Reconstruction talk:Proto-Ryukyuan/mukaze

Sources

@Kwékwlos Assuming Kunigami (Yonamine, Nakijin), Northern Amami-Oshima (Yamatohama), Okinawan (Shuri-Naha) were from dead but reopening in future Ryukyu-go Onsei Database, Miyako and Yaeyama from Nevisky. Where's your Oki-no-Erabu, Toku-no-Shima, and Yonaguni sources? Chuterix (talk) 19:52, 12 June 2023 (UTC)

Okinoerabu from https://www.jlect.com/entry/1695/mukaji/

Tokunoshima from https://www.jlect.com/entry/3401/nukade/

Yonaguni from https://core.ac.uk/download/pdf/223199558.pdf (Nkadi) Kwékwlos (talk) 20:01, 12 June 2023 (UTC)

Thank you @Kwékwlos; I forgot to check JLect; also I have various papers about Oki-no-erabu, Toku-noshima, Yonaguni, etc, by Uwano.

It is possible to use the classical dictionaries on google (but with snippet search; find the expected reflex or japanese equivalant. For instance searching 急ぐ (isogu, “to rush”) brings up ?isjogari ~ ?isjogaruri here, added to PR *isogi).

• Amami Hogen Bunrui Jiten (Northern Amami-Oshima; Yamatohama): https://www.google.com/books/edition/%E5%A5%84%E7%BE%8E%E6%96%B9%E8%A8%80%E5%88%86%E9%A1%9E%E8%BE%9E%E5%85%B8/lBGxAAAAIAAJ?hl=en&gbpv=1&bsq=%E5%A5%84%E7%BE%8E%E6%96%B9%E8%A8%80%E5%88%86%E9%A1%9E%E8%BE%9E%E5%85%B8&dq=%E5%A5%84%E7%BE%8E%E6%96%B9%E8%A8%80%E5%88%86%E9%A1%9E%E8%BE%9E%E5%85%B8&printsec=frontcover
• Okinawa Nakijin Hogen Jiten (Kunigami; Nakijin, Yonamine): https://www.google.com/books/edition/%E6%B2%96%E7%B8%84%E4%BB%8A%E5%B8%B0%E4%BB%81%E6%96%B9%E8%A8%80%E8%BE%9E%E5%85%B8/jfWwAAAAIAAJ?hl=en&gbpv=1&bsq=%E4%BB%8A%E5%B8%B0%E4%BB%81%E6%96%B9%E8%A8%80%E8%BE%9E%E5%85%B8&dq=%E4%BB%8A%E5%B8%B0%E4%BB%81%E6%96%B9%E8%A8%80%E8%BE%9E%E5%85%B8&printsec=frontcover
• Yoron Hogen Jiten: https://www.google.com/books/edition/%E4%B8%8E%E8%AB%96%E6%96%B9%E8%A8%80%E8%BE%9E%E5%85%B8/TRjs9Pc3NCQC?hl=en&gbpv=1&bsq=%E4%B8%8E%E8%AB%96%E6%96%B9%E8%A8%80%E8%BE%9E%E5%85%B8&dq=%E4%B8%8E%E8%AB%96%E6%96%B9%E8%A8%80%E8%BE%9E%E5%85%B8&printsec=frontcover

Kikai dialect dictionary can be found here, also see Kiki Gengo DB.

Today some random IPs were adding Proto-Japonic links here, so I've either created them (PJ *paka (“grave”)) or fixed bad IP mistakes (PJ *Enəti (“life”) and *məmə (“peach”)). I want to reconstruct all of Proto-Japonic; the activity on PJ has been dormant since starting from at least 2021 (after PJ *muk- (“to turn”)). I will now ask you more questions and details on your profile. Chuterix (talk) 20:16, 12 June 2023 (UTC)

Lenition

This is probably from a Kyūshū dialect that fortified from /-d-/ to /-z-/. C.f. Kagoshima mukaze. Chuterix (talk) 16:13, 14 November 2023 (UTC)

A shift from /d/ to /z/ would not be w:fortition, but rather sonorization as a specific kind of lenition, and w:assibilation.

