While wandering aimlesly through the King Lear section of Isaac Asimov's Guide to Shakespeare I encountered his stating that Cornwall was a contraction of the Latin "cor," horn, and "wal," Wales. Asimov writes at length about the many anachronisms in Lear, but this word seems anachronistic too. Don't we have a Latin word attatched to an earlier Celtic word? Did the Romans have a Latin word for Wales?

Did the Romans have a Latin word for Wales?
Did the Trojans have a Trojan word for Snails?
Did the Stygians at the Styx
Coin a phrase for fir toothpicks?
Modern Linguists chase such fascinating tales !

Well, the whole thing is wildly anachronistic, but as Cornwall is the modern name of a place, it's not anachronistic to use that.

I have my doubts about whether the name Cornwall contains the meaning 'horn'. The earliest Latin form is Cornubia, I think. The lenition from [ b ] to [v] and then [w] is a familiar process in the British language(s), so is enough to explain the later forms Cornovia, Cerniw, and modern Kernow, and the Old English compound Corn-wealas (wealas 'foreigners', which also gave Wales).

However, I can't see how the [ b ] could get there in the first place. The Latin for 'horn' is cornu, the Germanic is horn, so the British was presumably corn- too, perhaps even cornu-. The strengthening of an original [ u] to [w], then to [b], in that position is much less likely.

This suggests to me the original name of the country contained Cornub- or Cornob-, and the connexion with 'horn' is an early folk etymology.

Fascinating, Aput. I have a question about "lenition." I hadn't heard it before, but in looking it up it says that "lenis" is when a letter is articulated with relatively low pressure of the airstream below the glottis, as English (b) and (d) compared with (p) and (t); then for "lenition" it says, "To undergo an increase in sonority or become lenis. Said of consonant sounds, as when (p) changes to (b), (b) to (v), or (v) to (w)."

So "lenis" describes the low pressure of the airstream below the glottis, but "lenition" is the changing to "lenis"? Would that happen with different accents?

The terms lenis and fortis, weak and strong, are ones I consider of only marginal value in phonetics. Traditionally [p] and [ b] are described as voiceless and voiced. In English however, and most other Germanic languages, they're both mainly voiceless and it's other features that distinguish them (aspiration after them, pin v. bin, and vowel length before them, cap v. cab). In the search for a single neat feature distinguishing them, the terms lenis and fortis were drafted in, but they're pretty vague: I've never seen a good phonetic description of them that I clearly know how to use. Whereas aspiration and vowel length are clearly real.

Lenition, however, is also a vague term but definitely real, and a common process: any weakening of consonants between vowels. One classic example is Latin [t] in matre-, which is [d] in Italian madre, a weaker [ð] in Spanish madre, and the Old French [ð] has disappeared entirely in Modern French mère. In some Spanish accents it's also disappeared in some places, e.g. -ado pronounced [ao].

In current English the only lenition I can think of is the treatment of medial [t]. In American accents this is always a kind of [d] now, as it is occasionally in some others. I don't know if the term 'lenition' should cover similar changes, to [?] in most of England, to [r] in Liverpool, and to a kind of [θ] in part of Ireland. I wouldn't define lenition in terms of this unhelpful term 'lenis' though.

Well I looked this one up and found much the same as Aput.
Cornubia c705 and Cornwalas 891
The latinized form being Cornovii meaning "peninsula people"
Their origin varies slightly in that the root is Celtic KERNOU meaning "horn".

Right, well 705 clinches it. There appears to be no evidence at all that 'Cornwall' refers to 'horn' or 'peninsula'.

First, in Roman times there wasn't a separate Cornwall, the land being part of the kingdom of the Dumnonii. It was only with Anglo-Saxon incursors that Dumnonia was reduced to the peninsula west of the Tamar. So we have no early Roman name for Cornwall. The year 705 cited in the Oxford Dictionary of Place Names is obviously from Bede, and he, writing in Latin, not his native language, had to Latinize a name from another language that wasn't his. So we can't work out exactly what the original Celtic name was that he thought would look good in Latin as Cornubia.

Second, the Cornovii were nothing to do with Cornwall: they were up in Shropshire and Cheshire. There appears to be not a skerrick of evidence (that I can find on the Web) that the Cornovii gave their name to distant Cornwall. In fact I don't think there's any evidence that there was ever a tribe called the Cornovii in Cornwall. The bare assertion in the Oxford dictionary gives no evidence.

Nor is there evidence that 'Cornovii' comes from corn- 'horn', except for the obvious fact that they begin the same; nor is there any evidence that the so far unrelated name Cerniw/Kernow/Corn(wall) comes from corn- 'horn'.

Note also: if Cornwall was named for the historical Cornovii of Shropshire, that scuppers the 'peninsula, because it looks like a horn' theory, because Shropshire isn't a peninsula.

I don't think it's a coincidence that Kernow/Corn(wall) resembles Cornovii: probably the two Celtic names do come from the same Celtic meaning. Whatever that is.

I remember when I studied Spanish, there was an idiom that I now don't remember what it meant, but it was with the word "codo," meaning "elbow." I remember my Spanish teacher being very precise as to how we pronounced that "d" in "Codo;" it was to be a very soft "d," he said, and not like any of our letters. Perhaps that was lenition?

Yes, Spanish b/v, d, and g are all soft (lenited) between vowels. They're only explosive as in English when at the beginning of isolated words, or in the groups mb, nd, ng.

I found an online Gaelic dictionary which gives the following for horn;
Corn - a drinking horn
early Irish -Corn
Welsh -Corn
Breton -Korn
Latin -Cornu
English -Horn
Greek -Keras

Peninsula is listed as Tairbeart which is clearly not related.

So would that account for the transition of the Isle of Wight from its Roman name of Vectis (which still appears on the side of its buses)?

Another contemporary reference is the Annales Cambriae (Welsh Annals), written in poor Latin by Welsh monks. For the year 875 they record "Dungarth rex Cerniu, id est Cornubiae, mersus est" -- "Doniert king of Cernyw, that is Cornubia, was drowned". Cerniu is a Latinization of the Welsh name Cernyw, so evidently Cornubia was an established Latin form of the name by this time.

The actual copy we have dates from about 960, so it's close enough to be accurate.

The Annals contain the only two quasi-historical references to King Arthur (516, Battle of Badon, and 537, Battle of Camlann).

King Doniert was the last, or last significant, king of Cornwall. His tombstone still exists, by the side of a quiet road, on Bodmin Moor. You can park and read it: "Doniert progavit pro anima" - "Doniert ordered this for his soul". I can't explain why different Latin texts spell his name in two significantly different ways.

Vectis is Wight. But with these names you need to check the actual historical use to find whether it was (a) a real Celtic name that the Romans adopted in Roman Britain (and in this case the Anglo-Saxons adopted too); or (b) a late or mediaeval Latinization when they needed to write about these places in Latin, like Oxonia or Cantabrigia, which obviously don't go back to Roman and Celtic Britain.