Mieke Lens

Old French Epic Cycles in MS. Turin l. 11. 14: The Development of Old French Narrative Cycles and the Transmission of Such Cycles into Middle Dutch Epic Poetry

In this paper Iintend to examine the codex as a structural factor in the formation of narrative cycles. In this connection I would like also to demonstrate how important it can be, in instances where a manuscript contains more than one text, to examine the specitic context of the manuscript in which the texts are found , separately or as a group. One may presume th at in most cases texts have not been randomly grouped together but that they have been collected in one manuscript for a good reason. So, texts can be gathered in codices for a variety of reasons. They may for in stance be similar subject-wise, they may have the same form and/or they may belong to the same genre or they might even be said to reflect the taste of the person who commissioned them. Whenever a number of different texts are found together in one manuscript this can affect the way in which they are interpreted, used and handed down. If, for example, such texts were initially circulated as separate en ti ties they might well, from the moment they are placed together in one manuscript, be recognized as a cluster. In other words all, or a number of the texts in a given manuscript can, from that time onwards, be passed on together and/or in fixed combinations with other texts dealing with the same subject matter. If this happens we can say that a dossier has been formed land the question which then arises is: can we speak of cyclification having taken place? Texts that resem bie each other content-wise and have often been grouped together - thus forming a dossier - need not therefore automatically be seen as acycIe. Likewise separate texts/stories do not automatically constitute cycles because of the fact that, for whatever reason, they exist in the same manuscript. There are other more specific conditions th at have to be met if such texts are to be collectively regarded as acycIe. Texts may be said to form acycIe when content-wise they are (more or less) explicitly linked together and/or when "their contents are combined or arranged to form a comprehensive structure in such a way that together they present a new, more or less, coherent literary 1 See 1.B. Voorbij , ' Medieval dossiers and modern stemmas. An exploration into manuscripts of the Epistolà Alexandri ad Aristotelem.' This artic1e is to appear in M. van MuIken and P.Th . van Reenen , eds. , New Perspectives in Stemmatology (Amsterdam: Benjamins, 1994). The author discusses fixed combinations of texts (dossiers) for constructing more reliable stemmas. He shows us that " editors should use codicological information and study texts in relation to their transmission context(s)". In this connection it would be interesting to compare the stemmas of the different Turin MS . texts discussed here with one another.

