«09 
 
 UC-NRLF 
 
 *B bED E3M 
 
opular Studies in 
 Mythology Ro- 
 mance &> Folklore 
 
 6d. net each 
 
 ■'"^O. I O **< The Romance ^ 
 Cycle of Charlemagne and ] 
 his Peers. By Jessie L. Weston | 
 
 i 
 Second Edition ' 
 
 Published by David Nutt, at the 
 
 Sign of the Phoenix, Long Acre, London 
 
 1905 
 
p 
 
 opular Studies in My- j 
 thology, Romance I 
 and Folklore 1 
 
 ISSUED UNDER THE GENERAL 
 EDITORSHIP OF 
 
 MR. ALFRED NUTT 
 
 PAST PRESIDENT OF THE FOLKLORE SOCIETY 
 
 AUTHOR OF "STUDIES ON THE LEGEND OF THE HOLY GRAIL '' 
 
 AND ^' THE VOYAGE OF BRAN " 
 
 Each i6mo. 6d. net (yd. post free) 
 
 ^ This series is intended to impart, in a concise and 
 accurate, but thoroughly popular manner, information 
 concerning the subjects of which it treats. Each study 
 gives a broad survey of the theme, unembarrassed by 
 notes or references. Bibliographical appendices are 
 provided for those who wish to carry the study of these 
 subjects further. 
 
 The Editor ventures to point out that nowhere else can 
 the English reader obtain information given in a concise 
 and popular form, yet based upon and in accordance 
 with the latest researches of scholarship concerning the 
 subjects of which the Series treats. H e also ventures 
 to call attention to the fact that the most favourable 
 notices of the Popular Studies are those due to the 
 learned press of the Continent and America. 
 
THE ROMANCE 
 
 CYCLE OF CHARLEMAGNE 
 
 AND HIS PEERS 
 
 BY 
 
 JESSIE L. AVESTON 
 
 SECOND EDITION 
 
 londo:n^ 
 
 DAVID NUTT, 57-59 LONG ACRE 
 1905 
 
K 
 
 
 \^ S^t:JowLK . 
 
PREFACE TO FIRST EDITION 
 
 The following study has been designed as a com- 
 panion to the earlier essay on " King Arthur and 
 his Knights" No series dealing mth Mediaeval 
 Romance could claim to he in any sense complete 
 loere the great cycle tohich gave an impetus to the 
 evolution of European romantic literature omitted, 
 A complete account is as yet hardly possible however; 
 many of the p^^ncipal texts are still unedited, 
 and students are largely dependent upon travaux 
 d'ensemble completed many years ago. But the 
 process of editing texts is proceeding steadily, and 
 students who desire a closer a/^uaintance with the 
 cycle will see from the Bibliography appended that 
 a considerable amount of material is already 
 available; we may reasonably hope that a few 
 years will place vs in possession of critical editions 
 
 300675 
 
4 *"' ' ' ' ' PRESAGE 
 
 of all the leading texts of the Cha7'lemagne and its 
 subsidiary cycles, 
 Pakis, March 1901. 
 
 A few slight changes have been made in the text, i 
 
 and the Bibliography has been brought up to date, \ 
 
 othermse this second issue is unaltered, ^ 
 
 JESSIE L. WESTON. \ 
 
 Paris, September 1905. | 
 
THE 
 
 CHARLEMAGNE ROMANCES 
 
 " Ne sont que trois vuttUres d nvJ home entcndcmt 
 Be France, de Bretagne, et de Home le grant." 
 
 The Middle Ages were, as we know, the ages of 
 Romance ; Romance embodied in Prose — pseudo- 
 historic chronicles, pseudo-biographical accounts 
 of noted heroes; in Poetry — short lais, longer 
 poems (metrical romances as we call them), some 
 independent, the greater number falling into 
 groups round some one central figure, and in their 
 entirety forming what we call cycles of Romance. 
 To the mind of a writer of the twelfth century, 
 whose words are quoted above (Jean Bodel, author 
 of La Chanson des Sais7ies), there were three of 
 such cycles, and to them alone might the atten- 
 tion of a poet of that day be worthily directed ; 
 and of these cycles the respective centres were 
 Charlemagne, Arthur, and Alexander. 
 
 To-day this seems a somewhat inadequate 
 method of classification, ignoring as it does the 
 great mass of Northern tradition (Siegfried is 
 
6 THE CHARLEMAGNE ROMANCES ] 
 
 surely a hero worthy of attention), yet it pro- ] 
 vides those who pursue the study of mediaeval I 
 literature with a useful formula of designation | 
 for the two great bodies of French romance, the \ 
 cycles of Charlemagne and of Arthur; the Matieres \ 
 of France and of Britain. 
 
 In the introductory number of these studies i 
 the Charlemagne romances have been alluded \ 
 to, and incidentally discussed, but the subject 1 
 matter of the study, The Influence of Celtic \ 
 upon MedicBval Romance, was naturally far more \ 
 closely connected with the second branch of \ 
 Romantic literature, the Arthurian cycle ; with j 
 Charlemagne Celtic legend has but little to j 
 do. In its later stages, when the Matiere de \ 
 France came into contact with the Arthurian j 
 story, the very soul of which is Celtic, it j 
 borrowed certain features from the Matiere de \ 
 Bretagne, but even then the fairy element, ] 
 inseparable from the latter, presents itself \ 
 partially, at least, under a Teutonic form. It \ 
 was the Matiere de Bret ague rather than that \ 
 of France which was discussed in the opening ; 
 study of this series. In the fourth number the j 
 various romances constituting the Arthurian \ 
 cycle were described and classified, and in the 1 
 fourteenth the Grail romances, forming a dis- \ 
 tinct section of that cycle with which their \ 
 connection is late and artificial, was similaily ; 
 
 1 
 
 \ 
 
THE CHARLEMAGNE EOMANCES 7 
 
 treated ; but so far the Charlemagne cycle has 
 not received the notice which its importance 
 demands. 
 
 It is natural that alike to English writers 
 and English readers the cycle which Jean Bodel 
 reckoned second in value should stand first in 
 charm and attraction ; indeed, it may be doubted 
 whether those for whom he wrote did not judge 
 even as we do; in purely literary value the 
 Arthurian cycle is probably superior to that of 
 Charlemagne ; the latter can count on its roll no 
 such names as those of Chretien de Troyes, Hart- 
 mann von Aue, or Wolfram von Eschenbach.^ 
 
 So far as the French literary presentment is con- 
 cerned, the Charlemagne cycle is the elder, and 
 the poems composing it, though the versions that 
 have descended to us are not the earliest versions 
 of the tales they tell, are as a rule cast in a form 
 more primitive than that adopted by the writers 
 of the Arthurian cycle. The prevailing form of 
 the French Arthurian romance, one not found 
 before the twelfth century, is a poem of eight 
 syllabic lines, each pair rhyming ; whereas in the 
 Charlemagne Chansons de Geste we find laisses or 
 tirades of varying length distinguished by a mono- 
 rhyme, with, in the earliest copies, a vowel asso- 
 
 1 In saying this I do not ignore the high epic vahie of the 
 Chanson de Roland, I rather refer to a conscious effort after 
 perfection of literary form. 
 
8 THE CHARLEMAGNE ROMANCES 
 
 nance prevailing throughout the laisse. This 
 metrical form may be compared with the allitera- 
 tive verse which in the Germanic languages was 
 replaced by various forms of the rhyming couplet or 
 stanza. Yet, in so far as the subject matter is 
 concerned, dealing as it does largely with mythic 
 and pre-historic elements, the Arthurian cycle 
 may be deemed the older. 
 
 Before entering into a detailed discussion of the 
 romances dealing with the Matiere de France^ 
 it will, I think, be not unprofitable to make clear 
 to our own minds the distinctive characteristics 
 of these two great bodies of romance ; both of 
 them of extreme importance in the history of 
 literature, yet differing so widely the one from 
 the other that even where they have come into 
 contact the influence resultant has been of the 
 slightest and most superficial character. In 
 general terms we may express this difference by 
 saying that the groundwork of the Arthurian 
 cycle is mainly mythical, that of the Charlemagne 
 cycle mainly historical. This does not imply that 
 there are no historical elements in the former 
 cycle, and no mythical in the latter, or that, as 
 contrasted with the great Emperor of the Franks, 
 Arthur is a mere creation of the imagination. ' 
 On the contrary, in all probability the leading lines 
 of the Arthur-legend proper, the King's fights with 
 the Saxons, his betrayal by wife and nephew, and 
 
THE CHAKLEMAGNE ROMANCES 9 
 
 death in battle have a foundation in fact, while 
 the Charlemagne of legend is in many respects a 
 wide departure from the Charlemagne of history. 
 But the real charm and abiding fascination of the 
 Arthurian story lies in the realm of fancy and not 
 of fact — realms, perhaps we should rather say, for 
 the student of Arthurian romance is free of more 
 than one kingdom : the land of faery whose 
 horizon is lost in the mists of Celtic heathen- 
 dom, and the brighter, but no less elusive, land 
 where ideal chivalry has sworn a close alliance 
 with Christian mysticism. 
 
