Reconstruction talk:Proto-Japonic/pukoru

Odd medial vowel?
What is the reason for reconstructing this as pukoru?

Proto-Japonic /o/ should be reflected as OJP /u/ in non-final locations -- granted. However, Proto-Japonic /u/ should also be reflected as OJP /u/, in any location.

We don't have any Ryukyuan cognates in our entry here. Poking around in JLect, I don't see any Ryukyuan cognates with medial /o/. Thorpe's 1983 paper Ryūkyūan Language History here (see page 337) reconstructs Proto-Ryukyuan *pukore or *pukure, but again none of the modern reflexes seem to indicate medial /o/: where the second Proto vowel is reflected at all (and it hasn't been contracted out), the modern vowel is /u/, which would seem to necessitate a /u/ at the Proto level as well.

What have you all found? ‑‑ Eiríkr Útlendi │Tala við mig 23:24, 16 February 2023 (UTC)


 * Doesn't it follow the same scheme as, , atleast as far as the reflexes follow? Kwékwlos (talk) 00:01, 17 February 2023 (UTC)


 * Thorpe's 1983 dissertation paper lists *sirosi as number 128 on page 304. *pukore ~ *pukure is number 225 on page 337.  I thought I saw kusuri or some reflex in there somewhere, but I cannot find it at the moment.  The PDF is not properly digitized and cannot be searched using CTRL+F, and the bafflingly complete lack of any semblance of organization in how the terms are ordered makes any term search a chore.
 * Anyway, for *sirosi, Thorpe says:
 * "The second mora in this word could not have been *ru, from the Sakishima evidence."
 * Given his cryptic notation, I'm not sure which item is supposed to be "Sakishima", not least because "Sakishima" isn't even listed in his abbreviations at the beginning of the paper -- that said, all of the listed terms have ru as the second mora except for "Sh." (Shodon, Kakeroma) and "Ko." (Koniya, Ōshima) that have rju, so this comment doesn't make a lot of sense without more context that Thorpe fails to provide. Thorpe also states just two sentences later:
 * "This word should be given little weight as phonological evidence, apart from its accents."
 * ... which seems to argue against his firm statement above...?
 * As I mentioned at the top of this thread, I'm seeing lexical evidence in reflexes that the middle vowel at the Proto level might have been o, but also nothing to categorically rule out that it might just as well have been u.
 * FWIW, Thorpe says "It was probably originally derived from JR *siri 'to know, to recognize'." Monolingual JA sources just flat out state that it is cognate with ; I don't think there's any doubt about that, despite Thorpe's more tentative statement here.
 * I also had a look in Pellard's 2008 paper Proto-Japonic *e and *o in Eastern Old Japanese, but unsurprisingly (given the title), this doesn't explore any o ↔ u correspondences, and neither sirusi nor sirosi appears anywhere in that text.
 * Cheers, ‑‑ Eiríkr Útlendi │Tala við mig 18:32, 17 February 2023 (UTC)
 * (looking back...) @Eirikr: Sakishima refers to the Yaeyama dialects of Ryukyuan. Some distinction could be observed there. Chuterix (talk) 23:05, 17 July 2023 (UTC)


 * Update: Thorpe lists *kuso’ri as number 130, on page 305. I'll have to dig into that later -- again, his notation doesn't make a lot of sense.  ‑‑ Eiríkr Útlendi │Tala við mig 20:55, 24 February 2023 (UTC)