The other names of Arthur

Bear Hunter Part 2

In a post from last year I explored the most promising etymology for the name “Arthur” (as in “King”) as a product of Latinising tendencies in Britain’s sub-Roman period.

What I didn’t do at the time was discuss why the prominent alternatives to this etymology are probably wrong. Let’s do that now.

Contenders

The most frequently cited candidates for the origin of the name Arthur are:

1. Artorius, from the roman gens Artoria, a plebeian family that produced several high-ranking soldiers. Of particular relevance is Lucius Artorius Castus, who served as Prefect of Legio VI Victrix at some point in the late 2nd or early 3rd century AD.

2. *Artorīx, a Proto-Celtic name meaning “bear king”.

3. *Artowiros, a hypothetical Proto-Celtic name meaning “bear man”.

4. Arcturus, the Latinisation of the Greek name for the brightest star in the constellation Ursa Major.

5. A continuation of P.I.E. *h₂er-tor-, the same root from whence the most controversial word in the Indo-European family.

All five of these have severe shortcomings that our preferred candidate, Arthwys < *Artowissos/Artowissus does not.

Artorius: 3 centuries too early

The Artorii are mentioned only scantly in historical accounts, but many more are known from inscriptions. The family’s origin and the origin of the name itself are obscure, although it is most likely a relation of the Etruscan personal name Arnth, Arnthur “fierce, biting”.

There are major issues with an Artorius connection to Arthur. In the first place, the most prominent Artorius associated with Britain, Lucius Artorius Castus, was most likely stationed there at the tail-end of the 2nd century AD. The earliest attestation of any variant of the name Arthur doesn’t occur until almost 400 years later.

Artúr, its Irish version, is the name of the son Áedán mac Gabráin, a chieftain of Dal Riata. Artúr (or Arthuir, with which it appears to be interchangeable) likely inherited his name from his Brythonic mother’s family (two of Artúr’s nephews also bore British names). It’s very likely that this Artúr was actually named for the Arthur, placing him within the vicinity of the 6th or 5th centuries AD, far beyond the memory of a Roman officer who only briefly served in Britain in a position which likely never saw combat.

Aside from the historical problems of associating Artorius with Arthur, we also have to contend with insurmountable philological problems. Artorius, borrowed into the Proto- or Common Brythonic stage of insular Celtic, would give us *Art(h)ri, not *Arthur, which would have likely been borrowed into Old Irish as Artri, which happens to already be an Old Irish name (from *Artorīx, mentioned above). Speaking of which–

*Artorīx: already exists (in Irish at least) and looks nothing like Arthur

Proto-Celtic *arto-rīx (< *h₂rkto-h₃rēgs) Would be expected to yield **Arthrriɣ in the earliest layer of Brythonic, and **Arthri(g) in Old Welsh, and As mentioned above, it’s attested in its expected form in Old Irish, and such. a close name in Old Welsh would likely have been understood as being identical and not preserved in a different form.

*Artowiros doesn’t exist (and would be wrong if it did)

*wiros generally appears as the first element of Celtic names, not the last. If it did exist, its early Brythonic reflex would be **Arthwir, becoming *Arthwur in Old Welsh (not **Artgur as is sometimes suggested: *w > *gw only occurs word initially as we discussed last time). In turn, Arthwur would become Arthwr or possibly Arthwyr, both of which are problematic for metrical reasons (Arthur is always rhymed in Welsh poetry with words containing a long u vowel, rather than a short w or y).

Arcturus makes little sense on phonological grounds

All Latin borrowings into Brythonic occurred early enough that sound changes between Proto-Celtic and Proto-Brythonic were already complete, but early enough for Latin loans to be fully assimilated to the changes that took place between Proto-Brythonic and Old Welsh.

This means that Arcturus would have been borrowed as **Arktur, likely realised as **Archtur, which would have yielded **Arithur in Old Welsh.

*h₂er-tor- leaves us hanging

*h₂er-tor-, which most likely makes its way into Proto-Celtic as something like **artoros, would continue into Brythonic and later Welsh as **Arthor.

So we’re back to Arthwys

Last time, I suggested that Arthwys could easily end up being Latinised as Arthurus, but I’d like to expand on this to suggest that the name arose as an independent form (with Arthwys continuing as a separate name in Brythonic) thanks to being borrowed into Irish right before Primitive Irish made its final transformation into Old Irish.

In this scenario, the late Common Brythonic *Arθöwɨs would be borrowed into Primitive Irish as Artowis, with the following result in short order:

1. *Artowis

2. *Artuwis (i-affection raises *o to *u)

3. *Artujih (*w palatalises to *j in the presence of *i, *s > *h)

4. *Artui (loss of word-final *h from *s)

We’re still, however, missing an r. This is where I think we should call upon the Latinizing tendency.

It is entirely possible that the Brythonic speakers of the heroic age were just as keen on Latinising the names of their chieftains as they were in later periods. If this is the case, *Arθöwɨs or might have already existed in a Latinised form like Artowirus or Artovirus.

If *Arθöwɨs was borrowed into Primitive Irish as *Artowir, it would resolve naturally as Artúr or Artuir following half a century of phonological upset. This would be just enough to distinguish it from Brythonic Arthwys, or for the two forms to become confused – especially in the context of intermarriages between Irish and Brythonic aristocratic families which would doubtless have carried such a prestigious name between them (As seen in the case of Arthúr mac Áedán mac Gabráin above).