 It is true that not all chronicled in the Charle- 
 magne romances has its parallel in historic reality ; 
 myth has certainly played a part in the stories of 
 the hero's birth and early trials, but in its main 
 lines the character of these romances is determined 
 by historic facts. Such heroes as Huon de Bor- 
 deaux and Girard de Yiane may be creatures of 
 imagination, but the struggles of the feudal nobles 
 against their over-lord are facts of stern reality ; 
 Vivien may never have lived, and rashly vowed, 
 and shed his blood heroically at the gates of Aries, 
 but at least the varying fortunes of the contests 
 between Christian and Saracen for the fertile 
 lands of Southern France are as historical as the 
 fights of our ancestors with Saxon and Dane ; and 
 if Ganelon never betrayed his king and country, 
 yet Roland died at Roncevaux. 
 
10 THE CHARLEMAGNE ROMANCES 
 
 Again, the Charlemagne story has its super- 
 natural element, but it is not that of wizardry 
 and faery as in the Arthurian story ; there is no 
 shape-shifting Merlin, no treacherous Morgain, 
 or beneficent Lady of the Lake. Nor is it laden 
 with wondrous hints and revelations of divine 
 doctrines and mysteries as in the Grail romances. 
 The supernatural machinery is celestial and strictly 
 simple ; a guardian angel watches by the em- 
 peror's pillow, and shields him from his foes ; in 
 answer to his prayer a hart shows his army the 
 ford across the swollen stream ; at his death St. 
 James of Compostella is beheld in a vision cast- 
 ing into a balance, wherein the devil weighs the 
 emperor's good and evil deeds, the churches and 
 shrines Charlemagne has erected in his honour. 
 
 When, in the last stage of its develop- 
 ment, the fairy element enters the Charlemagne 
 cycle it is manifestly due to the influence of 
 Arthurian romance ; thus Huon of Bordeaux is 
 aided by Oberon, the fairy king (who is, indeed, 
 rather a Teutonic elf {alhe) than a Celtic fairy), 
 but Oberon is the son of Morgain, and the right- 
 ful heir to his kingdom, with whom Huon must 
 come to terms, is Arthur. Ogier and Renouart 
 alike live on in fairyland (though each is sup- 
 posed to have ended his days as a monk !), but 
 that fairyland is Avalon. The supernatural 
 element proper to the legend is presented under 
 
THE CHAKLEMAGNE EOMANCES 11 
 
 the simplest and most obvious form, that of direct 
 Divine protection. 
 
 That the characterisation of the Charlemagne 
 cycle should be more forcible than that of the 
 Arthurian is only what we should expect; the 
 authors of the Chansons were dealing with real 
 men and women, like to, if not of, themselves. 
 Charlemagne plays a far more important role than 
 does Arthur. The British king is, after all, little 
 more than a picturesque centre for a series of 
 adventures in which he himself takes no part. 
 He certainly leads his hosts to battle, but it must 
 be admitted that the wars of the Arthurian story 
 are its least interesting and most wearisome 
 portion ; otherwise, Arthur presides in a dignified 
 manner at feasts, and invites adventures, which 
 his knights achieve ; as a personality he is not 
 convincing. 
 
 And of his knights Gawain, with all his grace 
 and courtesy, has about him that note of elusive- 
 ness that makes one realise that his proper 
 destination is, like Arthur, the land of faery. 
 Lancelot is but a stage lover ; Galahad a painted- 
 window saint. Perceval and Tristan, as we meet 
 them first, are indeed human, very creatures of 
 flesh and blood, but the Arthurian story is not 
 content to leave them so, the former it turns into 
 a being scarcely less shadowy than Galahad, the 
 latter into a lover as conventional as Lancelot. 
 
12 THE CHAKLEMAGKE KOMANCES 
 
 But in the Charlemagne romances it is otherwise. 
 The old Emperor, with his long white beard, is a 
 majestic figure, which even the hint of years im- 
 possibly prolonged cannot rob of its reality. His 
 intense family affections, his uncontrolled temper, 
 violent fits of rage, savage revenge and unreason- 
 ing tyranny are all reaL) We feel the relation- 
 ship between him and Roland to be no mere 
 literary convention. The younger man, with his 
 fierce temper, indomitable pride, and reckless 
 courage, is exactly what we should expect Charle- 
 magne's next of kin to hej Oliver, equally brave 
 but less hot-headed, ready to temper his valour 
 with discretion, is quite as real as his friend. 
 Yery real, too, that doughty champion of the 
 church militant. Archbishop Turpin ; and Gane- 
 lon, whose treason is in truth the attempt of a 
 cowardly man to revenge himself upon one who 
 has thrust him against his will into a post of 
 danger. 
 
 *' Rollanz m'fors-fist en or et en aveir, 
 Pur que jo quis sa mort et sun destreit ; 
 Mais traisun nule n'en i otrei." 
 
 Convincing, too, is William of Orange ; now 
 battling valiantly against the overpowering force 
 of his Saracen foes ; now melting into tenderness 
 over his dying nephew ; and again wrathfully 
 demanding aid from his pious and peace-loving 
 
THE CHARLEMAGNE ROMANCES 13 
 
 brother-in-law, King Louis — who wishes himself i 
 
 otherwhere. The impression left upon us is that \ 
 
 if these heroes did not really live, they might well ] 
 
 have done so. We are not surprised that in his j 
 
 journey through the other world Dante beheld a \ 
 
 goodly group of souls of the Charlemagne heroes, J 
 while of the Arthurian he saw none, save Tristan. ^^Jj>fi^j^ ■ 
 Apart from theii* literary interest, the Charlemagne fy^^^^jF^t 
 cycle appeals rather to the student of History, the fjiJc^I 
 Arthurian cycle to the student of Folk-lore. ^^^r^f>^^"^ , 
 
 The literary development of the two cycles not I 
 
 only sets the above-noted differences in a vivid | 
 
 light, but illustrates their true nature, and enables ' 
 
 us to realise the history reflected in the Charle- :; 
 
 magne cycle. The great Emperor died in 814, \ 
 
 and with him died, as we can see, the conception | \ 
 
 of a France forming an integral portion of a vast | I 
 
 Germano-Roman Empire. The warriors who fol- ' 
 
 lowed Clovis and Dagobert, the companions of \ 
 
 Charles the Hammer and of Pepin the Little, had 1 
 
 in the course of centuries been putting off their 1 
 
 Germanhood, been differentiating themselves from ^ 
 
 their kinsmen across the Rhine. The popular j 
 
 songs commemorating the mighty feats of Mer- \ 
 
 wing and Karling kings, songs of which monkish \ 
 
 chroniclers have preserved us a few scraps in J 
 
 their barbarous verse, or of which they have 1 
 
 partially rendered the substance in their dreary '\ 
 
 prose — these songs, originally German, gradually \ 
 
U THE CHARLEMAGNE ROMANCES 
 
 passed into Roman, the language of the conquered 
 race, as the Germanic element weakened. After 
 Charlemagne's death his empire broke up, the 
 Roman portion was cut definitely loose from the 
 German-speaking world, and in less than a cen- 
 tury the last traces of German speech vanished. 
 The descendants of Frankish and Burgundian 
 conquerors became French, and every fragment 
 of German hero-song either put on a Roman dress 
 or else died out. Small wonder if in the process the 
 historic basis was shifted, if the deeds of earlier 
 chiefs and warriors got transferred to the great 
 Emperor. ( The fame and achievements of the 
 latter would indeed have sufficed to inspire popular 
 minstrels ; but he also inherited the renown of 
 many predecessors, and thus the earliest singers 
 of his glory found themselves from the outset in 
 possession of no inconsiderable stock of poetic 
 material. The songs accumulated during the 
 ninth century, and the decadence of the later 
 Carolingians, threw into stronger relief the prowess 
 and fame of Charlemagne-y 
 
 CAs early, perhaps, as the first- third of the 
 tenth century, certainly by the middle of the 
 century, Chansons de Geste, as distinguished from 
 the popular songs on which they were based, had 
 begun to appear, professing to narrate events of 
 Charlemagne's lifetime^ ^Throughout the tenth, 
 eleventh, and early twelfth centuries these Ghan- 
 
THE CHARLEMAGNE ROMANCES 15 
 
 sons were being enlarged, worked over, adapted 
 to cyclic requirements^ These two hundred years 
 were fertile in strong characters, in fierce passions, 
 in events and movements which transformed the 
 old Franko- Roman Empire into modern France. 
 The Carolingian polity decayed and passed away ; 
 the Capets, embodying the aspirations and ideals 
 of a new nationality, rose to power, and founded 
 a monarchy destined to last for eight hundred 
 years, and to incarnate, far more than was the 
 case in England, the national genius. The pangs 
 and throes which accompanied the birth of modern 
 France were fierce and prolonged ; Norman and 
 Saracen assailed from without; king and feuda- 
 tory grappled in deadly struggle within. All this y 
 we find mirrored in the Chanson de Geste, Itself ) 
 the record of a nation's formation, it exercised, , 
 we cannot doubt, a formative influence, the force I 
 of which it were hard to overestimate. Germanic 
 in its pristine essence as it was, Germanic as it 
 remained in many of its animating ideas, it is in 
 its highest moments a magnificent record of French ' . 
 patriotic feeling, an ardent fosterer of devotion to 
 the fair land apostrophised by Roland — 
 
 " Tere de France, mult estes dulz pais I " 
 
