The previous chapter showed evidence that a significant percentage of the material in the Arthurian stories came from Britain. In this essay, the focus will be on the main characters in the traditional abduction story. No one denies that the names of the chief players—Meleagant, Guinièvre, Keu, Gauvain, Arthur, and Lancelot—and their occurrence in the literary material present incontrovertible evidence for the ultimately British origins of the story.2 However, three topics must be discussed in order to fully understand their true nature: the strange imposition of Urien/Valerin as the alternate abductor, the exact role of each of the main characters in the presumed original tale, and the birthplace of the name Lancelot.
Lanzelet and Diu Crône employ an alternative Valerin/Urien as the antagonist.3 As all the other characters are consistently present and retain their functions in the different versions, it seems reasonable to assume that either Melwas or Urien is a later intrusion on the story, instead of deciding there might have been two. Common sense suggests that Urien is the later addition.4 The fact that Urien of Rheged was both memorable and active during the same heroic age as the Arthurian figures is the telling piece of information here. As a rule, such individuals tend to be drawn into the orbit of more popular leaders as their stories gain wider circulation, and this scenario may well be what happened here.5 The fact that no evidence of his presence exists in the tale may well mean nothing. However, etymology offers an alternative solution.
The earliest and overwhelming majority of the sources use mel for the first syllable of the antagonist. The second syllable tends to vary, but in its earliest form is was. Professor Sims-Williams translates melwas as “honey-youth.”6 The other theory, proposed by Dr. Bromwich, is that of Melwas<mael “prince” + gwas “lad,” “servant.”7
The translations “prince-lad” and “honey-youth” both sound like titles or nicknames. In the context of Ymddiddan Gwenhwyfar ac Arthur or The Dialogue between Gwenhwyfar and Arthur, such an appellation would be applicable as the moniker for a well-known character, such as Urien. On the other hand, it could simply be a poetic invention for a nameless hero as has been believed. Apart from this story and that found in the second vita of Gildas, Melwas is unknown in Britain.
While a Melwas/Urien theory may not have strong supporting evidence, it is viable and explains the later transposition on the continent. As will be seen below, the authors of both Lanzelet and Diu Crône had direct access to Insular source materials. It could be that their source named Valerin/Urien directly. However, no matter what route the antagonist took—as a nameless individual who was later supplanted by Urien or as Urien who during the golden age of the Arthurian literature was known by his nickname—the character in the poems was a consistent one.
Guinièvre seems like a very obvious and necessary element of the tale as well. She is present in all versions as both Arthur’s queen and the victim of a kidnapping. However, there is some negative evidence that this character was not an old one. Both Guinièvre and her progenitor Gwenhwyfar are unknown in any tale whose orthography dates before the twelfth century. She is not to be found in the oldest group of triadic material, either. Using any literature’s composition, one cannot with certainty date the introduction of Guinièvre to much before 1100. It seems at least possible that she is a superimposed figure who was not named in the traditional lore. The fact that the names of famed kings’ wives were rarely kept, and then only when they were the daughters of famous kings is telling here.8 For the record, Gwenhwyfar is given several fathers in the literature, each equally unlikely.9
On a more positive note, the often opposing elements in the four British versions of the abduction suggest that Gwenhwyfar was considered to be Arthur’s queen before the transference of the Arthurian material to the continent.10 However, it is possible that their agreement here is a direct result of the overwhelming influence Geoffrey of Monmouth had on the development of the Arthurian legend. Until more information has been collected on this character, no clear idea of her nature or history may be given for certain.
Keu/Cei’s role in the abduction is much simpler to determine. With the sole exceptions of the terse Vita Gildae and the heavily edited story in Iwein, he is always present. Further, the Welsh “Pa gur?” Ymddiddan Gwenhwyfar ac Arthur, and Culhwch ac Olwen indicate that Cei was originally a formidable warrior in Arthur’s troupe. If, as the majority of the abductions say, Gauvain and others were away when Melwas/Urien came to court, Keu would have been an acceptable substitute for him. It is likely he was an original element in the tale and acted as Guinièvre’s protector.