Mieke Lens

127

unit".2 In a dossier by contrast such arrangements or connections do not necessarily have to exist. ~ When individual texts turn into a cycle or become a part of a cycle, this will inevitably affect the way in which they are interpreted, received and used. In other words, the creation of a new literary unit unavoidably influences the individual component parts of that same entity. In conjunction with the present discussion on cyclification and on the influence that codices have upon this process I would like to examine the L.II.14 manuscript (otherwise known as the T manuscript) because it appears to have been compiled in a structured, that is to say, weil thought out way. This voluminous and richly illustrated manuscript dating from 1311 th at runs into 586 folios is kept in the University Library in Turin. 4 We do not know exactly whom the manuscript was written for but, bearing in mind its elaborateness it must have been made for a wealthy and/or prominent person. It might have been in the possession of the dukes of Savoy but this remains uncertain. 5 If we study the contents of the manuscript we see that it contains texts that have apparently deliberately been placed together because of the themes they have in common like, for instance, how the Christian faith was developed and perpetuated. 6 There is no doubt about the fact that the theme of warring crusading Christians and Saracens was prominent in the manuscript as a whoie. What is significant for this article is the fact that the manuscript we are examining contains a number of texts in which the adventures of Huon de Bordeaux and his descendants are related and so we are definitely dealing here with a narrative cycle. The events related and the people described in Huon de Bordeaux, a thirteen th-century chanson de geste, were definitely a source of inspiration and a basis for new stories about this hero. I will be focusing upon this cycle and its direct context (that is to say, its place within the manuscript) and seeing to what degree this referred to context affected the entire cycle. The cycle opens with the enfances of Auberon, the famous fairy king: the Roman d'Auberon (f. 283r-296v) 7 and continues with Huon de Bordeaux (f. 297rSee also the Organizing Committee's broad definition. A dossier can contain for instance, without any changes having been made, two or more different texts on the life of Alexander the Great that might overlap or even contradict each other. 4 E. Stengel in his publication Mitteilungen aus franzàsischen Handschriften der Turiner Universitäts-Bibliothek (Marburg, 1873), pp. 11-38 gives a detailed description of the contents of the manuscript. After 1904 when the manuscript was damaged by fire it was rebound and made into four volumes. A new description was made in 1934 by E.G. Wahlgren with the title 'Renseignements sur quelques manuscrits français de la Bibliothèque nationale de Turin ', in Studier i Modern Spräkvetenskap 12 (1934), pp . 79-124. 5 In her introduction to the publication containing the Huon de Bordeaux continuations Brewka tries to prove that the manuscript was once the property of the dukes of Savoy but she is unable to provide absolute proof of this. It is far from certain that the Huon MS. (no. 79) listed in the 1713 Index alphabétique is a) the same manuscript as that given in Tand b) that it was part of the collection of manuscripts that Victor Amadeus 11, king of Sardinia and duke of Savoy donated to the University Library in Turin some seven years later. See for the history of the T MS.: B.A. Brewka, ed., Esclarmonde, Clarisse et Florent, Yde et Olive I, Croissant, Yde et Olive 11, Huon et les géants, sequels to Huon de Bordeaux, as contained in L.II.14, the Turin MS. (Nashvillerrenessee, 1977), pp. 66-70 (univ. Microfilms Intern . Ann Arbor, Michigan USA, 1983). 6 See for the contents of the T MS. the appendix based on Stengel's 1873 description . 7 J. Subrenat, éd., Le Roman d'Auberon, prologue de Huon de Bordeaux (Paris/Genève: Droz, 1973). 2

3

128

Old French Epic Cycles in MS. Turin L. IJ. 14

354v)8 EscJarmonde and their progeny are central (f. 354v-460v).9 The Roman d'Auberon functions as a prologue to the cycJe, which ends with the Chanson de Godin, a tale in which Huon's son Godin is the protagonist. As far as we know the T MS. is the oldest manuscript containing a Huon cycJe. If one thinks of the Huon de Bordeaux as part of the cycJe one sees that the story in the T MS. text differs from that of the Tours MS. (M) which contains no continuations to the story. 10 The T MS. story en ding has been constructed in such a way that it merges imperceptibly with the following text, making it difficult to establish exactly where the Huon ends and its continuation begins. 11 The beginning of the Huon de Bordeaux has not been adapted; there is no smooth transition between the end of Auberon and the beginning of Huon yet content-wise the two texts are interrelated. 12 In Auberon various of the Huon de Bordeaux themes are developed and several times events that are later to be related in this same text are anticipated. Apart from having been passed down to us in the T MS. the Huon cycJe exists in two other manuscripts both of which date from the fifteenth century. 13 In all three manuscripts the cycJe varies in size and content but nowhere is the cycJe as extensive as in the T MS. Two of the texts incJuded in the T cycJe, the Roman d 'Auberon and the Chanson de Godin are peculiar to this manuscript. Is it perhaps possible that they were especially written for this manuscript? According to F. Suard the development of the cycJe probably started towards the end of the thirteenth century, in other words, before the T MS. was started. 14 He does not indicate when Auberon and Godin were added to the cycJe. Is it perhaps possible that the T compiler made the connection because he wanted to draw together all the texts known to him about Huon and his descendants in one manuscript? And is this perhaps the reason why the cycJe presented in the T manuscript is so extensive?15 The Huon text in the T MS . has not been published but may be looked up in the Tours MS. (M): P. Ruelle, éd., Huon de Bordeaux, Université Libre de Bruxelles, Traveaux de la Faculté de Philosophie et Lettres 20 (Bruxelles/Paris, 1960). 9 All the Huon continuations have been published: M. Schweigel, ed ., Esclarmonde, Clarisse et Florent, Yde et Oliue, Drei Fortzetzungen der Chanson von Huon de Bordeaux, nach der einzigen Handschrift zum ersten Mal veröffentlicht (Marburg, 1891); Brewka, Esclarmonde; F. Meunier, éd., La Chanson de Godin, chanson de geste inédite, Université de Louvain, Recueil de traveaux d 'histoire et de philologie, 42 série, 14 (Louvain: Presses Universitaires, 1958). 10 This manuscript is kept in the bibliothèque municipale in Tours. 11 The layout of the manuscript provides us with no c1ues. 12 The two texts are also presented as visual entities: Auberon ends on f. 296v and Huon commences on f. 297r. D The manuscripts are preserved in the Bibliothèque Nationale in Paris and their shelf marks are: f. fr. B.N. 22555 and f. fr. B.N. 1451. 14 F. ~uard, ' Le cycle en vers de Huon de Bordeaux: étude entre les térnoins français' , in E. Baumgartner et al., éds, La chanson de geste et Ie mythe carolingien: mélanges Réné Louis publiés par ses collèges, ses amis et ses élèves à {'occasion de son 75e anniversaire (Saint-Pèresous: Vézelay Musée Archéologique Regional, 1982), vol. 2, pp. 1035-50 and notably pp. 1046-47. 15 The texts referred to fit into the context of this manuscript very weil because of their themes. For instance the Roman d'Auberon was inspired by religious stories Iike the Histoire de Judas Machabée and La vie de Saint George. The T MS. is for a large part composed of such texts (see the description of the contents). Perhaps this was another reason why Auberon was added to the cycle in the T MS. See for the religious influence upon this story: Subrenat, Le Roman d'Auberon, pp. XLIX-LXVI. It would be interesting to study Auberon in relation to the context in which it has come down to us. 8

