ARTHWYS, ARTHUR AND THE IRTHING: MY FINAL ARGUMENT FOR A HERO IN THE NORTH

Birdoswald/Banna Roman Fort, Looking North Across the River Irthing

(Aerial Photo Courtesy English Heritage)

In my book THE ARTHUR OF HISTORY and subsequent blog posts here on "Shadows in the Mist", I very tentatively identified as my candidate for a Northern Arthur one Ceidio son of Arthwys, a chieftain found in the ancient Welsh genealogies for the 'Men of the North.'   In these sources I gave two possible etymologies for Arthwys.  One saw this as a fairly standard Celtic name meaning "Knowing Bear" or the like.  The other saw the -wys as the usual Welsh suffix denoting a region.  

From the Geiriadur Prifysgol Cymru:

-wys 1
[H. Gym. -uis (Linnuis), bnth. Llad. -ensēs]
trf. ll. e., e.e. Gwennwys, Lloegrwys, Monwys.

If Arthwys was, then, not properly a personal name, but a personified territorial designation, then given his relationship to other members of his family, the most logical geographical fix for him was the Irthing Valley.  Irthing has been derived by noted place-name scholar Andrew Breeze from a Cumbric word meaning "Little Bear."

The son of "Arthwys", Ceidio, displays a hypocoristic or shortened pet-name from a British *Catu-, 'battle', plus an original component that was later dropped.  There are several instances of *Catu- names which have meanings such as "Battle-leader".  I surmised that Ceidio had been given such a birth-name and that this accounted for the "dux erat bellorum" title ascribed to Arthur.

Birdoswald or Banna appeared to have at least one very important personage whom I had been able to confirm was associated with the fort: none other than St. Patrick, who was actually born there.
Not far away, also in the Irthing Valley, was Camboglanna, a Roman fort that bears a name that could well represent Arthur's Camlann.  A little further west along Hadrian's Wall was the Aballava/Avalana fort with its inscriptions to a Goddess of the Lake.  Myrddin or Merlin belonged just a little to the west in the area around Longtown, as did Myrdin's lord and Ceidio's son, Gwenddolau ('White dales', another possible personified place-name).

Birdoswald Roman Fort

In my new book THE BEAR KING, I presented my argument on why I was not entirely satisfied with this theory.  One of the two points that continued to nag at me was my inability to identify Uther Pendragon.  It seemed to me that I must find some importance for the use of the dragon descriptor, one that went beyond the usual dictionary sense of  'warrior' or 'chieftain.' My gaze shifted to NW Wales for this very reason.

Now, however, having realized, belatedly, that the Dacians with their draco were garrisoned at Birdoswald until the 4th century, I believe that a case can be made for a 5th-6th century leader there being referred to as the "Terrible Magister Draconum."  This would necessitate a survival of Roman military practice at the Banna fort or, at the least, a traditional vestige of that practice, or even of a sort of antiquarian honoring of it.

If I accept this as a strong possibility - unprovable, yes, like so many things in Arthurian theory - but a strong possibility, nonetheless, then I have left only the one stumbling block that has tripped up every researcher: Arthur's name is absent from the pedigrees that are not dependent on the fiction of Geoffrey of Monmouth.  In the North, I can make my case for Ceidio.  In the south, for Ceredig son of Cunedda.  Both can be associated with "bear" rivers.  Both have family members and/or regional designations containing the word arth.  While the Arthurian battles can be most perfectly placed in the North, they can, with some allowable linguistic manipulation (allowable if we accept the Nennius battles as Welsh attempts at battles listed in the ANGLO-SAXON CHRONICLE), be situated in the South as well. Arthur's final and fatal battle of Camlann was in NW Wales, as far as Welsh tradition is concerned.  But this could have been a relocation.  Medraut or Moderatus makes the most sense at Cambloglanna on Hadrian’s Wall, for we know of a Trajanic period prefect named C. Rufius Moderatus, who left inscriptions at Greatchesters on the Wall and Brough-under-Stainmore in Cumbria (CIL iii. 5202, RIB 1737, 166-9, 2411, 147-51). The name of this prefect could have become popular in the region and might even have still been in use in the 5th-6th centuries AD.  A Moderatus in NW Wales is not impossible, of course, and I've suggested that he may have been the original owner of Dinas Emrys, as Ambrosius's placement at that fort may in part have been brought about by the fact that the latter is called in Gildas a 'modest man' and the Latin words moderatus and modestus have, essentially, the same meaning.