Arthur is the king and husband to the victim. In Irish lore the abduction of a powerful king’s wife is a common phenomenon and thus an explanation of his role requires a review of the abduction theme in Irish myth. Typical abduction tales include a one-year delay between the kidnapping and the abductor’s death or the surrender of the wife (see below). This is so in Chrétien and the Vita Gildae. However, in the Irish tales the king is usually the rescuer. In La Charrette and most of the variants, Arthur does not have this role. From a literary standpoint this is intriguing. From an historical standpoint, Arthur’s role in the plot as it stands is inconsistent regarding the degree of his power and his representation as a warrior elsewhere. The author would suggest the Arthurian world has at one stage added Gwenhwyfar’s abduction from another heroic age source.11
Any thesis that explores the Celtic aspect of Le Chevalier de la Charrette must also address the question of Lancelot. Did he exist in the British tradition before the birth of Arthurian romance in the twelfth century? If so, why is he not visible in Geoffrey of Monmouth’s Historia Regum Britanniae, The Pictish King-lists, the oldest Welsh poems, or any other British tradition? Is he present in one of these works, but his name is so linguistically altered from something recognizable that one cannot distinguish him? To date, no one has linked him to any twelfth-century personage Chrétien may have wished to extol. Where did Chrétien get the information about Lancelot saving Guinièvre? Did he or Marie de Champagne invent the name? Lancelot does not rescue the queen in the other abductions—except for Lanzelet. Even then he is only one of many. That means that even if he has a history before 1100, Lancelot traditionally did not rescue the queen.
It would indeed appear that Chrétien simply invented Lancelot, but theories abound. Tatlock claims that Geoffrey’s book contains Anguselaus—L’Ancelot.12 He is Lleu Llaw Gyffes of the Mabinogion.13 He is Lleog/Lleminog of Preiddeu Annwn.14 Such theories are linguistically challenged, however, and the characters show no literary connections to Lancelot.
Take Rachel Bromwich’s perception of the issue. In her massive undertaking, Trioedd Ynys Prydein, she states persuasively that “because of the absence of any convincing resemblance in the name forms, I feel no hesitation in rejecting the derivation of “Lancelot”15
The examination will begin where Dr. Bromwich does. In a concluding remark following several well-researched and generally accepted Arthurian matches, Thomas Jones informs us that Middle Welsh Eliwlad is “a kind of rendering of Lancelot.”16 He gives no literary, historical, or referential support for this theory, nor does he explain the word change by oral or written means. The author is in total agreement that Jones’s argument is unsupported and weak.
In Celtic Myth and Romance Professor Loomis theorized that Lancelot was in reality the Irish god Lug descended through several men listed in the Welsh Arthurian poems.17 He continued this belief throughout the rest of his career. The Welsh intermediaries, he said, could be found in Culhwch ac Olwen, “Pa gur?,” and two versions of the same triad of TYP.18 In Loomis’s words:
The puzzle may be explained by the fact that the person and the name were borrowed from some foreign people, probably from the Irish since the Welsh are known to have borrowed a few figures famed in Irish saga and enrolled them among Arthur’s warriors, e.g. Manawidan ab Llyr in The Black Book of Carmarthen, and Cynchwyr mab Nes, Cubert mab Daere, Fercos mab Poch, Lluber Beuthach, Corvii Bervach, and Sgilti in Kulhwch.19
Thus, Lancelot’s original name was Llwch, transferable to Irish as Lugh. His last name was originally Lonnbemnech but, for some reason Loomis does not explain, the linguistically related and culturally tied Welsh could not understand or even translate it. For Loomis, this accounted for the varying appellations the hero has been given over the centuries. However, the theory does not explain the otherwise unmentioned Lleog/Llemenawc of Preiddeu Annwn. It can be conceded that the first syllable of the character, llem<llam meaning “the bounding or prancing one,” does have some affinity with the athletic Lancelot.20 However, there remain difficulties with accepting this proposal. For one, such arguments rely heavily on linguistic evidence, an uncertain tool at the best of times.21 Arthurian studies are hardly that, Old British was going through a period of dramatic changes during the mid- and late sixth centuries; most likely a great many names and places from that period simply disappeared. When the Cumbric language died around 1100, likely many more were lost.
