Reconstruction talk:Proto-Korean/patak

Japonic origins?
Vovin apparently traces this to Japonic wata. I find this odd in a number of ways.


 * There are only two terms extant in Japanese that are related to this root. Old Japanese, and currently largely obsolete, noun, and Old and modern verb .  If wata were a native Japonic term, I'd expect to see more words related to this root.
 * What I've read of Vovin's reasoning about the phonetics is that Koreanic had no, so a borrowing from a Japonic term with initial would have been realized as .  Except it looks like Old Korean did have , as we see in ?
 * A borrowing in the other direction, from Koreanic into Japonic, is problematic because Japonic had both initial and, and this distinction was maintained for many (most?) terms well into the historic era (albeit with initial  leniting to ).  Yet we have no Japonic attestation for any form pata or fata, only for wata.

I wonder if the Japanese might not reflect a borrowing from a dialectal term in Old Korean, or a related language like Baekje or Kaya / Kara / Gaya, that had initial and was cognate with the Old Korean ? ‑‑ Eiríkr Útlendi │Tala við mig 22:46, 26 October 2020 (UTC)


 * Old Korean did not have initial as far as we're aware.  is just  for.
 * Vovin actually considered this a borrowing from Korean into Japanese in his earlier work Koreo-Japonica (2010):
 * The comparison is valid, but within Japonic the word wata "sea" is confined to Western Old Japanese; there are no Eastern Old Japanese or Ryukyuan attestations. On the other hand, its doublet OJ umî "sea" (< PJ *omi) is attested in all branches of Japonic. Thus, I believe that WOJ wata "sea" is a Korean loan on the basis of its distribution pattern in Japonic. (p. 111)
 * By 2015 Vovin had changed his mind according to Whitman's argument that the OK root pata- should be reflected in Old Japanese as pata-, but you're right that wata- could well have been a dialectal Koreanic form. Some Koreanic varieties today irregularly sometimes have initial for historical  (and this is also a phenomenon that occurs in ideophones throughout the dialects, e.g. ) but it's hard to know since exactly when this has been the case.--Karaeng Matoaya (talk) 23:32, 26 October 2020 (UTC)
 * To formally close the loop, thank you for the additional details above. :)
 * (I confess I've long hated Yale romanization. I get what they were going for, but I strongly disagree with several of their choices.  Using  to denote, when the intended audience of English speakers often parse ㅓ as an open-o or schwa sound, and when the language also has both  and ?  That seems like a special kind of stupid...)
 * Re: wata, unless I run across further information to the contrary, my thinking is more in line with Vovin 2010 than Vovin 2015 -- the distribution in Japonic makes it hard to accept as a native term. The lack of any evidence that this was ever  in Japonic seems more a factor for how and when and where it was borrowed, rather than evidence for native-ness.
 * Cheers! ‑‑ Eiríkr Útlendi │Tala við mig 18:44, 3 November 2020 (UTC)