Faces of Arthur Index

Vortigen Studies Index

The undisputed authority on the Historia Brittonum is Profressor David Dumville of Cambridge. As such, anyone wishing to treat of the "Arthuriana" section of this source, with an eye towards shedding more light on either history or historical tradition, must first address the various expressed opinions of this authority which relate to the section being critically examined.

To accomplish this goal, we will confine ourselves to those statements made by Professor Dumville which directly relate to the Arthuriana of the Historia Brittonum account. It lies beyond the scope of the present study to attempt a recapitulation of all the facts and theories embodied in the various publications of Professor Dumville. For these, the reader is referred to the bibliography.

In "The Historical Value of the Historia Brittonum" (Arthurian Literature, 6, 1986, pp. 13-14), Professor Dumville has the following to say on the Arthuriana:

"It was proposed many years ago by the Chadwicks that this account of Arthur derived from a Welsh battle-catalogue poem such as those on Cynan Garwyn and Cadwallon, and this has (I think) received general acceptance. The point here, of course, is that such catalogues – especially when composed centuries after the event, like that on Cadwallon – bear no particular relation to the historical circumstances of the period of the subject. Such a poem is evidence only for the extent of the development of the Arthurian legend by 830 [the reputed date for the composition of the Historia Brittonum], and not for the historicity of Arthur or the alleged events of ca A.D. 500."

Professor Dumville then goes on to voice his expert opinion on the value of the above-mentioned 'battle-catalogue':

"In general, our ignorance of the political history of the British fifth century is almost total; in my view, it is not legitimate to seek to lighten this darkness by the use of unhistorical sources offered by a writer whose ignorance was complete and whose concept of history did not require him to distinguish between certain types of evidence, as we must do. We must seek alternative testimony which will introduce us to other areas of fifth-century culture."

Thus does Professor Dumville dispose of the Arthuriana of the Historia Brittonum.

An even more scathing denunciation of the battle-catalogue’s validity was made by Professor Dumville in his earlier paper, "Sub-Roman Britain: History and Legend" ( History, Nottingham Mediaeval Studies 62, 1977, pp. 187-88):

"I think we can dispose of him [Arthur] quite briefly. He owes his place in our history books to a ‘no smoke without fire’ school of thought. What evidence is there for his existence?… The totality of the evidence, and it is remarkably slight until a very late date, shows Arthur as a figure of legend (or even – as Sir John Rhys pointed out last century – of mythology)… The fact of the matter is that there is no historical evidence about Arthur; we must reject him from our histories and, above all, from the titles of our books."

I believe there are two flaws to Professor Dumville’s reasoning. In the first place, we cannot possibly know whether the "battle-catalogue" is worthless or not. Given the rigid nature of poetic compositional rules followed by ancient poets in Britain and elsewhere, and the emphasis placed on using all sorts of mnemonic devices to preserve the integrity of orally transmitted lore, it is not inconceivable that the very battle-lists which Professor Dumville so summarily dispenses with may, in fact, represent an intact record of military campaigns.

True, we cannot prove that this is so, but neither can be disprove it. There are, indeed, only two characteristics of the Arthurian battle-list which argue for alteration by the author of the Historia Brittonum, or an even earlier transmitter. The first of these is the simple fact of there being twelve battles. The number twelve has known mythological/zodiacal associations; one needs only think of the Twelve Labors of Hercules, for example. But if the list were an artificial creation of 12 battles, we would expect 12 individual battle sites. Instead, we have only 9 battle sites. Claiming that Arthur had four engagements at battle site No. 2, the mouth of the Glein, made up the difference.

The second characteristic has to do with obvious embellishments, which could have been added to the battle-list at any time. To the account of the Castellum Guinnion battle someone saw fit to provide Arthur with religious symbolism in the form of St. Mary. This is understandable, for is not W. gwyn (see Guinnion below) not only "white", but "blessed"? A supposed "Blessed Fort", quite naturally, would have led to the inclusion of St. Mary in this battle. In the Annales Cambriae, the sacred symbolism in which Arthur bore at Guinnion is found transferred to the Battle of Badon.

The 960 Saxons Arthur is said to have personally charged and slain in the Badon battle may derive from Tacitus’s Annals (IV, 73), wherein 900 Romans are slaughtered in the grove of the goddess Baduhenna. However, Thomas Jones (in "The Early Evolution of the Legend of Arthur", Nottingham Mediaeval Studies, 8, 1964) drew parallels between the 960 Saxons of the account of the Battle of Badon in the Historia Brittonum and similar numbers of slain enemies used in early Welsh poetry.

