Logo DBNL
Queeste. Tijdschrift over middeleeuwse letterkunde in de Nederlanden. Jaargang 2006 (2006)

Queeste. Tijdschrift over middeleeuwse letterkunde in de Nederlanden. Jaargang 2006

(2006)– [tijdschrift] Queeste–rechtenstatus Auteursrechtelijk beschermd

Vorige Volgende

At which point does intertextuality slip into speculation?
Bernadette Smelik

With reference to: Marjolein Hogenbirk, Avontuur en Anti-avontuur Een onderzoek naar ‘Walewein ende Keye’, een Arturroman uit de ‘Lancelotcompilatie’. Amsterdam/Münster: Stichting Neerlandistiek VU/ Nodus Publikationen, 2004. 200 p., ISBN 90-72365-79-8/3-89323-742-9. Price: € 22.50.

Intertextuality is, nowadays, often used as a means to get more insight in a medieval text, its meaning, its purpose and its (intended) public. It can also help in dating (and occasionally localizing) a text, but this is the case mainly when explicit references occur to other texts (which are presumably older than the text in question) in the examined text. By using the term intertextuality a scholar assumes that the author deliberately introduced connections with other texts in his own tale to bring about a specific reaction among his audience. Not only explicit references, as mentioned above, attract the attention of scholars, but also (or should I say especially) implicit ones. An implicit reference can be seen for example in a different use of a well-known motif or story structure, aimed at those in the audience who not only recognize the motif or structure, but also appreciate its different use. Such veiled references are by nature captivating for a public of connaisseurs, but less so for the occasional passers-by or youngsters. How intertextuality can procure more insight in medieval romances was demonstrated in 1993 by Bart Besamusca's study of three indigenous Middle Dutch Arthurian romances, two of which are preserved in the famous Lancelot Compilation.1 This study has generated a discussion on this specific method of research which has been summarized by Jef Janssens in his introduction to Op avontuur. Middeleeuwse epiek in de Lage Landen (1998).

Research using this method has its own problems even in apparently straight forward cases, in which an explicit reference is made to a source text. An example is the 17th-century Irish Arthurian romance Eachtra Mhacaoimh-an-Iolair (The Adventure of the Eagle-Youth). The author of the tale, a certain Brian Ó Corcrán, informs us that he ‘received the bones of this tale from a gentleman who said that he had heard it told in French’.2 In my article on this romance, I have demonstrated that its structure shows ‘startling similarities with the post-classical double-quest structure found in later Arthurian romances’. But a specific source is more difficult to pinpoint, in fact I have concluded that it is not possible. And yet it is tempting to speculate further on this topic. There is a tale which comes close in structure to this Irish story, the Middle Dutch Arthurian Roman van den Riddere metter Mouwen. In my article I have said that ‘[s]ince it is unlikely that this story was known in Ireland I will not explore this thread here’.3 But is it so unlikely? Could it not be that a Flemish merchant went to Dublin4 and listened there to stories told by Anglicized Irish story-tellers in a language common to both, viz French?5 And could it not be that he was challenged to tell himself a story and that he told the Flemish story of a knight with a sleeve in French? Another possibility is that an Irishman visited the Irish Franciscans fled to Louvain in 1607. Socializing with the locals, he could have heard a story

about a knight with a sleeve told in French.6 He then went to another place and retold this Flemish story in French disguise in Irish to our Brian. As this Brian can in all probability be identified with the Brian Ó Corcráin who wrote poems for Cú Chonnacht Maguire, an Irish lord who also fled to the Continent after the revolt against the English failed (with a probable first stop in Louvain),7 especially the latter chain of events is very tempting, opening up even more new areas of research. I have inserted this attractive set of speculations in order to demonstrate that anybody can think up all sorts of explanations and possibilities when seeing similarities in structure or motifs. I have not even taken into account here the theory that the Roman van den Riddere metter Mouwen could be based on a lost (French) story which in turn was based on the French Richars li Beaus ... The horizon becomes wider and wider. However, there is a very serious problem with this way of thinking, and with a thump we land on solid ground again: we do not have any proof that this hypothetical Flemish merchant or Irish visitor to Louvain either (or both?) listening to a Flemish tale told in French ever existed. We simply have no written accounts of these things happening; all of it is conjecture (very probable I add stubbornly, but still conjecture). In this case we very definitely have crossed the boundary of the intertextuality method when studying old romances.

