Module talk:ine-verbs
middle 3sg/pl endings
[edit]I noticed that 3sg of *ḱéytor is given as *ḱéytor, *ḱéyor (with 'new' ending first), but 3pl as *ḱéyror, *ḱéyn̥tor (with 'archaic'/'stative' ending first). So I would reverse the order of the middle 3pl endings, and I also doubt whether -rór/-ró should be given for thematic verbs. But for the optative I would only give 3pl -ró (which is what Ringe gives on Sanskrit evidence).
A further debate would be whether verbs like *ḱéytor should be given with the endings -tor/-n̥tor at all (given the attested 'archaic' endings in some languages). See for example the discussion in {{R:ine:HCHIEL|page=2144}}: "retention of *-or may have been due to high frequency, e.g., in a core vocabulary item like Ved. śáy-e ‘lies’".
@Victar @Mellohi! @Caoimhin ceallach Exarchus (talk) 19:57, 16 October 2024 (UTC)
- First part: I agree. Concerning ḱéytor and similar: I think bespoke inflection tables is getting too far in the weeds. Better to mention evidence for archaic inflection in a reconstruction note. —Caoimhin ceallach (talk) 21:20, 16 October 2024 (UTC)
- Garcia Castillero has written multiple papers dismissing the t-free endings, in one of which he proposes essentially that Insular Celtic -or (OIr. -ar) was clipped from -etor so that the passive (< middle) forms have the same number as syllables as their past passive participles (a situation originally only true in weak verbs). — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 21:49, 16 October 2024 (UTC)
- Without actually reading it, I'm not buying it. Trying to explain a bunch of suspiciously similar irregularities as independent innovations instead of as archaicisms is doomed to fail in my view. Also, he doesn't even cite Jasanoff's Hittite and the Indo-European verb (2003), which is the basis for the model we currently have. —Caoimhin ceallach (talk) 23:15, 16 October 2024 (UTC)
- One opinion about the t-less endings is that they were used for intransitive verbs. See for example Beekes' Comparative Indo-European Linguistics, page 268, or also Indo-European middle verbs formed with -r- as a phonetic archaism by Witczak. I don't pretend to know who is correct here.
- (as an aside: Witczak makes an error in giving Sanskrit 3pl. śénte on page 50, firstly because the actual form is irregular śérate but also because the regular form would be śáyate) Exarchus (talk) 08:23, 17 October 2024 (UTC)