Mieke Lens

129

Upon examining the Huon cyc1e in its T MS. context one discovers th at it is preceded by a substantially big section of another narrative cyc1e, the Geste des Lorrains and that it is followed by the song of the adventures of Bueve de Hantonne. Why then was the Huon cyc1e inserted between these two texts? What is most nota bie is that they are all chansons de geste or texts of a type related to this genre. Was it mere coincidence that these texts we re placed together or was it done deliberately? To answer this question properly we must examine the whole Huon cyc1e in its rightful context. In the T MS . the Geste des Lorrains is preceded by two prologues 16 the first of which is to be found in a completely different place in the MS. than the Geste itself and provides us with Garin and Fromont's family tree. In this genealogy mention is made of duke Seguin 17 but explicit links with the story of Huon de Bordeaux who had the same name as his father are lacking. In this respect the second prologue is more interesting, it immediately precedes the Geste des Lorrains, it creates a link between the Romans de Vespasien - which in turn precedes this second prologue - and the Geste des Lorrains and it connects the Geste with Huon de Bordeaux. The prologue tells of the battle th at Seurin, the brother of Titus and Helène, has against the Saracens. 18 Once Seurin has defeated the Saracens he marries Chedaire who bears him three sons. The poet then goes on to identify Seurin 's offspring mentioning th at Piere, Seurin's third son, marries Aelis. Then follows the allusion to Huon de Bordeaux. 19 The story goes as follows: Huon, son of duke Seguin is banished to Lombardy where he is employed by Guinemer, the son of S. Bertin. He falls in love with a girl who bears him a son named Henris. Huon is poisoned and dies and fearing that he too may be poisoned Henris subsequently flees to Metz. He marries in Digon and fathers five sons one of whom, Terzis, is later to be appointed provost of Metz by duke Piere. This is the point at which the stories of Seurin and Huon converge because Piere is Seurin's son. As we know, in the Huon de Bordeaux story Huon is not exiled to Lombardy but is sent instead on amission to Babylon by Charlemagne. Should we therefore presume, as some scholars have done, that we are dealing here with another older version of the Huon story?20 1 think not. Wh at seems more likely, in view of the context of the manuscript, is th at the author of the prologue made these changes in the story because he wanted to forge a link between the two narrative cyc1es that were adjacent to each other in the manuscript. It is c1ear from the prologue why these two cyc1es are found together. We are dealing therefore with an ad hoc connection. In making this link the poet and/or compiler created from two separate cyc1es one new literary entity with, as a central theme, the battle of Bordeaux which was the main theme of the Geste des Lorrains and an important theme in Huon de See appendix. " Bueue a la barbe et Ie franc duc Seguin . " 18 The legend of the 11000 virgins is woven into this story abou,t Seurin . 19 This passage is cited by M . Rossi in Huon de Bordeaux, et /'Evo/ution du Genre épique au xllle siècle (Paris: Champion, 1975), p. 45. The complete text of this prologue may be found in Stengel, Mittei/ungen , pp. 25-29. 20 See Rossi, Huon de Bordeaux, pp . 45-46 for an exposition on this prologue. 16