 From this, the creative period of the French epic, 
 
 we possess comparatively little in an authentic 
 
 and ungarbled form. Chief of what has come 
 
 down to us is the earliest version of the Chanson 
 
16 THE CHARLEMAGNE ROMANCES 
 
 de Roland, and even this, there can be little 
 doubt, represents a fairly advanced stage of de- 
 velopment. The great bulk of Charlemagne 
 romances belong to a period reaching from the 
 early twelfth into the fourteenth century. The 
 main outlines of the Chanson de Geste had been 
 determined, its leading types of character and 
 incident had been settled, it had acquired a pro- 
 digious stock of conventions, it still in a large 
 measure reflected the religious and social ideas 
 of the time ; thus it could not escape the hands 
 of the adapter, the rearranger, the hack writer 
 who thought more of dressing up time-honoured 
 stories according to the literary fashion of the 
 moment than of preserving their original spirit 
 and form. From the middle of the twelfth cen- 
 tury it was exposed to the competition of the 
 Arthurian stories — a competition against which, 
 as we have seen, it largely defended itself by 
 adopting the tone and style and temper of its rival. 
 How different was the fate of the Arthurian 
 romance on French soil ! It came into French 
 hands with a stock of incidents and characters, 
 above all with an aesthetic, and what, in default of 
 a better term, must be styled an ethical character 
 of its own, which persist despite the modifications 
 imposed by the alien French genius. Its period 
 of evolution is comparatively short ; in from fifty 
 to one hundred years it runs its full course ; its 
 
THE CHARLEMAGNE ROMANCES 17 
 
 development is not determined by nor does it 
 mirror the political situation or the political 
 changes of the period. Yast and far-reaching 
 social changes it does indeed herald and record, 
 but indirectly and symbolically, not, as is the case 
 with the Chanson de Geste, directly and realisti- 
 cally. The one body of literature is a monument 
 of French intellect and French artistry exercising 
 themselves upon an alien and imperfectly compre- 
 hended subject matter ; the other isCtlie nation 
 typified, recording as it does its fierce birth-pangs, 
 its wild and dour enfances, the exultant spirit of 
 its early manhood^ 
 
 It would be difiicult to exaggerate the popular- 
 ity, in mediaeval times, of the Charlemagne cycle, 
 or its importance as a factor in the history of 
 European literature. In Italy it was the parent 
 of a literature scarcely less extensive than that 
 from which it sprang ; indeed, the evolution was 
 more complete. Italy yields pseudo-historical 
 chronicles and metrical romances representing 
 the legend at every stage : from that of historic 
 reality, as typified by the rough -hewn figures of 
 Roland and Oliver at the portal of Verona Cathe- 
 dral, to that of pure fantasy, as in the Orlando 
 Amoroso and Orlando Furioso. In Spain the 
 Charlemagne story, as related in the Chronica of 
 Alfonso X., gave impetus to the formation of a 
 national cycle, the heroes of which — Bernardo del 
 
 B 
 
18 THE CHAELEMAGNE KOMANCES 
 
 Carpio, the children of Lara, and the Cid — should ' 
 rival in popularity the heroes of the earlier gestes. } 
 In Scandinavia and in Germany the romances ) 
 found translators and imitators, while in England j 
 we fail to realise that our Sir Bevis of Hampton is : 
 but an imitation of a French poem, and is reckoned 
 by scholars as an offshoot of the cycle ; while a j 
 nobleman and statesman like Lord Berners thought '\ 
 the translation of the tale of Huoii of Bordeaux a \ 
 task not unworthy of his time and labour. And \ 
 have not we here in England the honour of pos- j 
 sessing, in the MS. of the Bodleian Library, the ! 
 oldest known copy of the most famous song of the J 
 cycle, the Chanson de Roland ? \ 
 
 To undertake to give, in the small compass of ; 
 one of these studies, an adequate account of \ 
 so large and important a body of literature j 
 (M. Leon Gautier, in his Epopees Fran^aises, ] 
 reckons (eighty chansons as belonging to what | 
 he terms " la geste du roi " alone, without con- ; 
 sidering the subordinate cycle of the Narbonnais) I 
 would of course be impossible; the utmost that j 
 can be done is to describe the general character ; 
 of the cycle, the lines into which it falls, and note i 
 the romances which will best repay the attention j 
 of the ordinary student of literature. Those who ; 
 desire a more detailed account will do well to con- | 
 suit M. Leon Gautier's monumental work, Les i 
 Epopees FrangaiseSf or the shorter but no less \ 
 
THE CHARLEMAGNE ROMANCES 19 
 
 scientific, and, it may be, better arranged Histoire 
 Poetique de Charlemagne, by M. Gaston Paris. 
 
 The dividing line of the Charlemagne romances 
 is less individual than in the case of the Arthurian 
 cycle ; there, it is comparatively easy to classify the 
 romances according to the knight who is hero of 
 the tale. The leading heroes of the Round Table 
 form so many centres round which the romances 
 respectively group themselves, and the collective 
 mass of these smaller groups or subsidiary cycles 
 make up the great Arthurian legend. But with 
 Charlemagne and his peers this guiding principle 
 will no longer serve us. There are certainly many 
 romances borrowing their title from the hero of the 
 adventures they relate, but none of Charlemagne's 
 warriors save William of Orange, the Marquis au 
 court nez, have anything like such a body of 
 romance connected with them as have Gawain, 
 Perceval, or Lancelot, or can fairly be described as 
 hero of a "cycle." The Charlemagne romances 
 deal rather with families than with individuals ; 
 they are Chansons de Geste,^ lays dealing with the 
 
 1 The translation of the word Geste is somewhat difficult ; 
 the meaning appears to have been originally chronicles = feats, 
 then the feats or actions of a particular family = family or 
 race. In this sense M. Gautier employs it, but I incline to 
 think that the earlier meaning is the more correct; the con- 
 cluding words of the Chanson de Molind, "Ci fait la Geste 
 que Turoldus declinet," cannot possibly have the signification 
 of race or family. 
 
20 THE CHARLEMAGNE ROMANCES 
 
 feats of a race rather than of a person. The ten- 
 dency is to look upon qualities, not as the individual 
 characteristics of one member of a family, but as 
 the natural and inevitable inheritance of all. 
 Valour and loyalty, cowardice and treachery, alike 
 pass from father to son. Thus one group of poems 
 deals with the heroic virtues of the descendants of 
 Garin de Montglane, another with the treacherous 
 race of Doon de Mayence. 
 
 But the more convenient method of classification, 
 that followed by M. Gaston Paris, is to group the 
 romances according to their subject matter as re- 
 lating to the Emperor, for Charlemagne, as we have 
 noted above, plays a far more important part in his 
 cycle than does Arthur. Following these lines, we 
 shall find one group of poems dealing with the 
 personal history of the monarch, his birth, his 
 youthful adventures, his domestic trials, his fabled 
 journey to the East, and final coronation of his son 
 as his successor. A second and more important 
 group deals with his various wars, principally those 
 with the Saxons and the Saracens,^ and is connected 
 with the subsidiary but highly interesting cycle of 
 the Narbonnais, the heroic family of Aimeri de 
 Nar])onne, whose son, William of Orange, is the 
 
 1 For romantic purposes the wars with the Lombards prac- 
 tically do not count, as the authors of the romances have 
 largely confounded them with the Saracens. We shall refer 
 to this again in connection with Ogier le Danois. 
 
THE CHARLEMAGNE ROMANCES 21 
 
 champion of Christianity against the Moslem 
 invaders of the South of France. The third sub- 
 division includes the romances which relate the 
 internecine struggles of the great vassals with their 
 over-lord, and counts among its number some of the 
 most popular legends of the whole cycle. 
 
 While thus practically following historical lines, 
 the compilers of the chansons have, however, by 
 no means limited themselves to events occurring 
 during the reign of the great Emperor, but freely 
 transfer incidents from one period to another at their 
 pleasure, ascribing to Charlemagne's reign what really 
 happened under his predecessor, Charles Martel, or * 
 his successor, Charles le Chauve, and presenting the 
 heroes of the gestes as living now under Charle- 
 magne, now under his son Louis, thus involving the 
 attainment of a truly patriarchal age. According 
 to the author of the Chanson de Roland, Charle- 
 magne was over two hundred years old at the date 
 of Roncevaux, and, to rightly understand the his- 
 torical background of the cycle, we must bear 
 in mind that the condition?, social and political, 
 there represented actually obtained for some three 
 hundred years or so, and were by no means 
 limited to the period covered by the reign of the 
 son of Pepin. The anachronism exists, but it is 
 not of such a nature as to destroy the value of the 
 poetical representation. 
 
