Now it seems to me that there are two orders of ill-constructed romances, and that we shall do well to differentiate between them. In one case we have a number of incidents of secondary character, obviously borrowed or imitated from those occurring elsewhere, strung together more or less cleverly on the thread of a hero's individuality. The incidents are all to be found in other romances, and as a rule none of them have any suggestion of Celtic or mythic origin. The literary style is superior to the matter. Such romances are e.g. Rigomer, Torec, Le Chevalier à la Manche. A very favourable example is Méraugis de Portlesguez. These are all certainly late romances.
In the other case we have a romance even more ill-constructed, but consisting not of incidents but of whole short tales, manifestly independent of each other, and some of them of distinctly antique and mythic character: the literary style is poor and the whole is less a romance, properly speaking, than the material out of which a romance can be evolved. This, I believe, marks an early stage of development, and of this we have naturally but few specimens. The Lanzelet is, I believe, one.
If I mistake not, the groundwork is a series of lais, each complete in itself, and having no connection with what precedes or what follows it. It is in no real sense a biographical romance, though perhaps it might be called a tentative effort in that direction. The Mantle episode certainly formed a single lai; the Fier Baiser, now found with other adventures, probably originally did so.24 Certain of the episodes, too, possess a distinctly archaic character, e.g. the description of the fairy's kingdom as a isle of women where no man penetrates, a conception much older than the Fata Morgana of the prose Lancelot; and the description of Guinevere's prison, the magic slumber in a fair dwelling, ein wünneclichez haus, surrounded by a dense thicket infested with serpents, is the sleeping beauty story in its oldest 'other world' form.25 The position of Gawain in the story is that held by him in the earlier, pre-Lancelot romances.
I cannot accept the suggestion of a biographical Lancelot from which both the Lanzelet and the Charrette were drawn. If we remember that the first mention of Lancelot in Arthurian romance can only be traced to the second half of the twelfth century, it does not seem probable that by 1164 (when, or about when, Chrétien wrote his poem) he could have become the hero of a fixed biographical romance. Nor, the motif of his liaison with Guinevere once introduced into the story, is the compilation of such a version as the Lanzelet subsequently probable. Professor Foerster feels this difficulty, and suggests a solution, which a little more consideration would have shown him to be untenable. On page xlvi. of his introduction to the Karrenritter, he says, 'wenn wirklich Kristian zuerst den Ehebruch eingeführt hat, so ist doch die Annahme zulässig dass Verehrer Arturs und seiner Frau diese neue ehrenrührische Erfindung zwar gekannt, aber mit Entrüstung abgewiesen haben, um ja nicht des idealen Königs Ehrenschild zu beschmutzen.' But a few pages further on the writer himself refers to the story of Guinevere and Mordred as told by Geoffrey26 and Wace. He must therefore be well aware that there can be no possible question of Chrétien's having introduced the motif of Guinevere's faithlessness; that is one of the oldest and most original features of the Arthurian story. The question is not, 'Did the queen have a lover?'—that was answered in the affirmative long before Chrétien's day—but, 'When did Lancelot become her lover? Was it through the version of the Charrette?' a very different matter.27
Taking into consideration the construction of the poem, and the character of the contents, I think we are justified in considering the composition of Ulrich von Zatzikhoven as a collection of lais which have not yet been worked over or taken final literary shape. When the scattered Lancelot stories did this, it was under the influence of a motif foreign to the original legend, his love for Guinevere. How that came to be introduced into the legend is a matter for separate consideration, but I do not think there is room for doubt that it was this introduction which determined the final and literary form of the Lancelot story. All conflicting elements, such as the various love affairs, were rejected and only the original germ retained.
And what was this germ? Authorities will no doubt differ. Some perhaps will say it was the story of Guinevere's imprisonment and rescue, but they must remember that in the Lanzelet this is not the work of the hero. I think myself that the root of the Lancelot tale was simply a Breton lai, relating the theft of a king's son by a water fairy: this seems to be the one abiding and persistent element in the tale, all else is uncertain and shifting. Here the hero is Arthur's nephew; elsewhere he is but the son of an old ally; at one time his father is a tyrant, 'chassé' by his own people; again he is a noble king, the victim of treachery and a foreign foe. Sometimes Lancelot's mother lives to see him restored to his kingdom; sometimes she dies while he is yet in the care of the fairy, and never sees her son again. He has two cousins on the father's side, Bohort and Lionel, and a bastard half-brother Hector; he has no relations on his father's side, but is cousin to Gawain through his mother. He is Guinevere's lover; he is not Guinevere's lover. He is unmarried; he is very much married—three times at least! He has four children born in wedlock; he has but one son, the offspring of a liaison. He is the most valiant knight of Arthur's court; he is scarce worthy of mention. Among all this shifting tangle and contradiction, there is but one thing, and one only, fixed and certain, he is Lancelot du Lac. I do not see how we can avoid the conclusion that in this record of his youth we have the one fixed point of departure for all the subsequent unfoldings of romance.
Not that this story was always unvarying in its details, on the contrary we find in it marked divergences. Thus in the Lanzelet the motive of the theft is clear, the fairy desires a champion and protector for her cowardly son; the motive in the prose Lancelot is not apparent; probably it was a mere capricious fancy for a beautiful child.
And if the motive was not always clearly understood, still less so was the character of the fairy. In fact she seems to have considerably puzzled the mediæval romancers. In the first instance the story would be excessively simple, she would probably be such a water-fairy as we find in Tidorel, and Ulrich seems to have retained this idea when he calls her a Merfeine or Merminne, but as the lai gained popularity, and it became necessary to supply details as to her kingdom, etc., it would be supplemented from other legendary sources. Ulrich's own description, the land of ten thousand maidens where no man penetrates, is manifestly the Meide-land which in Diu Krône Gawain visits, and which is universally admitted to be a remembrance of the 'Isle of Women' of old Celtic tradition. It may have touched the Lancelot lai through the medium of the Gawain's story, but as a 'property' of old Celtic belief it may well have been known independently. I think it probable that this identification may explain a very curious passage in Diu Krône, where Kei reproaches Lancelot who has failed in the glove test in the following terms:
'er hât daz vil rehte erspeht,
daz iz di gotinne
verkurt an ir minne,
diu iu zôch in dem sê.'—ll. 24517-20.
Certain it is that while the queen of the 'Isle of Women' does not appear to be addicted to child-stealing, she does entice, or abduct, earthly knights to be her lovers. It is not impossible that a version of the Lancelot story, redacted by some one familiar with the real character of the kingdom, may have represented him as the queen's lover. It is also not impossible, were this the case, that the story of the imprisonment of Guinevere