17

130

Old French Epic Cycles in MS. Turin L. 11. 14

Bordeaux and its continuation Esclarmonde. 21 It would therefore appear to be no mere coincidence that these two cycles were placed together and linked together in the Lorrains prologue. Would it be going too far to assert th at we are dealing here with a new cycle (the " Bordeaux" cycle)? Bearing in mind that the definition for cyclification commonly adhered lo is a broad one I would feel inclined to answer this question in the affirmative. In this case the codex was a stuctural factor in the formation of the cycle. Were the texts also seen as acycIe by contemporary poets and readers? Did medieval poets and readers of these texts maintain that the texts belonged together or is this purely a modern interpretation? I shall return to this point when I co me to discuss the transmission of French epic texts into Middle Dutch but first I want to look at the relationship between the Huon cycle and Bueve de Hantonne. As has been stated the Huon cycle in the T MS. is directly followed by Bueve de Hantonne. Though there is no explicit connection between the Huon cycle and the Bueve de Hantonne it would a ppear to be no coincidence that the latter text comes immediately after the former text in this manuscript. Huon and Bueve 22 have many themes in common like for instance, the banning of the hero and his subsequent journeys to and through lands far away and over the seas. The most striking resemblance though between the Huon cycle and Bueve de Hantonne is the arrival on the scene of a " new " personage in the form of the pagan king Yvorin de Monbrant. It is probable that this king was first introduced in Bueve de Hantonne and that the Huon de Bordeaux poet borrowed this character from that text. 23 Apart from appearing in Huon de Bordeaux Yvorin features quite prominently in the Chanson de Godin , the last text in the Huon cycle which, in the T MS. is immediately followed by Bueve de Hantonn e. Bueve probably did immediately follow Godin, i.e. the Huon cycle because the Saracen king features in both texts. We can say that these texts belong to a new " matière " (the " matière de Monbrant "). 24 This Saracen king also enjoyed a degree of fame outside of the French speaking world. 25 Yvorin (Yvorijn) not only figures in the Middle Dutch version of Huon de Bordeaux but also in other Middle Dutch texts 26 which 2 1 In Huon de Bordeaux Gérart usurps the city of Bordeaux in the absence of his brother Huon . In Esclarl1londe Bordeaux is attacked by the German king who linall y besieges the city beca use he wants to make Esclarmonde his queen. 22 In particular, the second continenta l version of th is story. 23 R ossi, Huon de Bordeaux, pp. 71-72 . 24 In this case there is no evidence of acycIe. The texts have a similar story personage (Yvorin de M o ntbrant) and/or contain a commo n location where some of the action occurs but one canno t claim that the stories a re in a ny way intertwined or ada pted to each other. 25 In my paper read at the XIIth congress of the Société Rencesvals in August 1991 I also discussed this " ma tière " . Cf. ' Huge van Bordeeus, un roman d'aventures du cycle de Charlemagne ', in Philip E. Bennett, Anne Elizabeth Co bby and Gra ha m A. Runnalls, eds., Charlel1lagne in the No rth. Proceedings of the TlI'eljth International Conference of the Société Rencesvals, Edinburgh 4th to 11th August 1991 (London : Grant and C utier, 1993), pp. 53-56. 26 Yvorijn a nd/or Monbrant (also spelt as Mombrant) a lso appea r in: Madelgijs, Buevijn van Austoen, Limborch, Florigout a nd in a n as yet unidentilied tex t kept in the State University Library in Ghent a nd numbered 173 . See in this connection H. Kienhorst's De handschriften van de Middelnederlandse ridderpiek. Een codiologische beschrijving (Deventer: Sub Rosa, 1988), vol. I, p.238.