My speculations have been provoked by a recently published study on the presumably indigenous Middle Dutch Arthurian romance Walewein ende Keye, one of the seven inserted romances in the Lancelot Compilation. No direct source text, whether in Dutch or in any other language (not even a reference to the possible language of a possible source text), is known to this date, and this of course determined the way in which Marjolein Hogenbirk has approached the study of this short but interesting romance. Her book ‘focuses on “Walewein ende Keye” as a text which combines a variety of conventions and common characteristics which in the course of the thirteenth century became part of the genre of Arthurian romance’ (p. 173). Hogenbirk's aim is to provide an insight in the way in which these conventions and innovations have been combined ‘to form a new, and, as we shall see, remarkable creation encompassing themes which are entirely its own’ (p. 173). To study a romance for which no source text is known, to see how conventions and innovations encompassing an entire genre have been used, is a difficult task. For confronting such tremendous problems Hogenbirk is to be lauded, but unfortunately she has not always been aware of the dangers of her chosen approach.

Her first premise (p. 21) is that in most Arthurian romances a more or less carefully created connection between adventures is to be recognized. Analogies, contrasts, parallels and the doubling of motifs all form clues for the interpretation of texts. Thus, there is always a connection between conjointure and the sen of the tale.8 She then asks herself the question whether or not this is also the case in Walewein ende Keye. To get an answer she analyses the structure of the adventures and the connections between the events. However, how can one assume a connection between conjointure and sen when all we have is a secondary reception of the text in question? Given the fact that Hogenbirk readily assumes Walewein ende Keye as transmitted in the Lancelot Compilation to be a secondary reception of a primary text which has not survived. Can deviations of conjointure and sen in the surviving tale be taken as evidence for the existence of a better structured original? What happens when the author of the hypothetical original source text himself was not capable of creating deliberate connections? We have to be careful, as Norris Lacy has pointed out a few years ago, stories can be just badly composed.9 A discussion or even an acknowledgement of this problem would have made Hogenbirk's case a lot stronger.

Her second premise is found two pages further on when she postulates that variations in the characterization of a figure can be important and significant. This is undoubtedly true and Hogenbirk therefore studies the characterization of the dramatis personae and questions whether or not the author fits in with or deviates from the conventions and why he does so. This is intertextuality in character portrayal, although she does not call it so and presents intertextuality as a third premise (p. 24). Here we encounter another problem: with which Walewein or Gauvain is the main character of this story compared? Walewein's character undergoes a change from the earliest time onwards in Welsh texts to the dramatic negative portrayal in later French romances. In her introduction Hogenbirk states that Walewein is presented as the perfect knight and described without ironic aloofness (p. 24, but see my problems with this statement further down). This she demonstrates, for example, with an analysis of the adventure in which Walewein is confronted with a jealous knight who has put his lady in a well because she considers Walewein a better knight than her own lover. Hogenbirk explains that similar confrontations are described in Old French romances, but that the damsels there always fall in love with the image they have created for themselves of Gauvain, without knowing him in person and often not readily accepting that the knight they have in front of them is the Gauvain they seek (after which they of course share his bed). According to Hogenbirk this scene in Walewein ende Keye needs to be read against this backdrop. The author of the Middle Dutch romance has changed this motif by uncoupling the esteem for Gauvain from love for or by him: his lady is not in love with Walewein, and Walewein does not want to share his bed with her (p. 48-49). This presumption cannot be denied, but is it the complete picture? If the author really deliberately constructed this connection between the French genre and the Middle Dutch romance, he must have had a great knowledge of Arthurian romances.10 And the problem is that we cannot know which texts the author knew. In fact this discussion says more about Hogenbirk's wide reading than about the author's familiarity with other Arthurian texts. If one looks a little further, other possibilities come afore.

