Zompist Bboard Again

BGMan wrote: ↑Sat Oct 14, 2023 8:16 am I've wondered for some time if Old Japanese had a more Korean-like vowel system?

An interesting discussion, but could we perhaps have it outside the ID subforum please?

(Unless you can transcribe a paper on the topic, that is!)

I've been a bit busy and focused on other hobbies when I've had time, but I finally went back to the original post to add the consonant inventory Martin was working from. I'll try to add some info from "Classical Japanese: A Grammar" by Haruo Shirane in the near future, as my schedule and energy allows.

As the name implies, "Classical Japanese: A Grammar" is more concerned with the grammar of Classical Japanese, however H. Shirane does mention some onbin sound changes, including when they first took place.

Table 3 on page 497 titled sound changes gives the following:
Verbs
i-sound change: Ki, gi, and shi in yodan become i; EX: 急ぎて -> 急いで
u-sound change: Hi, bi, and mi in yodan become u; EX: 歌ひて -> 歌うて
Nasalized sound change: Bi and mi in yodan and ni in nahen* become n followed by a voiced particle (de) or auxiliary verb (dari); EX: 学びて -> 学んで
Compressed sound change: Chi, hi, and ri in yodan and ri in rahen become tsu; EX: 失ひて -> 失つて -> 失って
Adjectives
i-sound change: Ki in ku and shiku adjectives becomes i; EX: 白き雲 -> 白い雲
u-sound change: Ku in ku and shiku adjectives becomes u; EX: 若く -> 若う
Nasalized sound change: Final ru of adjectives in rentaikei and adjectival verbs becomes n when followed by auxiliary verbs like meri or nari 多かるめり -> 多かんめり

It should be noted that some of these changes are at odds with each other and one change might be more common in Western Old Japanese while others were more common in Eastern Old Japanese.

In chapter 2 section 2 (2.2), Shirane mentions that the difference between unvoiced and voiced consonants (clear/muddy in Japanese, and really unnasalized and nasalized) p/b, t/d, k/g and s/z existed as early as the Nara period, but weren't written differently until much later.
2.3 tells how the syllabic n (or N) emerged in the Heian, but was usually written as whatever syllable had become N. So the nasalized sound changes about can be pinned down to the Heian.
4.4 also states that the verb sound changes above occurred during the Heian.
In regards to the adjective sound changes, a note in chapter 5 (5.3.2A) does not mention when these first appeared. But appendix 2 starting page 339 states that most of these sound changes began in the Heian and does not differentiate verb versus adjective, and in fact A2.1.4 mentions that these changes also occurred sporadically in nouns (EX: きさき "empress" -> きさい "same"), adverbs (まして "all the more" -> まいて "same") and other word groups. This is then followed by two notes: One mentioning that in the latter half of the Edo Period, り -> い in some verbs before -ます. The second is a Historical Note that the き/ぎ -> い is the earliest of the onbin changes, and found in the Man'youshuu (after 759 AD but before the end of the Nara Period.

*corrected for spelling from the text

Shirane wasn't really concerned with the phonology, but he seems to be starting from something like:

/m n/
/p b t d k g/
/s z (h)/
/w r j/

/a i u e o/

I'm not sure how helpful this is, but I figured it's still a helpful data point.

This site may be of use to you'll

https://www.jlect.com/resources.php

Neonnaut wrote: ↑Thu Aug 29, 2024 12:52 pm This site may be of use to you'll

https://www.jlect.com/resources.php

That is fantastic! Thank you

Jlect is super exciting until you realize how little breadth it has -- it's really not usable except as an online user interface for a selection of dictionaries -- standard language, a lot of Ryukyuan langauges, and a very patchy handful of topolects.

Unfortunately and obviously, the non-appearance of forms in JLect is used on the internet to prove that that form doesn't exist.

fusijui wrote: ↑Thu Aug 29, 2024 11:34 pm Jlect is super exciting until you realize how little breadth it has -- it's really not usable except as an online user interface for a selection of dictionaries -- standard language, a lot of Ryukyuan langauges, and a very patchy handful of topolects.

Unfortunately and obviously, the non-appearance of forms in JLect is used on the internet to prove that that form doesn't exist.

