In about 1136, a Welshman named Geoffrey of Monmouth wrote an ambitious composite of British literature and pseudohistory known as Historia Regum Britanniae. It featured a famed British figure named Arthur, who had been hitherto obscure to the continent.
Arthur was an immediate hero in medieval Western Europe, and as a result Geoffrey’s book would take a back seat only to the Bible in terms of copies made in the Middle Ages. His main character would even go so far as to be named one of the three Christian Worthies by antiquarians within a few centuries.
Arthur’s fame was fortuitously supported by the political landscape of the time as well. Henry II was the dominant force in the Celtic lands at this time. He controlled a large portion of the Welsh population, Scotland, England, and half of France. His political control allowed for the unimpeded spread of the Arthurian legend to France. From there it spread to all of the literary and musical minds of the French musicians/poets known as the trouvères and troubadours. Because of their efforts, the legend would spread to all of Europe.
It was not long before other British heroes were associated with Arthur.1 The trouvères and troubadours formulated courtly personalities for them based loosely on their legends. In the twelfth century, several romance poets would pull stories out of this fast-growing body of information and write on them. The greatest of these was Chrétien de Troyes, who did much to create a new genre of literature with his Arthurian romances. He successfully blended a British plot with the concept of courtly love and a broad knowledge of ancient and contemporary literature. The subject matter introduced by Geoffrey of Monmouth and the manner of presentation brought to fruition by Chrétien de Troyes influenced and at times served as an inspiration for thousands of medieval and modern writers. Their authority continues to do so to the present.
However, the British sources of information for Geoffrey, Chrétien, and many of the early romance writers remain in doubt. Geoffrey undoubtedly drew on Welsh genealogical tracts and Bede,2 but beyond a few specific examples, much of the source material he used remains uncertain. With Chrétien and many early romance writers, nothing but a British plot and characters’ names have been distinguished.
The trends in Celtic scholarship would also favor caution about pinpointing a specific source or sources for the romances. Every major British written source for this period has come under scrutiny in the past forty years. Gildas’ historical prelude to his denunciation of several contemporary British kings in De Excidio Britanniae (c. AD 535) is often vague. He is often demonstrably historically incorrect when he does manage to be specific. In several major articles Professor Dumville has demonstrated that Historia Brittonum is a composite, pseudohistorical book with only limited uses as an historical source.3 Kathleen Hughes’s study of Annales Cambriae has demonstrated that the original document was compiled in St. David’s beginning in AD 795.4 She has also shown that from the years AD 613 to 795 it drew on Irish chronicles and the Northern British Chronicle.5
Professor Jackson, Dr. Jarman, and Professor Dumville have advocated a more tentative stance to the poem Y Gododdin concerning its historicity and reliability as a literary source of the sixth century.6 In addition, Professor Sims-Williams has stressed that few of the literary materials may be dated more specifically than pre–1150. Thus, much of the native Welsh material cannot be proven to be free of Geoffrey’s influence, especially when it comes to Arthur.7
The present work is in part a response to their labors. The above-mentioned scholars have cumulatively demonstrated the seeming futility of reconstructing fifth-century British history with the limited resources available and therefore have inferred the desire for more material with which to work. This is the service I am attempting to render here. The conclusions I have come to may well be shown incorrect in the future, but a better understanding of the period must be the result of such labors. I believe it more useful to propose a new and possibly incorrect thesis than never to make the attempt.
It is my contention that other, ancient literary sources of information did exist in Britain at one point. These oral and literary records dealt mainly with fifth- and sixth-century figures and were the rudimentary beginnings of the King Arthur of the medieval literary world. Later these records were transferred to the continent as tales or, less probably, as sagas.8
Further, I believe that four of the five extant Arthurian poems written by Chrétien were based on one or several of these sources of information from Britain. His poems in their turn became the inspiration for variations of his plots in particular and the Arthurian corpus in general. For this reason, it seems necessary to start with him.
This book will explore only two of Chrétien’s four romances: the abduction of Guinièvre and the grail. They were chosen because they have not been given the one-dimensional sovereignty label of Yvain and Erec. Because their nature has never been widely agreed upon, the author needs only to focus on the evidence and where it leads instead of breaking down accepted theories point by point. It is good fortune that these two are also the most deeply enriched in the Celtic tradition and have the most and earliest versions. Because of these facts, they will yield the strongest findings.
The contentions I begin with are twofold and may seem radical at first. However, the pages below will bear them both out. First, it is possible to reconstruct the British prototypes of the abduction and grail stories with the use of several parallel versions also dependent on British sources. Second, one might date these prototypes safely to the ninth century and tentatively as early as the sixth century. The idea that the Arthurian romance writers of the late twelfth and early thirteenth centuries were historians in any sense is not suggested. Certainly no sixth-century warrior crawled across a sword, as Lancelot does in La Charrette. I would, however, suggest that the persons who wrote on these stories of the abduction and the grail accessed a body of sources deriving from the sixth century.9
Central to the tales under study here will be Chrétien, the first and most imitated court poet of the continent during the Middle Ages. It is through an understanding of him that a clear picture of the original version of these tales is achieved. The transition between heroic tale and romance is also most clearly seen through him. Every scene and detail that is out of place or unnecessary in his poems will be scrutinized for such anomalies.
Having determined them, the question of where such oddities might have been derived will be posed; here the work of previous scholars has suffered because there is only a paltry amount of knowledge about early Britain. Very little was written and even less preserved from the Arthurian period. So, the theory of Hakam’s Razor will be employed to postulate a most likely cultural scenario for the timeframe in question. Knowledge of the pre–Roman British culture will be derived from Roman, Greek, and the occasional native sources. The law texts and the observations of eleventh- and twelfth-century writers regarding contemporary Welshmen will give a strong picture of medieval society. The most reasonable transitional culture between the societies of Roman Britain and historical Wales will be assumed, guided by what little contemporary evidence exists for the period in question.10
Chrétien’s influences will also be discussed at length, most notably his two patrons. Their hand will also be looked for in both poems, and reasons for their influence on certain aspects of the poems will be examined. In these scenes and details Chrétien has in some cases apparently been asked to edit or disguise an established line of the plot because it ran counter to the patron’s wishes. In these cases, Chrétien inevitably replaced the offending scene or detail with something especially palatable to his patron. Knowing the motivations of both Marie de Champagne and Philip (of Flanders), it is possible to identify and retrace these instances. I will then compare the segments where these instances occur to what can be discovered of sixth-century Britain using the meager sources at hand. Among these elements a significant number will be uniquely tied to British culture before AD 900. A larger number of these elements can be placed there more easily than in the twelfth century.
However, Chrétien will not be the only author studied here. With both stories, all versions showing some independence from him and contain traditional British information are discussed to determine the earliest possible date at which their common original must have existed.
Parallel to developing a more accurate period of origin for the source materials for both stories is an evolution of our understanding of the characters and their predictable roles in a reconstructable original. The analysis of all analogous material is employed to fine tune this method.
The process outlined here is designed to reveal the most historically accurate elements in literature that was entirely created as fiction. This approach will no doubt raise eyebrows among specialists in British historical and archaeological circles. But the study is not purely one of historical or archaeological research. It is, by the very nature of the evidence at hand, a multidimensional one: the techniques of each field must be applied. It must be kept in mind that the various approaches are just as valid within their own fields as the historical and archaeological approaches are within theirs.11