Hogenbirk restricts herself to scenes in which Gauvain/Walewein plays a role in French and Dutch romances. But why not compare this scene with other similar scenes in which Lancelot plays the leading role in the Lancelot proper,11 which is transmitted in part in the same manuscript as Walewein ende Keye (and which was translated into Middle Dutch at least three times)? Such a comparison could also have led to interesting conclusions about the rehabilitation of Walewein in the Middle Dutch Lancelot Compilation: as is well-known, attempts by Middle Dutch adaptors to rehabilitate Gauvain are a favourite explanation by modern scholars accounting for Walewein's presentation as the ideal knight in Middle Dutch romance.12 No doubt other characters in the Arthurian genre can be found in similar circumstances and can be compared with the Walewein portrayed in Walewein ende Keye. Can one limit oneself to adventures experienced by one and the same character when one looks for intertextuality in medieval romances? I do not think so. Of course such scenes have to be taken into account,

but any medieval author could easily have modelled his own scene featuring Gauvain as protagonist on an adventure a different character experiences in another text. As long as there is no direct borrowing in phrases between two scenes, such comparisons remain a little bit speculative. We do not know anything about the author of Walewein ende Keye, so speculations about what he must have known need to be presented in a very delicate and carefully worded manner. And although Hogenbirk sees interesting connections with a whole range of (mainly Old French) romances,13 the presentation lacks in understanding the underlying problem of intertextuality.

The problem becomes more complicated when she draws in stories belonging to the epic poetry around Charlemagne. For example, scenes portraying Keye as a traitor are traced back to this genre, and parallels are adduced in which traitorous seneschals appear to deceive credulous kings (p. 36-39). But in this instance there is no need for an excursion into Charlemagne material, as a traitorous seneschal also appears in the beginning of the Lancelot proper, where king Ban (Lancelot's father) is betrayed by his seneschal. The betrayal results in the kidnapping of Lancelot, the capture of Lancelot's nephews, Bohort en Lyonel, and Ban's death.14 Another influence from the chansons de geste is seen by her in Walewein's dream in which he is attacked by a lion (p. 39-41). However, an opponent appearing as a lion in a dream can also be found in the Lancelot proper.15 So why assume per se an influence from this totally different genre? One would expect a little elaboration, motivating why the author should prefer a borrowing from a different genre over an Arthurian text containing the same motif, dang lamor at least substantiating that the chansons de geste literature was popular in circles in which the author moved, but such a foundation of her argument is lacking. In fact she never explicitly explains with which texts she compares Walewein ende Keye and why she does so. According to p. 28, her preferred source texts belong especially to the Old French and Middle Dutch tradition. She is not specific, however, whether she compares mainly with romances in verse, or also with prose texts such as the Vulgate Cycle (although the above-mentioned parallels suggest that she did not consider this cycle). The comparisons with the chansons de geste are in flagrant contradiction with her alleged intention that she will compare this Dutch tale to the genre of Arthurian romance (p. 10), and that she will not explore more genres; the comparison with the chansons de geste genre comes rather as a surprise.

Hogenbirk's comparison between Walewein ende Keye and Chretien de Troyes' Yvain ou le Chevalier au Lion rests on more solid ground (p. 71-75). This supports the observation of Keith Busby that Yvain had already become ‘part of a supra-regional canon’ by the time Walewein ende Keye was written.16 More interesting are her observations that not only Yvain, but also L'Âtre Périlleux, transmitted more than once together with Yvain, could have been known to the author of Walewein ende Keye (p. 50-52).17 Although Hogenbirk mentions this combination, she does not speculate on its significance for the similarities of Walewein ende Keye with precisely these two texts. It could be possible that the author of Walewein ende Keye heard (or read) both texts because they were present in the same manuscript (to which he then had access). In this respect I refer the reader to Hogenbirk's lucid article ‘Gauvain, the Lady and her Lover’.18 Also, her arguments that the author could have known the Middle Dutch Fergus translation, Ferguut, are very reasonable. She compares lines in Ferguut with lines in

Walewein ende Keye (p. 63). In Walewein ende Keye it is said that it was raining ‘starc ende groet’, as it does in Ferguut (‘herde groet regen’). She could have substantiated her argument that here the author of Walewein ende Keye betrays knowledge of the Ferguut and not its source, by citing the French text in which it only drizzles (l. 904-905: ‘Il commencha a plovenir/Une pluiete molt menue.’).19