 In the romances dealing with the youth of 
 
22 THE CHAKLEMAGISrE ROMANCES j 
 
 Charlemagne^ we are on mythical rather than on \ 
 historical ground. The story of his mother, Berte \ 
 aux grans pies, is the familiar and oft-told tale of i 
 "The False Bride," the waiting-maid substituted j 
 for her mistress, and as such belongs to the domain i 
 of Folk-lore. Equally the tale of his youthful 
 adventures, when he flies from the death by poison 
 prepared for him by the sons of the false maid, ' 
 and under the name of Mainet takes refuge with ] 
 the King of Spain, frees him from his enemies, 
 and marries his daughter, is a creation of fiction, \ 
 and has no historical basis. The chronicle of \ 
 . Eginhard distinctly states that nothing definite j 
 concerning Charlemagne's youth was known, and \ 
 the author therefore judged it inadvisable to write \ 
 of it. But it seems doubtful whether the name : 
 by which the great Emperor is known did not 1 
 take its rise in this popular fiction, and Oharle- ■ 
 magne be not derived from Charles Mainet, the \ 
 two names being often coupled together, rather \ 
 than from Garolus Magnus. In any case it is to \ 
 be regretted that so many English writers of the \ 
 present day substitute the common-place transla- I 
 tion Cliarles the Great, for the time-honoured and \ 
 far more impressive Charlemagne. \ 
 
 These tales, and other scattered legends relating j 
 
 \ 
 1 I only refer in the text to the more important members of j 
 each subdivision ; they will be found fully enumerated in the \ 
 bibliographical appendix. 
 
THE CHAELEMAGNE EOMAls^CES 23 
 
 to the personality of the great Emperur, are to be 
 found in the vast compilation of the Venice Library, 
 consisting of a number of the Chansons collected 
 together under the name of Charlemar/ne ; also in 
 the Icelandic Karlomagnus Saga, which latter, how- 
 ever, begins the record of his adventures at a rather 
 later date. A German poem, Karl Meinet, has pre- 
 served the account of his residence in Spain. 
 
 Purely fabulous, too, are the accounts of Charle- 
 magne's journey to Jerusalem, accompanied by his 
 twelve peers, and the extravagant feats they per- 
 form in fulfilment of their gabs or boasts ; and of 
 the false accusation of his Queen Blanchefleur, by 
 the traitor Macaire — a version of which, under 
 the title of La reine Sihille, enjoyed a widespread 
 popularity. 
 
 The real interest of the legend lies not in Charle- 
 magne's domestic life, but in his public actions; 
 the energy with which he defended Christianity, 
 and consolidated the Empire. This is, as we have 
 said above, the historical element of the legend 
 which, reflected in the romances, constitutes the 
 distinctive feature and real importance of the cycle. 
 The Emperor's object was obtained only at the 
 cost of wars, foreign and domestic, and with such 
 struggles the majority of the romances are con- 
 cerned. 
 
 For poetical purposes the foes of the Emperor 
 beyond his border were the Saxons and the 
 
24 THE CHARLEMAGNE ROMANCES 
 
 Saracens ; both were alike enemies of God and 
 of Holy Church, but the poems dealing with the 
 latter are not only more numerous, but strike a 
 stronger and a truer note. This may, of course, 
 be largely owing to the fact that much of the 
 struggle was fought out on the soil of France, and 
 the reality of the contest was thus more forcibly 
 brought home to the imagination of the writers. 
 The gulf of nationality, too, was wider ; there was 
 less difference between the barbarous Saxon and 
 semi-civilised Frank, both white races, than between 
 the latter and the dusky hordes that swarmed from 
 Africa through Spain into Southern France, even 
 though these latter might be representatives of a 
 civilisation older than that of the West. Wotan 
 and Thor, barbarous as were their rites, never seem 
 to have raised half as much horror and antagonism 
 in the minds of mediaeval Christians as did the 
 fabulous gods of the Saracens, Mahmoud, Terma- 
 gant, and Apollo ! In mediaeval romance the 
 iconoclastic followers of Islam are represented as 
 idolaters of a monstrous type, a quintessence of all 
 the evils of paganism and heathenism, and they are 
 provided with a motley pantheon borrowed from 
 classic tradition, supplemented by the fertile ima- 
 gination of romancers. Over and over again these 
 heathen hordes are represented as besieging Rome, 
 sometimes as having gained possession of the Im- 
 perial city, and holding in their power the most 
 
THE CHARLEMAGNE ROMANCES 25 
 
 precious relics of Christendom. More than once 
 Charlemagne marches to the relief of VApotre — as 
 the Pope is generally termed in French romance — 
 which relief is as a rule effected by a single combat 
 between one of the Christian Paladins, and a giant 
 more or less malicious, more or less willing to be 
 converted, representing the Saracen host. It is 
 the "motif" of David and Goliath repeated ad 
 nauseam. 
 
 For such romances as Aspremont, Les Knfances 
 Of/ier, and FierahraSj there is no real foundation in 
 history. The most that can be said is that they 
 represent a distorted reminiscence of the siege of 
 Rome by the Lombards. 
 
 But when we come to the group of poems dealing 
 with Charlemagne's expedition to Spain, and culmi- 
 nating in the Chanson de Roland, we are on surer 
 ground ; history has indeed been modified under 
 the influence of the Saracen nightmare, but we are 
 dealing with modification, not with invention. 
 Briefly related, the facts as chronicled by Eginhard 
 and others are these: — In 778 two Moorish emirs 
 from Spain presented themselves before the Em- 
 peror and declared their desire to become his 
 vassals. Encouraged ])ythis, Charlemagne marched 
 with a large army into Spain, besieged and took 
 Pampeluna, and laid siege in vain to Saragossa. 
 (This part of the expedition is found, much em- 
 bellished, in the following romances : U Entree en 
 
26 THE CHARLEMAGNE K0MA:NCES 
 
 Espagne, La Prise de Pampelune, and Chii de 
 Bourgogne.) On the return of the array to France, 
 the rear guard was surprised by the Gascons in the 
 defile of Roncevaux, in the Pyrenees, and practi- 
 cally exterminated, Roland, prefect of the marches 
 of Brittany, being among the slain. M. Leon 
 Gautier remarks that this defeat must have been 
 of far more importance than the chroniclers care 
 to admit. Certain it is, that they pass over in but 
 few words an event which has left an indelible 
 impression on the popular mind, and the echoes 
 of which can be caught at every subsequent 
 period of French history. It is probable, too, that 
 the Saracens lent a helping hand to the Gascon 
 ambuscade. It is certain that tradition has for- 
 gotten the real authors of this shattering blow to 
 the Emperor's prestige, and attributes it to the 
 hereditary foe of Christianity, the Moslem. 
 
 But whatever be the true history of Roncevaux, 
 the legend is the culminating point of the Charle- 
 magne tradition. French scholars have vied with 
 each other in praise of the Chanson de Roland, its 
 dignity, its simplicity, and the lofty tone of 
 courage, devotion, and patriotism which inspires 
 it, and any unprejudiced critic must largely agree 
 with them. It is not the work of a finished poet 
 like Chretien de Troyes, it has not the easy literary 
 grace which marks the lais of Marie de France, but 
 the force and directness of its language, and the 
 
THE CHARLEMAGNE ROMANCES 27 
 
 universality of the feelings to which it makes 
 appeal, can never fail to awnke a response. We 
 sympathise alike with Roland in his desire to fight 
 unaided the unequal combat; with Oliver in his 
 calmer appreciation of the overwhelming odds 
 against them, and his vain attempts to induce his 
 headstrong friend to realise the truth; with Arch- 
 bishop Turpin as he solemnly absolves the doomed 
 army, and, having thus performed his duty as a 
 Christian and cleric, gives valiant account of him- 
 self as man and warrior. All alike are inspired by 
 one spirit, by the desire that none shall hereafter 
 sing male cangun regarding theii* end. 
 
 Perhaps the most impressive and affecting part 
 of the poem is the lament of Charlemagne over the 
 dead body of his heroic nephew, when in pathetic 
 words he paints the picture of his return to France, 
 how he shall sit throned in the hall of Laon, 
 and the representatives of the races subdued by 
 Roland^s aid shall come before him and ask tidings 
 of the valiant captain of his host, and he must 
 needs answer, " In Spain he lieth dead ! " Then 
 they, taking courage at the tidings, shall rebel 
 against him, and who shall put them down? The 
 poem might well have ended here, as indeed, in 
 the earlier versions, it doubtless did. The defeat 
 of the Saracen army, and the punishment meted 
 out to Ganelon and his race, come somewhat as an 
 anti-climax. 
 
28 THE CHARLEMAGNE ROMANCES 
 
 It may be worth while to ask here, what is the 
 historic foundation for the heroic character of 
 Roland ? The chronicle of Eginhard, relating the 
 catastrophe of Roncevaux, simply says : " Ansel- 
 mus comes palatini, et Hruodla7idus Britanniei 
 limitis prcefediis, cum aliis compluribus inter- 
 ficiuntur." ^ Thus, here, Roland is simply prefect 
 of the marches of Brittany, and no word is said ^ 
 of his relationship to the Emperor. It may seri- 
 ously be doubted whether such a relationship did, 
 in fact, exist. History records that Charlemagne 
 had but one sister, who early became a nun, and 
 thus could not possibly have been the mother of 
 Roland. The relationship of uncle and nephew, 
 as subsisting between the royal centre of an epic 
 cycle and the hero of that cycle, is so general {e,g. 
 the instances of Conchobar and Cuchulinn, Finn 
 and Diantiid, Mark and Tristan, Arthur and 
 Gawain), that it does not seem improbable that 
 the Charlemagne legend may have been affected 
 by the prevailing tradition. This supposition is 
 strengthened by the fact that certain twelfth 
 century texts represent Roland as not merely the 
 nephew, but also the son of the Emperor, a feature 
 manifestly borrowed from tradition, and highly 
 primitive in character. Such is the relation be- 
 tween Sigmund and Sinfiotli in the Yolsunga- 
 
 1 Eginhard, Vita Caroli IX., quoted by M. Gautier, 
 Epopees Fran Raises, ii. p. 363. 
 
THE CHARLEMAGNE EOMANCES 29 
 
 saga, Arthur and Mordred (in the first instance 
 it was probably Gawain), and, in some versions 
 of the story, between Conchobar and Cuchulinn. 
 Thus, while we may take it as settled that history 
 determined the character and fate of Roland, it 
 yet seems probable that his relationship to Charle- 
 magne was due to the influence of mythic tra- 
 dition. 
 