With a front vowel like /e/ as the catalyst for palatalization, a shift from /de/ to /d͡ʒe/ to /ze/ would not be all that unusual.

Given the usual progression of such sounds diachronically, I suspect your move was premature -- surely mukade would be the older form, with mukaze a later innovation. ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:51, 14 November 2023 (UTC)

@Eirikr: How come all Ryukyuan forms show reflexes of mukaze and not mukade? For instance, Shuri should not trigger palatalization for mukade; *nkadi would be expected. Yonaguni fortified *mukaze. C.f. pR *kaze.

I'm saying it's intermeditary form to go to mukantay > mukade > mukaze > modern Ryukyuan reflexes.

There are words that exist in Kyushu but not Standard Tokyo Japanese, that has Ryukyuan cognates. See here for a Kagoshima term meaning 'mushroom', found throughout Ryukyuan: Yamatohama has naba as a prefix or suffix, e.g. ナバヤマ (nabayama, “mountain of mushrooms”), Yoron ナバ (naba, HH), Nakijin ナバー (nabā, HLL), Shuri ナーバ (nāba, LLL), Ikema ナバ (naba, “filth”, LH=HH LL), Ishigaki ナバ (naba, HH), Yonaguni ナバ (naba, LF), etc. Chuterix (talk) 00:32, 15 November 2023 (UTC)

I'm afraid I don't understand much of your post here.

• "Yonaguni fortified *mukaze. C.f. pR *kaze."

What do you think is "fortified" about Yonaguni *mukaze?

Proto-Ryukyuan *kaze is reflected in Yonaguni with the Proto-Ryukyuan /z/ appearing as fortified /d/ in Yonaguni kadi.

• For Proto-Japonic *mukantay, there is no sensible reason for this to become mukaze in Proto-Ryukyuan, then for the assibilation of the /d/ to proposed /z/ to reverse course and become variously /d/ or /d͡ʑ/.

Going from Proto-Japonic *mukantay to Proto-Ryukyuan *mukade, and from there to forms like Kikai and Yoron mukadi, or Kikai and Okinoerabu mukaji, is much more phonologically reasonable. Otherwise you're proposing -nt- → -d- → -z- → -d-, which doesn't make much sense.

‑‑ Eiríkr Útlendi │^{Tala við mig} 02:42, 15 November 2023 (UTC)

• What do you think is "fortified" about Yonaguni *mukaze?

The last syllable. It's why I compared pR *kaze (“wind”) there. In addition Yoron also fortified the last syllable there.

• there is no sensible reason for this to become mukaze in Proto-Ryukyuan

This is why I proposed it gone under Kyūshū influence, where the terms mukaze exist.

It's where all the stuff like pR *iwo (“fish”) came from.

Talk about an early loan from an ancient Kyūshū dialect. Chuterix (talk) 08:16, 15 November 2023 (UTC)

1. Yonaguni /ze/ is not fortified. See again the article at w:Fortition.
2. If you're postulating a borrowing from a Japanese dialect, then by definition it cannot be Proto-Ryukyuan.

‑‑ Eiríkr Útlendi │^{Tala við mig} 18:45, 15 November 2023 (UTC)

I say "Early borrowing"; this might be Kyushu Japanese substratum.

It's /di/ fortifies. Not */ze/ Chuterix (talk) 20:23, 15 November 2023 (UTC)

Thank you for clarifying, I was confused by your notation above that suggested that the Yonaguni term is mukaze.

