Reconstruction talk:Proto-Japonic/intomi

Reconstruction of medial /o/?
JA sources derive modern izumi from idumi, in turn from +.

w:Proto-Japonic lays out that Proto-Japonic should be reflected as Old Japanese, "except word-finally", where it would be reflected as. I'm not aware of any evidence for any OJP term, which is what we would need as the standalone version of the initial part of this compound. Meanwhile, we do have two other instances of appearing as the first part of a compound, in  and the proper noun.

@Chuterix, what reason do you have for reconstructing medial instead of ? Are you proposing that the idu- initial in these Old Japanese compounds is not ? ‑‑ Eiríkr Útlendi │Tala við mig 20:19, 7 December 2023 (UTC)


 * I just put because scarcity of Ryukyuan cognates.
 * Worth noting is that the initial *i- palatalizes the preceding syllable; that's why it looks like izumi and not idumi in the descendants. The pJ attributive would be *-o (if you need evidence just ask; or c.f. EOJ attributive -o), and Shuri (Okinawan) *ijimi/*izimi and Yamatohama (Northern Amami-Oshima) *izïmi would be expected.
 * Thus, the derivation would be *int-o-mi (go.out-ATTR-water). To account for *meNtu we could hypothetically say *-mi and *me- are from *mer, but this is overly speculative and should be taken no more than a,grain of salt. Chuterix (talk) 20:36, 7 December 2023 (UTC)
 * Where does *mer come into any of this? Your mention here is the first time I've encountered any mention of *mer for "water" in Japonic. ‑‑ Eiríkr Útlendi │Tala við mig 22:30, 7 December 2023 (UTC)
 * Separately, sounds like what you're proposing for this Proto-Japonic entry is still consistent with a compound of idu + mi, just with idu reflecting a yodan conjugation paradigm and the (presumably older) -o₁ adnominal form. Is that correct?  if so, should we add an etym section? ‑‑ Eiríkr Útlendi │Tala við mig 22:43, 7 December 2023 (UTC)
 * Yes, we should add etym section here. Chuterix (talk) 23:04, 7 December 2023 (UTC)