 The twelve peers, Roland's companions, who, 
 according to the poem, shared his fate at Ron- 
 cevaux, owed their origin, M. Gautier considers, 
 to Germanic custom. Among primitive German 
 tribes it was the rule for certain warriors to asso- 
 ciate themselves closely with the chief of the clan, 
 to share with him his dangers and his spoil. 
 They were his pairs. Hence, M. Gautier ^ thinks 
 the douze pairs, their number being an imitation of 
 that of the Apostles. M. Gaston Paris is, however, 
 inclined to consider the institution of later date. 
 The names of the peers vary in different poems, 
 and two of the most famous of Charlemagne's 
 warriors — Naimes de Baviere and Turpin — do not 
 appear to have belonged to this body. According 
 to Girard de Viane, it was Naimes who persuaded 
 the Emperor to institute the order, as a kind of 
 superior tribunal of judgment (cf. supra). The 
 extreme popularity of the peers is shown by the 
 introduction of their title into English mediaeval 
 ^ Epopees Frant^aiseSt vol. ii. p. 173. 
 
30 THE CHARLEMAGNE ROMANCES 
 
 romance, where we often find the word " dosypere " 
 as equivalent for a valiant knight. 
 
 Compared with the Chanson de Roland, the 
 poems dealing with the Saxon wars, Guiteclin and 
 La Chanson des Saisnes^ are far inferior in interest, 
 marked by inordinate length and wearisome repeti- 
 tion of incident. 
 
 The third group of romances, those which 
 relate the story of the Emperor's struggle with 
 his rebellious vassals, is, taken as a whole, the 
 most interesting of the three. Two among the 
 number, Renaud de Montauhan (les quatre jils 
 Aymon) and Huon de Bordeaux, were in all pro- 
 bability the most popular and widely known of 
 the Charlemagne romances, and have more or less 
 retained that popularity to our own time. 
 
 Good mediaeval translations of both are published 
 by the Early English Text Society. The former 
 gives a very fine picture of the relations between 
 a vassal and his feudal lord, and the manner in 
 which, among the nobler natures of the time, the 
 obligations imposed by feudal service were realised 
 and fulfilled. Charlemagne is entirely in the wrong 
 in his treatment of the four brothers, but the old 
 knight, Aymon, feels himself compelled by his oath 
 of fealty to extend no aid or countenance of any 
 kind to his sons. When in dire need they throw 
 themselves upon the protection of their mother, 
 who receives them with open arms, Aymon leaves 
 
THE CHARLEMAGNE ROMANCES 31 
 
 the castle at their disposal and goes forth ; he will 
 not break his vow by aiding them, nor will he 
 forbid his wife to follow the instincts of natural 
 affection. Renaud, the principal hero of the tale, 
 has as keen a sense of honour as his father ; when 
 his clever and resourceful cousin Maugis, whose 
 wiles have been the salvation of the brothers, casts 
 the Emperor into a magic slumber, and thus con- 
 veys him into the castle he has been besieging, 
 Renaud refuses to profit by what he deems a 
 disloyal action, and sends Charlemagne again to 
 his host in safety : a forbearance which, it must 
 be owned, the Emperor's conduct does not jus- 
 tify ! The four sons of Aymon and their gallant 
 steed, Bayard, were deservedly popular; indeed, 
 in folk tradition Bayard still roams the forests of 
 Ardennes. 
 
 Here we may point out the gradual declension 
 which the character of the Emperor, as repre- 
 sented in the romances, undergoes. In the 
 Chanson de Roland he is a venerable but an im- 
 posing and dignified figore ; in Renaud de Mon- 
 tauhan and Huoii de Bordeaux he is capricious, 
 tyrannical, given to fits of senile rage, cruel and 
 unjust in the highest degree ; his barons openly 
 flout him, and the authors do not hesitate to stig- 
 matise him as un vieil radote. How are we to 
 account for so fundamental a change of concep- 
 tion ? It seems clear that it was due to historic 
 
32 THE CHARLEMAGNE ROMANCES 
 
 causes, and was the outcome of a radical change 
 in the relations between sovereign and subject. 
 Under the feebler rule of the great Emperor's 
 successors the power of the feudatory barons be- 
 came increased to an alarming extent. The later 
 romances, faithfully reproducing the character- 
 istics of their age, have shifted the point of 
 interest from the feeble and vacillating monarch 
 to the rebellious but powerful vassal. 
 
 If the authors had maintained throughout 
 the identity of the king during whose reign the 
 romance was compiled, or remodelled, the pic- 
 ture would have been complete; but the posi- 
 tion of Charlemagne as centre of the Matiere 
 de France was so firmly grounded, that they con- 
 tinued to retain him as representative of a system 
 entirely alien to his methods. The relations be- 
 tween William of Orange and King Louis, in 
 Aliscans, are quite possible, and a legitimate and 
 artistic presentment of the situation as conceived 
 under the reign of that king ; postulated of 
 Charlemagne they are incorrect and misleading. 
 The character of the Emperor has really suffered 
 from the continued popularity of his cycle, and 
 the need of adjusting the romances to contem- 
 porary social conditions. 
 
 The romance of Ogier le Danois, consisting of 
 no fewer than twelve branches, belongs, in so far as 
 the older portion is concerned, to the earlier and 
 
THE CHARLEMAGNE ROMANCES 33 
 
 better period of the Charlemagne cycle, but it is 
 somewhat marred by the barbarous fierceness and 
 savagery of the hero. 
 
 Nevertheless, certain portions of the story have 
 an epic force and vigour which raise them to the 
 first rank of romantic legend. Such is the account 
 of the prolonged siege of Chastelfort by the 
 Emperor, a siege lasting for over seven years, 
 during the progress of which all Ogier's men are 
 slain ; but the undaunted hero makes figures of 
 wood, and clothing them in the armour of the 
 dead knights, succeeds in deluding his foes into 
 the belief that the castle is fully garrisoned. 
 Also the charming story of the recognition of 
 Ogier, after many years' imprisonment, by his 
 faithful steed, BroiefoH^ which has been made the 
 draught-horse of the neighbouring monastery, but 
 retains sufiicient spirit to carry its aged master to 
 victory once more. 
 
 This story of a hero and his faithful steed was 
 extremely popular in mediaeval times, and we find 
 it ascribed to Walter of Aquitaine (Waltharius) 
 in the Chronica Novalense, an interesting monkish 
 compilation of romantic legend ; to Heimi, in the 
 Thidrelc Saga ; and to William of Orange, in the 
 Moniage Guillaume ; in the two first cases the 
 hero being a monk, and in the third a hermit — 
 not a prisoner, as Ogier. The monastic version 
 M. Gaston Paris holds to be the earliest form of 
 
 C 
 
34 THE CHARLEMAGNE ROMANCES 
 
 the story, of which the Chronica probably gives 
 the oldest extant version, the Thidi^ek Saga being 
 probably borrowed from a Lombard chronicle. 
 
 In the older parts of the Ogier romance we have 
 an account of Charlemagne's war with the King 
 of the Lombards, at whose court the hero seeks 
 shelter. This is an historic feature, and of the 
 more value in that, as we have noted above, the 
 tendency of the later romances is to ignore the 
 wars with the Lombards, and in the traditions 
 relative to the siege of Rome to replace them by 
 the Saracens. 
 
 The character of the hero appears to be more or 
 less founded on fact ; there was certainly at Charle- 
 magne's court a valiant soldier of the name of 
 OggeriuSy or Otkar, but his nationality is doubtful. 
 Certain chronicles speak of him as of the family 
 of Pepin, in which case he would, of course, be a 
 kinsman of the Emperor. The title Danois is by 
 some modern scholars held to be a misreading of 
 the original Ardennois, and Ogier is thought to 
 have been of the Ardennes rather than of Den- 
 mark. In its final stages the tale shows distinct 
 traces of Celtic influence ; and this modern scholars 
 have strongly felt. Mr. Nutt, in the study already 
 referred to, remarks that Huon and Ogier are 
 '* Arthurian heroes who have strayed by accident 
 to the court of Charlemagne ; " and the late William 
 Morris, in that fascinating collection of legendary 
 
THE CHAELEMAGNE ROMANCES 36 
 
 tales, The Earthly Paradise, gives the story of Ogier 
 the Dane to a Breton sailor. Nevertheless, in its 
 essential spirit the tale appears to be Germanic 
 rather than Celtic. 
 
