"The very tasty carrot."
Translation:An curran glè bhlasta.
The extra h is a case of lenition, where the starting consonant sound changes under certain conditions. Glè and cho are both intensifying adverbs, but only some intensifying adverbs trigger lenition. Glè is one that does trigger lenition, while cho is one that does not trigger lenition. See here for more info.
The origin of cho is not known for certain. MacBain points out that the Irish is comh and the Welsh cyn. He also says it comes from com 'with'. I presume he means Old Irish com(-) as Latin com/cum would be nonsense. This makes sense as words like this are used as equative prefixes in several languages - so they all basically mean the same as cho.
Wiktionary agrees, which is not surprising as it may well be an uncredited copy. It suggests the Irish cognate is chomh, as do other modern dictionaries. Dineen says it is 'cómh generally chómh'. So this suggests that MacBain had the older version and that the version with an h is more modern. So putting all this together, it is clear that the word used to start with a c and the change to ch (which is lenition) is a development in both Gaelic and Irish. There is of course no clue as to why this lenition has occurred.
My contention was that words that are themselves lenited like this seem, for no obvious reason, not to cause lenition themselves. The most blatant case is where where we have two forms of the same word, do, that causes lenition, and dha that doesn't. I would welcome any examples or counter-examples (of words that are inherently lenited, not ones that just happen to be lenited because they are after some other word that causes lenition). D
hhzhang, Terminology here is extremely confusing. To any non-Celtic linguist, lenition means any sound change that could be classed as 'weakening' from the Latin lenis weak. But what does this mean? Some phoneticians say that a sound is weaker if the air pressure is lower. This happens if you let the air out more easily (/b/ → /v/) or if you obstruct your vocal chords (/p/ → /b/), for example. This may happen diachronically (over time) as when a /t/ sounds more like a /d/ in some American dialects, or synchronically (due to a change in circumstances not a change in time) as when the s in house voices to what is pronounced as houze when the word is used as a verb.
Within Celtic linguistics, we tend to use the word for the more specific changes that are found, for example after feminine nouns. But changes on this list can occur diachronically as well - for example tu is usually now thu. And that is the sense in which I used the word here - a change that is similar to what happens after a feminine noun, but a permanent change, for unknown reason.
Yet further confusion is caused by the changes being different in different languages. For example, in Gaelic, c → ch is lenition, but c → g isn't (and in fact c → g is nasalization in Irish , whereas in Welsh, c → g is lenition, but c → ch is aspiration. In other words, the sound change that happens in one language in one situation happens in another language in another situation.
Yet further complication is caused by spelling conventions. We tend to think of lenition as 'adding an h', but there is no such convention in Welsh - indeed in one case (rh → r) you actually remove an h.
Part of the confusion is caused by the fact that all the sound changes that we regard as Celtic mutations are quite natural, so often occur diachronically. The example with cho may be one of these. Another example is the change /s/ → /h/. We are familiar with this in mo sheanair but it is a diachronic change in Welsh as sean has simply changed to hen.
Altogether a right mess. D
The etymology that you bring up seems solid. I guess when I think of lenition, I think of it only applying to a word that has a non-mutated form as well, which is why I said that cho is not lenited. But I can easily see your point as being valid as well, that cho is a lenited version of a historical word that lacked the h.
Thanks for the informative response. As a relative beginner in Celtic languages, I've only encountered mutations in the synchronic (synchronical?) sense, since knowing about diachronic mutations isn't generally needed to learn the modern language. But with the additional information you provided, I definitely agree with you now that if cho does have the etymology you detailed, it's a case of diachronic lenition. Also agreed that it's all a bit of a mess - one of the peculiar...charms? of Celtic languages in general.
Regarding your original thought on the existence of other words that trigger lenition which are themselves lenited - perhaps the preposition bho? It starts with a lenited consonant sound, but it's not clear at all to me whether this is actually a case of diachronic lenition based on the etymology given on Wiktionary.