Hogenbirk's discussion of the dominating theme of Walewein ende Keye, which is the contrast between hubris and humility, is carefully done (p. 107-126). She shows that all different layers of meaning of Walewein ende Keye Walewein's epithet ‘der avonturen vader’ are present in this tale and she shows that Walewein's adventures form a coherent unit with analogies, parallels and an increase in complexity. This chapter is a pleasure to read, and although many of the main observations have been made by others and herself already, she digs deeper and presents all possible arguments. All in all Hogenbirk is strong when she limits herself to the internal structure of Walewein ende Keye, although a more balanced discussion of the jocular scenes with Walewein would have been welcome. It is regrettable that she was not able to use Veerle Uyttersprot's discussion of Walewein in Roman van Walewein, where this character is treated with irony, although he is also presented as the shining example for all knights both by the narrator, and by the characters in this romance.20 I cannot help but thinking that this might be the case in Walewein ende Keye as well, for example in the initial scene, in which Keye accuses Walewein of boasting that more adventures would befall him in the space of a year than would all those present at court (l. 46-53). The reaction of Walewein is multi-staged; first, he says that he is an imperfect man (l. 119), so why would he boast. But when knights of the court have implored him to set aside his anger, Walewein announces that he shall only return to court after he has fulfilled the boast Keye has accused him of (l. 155-160). This comes as a surprise. If Walewein is indeed portrayed here as the example of humility, why does he not put aside his anger? Now he brings the court in disarray, giving the seneschal exactly what he wants, and with that he shows he is not beyond the sin of pride. Also, the humorous scene in the opening of the romance, in which Walewein falls out of bed after a fight with a lion attacking him in his dream, shows that Walewein is not the perfect hero Hogenbirk thinks him to be. In Hogenbirk's view (p. 126), Walewein's fall makes him human. By portraying him as a human being, the author makes it possible for youngsters among the audience to identify themselves with their exemplary hero. But could this scene not refer to someone praying prostrate on the ground? In another scene (l. 734-767), Walewein puts his head in a reliquary (scrine).21 He does so to help a damsel, who had bound herself by a promise to capture his head in a reliquary. Can this scene be explained as a portrayal of Walewein as a mere human being? I do not think so. In fact this scene is humorous because of the nature of the box in which Walewein has to put his head. The function of a reliquary is to shelter the remains of a saint. 22 So in this scene, in the eyes of a medieval public, Walewein is compared to a saint! If one remembers that Walewein's first appearance on the scene is when he is praying in church after he had an disturbing dream about an attacking lion, I cannot but wonder whether or not the author of Walewein ende Keye plays a game with the more religious tinted Arthurian tales (and not completely in Walewein's favour). It seems to me that in Walewein ende Keye we can see the same mixed portrayal of Walewein as in Roman van Walewein and that Hogenbirk's claim that Walewein is the ideal knight has to be modified.

Her last premise (p. 26) is based on studies on other interpolated romances in the Lancelot Compilation, in which it has been shown that the compiler of this manuscript has used existing tales and has modified them to suit the context in which they were put. According to Hogenbirk, it seems likely on the basis of these parallels that the extant version of Walewein ende Keye is an adaptation of an earlier tale, in all probability a Walewein-tale (without a substantial part of Keye). In chapter 7, she evaluates the compiler's editorial interventions elsewhere in the compilation, and she compares these interventions with the state in which Walewein ende Keye has come down to us. In this tale, a number of adventures shows ‘such a standardized, rapid development of events that it has occasionally obscured the course of events’ (p. 178-179), while on the other hand other episodes ‘do not seem to have suffered much abbreviation, possibly because they are central to the overall theme’ (p. 179). In this part of her study (p. 127-161) her observations are quite acute. There is no doubt that there are great differences to be seen in the way episodes have been treated, implying interferences from a person other than the author. Her arguments that the adventure of Keye has been inserted by the compiler and was not originally part of Walewein ende Keye seem sound to me. The structure of Keye's adventure does stand apart from the other adventures. Of course one could argue that the author did this deliberately in order to create a contrast between the two major characters, but I think Hogenbirk is right in arguing that not only the structure of his adventure stands apart, but that also the portrayal of Keye is not consistent through the story as a whole. However, although I do believe her when she says that Keye's last appearance in the last scene of the story betrays the compilers hand, I cannot accept her interpretation of the chapter heading in the manuscript. The original title ‘Hoe si alle quamen te hove die hi verwonnen hadde’, has been supplemented with an additional phrase ‘ende van Keien’ (p.138-39). This ‘ende van Keien’ has indeed been added, but the addition is not in the same hand. Thanks to an illustration (p. 139), one can observe that these words have been added by a different hand. The colour of ink is not important here, but the shape of the letters, notably ‘e’ and ‘a’, is. This addition could therefore have been written (much) later, and it is not an argument that the last part of Walewein ende Keye has been written by the compiler to fill the last leaf of a separate quire. But her arguments that Walewein ende Keye is a reworking of an original (but lost) Walewein-romance did convince me.