That said, if it's a borrowing -- even from an early Kyūshū dialect -- then it's a borrowing, and not an inherited Proto term. Proto-Ryukyuan inheritances must be from Proto-Japonic, not from a later established Japanese dialect. ‑‑ Eiríkr Útlendi │^{Tala við mig} 21:57, 15 November 2023 (UTC)

@Kwékwlos, see above -- what grounds do we have for reconstructing any /mukansai/ form? ‑‑ Eiríkr Útlendi │^{Tala við mig} 02:13, 14 February 2024 (UTC)

I based it on the fact that the PR form was *mukaze, evidenced by Okinawan having a ji < *ze instead of di < *de, and in some Japanese dialects where the form is mukaze (https://www.jlect.com/entry/560/mukaze/ and https://www.jlect.com/entry/1234/mukaze/). Either *nt > *d must have arisen from contamination with *tai (“hand”), or *ns > *z from spontaneous affrication (in Kyushu-Ryukyuan and Northern Japanese). Kwékwlos (talk) 02:21, 14 February 2024 (UTC)

@Kwékwlos – What of the multiple modern Ryukyuan reflexes that have /d/, then? Are you postulating that Proto-Ryukyuan /z/ underwent fortition? Or that there were both forms /mukaze/ and /mukade/ at the Proto-Ryukyuan stage? ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:54, 17 February 2024 (UTC)

By comparing this with *kaze, in Kikai, the reflex of *ze can either be di or ji, but mukaze or mukade are found instead, pointing to a likely Japanese borrowing. Tokunoshima has nukade (*kaze > kazï, kadï), but like Kikai, borrowing may be posited on basis of the vowel. Yoron has mukadi (*kaze > hadi) with probably regular fortition. Yonaguni nkadi (*kaze > kaji (1930) > kadi) shows regular fortition. But in any case they can go back to a *z form. Kwékwlos (talk) 13:21, 17 February 2024 (UTC)

@Kwékwlos

Hmm, hmm, ya, thank you for that! I take your point. 😄

That said, I have trouble reconciling the proposed "alternative form" setup as we currently have it. /t/ or /ⁿt/ and /s/ or /ⁿs/ are distinct phones, as far as I'm aware, and I haven't encountered any description of these as interchangeable or otherwise related in a Japonic context.

Might we be dealing with two independent etyma?

FWIW, per JLect, the mukaze form in mainland Japanese is found in Fukushima and Tottori prefectures, north and southwest respectively. This form is not recorded in any "standard" Japanese reference I have to hand, not even as a rare alternative, leading me to think that this form has not been part of "central" Japanese. Meanwhile, the NKD entry cites mukade to the Kojiki of 712, but only with the non-phonetic spelling 呉公, ostensibly glossed as mukate. I have no idea which manuscript of the Kojiki includes the ruby gloss for this.

@Chuterix, you have access to the fuller NKD content, does that have anything helpful to say? (I'll keep an eye on Discord.) ‑‑ Eiríkr Útlendi │^{Tala við mig} 21:58, 19 February 2024 (UTC)

@Chuterix replied on Discord with some interesting data on dialectal variations.

ハガジ	〔千葉〕
ヒカデ	〔岩手〕
ミカゼ・ンカセ	〔島根〕
ムカーデ	〔岡山〕
ムカエデ	〔周防大島〕
ムカジ	〔岩手・八丈島・石川・福井・岐阜・静岡・志摩・南伊勢・大和・紀州・和歌山県・伊予〕
ムカジェ・ムカズィ	〔鹿児島方言〕
ムカゼ	〔岩手・茨城・栃木・埼玉方言・神奈川・新潟頸城・石川・福井・福井大飯・岐阜・飛騨・静岡・愛知・南知多・伊賀・志摩・京言葉・大阪・神戸・和歌山・鳥取・島根・広島県・愛媛周桑・伊予・瀬戸内・福岡・佐賀・長崎・壱岐・壱岐続・島原方言・豊後・鹿児島・鹿児島方言〕
ムカゼー・ムガゼー	〔鹿児島方言〕
ムガゼ	〔山形小国・福島〕
ムカヂ	〔山梨奈良田・淡路〕
ムカン	〔新潟頸城〕
ムカンデ	〔伊予〕
モカゼ	〔石川・鳥取〕
モカデ	〔鳥取・島根〕
モガテ	〔茨城〕

Mukaze is very common indeed, making me wonder why on earth this isn't even mentioned in JA dictionaries. There are also the two fascinating outliers, hagaji and hikade.

Anyway, food for thought, at least.  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 01:08, 20 February 2024 (UTC)