None of these embellishments to an earlier "bare-boned" battle-list detract from the potential historicity of the list as it now exists in the Historia Brittonum. Ironically, quite the opposite is actually the case. Arthur at Guinnion is not nearly so dramatic as Arthur at Guinnion with St. Mary. The same is true of an Arthur who single-handedly slays 960 of the enemy at Badon. Such "poetic license", while totally lacking in historical value in and of itself, would help prevent the battles in question from being forgotten.

In my opinion, the most damaging of Professor Dumville’s academic legacies is that by throwing out the Historia Brittonum’s battle-list, and refusing to further investigate the possible whereabouts of the battles themselves – which for him and many who subscribe to his school of thought is a futile and even dangerous exercise - he has essentially come to two erroneous conclusions: 1) there can be no history in the battle-list, so we can learn nothing about Arthur from it and 2) even if there is no true history in the battle-list, a correct identification of the battle sites cannot tell us anything about how the 9^th century saw Arthur.

These are not only astonishing claims, but have done much to retard or even halt serious research into the nature of the battle-list sites. If we wish to make progress again in this arena – being careful, of course, to avoid the pitfalls of the past – we must begin operating under a new set of premises. I would set forth these premises as follows:

We may conditionally assume, for the sake of any positive results, which may be forthcoming, that there MAY BE some historical relevancy to the battle-list poem of the Historia Brittonum and the two Arthurian entries in the Annales Cambriae.

"Historical relevancy" is here defined two ways:

The battle-list may actually preserve a historically accurate series of battles that were fought by an Arthur in the 5^th-6^th centuries AD.

Or the battle-list may represent a series of battles intended by a 9^th century cleric or his source to represent those places where Arthur was thought to have defeated his enemies. Such a portrait of Arthur would not be history, but a traditional account that sought to provide a geographical context to the sphere of Arthur’s military activity. A traditional account is of immense value historically, in that it can tell us a great deal about the socio-political thinking of the time.

The very nature of the material at hand will quite possibly preclude us from being able to determine whether a) or b) ultimately holds true. However, if the battle sites prove to be, once adequately identified, of widely divergent locations, with no recognizable pattern or regional significance, then opting for b) would be preferable. On the other hand, if a pattern or regional significance does emerge, and this itself conforms with what we know of 5^th-6^th century Britain from other ancient written sources and archaeology, then a) would doubtless be a better choice.

Professor Dumville’s focus has also perhaps been too narrow. He complains about the "’smoke without a fire’ school of thought", yet does not seek to account for the known historical 6^th-7^th century Arthurs. It apparently has not occurred to him to ask why there are four Arthurs of Scottish Dalriada, with the name being found no where else (I will make a case below for Arthur son of Petr/Petuir being an importation into the Dyfed genealogy of the Kintyre Arthur son of Bicoir "the Britain"), only a generation after the 5^th-6^th century Arthur of the Historia Brittonum and the Annales Cambriae. These Arthurs are not so much a "smoke with a fire" as a "smoking gun". Their very existence demands the existence of the Arthur Dumville would have us eschew entirely.

To quote from Rachel Bromwich ("Concepts of Arthur", Studia Celtica, V. X/XI, 1975/76):

"If these four [Arthurs] were all named after ‘the historical Arthur’, it would be a type of commemoration for which Celtic tradition offers no other parallel, so far as I know. That a son (or grandson) of Aedan of Dalriada should have borne a British name is improbable in itself, owing to the other British family connections which are attributed to Aedan, and in particular to the long-standing relations (whether friendly or hostile) between him and Rhydderch Hael, the ruler of the neighboring kingdom of Strathclyde."

By demanding that we ignore the Arthurian battle-list of the Historia Brittonum and the Annales Cambriae Arthurian entries, Professor Dumville is seeking to deny us the only means we may have at our disposal to perhaps discover the Arthur whose name and reputation was already so famous, only a generation after his traditional floruit, that Dalriadan Scots with British wives and mothers named sons after him.

Professor Dumville is also refusing to acknowledge that Dalriadan Arthurs would, it is logical to assume, have been named from an earlier Northern British Arthur, and not from a Southern one. Rather than seeing this as a potential clue, something that may point a qualified battle-list researcher in the right direction, Professor Dumville would have us assess the value of the battle-list itself as historically worthless PRIOR to its having been properly assessed.

Scholars owe it to themselves, their students and to the field of Arthurian Studies to ascertain, once and for all, whether the battle-list of the Historia Brittonum has any historical veracity. To do otherwise, to turn the blind eye to what may be our earliest evidence for a historical Arthur, would be to fall victim to a nihilistic approach unworthy of true scholarship.

Dumville and the battles of Arthur is Copyright © 2002, August Hunt. All rights reserved. Used with permission.

Comments to: August Hunt

VortigernStudies is copyright © Robert Vermaat 1999-2009. All rights reserved