Mieke Lens

131

brings me to the last subject that I want to touch on in connection with with cyclification in the T MS . I want to consider the way in which medieval poets and writers viewed and used texts of the kind to be found in the T MS . Can we find evidence to support observations made about the connections between the texts examined? In this connection I would like to see how the Huon cycle was viewed and implemented in other languages. It is interesting to find out whether or not such cyclification extends beyond the barriers of language and, if so, to discover how the Huon cycle was translated into Middle Dutch and/or adapted . The Middle Dutch poem Huge van Bordeeus is a free adaptation of Huon de Bordeaux. Though the main characters and events included in the chanson de geste also figure/occur in the Middle Dutch 27 poem it is equally true to say that just as many of the events and characters included in this poem are not present in the French Huon texts. Apart from this the things related in the Middle Dutch poem that mayalso be replicated in French texts are usually told totally differently: there is little or no evidence of literal repetition . One explanation for this discrepancy might be sought in the influence of Huon continuations upon Middle Dutch poetry. 28 It would appear that the poet in question borrowed elements from his Middle Dutch Huon cycle adaptations in order to enrich and extend his new poem. 29 If this was indeed the case the poet did not translate or adapt the individual poems in the cycle as did for instance the reviser of the Middle Dutch Geste des Lorrains but instead he drew upon elements from different poems in the cycle and combined them to form one new poem. Apart from taking extracts from the Huon continuations he probably also borrowed components from other Middle Dutch stories like for in stance the Roman der Lorreinen (an adaptation of the Old French Geste des Lorrains). He probably chose this text because of the Bordeaux adventures and the portrayal of the traitor Guweloen. It is interesting to see that it was precisely this text th at seemed to influence the Middle Dutch Huon adaptation. This would support the notion th at these texts are indeed related to each other content-wise at least. To my mind it would seem logical to presume that a medieval poet would first of all select matter from text fragments that bear some relationship to his own text. By contrast it seems unlikely that he would sift through a text like for instance the Ferguut in search of material th at might just add colour to the epic battle of Bordeaux if he had at his disposal a text like the Lorreinen in which this same theme was prominent. The texts drawn upon by the Huge van Bordeeus poet would reinforce the notion th at the texts discussed in the T MS. may be viewed as related and therefore as grouped together for a reason . In the case of Huge van Bordeeus there is no question of cyclification; it was rather so that components from the French cycle were used to create a new story. 30 27 0ne should realize that only parts of this poem have been preserved. 28 In this article it is not possible to go into details about the relationship between the Middle Dutch fragments referred to and the French tradition. The reader is therefore referred to my dissertation research. 29 For instance, the crowning of Huge as king of Aragon in the Middle Dutch poem might weIl have co me from a combination of elements derived from the Huon continuations. 30 lt seems unlikely that the Middle Dutch poet availed himself of the T MS.