It's still a resource I did not know about despite researching for a long while in this sphere, so I'm happy to have it to supplement others I already have.

You're right, and I apologize for jumping on it with both feet so quickly. It is an easy-to-use resource, and I also would guess that most potential users are mostly interested in Ryukyuan, which it's got fine coverage of. And for that matter, it's a pretty decent online dictionary for the standard language, I think.

Beginning my efforts of going through Thorpe (1983), Ryūkyūan Language History.

Abbreviations Geographical affinities
PR SYNCHRONY

• [+cons +syll -son] [-son] > [±ant ±for ±vc ±cont] [±ant ±for ±vc ±cont]
• [+cons +syll +son] [+cons] > [±ant ±cor] [±ant ±cor]
• Palatalization in three stages: *t *d > ts dz before high vowels; *t *d *s *z palatalize before *I
  1. [-son +cor -cont] > [+strid] / _ [+hi]
  2. [-son +cor] > [±ant] / _ [+hi ±bk]
  3. [-son +cor] > [+ant] / else
• [+syll +hi +bk] > [-ro] / [-syll -son +cor] _
• [-son] > [+vc] / [+syll] [+cons -syll] [+syll +vc] ( [+syll] ) _
  Obstruents voice “in a third or subsequent mora, provided it is not preceded directly by a de-voiced (high) vowel”; a morphological rule

VOWEL SHIFTS

• KIKAI
  1. *e > i / {#, w, j} _ (A, Shi.)
     *e > ɨ / s _ (A, Shi.)
     *e > ɯ / else (A, Shi.)
  2. *ɯ > ɨ (Shi.; ongoing, “at least for some speakers”)
  3. [+syll +hi] > Ø / # [+cons -vc] _ [+cons -vc]
• ŌSHIMA & KAKEROMA
  1. *e > *ɨ / [-son +cor -cont] _
     *e > *ɯ / else (Ko., N, Sh., Ya., Yu.)
  2. {*ɨ, *ɯ} > i / ! w _ (Ko.)
     {*ɨ, *ɯ} > i (N, Sh.)
  3. *wɯ > u (C, N. Okinawa)
  4. *e > {ɯ, ʔje} / #_ (Ya., Yu.)
  5. *ɯ > ɨ / [-son +cor -cont] _ (Ko., Sh.)
  6. [+syll +hi] > Ø / unless an illegal cluster forms (Ko., Sh.)
  7. [+syll +hi] > Ø / # [+cons -vc] _ [+cons -vc]
• TOKUNOSHIMA
  1. *i > ɨ / [-syll -grave] _ (In., Ka.; *j is considered [-grave] here)
  2. *e > ɯ / {k, g} _
     *e > i / else (In., Ka., M)
  3. *ɯ > ɨ (In., Ka., M)
  4. *ɨ > i / [+cor] _ (M)
  5. [+syll +hi] > Ø / # [+cons -vc] _ [+cons -vc]
• OKIERABU
  1. East Okierabu
     1. *e > ɨ / [+cons -grave] _
        *e > i / else
     2. *ɯ > ɨ
  2. The rest of the Okierabu dialects (Te., Se., Kam, Tamina) end up with “the simple Okinawa vowel system” (q.v.)
  3. [+syll +hi] > Ø / # [+cons -vc] _ [+cons -vc]
• YORON
  1. *i > ɯ / [-son +cor] _ (Mugiya)
     Noted as “a recent change”
  2. “. . .in C and elsewhere on the island the simple Okinawa vowel system is found.”
• OKINAWA & NEIGHBORING ISLANDS
  1. *{ɯ, e} > i
     This leaves a basic tripartite triangular vowel system, though secondary developments (which Thorpe doesn't get into) have resulted in [e] and [o]
  2. [+syll +hi] > ʔN / _ [+nas]
     [+syll +hi] > ʔN / _ [-son +vc] (Na., S.)
  3. [+syll +hi] > Ø / # [+cons -vc] _ [+cons -vc]
  4. [+syll +hi] > Ø / # [+cons -vc] _ [+cons -vc]
• SAKISHIMA
  1. *i > ɯ / ! {#, N} _ (all except T)
  2. *e > i (all; T keeps [ɯ] distinct, which Thorpe notes as remarkable)
     The above implies *ii > *iɯ initially after a nasal (thence > ii, ee, &c.)
  3. *i > ɯ / # _ [+son +cont] (O, Yo., U)
  4. *ɯ > i / ! [-son +cor] _ (Ik., Sa.)
     Noted as a true “reversal” of the first change listed as opposed to a retention
  5. *ɯ > {i, u} / ! [+cons +cor] _ (Ob.; occasionally this [ɯ] rounds to [u] anyway)
  6. [+syll +hi] > N / _ [+nas]
     [+syll +hi] > N / _ [-son +vc] (O, Yo. U)
  7. [+syll +hi] > Ø / # [+cons -vc] _ [+cons -vc] (Thorpe says this was incompletely done in Ob. but doesn't give further specifics)
  8. *ɯ > i / ! s _ [+cons -vc], r _ (Ku.)
  9. *ɯ > u / r_
     *ɯ > i / else (Ha.)
  10. [+syll +ro] > i / # _[+cons +syll]
  11. *ɯ > i (So.)
  12. *u > Ø / # _ b (“sporadically” and “a rather recent local change”) (H)
• YONAGUNI
  1. {*e, *ɯ} > i
     Appears to have been a two-stage process:
     1. {*ɯ, *ɨ} > X
        *e > Y
     2. {X, Y} > i
     Thorpe sugests that either X = *ɯ and Y = *i or X = *i and Y = *ɨ.
  2. [+syll +hi] > N / _ [+nas]
     [+syll +hi] > N / _ [-son +vc]
  3. *i > Ø / # _ [+son]
  4. [+cons -son] [+syll +hi] > N / _ [+cons +vc]
     [+cons -son +vc] [+syll +hi] > N / _ [+cons -vc]
     “A somewhat rarer change”
  5. Two-stage glottalization:
     1. [+cons -son -vc] [+syll +hi] > Q / _ [+cons -vc]
     2. QC > Cˀ / # _
        QC > C / else
• SPECIAL CHANGES
  • “Assimilations of one kind of high vowel, original or otherwise, to another are exceedingly common”
    *ɯ > i / iC _, _ Ci
    *ɯ > u / UC _, _ Cu
    “. . .and so on. . .”
  • [+cons +vc] [+syll +hi] > N / _ # “[i]n quite a few modern dialects[‘] noun stems. . .”