Even if the etymologies of Jones or Loomis are correct and there is a connection to one of the legendary or mythic characters of Irish or pan–Celtic literature, this still proves nothing. The character Lancelot does not need to have been euhemerized; an etymological connection to a deity does not preclude him having been an historical figure. In many instances people have been named after deities of an extinct religion. Current religions occasionally have followers named after one of their deities. The name Bridget among the Welsh and Irish has been quite popular over the centuries, as have Thor or Freyr among the Germanic countries and names beginning with Os among the English.22 In Arthuriana, the textual variant Arcturus has connected Arthur to Ursa Major,23 yet no one would use this isolated case to make any connections between him and any divinities.
Thus the two most prominent theories cannot stand up under scrutiny. Tatlock’s theory, however, seems much stronger at first sight. Tatlock was building off of a comment the Viscount Hersant de la Villémarque made in an 1842 discussion on Le Chevalier de la Charrette. He had opened with a short commentary on a long-standing problem in French manuscripts—the L’. When this nuisance is applied to the Chrétien poem, the L’ could very well mean that Lancelot was originally L’ Ancelot, as several manuscripts read.24
Ancelot could possibly be a name found in the Insular sources, and Tatlock’s Anguselaus seems the most likely target. Dr. Pope’s From Latin to Modern French with Especial Consideration of Anglo-Norman is of great help here.25 Dr. Pope suggests that the name Anguselaus would normally lose its middle syllable in translation from Latin to Old French if it were in a weak and unaccented position, which gu is. This change would leave Anselaus, a name corrupted into Ancelot more easily than either Eliwlat or Lluch.
However, it should be noted that the author here speaks of the alteration of words that occurred over a period of a thousand years. Tatlock speaks of an outright translation from Latin to Old French. Still, the adjustment of Anguselaus into Old French Anselaus and from there to Ancelot is linguistically plausible. From here, Chrétien’s talent and the fame of his poems must have controlled the fate of the name. From a literary standpoint, ancelot in French means “servant,” giving L’Ancelot “the servant.”26 Chrétien would have enjoyed the duality of meanings in the name, his hero was now both a king and a love slave, and the fact that several manuscripts have the form L’Ancelot supports the idea that this was his intended meaning.
Anguselaus’ British form would be Unguist.27 This name matches two figures in The Pictish King-Lists. The more likely candidate to have been drawn into the Arthurian milieu is the son of Forgus, who ruled in the early sixth century. Forgus succeeded in conquering Dalriada and reigning approximately thirty-one years.28 He was a very powerful and warlike king and most likely inspired heroic stories that could then have been brought into Arthur’s literary court. Of course this does not preclude other possibilities; Unguist could represent the cognomen of any British hero or king who existed between 400 and 1200.
With the possibility of Unguist/Anguselaus/Lancelaus/L’Ancelot one has a scribal solution. There are, though, two problems with the theory. As Professor Tatlock and others have made note, there was an Angus of Moray killed in 1130 who claimed to be the King of Scotland, just as Geoffrey’s Anguselaus of Albany does. It has been argued Geoffrey could have been using the name of Anguselaus to introduce a humorous element in the story—he along with everyone in the English court were laughing at their contemporary’s impertinence.
