Arthurian linguistics are unpredictable and often don’t follow predictable patterns. Explain how Gwalchmei became Gawain or Gereint became Erec. Camelot is no different. I have heard it twisted into “Castle of the Bear” after the first syllable in Arthur’s name. Southern England has claimed South Cadbury was his home for centuries, and that transition is even more difficult to follow. My personal favorite is that it is a simplified version of “Caermalleator”, Castle of the Hammerer. Arthur’s reputation in the earliest poetry is that of a punishing raider so that possibility at least sounds good.
I don’t have the linguistic background to make an argument though, so I’ll let that rest. Most scholars have tried to locate Arthur based on what they can find in the literature and what they know of the period – very limited in both cases. Dr. Bromwich once compared the extant British literature to the part of an iceberg sticking out of the sea. We don’t even know the names of too many battles much before 570, let alone the politics behind them. It’s pretty hard to work out a strong theory without that.
Our assumptions run deeper than that, too. We have always assumed that Arthur was a Saxon killer based on what we read in Nennius. Now that you have seen how ahistorical that is, you can see how flawed the assumption is.
All we can really go off of are associations. There are local legends about Arthur throughout the island and even into Brittany, but he is connected to Edinburgh in one of the oldest poems, “Pa Gur?” Mabon is mentioned there and elsewhere; he he was a god connected to the area between the walls. Hueil, brother to Gildas, is named in several early sources. He is also linked with the north. His father Caw is said to have lived beyond the Bannawc Mountains, in other words Scotland.
Belatacudros is mentioned as Beli, Pelles, Pellinore, and many other forms once in the earliest British literature and figures prominently in grail literature. He is named in many Welsh families, but the original worship of Belatacudros focused on Hadrian’s Wall.
All that gives us a consistent view of Arthur’s geography – between or at least near the great Roman walls. And if we list all the places named as his capital in the poems, stories, and even romances and delete all of them that were put there with any sort of bias (saint’s lives tend to place kings near the monastery, political histories tend to place him where they can support a contemporary agenda, etc.), strangely enough only one site really emerges consistently – Carlisle.
That might seem a little misleading at first though. There are actually two Carlisles in Britain, an Old Carlisle that may have served as the supply base for Hadrian’s Wall and modern Carlisle which is a few miles away. Once that correction is made, however, it seems to make sense. It fits most of the associations. It is not in one of the coastal areas that were conquered immediately. Carlisle was not in the middle of the Germanic territories, but then again it doesn’t have to be.