Reconstruction talk:Proto-Celtic/anextlom

*g or *x

@Mellohi!, Mahagaja Is there a way to explain the /g/ in Old Irish from *x? I guess that's why Matasović has *g. But that's difficult too, because even if we assume the suffixation happened after *g > *x/_*s or *t in Pre-Proto-Celtic, it should have become [x] anyway, as we see in the preterite anacht. —Caoimhin ceallach (talk) 14:50, 9 February 2024 (UTC)

anacol has a /k/, actually, not /g/. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 15:31, 9 February 2024 (UTC)

I don't think there's any way to get Old Irish anacol directly from *anextlom. The intervocalic /k/ suggests rather an *anekk(V)lom. —Mahāgaja · talk 15:38, 9 February 2024 (UTC)

You're both right. I should have looked it up before asking. It's in GOI (p. 113). I guess first you had assimilation in a cluster *xtl > *xkl and then regular delenition of homorganic consonants. —16:29, 9 February 2024 (UTC) Caoimhin ceallach (talk) 16:29, 9 February 2024 (UTC)

Ah I forgot what actually led to this question: McCone (1997, p. 209) seems to believe anacol and innovatory Middle Irish forms -ainic, -aincfe have /g/. But that might be a mistake. Caoimhin ceallach (talk) 14:13, 10 February 2024 (UTC)