 According to the testimony of the chroniclers, 
 Ogier was one of the most popular of mediaeval 
 heroes, but the numerous romances connected with 
 his name have not retained their popularity as liave 
 Renaud de Montanhan and Huon de Bordeaux. 
 To-day most of us probably only know him through 
 the medium of Hans Andersen's tales, though but 
 few realise that the slumbering Danish hero 
 Olge Danske is identical with the paladin of 
 Charlemagne. 
 
 Both the romance of Ogier le Danois and that 
 of Renaud de Montauban, if classified according 
 to the family method suggested by M. Gautier, 
 would belong to the Geste of Doon de Mayence; 
 but the fact that that scholar himself was obliged 
 also to include them in the Geste du roi seems 
 an argument for the simpler method adopted by 
 M. Gaston Paris, and followed in these pages. ^ 
 
 The tale of Huon de Bordeaux is less cliaracter- 
 
 1 The romances belonging to the Doon family are found in a 
 collected form in a MS. of the Montpellier Library. They are 
 the following : — Doon de Mayence^ Gaujrcy^ Les Enfances 
 Ogier, La Ghevalerie Ogier, Aye de Avignon^ Qui de Nanteuily 
 Parise la Duchesse, Maugis d'Aigremontj Vivien Vamachour de 
 MonbranCf Renaud de Montauhan. Of these only the two 
 mentioned above are of the first rank. Parise la Duchesse is a 
 variant of the Berte and Macaire stories. 
 
36 THE CHAELEMAGNE ROMANCES 
 
 istic of any special age. The interest lies rather in 
 the marvellous adventures of the hero, and- the aid 
 and protection extended to him by the fairy king 
 Oberon. It is a tale of faery not only in the loose 
 but in the strict sense of the word, Huon's adven- 
 tures reproducing closely those of the hero upon 
 whom a task is laid which he can only accomplish 
 by supernatural aid familiar to us in so many fairy 
 tales, and as such it is one for all time. From it 
 Shakespeare borrowed his fairy king, and Weber 
 the libretto of his opera. 
 
 Among the romances of this class Girard de 
 Viane, one of the oldest, is interesting as giving 
 the account of the first meeting between Roland 
 and Oliver; they fight themselves into friendship 
 beneath the walls of Viane. 
 
 In most of these tales Charlemagne is represented 
 as in extreme old age, in fact, as we have shown 
 above, an unreasoning dotard, un vieil radote ! We 
 therefore feel that the situation depicted in the 
 Couronnement Looys is natural and inevitable ; it 
 may be considered as practically closing the cycle 
 of Charlemagne and opening that of William of 
 Orange, though there are, of course, poems dealing 
 with the earlier history of that hero. 
 
 In the Couronnement we find the aged Emperor 
 laying aside his crown in favour of his young 
 son Louis, who, gentle and timid in disposition, 
 shrinks from the responsibilities awaiting him. A 
 
THE CHARLEMAGNE ROMAjNCES 87 
 
 certain Hernaud, of the traitorous race of Ganelon, 
 comes forward with an offer to rule the kingdom 
 till Louis feels himself prepared to take up the 
 reins of government, but William Fierahras, de- 
 tecting the traitorous purpose concealed beneath 
 the ofi'er, fells the traitor to the ground, and 
 announces that he will be the. protector and 
 champion of the young king, a task he loyally 
 performs. 
 
 The historic personality underlying the epic 
 figure of this William, the hero of the important 
 cycle of the Narbonnais, is not clear. Investiga- 
 tion discloses even more forcibly than elsewhere 
 in the cycle of how composite a texture it really 
 is, and how it welds together in one picture, 
 periods separated by the stretch of centuries, 
 regions separated by the width of France. 
 Monsieur Gaston Paris considers that the legen- 
 dary hero represents a reminiscence of the feats 
 of at least four historical Williams, i.e. William 
 Fierabras, William au court nez, William of 
 Toulouse, and William of Orange. Of these 
 William Fierabras (who may, but this is doubt- 
 ful, have borne the appellation au court nez) and 
 William of Aquitaine, later of Orange, were con- 
 temporaries of the great Emperor, but the one 
 belonged to northern, the other to southern 
 France. William of Toulouse, undoubtedly an 
 historical character, and one of whom we possess 
 
38 THE CHARLEMACxNE ROMANCES 
 
 a fair amount of authentic record, belonged to 
 the tenth century. His was a striking personality, 
 and he seems to have attracted stories belonging 
 to the earlier William of Orange. Moreover, in 
 the epic, William, when old, turns monk, and 
 here would seem to have borrowed traits from 
 two southern French saints, S. William du Desert 
 and S. William le Pieux, Certain it is that the 
 titles Fierahras, au court nez^ and d^Orange, are all 
 applied to one hero. But in those days William 
 was the most common of Christian names. In 
 a certain assemblage of nobles out of five hundred 
 present three hundred and eighty were Wil- 
 liam ; a fact which goes far to explain any confu- 
 sion of identity which may have crept into the 
 legend. 
 
 According to the romances, however, the parent- 
 age and personality of this William, though his 
 surnames may vary, are distinct enough. He is 
 the son of Aimeri de Narbonne, a descendant of 
 that Garin de Montglane who, as we mentioned 
 at the outset of our study, represents the heroic 
 family or geste of the cycle. An extensive Italian 
 compilation of the fourteenth century, under the 
 title of Storie Nerhonese^ recounts all the doings of 
 the valiant family of Narbonne. 
 
 M. Gautier reckons twenty-four chansons com- 
 posing this cycle, but here we need only enumerate 
 those directly connected with the life and deeds of 
 
THE CHARLEMAGNE ROMANCES 39 
 
 the hero, Les Enfances Guillaume,^ Siege de Nar- 
 
 bonne, Le ChatToi de Nimes, La prise d' Orange. 
 
 These trace the history of William from his earliest 
 
 years to his establishing himself as lord of Orange, 
 
 from which city he has driven out the Saracens, 
 
 and married the wife of their king. The loves of 
 
 William and Orable, who in the later poems is 
 
 known by her baptismal name, Guibourc, occupy 
 
 a great portion of the story. 
 
 At this point another hero appears upon the 
 
 scene, Vivien, the nephew of William, whose 
 
 valiant deeds and untimely death appear to have 
 
 been intended as a parallel to Roland in the earlier 
 
 story. Tlie poems directly connected with this 
 
 young hero are Les Enfances Vivien^ Le Covenant 
 
 Vivien^ and the famous Aliscans; this last being 
 
 the crown and centre of the " William " cycle, even 
 
 as the Chanson de Roland is of that of Cliarle- 
 
 magne. Both poems relate the defeat of the 
 
 Christians by the Saracens, and in both the 
 
 catastrophe is due to the rashness of the youthful 
 
 hero. 
 
 In the Covenant Vivien we learn how the youth, 
 
 on receiving knighthood, makes a solemn vow never 
 
 to retreat before the Saracens, a vow which even 
 
 1 Perhaps it may be well to explain that Enfances is a 
 technical term ajjplied to the account of the deeds of a hero 
 before he receives knighthood. The tendency of later research 
 is to prove that the "William" cycle was of greater import- 
 ance than generally supposed. 
 
40 THE CHAKLEMAGNE ROMANCES 
 
 his uncle William, no model of prudence, condemns 
 as unduly rash. A large Saracen fleet appears on 
 the river beyond the plains of Aliscans (Aliscans- 
 Aliscamps = Elysian Fields, and is the name of 
 the famous cemetery beyond the walls of Aries). 
 Vivien urges on his young comrades to attack them, 
 which the lads do with an wholly inadequate force, 
 and are put to the worse. Under pressure Vivien 
 allows one of his cousins to ride to his uncle 
 William and demand aid, and the poem of Aliscans 
 opens at the conclusion of the fatal struggle. 
 William has seen all his men slain, his young 
 nephews, his brother's sons, taken captive, and is 
 compelled to fly from the field. But first he must 
 know the fate of Vivien ; he seeks him at immi- 
 nent risk to himself, and at last finds the lad 
 mortally wounded, and at the point of death, beside 
 a spring. The scene that follows is exceptionally 
 fine; the count at first yields to a natural out- 
 burst of grief at the death of one so young and 
 valiant, but suddenly he recalls himself to a sense 
 of his duty. No priest is at hand; Vivien is 
 dying fast; it devolves upon William as nearest 
 of kin to render the consolations of religion. The 
 warrior becomes a priest ; taking the dying boy in 
 his arms he rests his head against his breast and 
 bids him confess his sins. Vivien can think of 
 nothing save that he has broken his vow, and 
 retreated before the enemy. William pronounces 
 
THE CHAKLEMAGNE KOMANCES 41 
 
 absolution ; for the first time gives him lepain bent, 
 and commends his soul to God. 
 
 " Dex re9oif s'arme par ton digne commant 
 Qu en ton sierviche est mors en Aliscans I " 
 
 This scene of the first communion and death of 
 Vivien has been held by critics equal, if not 
 superior, to that of the death of Roland. 
 
