contributo per una nuova edizione, Medioevo Romanzo 17 (1992), 197–199. Moreno, Paola, La tradizione manoscritta del «Foucon de Candie», Napoli, Liguori, 1997. Moschini, Giovanni Antonio, Della letteratura veneziana del secolo XVIII fino a’ nostri giorni, vol. 2, Venezia, Dalla Stamperia Palese, 1806. Murano, Giovanni, Opere diffuse per «exemplar» e pecia, Turnhout, Brepols, 2005 (Textes et Etudes du Moyen Age, 29). Novati, F., I codici francesi de’ Gonzaga secondo nuovi documenti, Romania 19 (1890), 161– 200. Paccagnini, Giovanni, Mantova: Le Arti, vol. I, Mantova, Istituto Carlo d’Arco per la storia di Mantova, 1960. Palumbo, Giovanni, Per la Datazione del «Roland rimato», Medioevo Romanzo 27 (2003), 353–412. Paris, Gaston, [article on the 1407 catalogue in Mantua], cf. Braghirolli/Meyer/Paris 1880. Paris, Gaston, Mélanges linguistiques: Latin vulgaire et langues romanes, langue française. Notes étymologiques, Paris, Champion, 1909. Perriccioli Saggese, Alessandra, I romanzi cavallereschi miniati a Napoli, Napoli, Società editrice napoletana, 1979. Pescasio, Luigi, Sordello da Goito, Mantova, Bottazzi Suzzara, 1993 (Storia di Mantova, 5: Profili). Pope, Mildred K., From Latin to Modern French with special consideration of Anglo-Norman Phonology and Morphology, Manchester, Manchester University Press, 1934.

624

Bibliography

Prou, Maurice, Manuel de paléographie latine et française, accompagné d’un album de 24 planches, Paris, Picard, 41924. Rajna, Pio, La geste francor di Venezia, codice marciana XIII della serie francese, ed. Biblioteca Marciana, Milano, Bestetti, 1925. Rejhon, Annalee C., Laon, centre religieux, et la tradition rolandienne, in: Philip E. Bennett et al. (edd.), Charlemagne in the North: Proceedings of the Twelfth International Conference of the Société Rencesvals, Edinburgh, 4–11 August 1991, Edinburgh, 1993, 197– 207. Robertson-Mellor, Geoffrey, Langue et littérature franco-italiennes; état des études et problèmes, Table-ronde co-ordonnée par G. Robertson-Mellor, in: Essor e fortune de la chanson de geste dans l’Europe et l’Orient latin, Actes du IXe Congrès international de la Société Rencesvals pour l’étude des épopées romanes, Padoue-Venise, 29 août–4 septembre 1982, vol. 2, Modena, Mucchi, 1984, 785–807. Romances et Pastourelles des 12e et 13e siècles, ed. Karl Bartsch, Leipzig, Vogel, 1870 (Darmstadt, Wissenschaftliche Buchgesellschaft, 1967). Roman de Troie. Par Benoit de Sainte-Maure. Publié d’après tous les manuscrits connus, ed. Léopold Albert Constans, Paris, Société des Anciens Textes Français, 1904–1912. Rosellini, Aldo, Di nuovo sul valore della traduzione della «Chanson de Roland» del ms. di V4, Studi Francesi IV (1960), 1–10 (= 1960a). Rosellini, Aldo, Étude comparative des manuscrits de Châteauroux et de Venise VII de la Chanson de Roland, Moyen Age LXVI (1960), 259–300 (= 1960b). Rosellini, Aldo, Rolandiana Marciana, Firenze, Olschki, 1962. Rosier, J., L’évolution du Roland (thèse de doctorat, 1958), Bibliothèque de la Sorbonne, Réserve ref. W 1958 I09(1–2), 4o (unpublished). Rossi, L., L’œuvre de Jean Bodel et le renouveau des littératures romanes, Romania 112 (1991), 312–360. Rossi, Vittorio, La biblioteca manoscritta del senatore veneziano Jacopo Soranzo, in: Il libro e la stampa, vol. I, Milano, Società Bibliografica Italiana, 1907. Rossi, Vittorio, I codici francesi di due biblioteche veneziane del Settecento, in: Scritti di Critica Letteraria, vol. III, Dal Rinascimento al Risorgimento, Firenze, Sansoni, 1930. Rychner, Jean, La Chanson de Geste – Essai sur l’art épique des jongleurs, Genève, Droz/ Lille, Librairie Giard, 1955. Schultz-Gora, Oskar, Einleitung, Folque de Candie nach den festländischen Handschriften, vol. IV, ed. Ulrich Mölk, Tübingen, Niemeyer, 1966 (Beihefte zur Zeitschrift für romanische Philologie, 111). Segre, Cesare, La tradizione della «Chanson de Roland», Milano/Napoli, Ricciardo Ricciardi, 1974. Steele, Robert, The Pecia, London, The Bibliographical Society, 1938. Talos, Ion, Mioriţa şi epica medievală franceză, România literară 18 (1992), 7. Tobler, A., Li proverbe au vilain, Leipzig, Hirzel, 1895. Tobler, Alfred/Lommatzsch, Ernst (edd.), Altfranzösisches Wörterbuch, 11 vols., Stuttgart, Franz Steiner Verlag, 1925–1989. La versione franco-italiana della «Bataille d’Aliscans», Codex Marcianus FR.VIII (= 252), testo con introduzione, note e glossario, ed. Günter Holtus, Tübingen, Niemeyer, 1985 (Beihefte zur Zeitschrift für romanische Philologie, 205). de Voyer d’Argenson, Marc-Antoine-René, marquis de Paulmy, La bibliothèque universelle des romans: ouvrage périodique dans lequel on donne l’analyse raisonnée des romans anciens et modernes, françois ou traduits dans notre langue; avec des anecdotes et des

Catalogues

625

notices historiques et critiques concernant les auteurs ou leurs ouvrages; ainsi que les moeurs, les usages du temps, etc., Paris, Démonville/Lacombe, 1775–1789 (Paulmy was contributor to the issues of November and December 1777 and March 1779). Zorzi, Marino, La libreria di San Marco: il lascito Recanati, Milano, Mondadori, 1987.

V. Catalogues Catalogo de’ Libri che saranno posti in vendita nella città di Padova l’anno 1780, 3 vols., held in Libreria di San Marco, Marciana, Reference: 118.C.171–173. Catalogue des bibliothèques nationales établies à Versailles dans les dépôts de la Commission des arts du département de Seine-et-Oise et cataloguées par Citoyen BUFFY (handwritten), vol. IX, Ref. 5389: Catalogue des livres provenant de chez LOUIS CAPET, condamné…, 1792. Inventaire du Cabinet du roi Louis XVI à Versailles, établi en 1775, held in the Bibliothèque Nationale de l’Arsenal, Paris, ref. nouvelles acquisitions françaises 2622.