 

This study of Walewein ende Keye brings home to us that intertextuality can be a risky business. When can it be assumed that a source text was known to an author (which is, of course, a prerequisite for intertextuality)? The most obvious assumption is that this is definitely the case when an explicit reference is made to a source text, either because the name of the text is mentioned or because a character appearing in one text is explicitly referred to in another tale, although one must hastily add that the mention of the name of a tale does not imply that the author knows the contents of that tale. But when the mention of the tale is accompanied by similarities or contrasts in contents, one can assume that the contents of the mentioned tale were known to the author. Referring to a character is even more complicated, since some figures are appearing in more than one tale, or even belong to a standard entourage.23 So while the Brangien reference in Chrétien's Erec et Enide can safely be interpreted as a deliberate connection made by Chrétien to the Tristan story,24 a reference to Gauvain cannot be interpreted in any other way than as a reference to the complete Gauvain-complex, from the ideal courtly knight to the one who will fail terribly in the Grail-quest because of his womanizing. This latter could be called generic intertextuality as in the words of Bart Besamusca.25

Implicit references are even more difficult to cope with. References to certain episodes can be expressed in several ways. In fact the possible connections between adventures within a tale (the conjoin-

ture and sen discussion), have with the help of intertextuality been transferred to connections between adventures in different tales. So in intertextuality research one finds an abundance of analogies, contrasts, parallels and motif doubling. But adventures are easily invented, especially if one understands the basic structure of adventures in Arthurian tales: Information, Provocation, Information, Determining the Hierarchy (or Fight) and Information.26 With the fixed clustering of function and type of character, all adventures can be classified in two basic types of ab ovo and in medias res.27 I have argued that these basic types help the public in following the tale.28 But it also aids the author. I once started a paper with a summary of an adventure. Readers of that (unpublished) paper accepted that adventure as existing in the Arthurian romance I was studying at the time. Afterwards, I revealed that I had made up that adventure. It was not based on any particular one in the romance, but one of the readers immediately saw connections to several other episodes, episodes I knew, but had not thought of when writing the adventure ... Implicit references can be misleading for intertextuality, provided that we define intertextuality as the study of deliberately introduced connections between tales. There is a gliding scale of possibilities between a near certainty that an author deliberately made the connections, to the conviction that there is no intertextuality to speak off. We cannot just simply compare an episode to all sorts of other episodes and conclude that a given episode has been the source for any given other episode, the only argument being that the earlier episode resembles the other one. Only when more than one episode (or the complete romance) seems to be modelled on another text, we can, with all caution, posit on the basis of implicit references that the author of the chronologically later text knew the earlier text. Hogenbirk has proceeded without sufficient safety measures when using the in itself valid methodology called intertextuality in her study Walewein ende Keye against the back-drop of the genre of Arthurian romance. Her observations on which romances the author of Walewein ende Keye probably and possibly knew have to be treated with the utmost caution. In this instance all we can say is that the author of Walewein ende Keye knew Arthurian lore and invented his own romance (not ‘providing the most idealized, and therefore least human Walewein portrait of Arthurian romance’).29 In this the author posits himself clearly in the Dutch Walewein tradition, as modified by Uyttersprot in her doctoral thesis, but it says nothing of the specific texts he knew. We will never get a satisfactory answer on questions of intertextuality, but it does not mean that we have to chuck in its whole conception. We just have to be careful in what we do, how we do it and how we present our case. If we do not do that, intertextuality slips into speculation.

 

Adres van de auteur: Kan. Venboetstraat 4, nl-6525 tt Nijmegen, bsmelik@ru.nl


Vorige Volgende

Over dit hoofdstuk/artikel

auteurs