 The author of Aliscans is not a litemry artist, he 
 repeats himself, indulges in lengthy description, 
 but the subject-matter with which he is dealing in 
 the first half of the poem is exceptionally good, 
 and he rises to the occasion. Very fine is the de- 
 scription of the arrival of William, a fugitive, and 
 alone, disguised in the armour of a dead Saracen, 
 at the gates of his own city of Orange. His wife 
 does not know him in such guise ; William would 
 never have returned without the lads he went to 
 succour ; and not till the pursuers are close on 
 his heels does Guibourc recognise and admit her 
 husband. Then she shows herself the stronger of 
 the two ; it is she who coui forts the Count, broken 
 down by the disaster which has overtakBn his 
 house, and bids him hasten at once to demand suc- 
 cour from King Louis ; she and her maidens dressed 
 in armour, will delude the Saracens into the belief 
 that Orange is fully garrisoned, and keep them at 
 bay. William goes to the court of the king whom 
 he has protected and aided, and who has wedded 
 
45 THE CHAKLEMAGNE BOMANCES 
 
 his sister, only to be treated with scorn and con- 
 tempt. This is one of the finest parts of the 
 romance. Eventually Louis yields, in sheer terror 
 of his truculent brother-in-law, and William is 
 provided with a new army — and the poem with a 
 new hero in the person of Kenouart, the gigantic 
 brother of William's wife, Guibourc, who, stolen 
 from his people in early youth, is acting the part of 
 scullion in Louis' kitchen. 
 
 The latter part of the poem is taken up with the 
 recital of the valiant deeds of Renouart in the 
 second battle of Aliscans, which results in a crush- 
 ing defeat of the Saracens. Finally Eenouart 
 marries Aaliz, the king's daughter, and in the 
 Bataille de Loquifer (a poem probably by the 
 author of Aliscans, but much inferior to that 
 work) is carried off to Avalon, where he combats 
 the monster Chapalu in the presence of Arthur, 
 King of Avalon. 
 
 The poem of Aliscans undoubtedly rests upon 
 historical tradition ; in the opinion of M. Leon 
 Gautier, it represents the welding together of two 
 widely separated events — the defeat of William of 
 Aquitaine by the Saracens at Villedaigne in 793, 
 and the defeat of the Saracens by William I., of 
 Provence, in 976. The leading *' motif" of the 
 geste of the Narbonnais, the long-continued struggle 
 between Christian and Moslem for the South of 
 France, is genuinely historical. 
 
THE CHARLEMAGNE ROMANCES 45 
 
 The end of William's career is related in two 
 romances, both bearing the same name, Le Montage 
 Guillaume, and generally referred to by scholars as 
 Montage I. and Montage II,, in which we read how 
 the hero eventually quitted the world and retired 
 to the hermitage, where (after again issuing forth 
 to combat the enemies of his country) he died in 
 the odour of sanctity. A similar romance bears 
 the name of Renouart, and relates how that hero 
 also became a monk ; but it is impossible to take 
 any real interest in a figure so completely the crea- 
 tion of imagination. Renouart is never more than 
 a serio-comic character, and distinctly out of place 
 beside so strenuous a hero as William ; neverthe- 
 less he appealed to the fancy of the Middle Ages, 
 and was certainly a more living and persistent 
 element in folk-tradition than the far more sym- 
 pathetic Vivien. 1 
 
 With the battle of Aliscans it seems fitting that 
 we should close this brief sketch of the great French 
 cycle. That the Charlemagne romances will ever 
 offer to English students so tempting a field of 
 inquiry as that of the Arthurian legend is doubt- 
 ful. The subject-matter of this latter, consisting 
 as it does largely of the mythical elements which 
 lie at the root of all history and all belief, must 
 
 ^ The story of scullion turned hero seems to have been 
 popular in mediaeval times. There is a version of it in the 
 Low German Thidrek Saga. 
 
44 THE CHARLEMAGNE ROMANCES 
 
 always make au appeal to a wider circle than that 
 represented by professed students of history or 
 literature. We have adopted Arthur as a national 
 hero, and as such take a pride in his name and 
 fame, but the interest of the Arthurian story lies 
 deeper than our interest in Arthur the King, and 
 the ideas he symbolises are not those which in- 
 spire the Matiere de France. The Charlemagne 
 legend, on the other hand, is of direct national 
 interest; it appeals above all to the children and 
 lovers of La douce France. Nor from a literary 
 point of view is it of equal value. Probably the 
 four best romances, in the opinion of literary 
 scholars, would be reckoned to be the Chanson 
 de Roland, Renaud de Montauban, Huon de Bor- 
 deaux, and AUscans ; but not one of these could 
 bear comparison, as a piece of literature, with any 
 one of the masterpieces of Arthujian romance. 
 I have suggested above that it may be owing, in a 
 great measure, to this deficiency in literary form, 
 and consequent failure to satisfy the more ex- 
 acting literary taste of the twelfth century, 
 that the Charlemagne cycle was superseded so 
 completely in popular favour by the Arthurian 
 romances. If, however, we distinguish content from 
 manner, the Matiere de France, as an epic cycle, 
 ranks above the Arthurian, which is not strictly 
 epic ; regarded in this light, the Chanson de Roland 
 has few rivals. And were it only as a picture of 
 
THE CHARLEMAGNE ROMANCES 45 
 
 the impression produced by a great man upon the 
 minds and imaginations of the people of his day, 
 an unrivalled collection of documents showing 
 how fancy deals with facts, and history becomes 
 folk-tale, the Matiere de France would be well 
 worth our study ; our fathers found pleasure in 
 these old stories, and we shall not do ill if we 
 follow their example. 
 
BIBLIOGRAPHICAL APPENDIX i 
 
 [r am mdehted to Mr. Nutt for revising and bringing 
 the Biblio'/rap/iy up to date.] 
 
 The standard works upon the Charlemagne legend as 
 a whole are the two mentioned in the text, Les Epopees 
 Frangaises, by M. Ldon Gautier (4 vols., 1878-82), in- 
 complete, the poems dealing with Doon de Mayence not 
 being included, and LHistoire po6tique de Charlemagne 
 (1865), by M. Gaston Paris. Of the latter, a new edition 
 by M. M. Paul Meyer and J. Bddier has just been pub- 
 lished (17s. 6d.). Les Epopees is out of print, and costs 
 about £5. An indispensable supplement for serious 
 students to M. Gautier's great work is his Bibliographie 
 des Ckansons de Geste, 1877. 
 
 The best popular account in English is still that in Mr. 
 Ludlow's Popular Epics of the Middle Ages (Macmillan, 
 1 865), which is also out of print. A most admirable survey 
 of the cycle by M. L^on Gautier may be found in vol. i. 
 of Petit de Julie ville, Hlstoire de la Langue et de la Litt6ra- 
 ture fratK^aise, 1896 (17s. 6d.). The section in M. Lanson's 
 Histoire de la Litterature frangaise is, like the rest of his 
 work, masterly in the extreme. 
 
 In the list which follows the dates in brackets are those 
 of the recension which has come down to us, as fixed by 
 M. Ldon Gautier in his Epopies Fran^aises. 
 
 ^ Of the poems marked by an asterisk new editions are 
 being prepared. ^g 
 
BIBLIOGRAPHICAL APPENDIX 47 
 
 The romances personally treating of Charlemagne are : — 
 Berte aux grcms pUs (c. 1275), ed. Paulin Paris, 1832. 
 
 Scheler, 1874. 
 Charlemagne (by Gerard d'Amiens). (o. 1300.) 
 Karl Mainet (13th century), ed. \ 
 
 A. V. Keller, Stuttgart, 1858 I German adap- 
 Karl (by der Strieker) (late 13th j tations. 
 
 century), ed. K. Bartsch ) 
 
 Voyage a Jerusalem (c. 1115), ed. E. Koschwitz, 1883 
 
 (4s. 6d.). 
 Ma^adre (late 13th century), pub. M. A. Mussafia, 
 
 1864 (7s. 6d.). 
 Couronnement Looys (late 12th century), ed. E. Lang- 
 
 lois, 1888 (12s. 6d.). 
 Cf. the mediaeval English Lyf of Charles the Orete, 
 
 E. E. T. S., 2 vols! (£1, lis.). 
 
 Compilations and pseudo-Chronicles— 
 
 Charlemagne^ Library of Venice (13th century), com- 
 prising : Beuves d^ Hanstonne ; Enfances Charle- 
 magne; Enfances Rolani ; Enfam.c€s et Chevalerie 
 Oyier le Danois ; Reine Blanchejieur or Macaire. 
 
 Reali di Francia — Italian rifacimento compiled by 
 Andrea da Barberino at the beginning of the 
 15th century. Edd. P. Rajna and G. Vandelli, 
 vols. 1. ii., in 3 vols., Bologna, 1872-1901 (£1, 14s.). 
 
 Karlamagnus §aga^ Old Norse translation of the 
 late 13th century, ed. by C. Unger, Christiania, 
 1860 (8s.). 
 
 Chronicle of pseudo-Turpin (early 12th century), ed. 
 
 F. Castets, 1880. Out of print, and scarce. 
 
 Wars of Charlemagne — 
 
 Aspremont (late 12th century). 
 Simon de Pouille (13th century). 
 
 Acquin (conquest of Brittany) (late 12th century), el 
 Jouon de Longrais, Nantes^ 1880 (8s.). 
 
48 BIBLIOGEAPHICAL APPENDIX 
 
 Fierabras (late 12th century), pub. in 1860 in Recueil 
 
 des Ancien Poetes de la France, vol. iv. (4s. 6d.). 
 