132

Old French Epic Cycles in MS. Turin L. 11. 14

The fourteenth-century Flemish poem Florigout provides us with an example of cyclification that cuts across all language barriers. 31 This poem was neither a translation nor an adaptation of an existing French text but was probably a kind of continuation to the second continental version of the Bueve de Hantonne story. 32 In this fragmented poem (which judging from the fragments we have must originally have been quite long) we are told how a group of people migrated from the city of Monbrant that was in the hands of Christians. During the course of the journey the members of the group become separated from each other and one by one the adventures of the main characters are told. Presumably they are all reunited at the end of the poem but this part has not been preserved. Just like Huge van Bordeeus Florigout was part of the new " matière de Monbrant " th at was at the time quite popular in Flanders. 33 Both of the texts have much in common, not only as regards their contents but also from a codicological point of view. 34 It is quite possible th at they originally belonged to the same codex and if this is indeed true then this is another example of a collection of texts contained in a single manuscript th at handles the same " Monbrant " subject matter and it is evidence possibly also of the codex as a structuring factor in the cyclification process . However this last assertion remains purely speculative because only fragments of both the poems have come down to us. When one studies the Middle Dutch Huge van Bordeeus poem and other such texts it is c1ear th at the way in which the chansons de geste are combined and arranged in the T MS. is by no means unique and this in turn supports the cyclification theory. I am weil aware that much more remains to be said about the texts that have been discussed here and about their transmission but that is impossible in the context of this paper. I only hope th at this examination of the T MS. and particularly of the context of the Huon cycle will have provided some new insight into the matter of how narrative cycles evolve. University of Groningen

3 1 K . Heeroma, ed., Florigout, fragm enten van een 14de-eeuws ridderverhaal, Maatschappij der Nederlandse Letterkunde 3 (Leiden : Brill, 1962); J.W .E . Klein, ' Een onbekend tekstfragment van de Florigout ', in Spektator II (1982), pp. 480-95 . 32 See Heeroma, Florigout , p. 2. The different versions of Bueve de Hantonne are published by A. Stimming in Der Festländische Bueve de Hant0l1l1e etc., 5 vols., Gesellschaft f. rom. lit., 25, 30, 34, 41 , 42 (Dresden, 1911-1920). The second continental version is the only Bueve version where the Saracen king Yvorin is converted to Christianity and Monbrant becomes a Christian city. 33 See in this connection note 24. 34 For a codicological description of the fra gments of this Middle Dutch epic poem see Kienhorst, De handschriften van de Middelnederlandse ridderpiek , vol. 1, pp. 49-52 .

Mieke Lens

133

Appendix Contents of the Turin MS . L-I1-14 (olim g.II.l3; MS . f. XXXVI ; R 1639 in the Inventari of G. Mazzatini, vol. XXVIII) , largely based on E. Stengel's 1873 description, pp. 11-38. f. I: Roman de Sapience , Hermann de Valencienne [f. 1Orl-v I]: introduction to the Loherain (comes before folio 2) f. lOvl-2 and f. 2-3vl: Li Roumans de Vespasien (according to Stengel the title is not correct) f. 3vl-12rl: Virgilius: from Adam's expulsion from paradise to Abraham f. 12rl-24v2: extract from the Roman de Sapience by Hermann de Valencienne f. 25rl-47r2: Vie de la S. Vierge Marie f. 47rl-49r2: extract from the Roman de Sapience by Hermann de Valencienne f. 49r2-79rl : probably another section from the poem: Vie de la S. Vierge Marie f. 79r2-83rl : a kind of past history of the Venganche Jesu-Christ (th is history contains details on Mohammed's life) f. 83r2-102vl: beginning of the Romans de Vespasien (also known as La Venganche Jesu-Christ or La Destruction de Jérusalem) f. 103r-l04v2: second prologue to the Loherain also containing the legend of the 11000 virgins f. 105rl-l77r2: first part of the Geste des Loherains: the Chanson de Heruis de Metz f. 177r2-282r2: Chanson de Garin Ie Loherain and part of Gibert de Mez f. 283-296vl: Roman d'Auberon (prologue to Huon de Bordeaux ) f. 297rl-354v2: Huon de Bordeaux f. 354v2-401vl: the Huon continuations: Esclarmonde, Clarisse et Florent , Yde et Olive, Huon et les géants f. 401 v-460v2: the Huon continuation Godin f. 461r-577r2: Beuve de Hantonne f. 578-579: La vie de Bonce Py late f. 579vl-583vl: La Vie de Judas f. 583vl-585vl: Chi commence del unicorne f. 585vl-586vl: Chest de la houce (a fabliau)

134

Old French Epic Cycles in MS. Turin L. 11. 14