The Bear Hunter

Philological commentary on the name "Arthwys"

In his recent article The Northern Arthur , my good friend Aurochs makes a compelling proposal for the origins of the Arthur myth in a figure called Arthuis ap Mar ap Ceneu ap Coel ap Tegfan.

I’ll let readers be the judge of the strength of his argument as a whole, and content myself with a small footnote to Aurochs’s discussion of Arthuis’s name. He says:

Let’s examine the name Arthuis, the origin of which is disputed, much like the name Arthur itself, which no one is able to give a dead to rights most likely origin. I have been rendering the name with the alternative spelling Arthuis, but it is also rendered as Arthwys. The name appears as Athrwys by the 7th century, suggested to be derived from Athro, a root for ‘master’, though Arthwys seems to be the earlier name. Arthwys, sometimes rendered Arthuis, may be derived from Arth (Bear) *uis, from Proto-Brythonic *gwɨs (Knowledge) Arthuis then may mean something like Smart Bear, or Bear of Knowledge. With this same logic the later version Athrwys may mean something like Master of Knowlege, or Teacher of Knowledge. The name Arthwys itself may have even become Arthur over time from copyists errors. Arthuis being a common spelling, it would be easy to see a smudged is becoming an r, rendering Arthuis to Arthur. This is of course conjecture, but, it is always a possibility for how a man named Arthuis ends up being referred to as Arthur years later.

I think it’s possible to make a compelling case that Arthur is actually a name that arises entirely from the Brythonic name Arthwys. The argument is not particularly complex, but slightly too involved for a Twitter thread. Happily, for such things, we now have Substack.

Bear necessities

Aurochs’s advance of an etymology from *arth (Bear) *uis, from Proto-Brythonic *gwɨs (Knowledge) with a proposed meaning of “Smart Bear” or “Bear of Knowledge” gets us 90% of the way to a satisfactory conclusion. For the remaining 10%, we must delve into the murky realms of Proto-Celtic.

Proto-Brythonic *gwɨs descends from Proto-Indo-European *weyd- (“see”, whence also Veda in Sanskrit, videō in Latin, and of course wit in English) via Proto-Celtic *wissus (< *wéydtus), “discovery”).

We would only expect a form with *gw- to arise in initial position (c.f. P.Celt. *windos = Welsh gwyn). In Arthwys, P.Celt’s original *w is maintained because the element occurs as the second part of a compound.

Here is how we would expect this word to develop from Proto-Celtic, according to established sound changes:

1. *Artowissus

2. *Arθowissus (spirantilisation of *t after *r)

3. *Arθöwɨssus (*i > *ɨ, internal i-affectation of *o)

4. *Arθöwɨs (apocope of all final syllables in Proto-Brittonic)

5. *Arθwys (syncope/loss of *ö in Old Welsh, *ɨ > *y in Old Welsh)

*Artowissus probably means something like “bear wisdom” or “bear hunter”. Artognou, also cited in the article, would be a near-synonym for this name, although it must represent either a very early stage of Brythonic (we would expect *Arθgnọw, whence Breton Artgnou a few centuries later), or represent a deliberately archaising form. *-gnow, itself derived presumably from a P.Celt *gnāwos, is attested only in this name and in Brythono-Latin name Uirognous, < ultimately *Wirognāwos, for which we would expect something like *Gwɨroɣnọw.

Since Artognou is attested from a find in Cornwall, and is known thereafter only in Breton, it’s tempting, as Aurochs does, to call this a Southwestern dialectical formation of Arthwys - in fact, I’d go so far as to suggest it is a Southwestern attempt to calque *Arθöwɨs that isn’t quite right. It could also be an attempt to render a slightly different P.Celt. morpheme, *wissos, “known”, which would have had the same outcome phonologically in Arthwys. “Bear Fame”, perhaps?

As for the transformation from Arthwys or Arthuis (a spelling variation which may be down to scribal preference or to slightly different realisations in vowel quality) to Arthur, this seems to me explicable as a case of marginal rhotacism of a kind common to many varieties of Vulgar Latin, and mirroring a classic shift seen between Old Latin and Classical Latin.

Arthwys, Latinized as Art(h)usus could very easily be realised as Art(h)urus, especially by a speaker familiar with the remarks of Varro on this phenomenon and of a “classicising” bent.