Second, no connection has ever been made with any thing or any one who is listed in the pertinent manuscripts. And while Goodrich’s theory that Lancelot was a noble Pict (and therefore not of interest to the British) might be used to account for this phenomenon, it seems unreasonable to simply assume Lancelot existed within the Arthurian orbit before Chrétien introduced him. After all, several Picts, Irish, and even a few Germanic warriors are mentioned in them.
In a recent article, Dr. Lloyd-Morgan summarized the argument concisely; Lancelot did not have a past before Chrétien invented him in Cligès.29 However, she does add a single, potentially damaging corollary to her conclusion. Tristan was not a prevalent figure, either. Dr. Lloyd-Morgan suggests the two may not have been common in Welsh lore because their themes both undermine beloved kings.30 Thus, the very argument she uses allows for a reason why Lancelot was not a popular character while stories that would have been associated with him were circulating in Wales. The window is still open, and Lancelot may have been a part of Welsh lore at one time.
And it would seem odd if he were an entirely made-up character. If that was the case, he and Cligès would be the only non–Celtic figures in the first round of Arthurian romances. It also seems unlikely to this writer that Chrétien or his translator would have created him for a bit-part in Cligès and Erec,31 and that this same invented character would then have been used at around the same time some twenty or thirty years later in both Le Chevalier de la Charrette and Lanzelet. If that were so it would mean that two men living in different kingdoms during the medieval period would have written independent variants of the same story and starring the same minor personage twenty or so years later. Such a scenario is highly unlikely, especially given the line of tradition we can see attached to other popular figures. The simple fact that Lancelot’s education is by fairies in Lanzelet is evidence that the German version of the story was in an oral environment for an extended period of time.32 One can safely say the Lancelot character had a predecessor before his inclusion in Cligès and Erec and therefore before 1160. However, as has been seen, finding a reasonable predecessor has proven problematic.
Proving that a saga figure lived at all is difficult. However, it is possible. Research on other heroic ages has shown that, if a figure did exist, he is to be located in the same period and general location as the central figure of the saga.33 On the other hand, it would seem that fictional incidents can be attached to a character as soon as they die34 or even within his or her lifetime. Professor Chadwick’s philosophy on the matter has been borne out over the decades.35
The fact is that Lancelot is a mystery as a literary and as an historical character. It is difficult to say whether he became confused later with Arthur or was initially a part of the early version of the legendary king’s band. He probably was not originally associated with saving Guinièvre from Melwas. There are no concrete or even tenable grounds in the Welsh oral sources to make such an assumption. He may, however, have been a Pictish or British king. He may have helped steal a cauldron under the appellation Llenlleawg, as Loomis suggested, but there is no tangible proof that he did. Whether or not any of this is so, the literary figure Lancelot probably existed in some form before 1100. Dr. Bromwich eventually followed Loomis in suggesting an eleventh-century Lancelin of Brittany as a prototype,36 and the author would agree that this name is the most important clue about Lancelot’s beginnings.
It can be assumed that the abductor, Cei, the abducted, and the hero (likely Gauvain or his Welsh antecedent) were in the sources of all the versions of the abductions.37 One can also be fairly certain they performed the same basic functions there. This, as Dr. Bromwich has often pointed out, is remarkable.38 Such a faithful preservation of a foreign culture’s heroes with or without that region’s stock features is, to the current author’s knowledge, unique in ancient and medieval world literature, both oral and written. Yet there it is, present in French, German, and Italian, and later in Dutch, Spanish, Icelandic, Irish, and Scottish romances, with all the accouterments of Celtic literature. This presence is because Glas was probably a written source39 that each progressive author felt a need to reinterpret, not remake. The abduction story, however, has more to offer an inquisitive scholar, as I hope the next three chapters will demonstrate. A penetrating examination of the first full-length version of the abduction will reveal elements in the story that could be historical, followed by a chapter devoted to certainly historical material. The aim of these chapters will be to establish a rough date for the earliest elements of the abduction in Arthurian literature.