So how do I decide between a Northern or a Southern Arthur?  Can I decide?  Or, perhaps more importantly, should I decide?  It does not seem likely that any additional evidence one way or the other is going to surface anytime soon.

Well, as always, any viable theory must deal with the problem of our hero's name.  And it is on this point that I've finally been able to make my decision.

For Ceredig son of Cunedda, whom I've also identified with Cerdic of Wessex, we must assume a British or Irish name or title meaning 'Bear-king'.  This name or title would have been given him because of the Afon Arth of Bear River, in the central part of his kingdom of Ceredigion.  Arthur, however, is beyond any doubt a Welsh form of the Roman/Latin name Artorius.  It cannot be derived linguistically from either British or Irish names for 'Bear-king.' We must propose that Arthur as a decknamen was substituted for a British or Irish name because it was either perceived as having the same meaning or because it closely resembled the Celtic.  We must, however, make a secondary and much more dangerous assumption: that the name Artorius, a rare name even among the Roman, was for some reason known by Irishman or Hiberno-Irishmen in NW Wales.  To add insult to injury, none of Ceredig's immediate descendents with bear-names adopted Roman names to conceal their "Welshness."

We do not have these problems (and they are major hurdles) in the North.  Ceidio son of "Arthwys" is related to Eliffer/Eleutherius of York.  The fort of Uxellodunum/Stanwix at the west end of the Wall was the largest cavalry fort in all of Britain, was a command center in and of itself, and was in close communication with York.  Peredur ("Praetor") and Gwrci, sons of Eliffer, are twice involved in major conflicts on the western half of Hadrian's Wall. And we know that a high-ranking Artorius was at York in the 2nd or 3rd centuries A.D.  Here the name could well have survived, being passed down among subsequent generations of Britons who traced their descent from the Romans or from Romano-Britons.  In my book THE ARTHUR OF HISTORY, I mentioned that the Arthur Penuchel made a son of Eliffer of York in a corrupt Triad was an interesting error.  Eleutherius means 'the Liberator/Deliverer', a known title for Constantine the Great, whose father died in York and who was proclaimed emperor in that city.  Welsh tradition applied their word Waredwr or 'the Deliverer' to the Constantine made a grandfather of Arthur in the Galfridian pedigree.

I simply cannot come up with any good explanation why the name Artorius/Arthur would be found in NW Wales.

Fergus Mor/Mar of Dalriada is found in the Northern genealogies.  He is even made the father (as 'Mar') of Arthwys father of Ceidio.  If there is any truth to the intrusion of Irish blood into the ruling families of the Men of the North, then we can account for the fact that the Arthurs in the next generation all belonged to Irish-descended dynasties in Britain.  We do not have to resort to Ceredig son of Cunedda as Arthur.    

Posts Marking the Site of the Dark Age Hall at Birdoswald

Artist's Reconstruction of the Dark Age Hall at Birdoswald

But if Arthur belongs on Hadrian's Wall, what to do with Ceredig of Ceredigion?

In my opinion, one of two things appears to have happened here.  Either a) Arthur and Ceredig became hopelessly confused in Welsh tradition or b) Arthur of the North was put forward as the preeminent British hero in purposeful contradistinction to Ceredig.  After all, Ceredig/Cerdic is offered to us in English tradition as the founder of the English kingdom of Wessex.  He was an Irishman or a Hiberno-Briton mercenary fighting alongside the English against presumed British enemies of the high-king of Wales at Wroxeter/Viroconium.  As such, he was far from being the champion of the British against the Germanic barbarian invaders.  On the other hand, Arthur of the North may well have only fought Saxons along what appeared to be the frontier zone up and down the Roman Dere Street.

Other than this, I can really say no more.  Once again, I leave it up to my readers to decide what Arthur they want, and why.

A NOTE ON THE ETYMOLOGY OF ARTHWYS

This from Dr. Simon Rodway of The University of Wales on the -wys element of Arthwys:

"The gwys that you refer to from GPC is a verbal form, as is clearly stated there – it does not mean ‘knowing, knowledge’ and it cannot possibly be an element in a name. There is an element –wys found in a number of words of obscure meaning and derivation which could be present in Arthwys, cf. doublets like mam ~ mamwys, neuadd ~ neuaddwys (Ifor Williams, The Poems of Taliesin, trans. J. E. C. Williams (Dublin, 1968), p. 51)."

Or, as I've mentioned in the past, -wys could be the Welsh suffix comparable to Latin -enses, which would designate a region named for Arth, "Bear."