 English translation, E. E. T. S. Charlemagne 
 
 Romances, No. 1. (15s.). 
 Otind (c. 1250), pub. in 1858 in Recueil des Anciens 
 
 Poetes de la France, vol. i. (4s. 6d.). English 
 
 translation in same volume as Fierabras. 
 * VEntrie en Espagne (c. 1315). 
 Prise de Pampelune (c. 1315), pub. by M. A. Mussafia, 
 
 1864. 
 Qui de Bourgogne (12th century), Recueil des Anciens 
 
 Poetes de la France, vol. i. (4s. 6d.). 
 Chanson de Roland (11th century), ed. by M. Leon 
 
 Gautier, with modern French translation (4s.). 
 Cf. The Song of Roland. A Summary for the use of 
 English readers with verse renderings of typical passages, 
 by A. Way and F. Spencer, 1895 (Is.). 
 
 Oaydon (13th century), pub. in Recueil des Anciens 
 
 Poetes, 1862, vol. i. (4s. 6d.). 
 Anseis de Carthage (13th century), ed. J. Alton, 
 
 Tubingen, 1892. 
 Chanson des Saisnes, written by J. Bodel at the end of 
 
 the 12th century, ed. F. Michel, 1835, 2 vols. 
 
 (£1, 5s.). 
 
 Wars with Vassals— 
 
 Oirard de Viane (early 13th century), ed. P. Tarb6, 
 
 Reims, 1850. 
 Renaud de Montauban (12th century), ed. H. Michelant, 
 
 Stuttgart, 1862. English translation, "The 
 
 Four Sons of Aymon," E. E. T. S., 2 vols. 
 
 (£1, 15s.). 
 Ogier le Danois. The only part of the Ogier cycle 
 
 that has been edited is Raimbert's 12th-century 
 
BIBLIOGRAPHICAL APPENDIX 49 
 
 m- Chevalerie Ogier, ed. J. Barrois, 2 vols. 8vo, Paris, 
 
 1842 (£1, 5s.). 
 
 Jehan de Lanson (13th century). 
 
 Huon de Bordeaux (c. 1240), Recueil des Anciens PoHeSj 
 vol. V. (4s. 6d.). Sixteenth-centurv English 
 translation by Lord Berners, E. E. T. S., 4 vols. 
 (£2, 10s.). A charming rendering into modern 
 French was published by M. Gaston Paris in 
 1898 (12s. 6d.). 
 
 Cycle of William of Orange. 
 
 The study of this cycle is at present the subject of 
 much careful work both in France and Germany. In the 
 latter country Professor Becker has published a "Travail 
 d'ensemble," Vie alt-franzosische Wilhelmsage. Halle, 1896 
 (4s. Gd.). In 1854, the Dutch scholar, M. Jonckbloet, pub- 
 lished a selection of the " William " romances, translating 
 some and giving others in their original form, under the 
 title: Gudlaume d' Orange, cJtanson de geste des 11* ct 
 12^ siecleSf 2 vols. 4to. 
 
 Vol. iv. of M. Gautier's J^pop^es contains full summaries 
 of all the works of the cycle. 
 
 *£nfances GuiUaume. 
 
 Departement des Enfans Aimerij published under the 
 
 title, Les Narhonnals, by the Society des Anciens 
 
 Textes Frangais, 1899, ed. H. Suchier, vols. i. ii. 
 
 (17s. 6d.). 
 La Mort Aymeri de Narbonne, ed. J. Couraye du 
 
 Pare, published by Socidtd des Anciens Textes 
 
 Fran9ais (8s. 6d.), 
 SUge de Narbonne. 
 
 *Le Charroi de Ntmes (Jonckbloet, 1854). 
 * Prise d* Orange (Jonckbloet, 1854). 
 Enfances Vivien, edited by Wahlund-Feilitzen, 1895. 
 Le Covenant Vivien (Jonckbloet, 1854). 
 
 D 
 
50 BIBLIOGRAPHICAL APPENDIX 
 
 Aliscans, Recueil des Anciens Poetes de la France, vol. x., 
 
 1873 ; critical edition by E. Wienbeck, &c., 1903 
 
 (14s.). 
 
 In 1904 an earlier draft of Aliscans turned up in England, 
 
 and was printed under the title La caru^un de Willame. 
 
 Cf. M. Paul Meyer's articles Romania, vols, xxxii., xxxiv. 
 
 WiUekalm, translation of Aliscans by Wolfram von 
 Eschenbacb, in complete edition of Wolfram's 
 works by Lachmann, 5th edition (8s.). 
 
 Storie Nerhonese, Italian compilation, published by 
 M. I. G. Isola, 3 vols., 1887-87. 
 
 *Le Moniage Guillaume J. 
 
 Le Moniage Guillaume II, 
 
 Le Moniage Renouart. 
 
 La Bataille de Loquifer, 
 
 Printed by Ballantyne, Hanson &* Co. 
 Edinburgh &= London 
 
DAVID NUTT, 57-59 LONG ACRE 
 
 THE SONG OF ROLAND 
 
 A SUMMARY FOR THE USE OF ENGLISH 
 READERS 
 
 JVith Verse Renderings of Typical Passages 
 
 ARTHUR WAY 
 
 AND 
 
 FREDERIC SPENCER 
 
 Trofessor of French and Romance Philology, Univeruity 
 College of North Wales, Bangor 
 
 Crown 8vo, 1895, 64 pp., Is. (Is. Id. post free) 
 
 SOME PRESS NOTICES 
 
 Journal of Education (Boston).— "Two thorough scholars, 
 with a fine sense of literary and historical values, have here 
 gone to work. . . . Most valuable and most pleasurable reading." 
 
 The Independent (New York).— "Very effective and fairly 
 accurate metrical remleriugs . . . will give the general reader 
 a good knowledge of the stirring ejuc." ,^ 
 
 Literary World. — " We are only required to look upon this 
 volume as a means of stimulating interest in the first Christian 
 epic, and those will be strange people who, reading this 
 judicious mixture, are not driven to ask for a complete 
 rendering of the poem from the pens of Mr. Way and Mr. 
 Spencer.'* 
 
 Speaker. — "Gives some spirited verse renderings of the 
 great passages of the epic, and the narrative is connected by 
 a running prose summary of the story." 
 
 Critic (New York). — "The authors have succeeded in pre- 
 senting a good summary of the action of the poem." 
 
 Literary World (Boston).— " The story is retold in clear and 
 vigorous prose, and the more stirring episodes are given in 
 somewhat ruggedly spirited verse." 
 
m 
 
 tfv Rnmnncj 
 
 J iu jd or are in the 
 
 press, November 1905: 
 
 No. 1. GET TIC AND MEDIAEVAL ROMANCE. By 
 
 No. 2. LORE : WHAT IS IT AND WHAT IS 
 
 Tfi > OF IT ? By E. S. Hartland. 
 
 No. 3. ^T AND THE OSSIANIC LITERATURE 
 
 CC ^ dlD WITH HIS NAME. By Alf^ Nutt. 
 
 No. 4. Y ARTHUR AND HIS KNiGilTS. A 
 Su ^ Arthurian Romance. By Jessie L. v^eston. 
 
 No. 5. g POPULAR POETRY OF THE FINNS. 
 
 B^ y 'S J. Billson, M.A. 
 
 No. 6 ^ FAIRY MYTHOLOGY OF "ttakE- 
 SI ^ By Alfred Nutt. 
 
 No. 7 r HOLOGY AND FOLKrAL^S Their 
 R r AND Interpretation. By E. ?v.. Haktland. 
 
 No. 8 ' lULAINN, THE IRISH ACHILLES. By 
 A /utt. 
 
 No. 9 RIGYEDA. By E. Ym.- on Arnold, Litt. ] ) 
 
 No. ^E ROMANCE CYCj E OF .HARLl 
 
 ]V AND HIS PEERS. By ' ilssie L. vVeston. 
 
 No. 1 ^ MABINOGION. By I--^ B. John, M.A., 
 
 1 C *^ University of Wales. 
 
 No. ] ^ EDDA: I. THE DIYINE MYTHOLOGY 
 C ^ NORTH. By Winifred Faraday, M.A. 
 
 No. T EDDA: IL THE HEROIC MYTHO- 
 
 ] K THE NORTH. By Winifred Faraday, M.A. 
 
 No. P E LEGENDS OF THE HOLY GRAD 
 
 : D Nutt. 
 
 No. ^] CYCLE OF DIETRICH OF BERN. 
 
 ANDBACH. 
 
 Second Series began with No. 13 
 
 Bacb number, neatly printed in attractive cover, 
 6d, net: 7d, post free 
 
=^ETURN TO the circulation desk of any 
 University of California Library 
 or to the 
 
 ^JORTHERN REGIONAL LIBRARY FACILITY 
 3ldg. 400, Richmond Field Station 
 Jniversity of California 
 =tichmond, CA 94804-4698 
 
 L BOOKS MAY BE RECALLED AFTER 7 DAYS 
 
 lonth loans may be renewed by calling 
 
 ;44^642-6753 
 
 ear loans may be recharged by bringing books 
 
 to NRLF 
 
 lewals and recharges may be made 4 days 
 
 Drior to due date 
 
 DUE AS STAMPED BELOW 
 
 PR29l993^