I also found this (PDF warning). It may be able to fill in the gaps Thorpe left.

For the record…the distinctive features Thorpe used in his 1983 thesis:

bradrn wrote: ↑Sat Oct 14, 2023 8:19 am

BGMan wrote: ↑Sat Oct 14, 2023 8:16 am I've wondered for some time if Old Japanese had a more Korean-like vowel system?

An interesting discussion, but could we perhaps have it outside the ID subforum please?

(Unless you can transcribe a paper on the topic, that is!)

According to Francis-Ratte ( https://altaica.ru/LIBRARY/KOREAN/franc ... tation.pdf ), proto-Japanese had /i e a o u ə/, and proto-Korean had /i e a o u ɨ ə/ (which is unchanged from proto-Korean-Japanese).

jcb wrote: ↑Wed Jan 08, 2025 2:40 pm

bradrn wrote: ↑Sat Oct 14, 2023 8:19 am

BGMan wrote: ↑Sat Oct 14, 2023 8:16 am I've wondered for some time if Old Japanese had a more Korean-like vowel system?

An interesting discussion, but could we perhaps have it outside the ID subforum please?

(Unless you can transcribe a paper on the topic, that is!)

According to Francis-Ratte ( https://altaica.ru/LIBRARY/KOREAN/franc ... tation.pdf ), proto-Japanese had /i e a o u ə/, and proto-Korean had /i e a o u ɨ ə/ (which is unchanged from proto-Korean-Japanese).

I thought Korean-Japanese was not considered plausible?

Man in Space wrote: ↑Wed Jan 08, 2025 4:42 pm

jcb wrote: ↑Wed Jan 08, 2025 2:40 pm According to Francis-Ratte ( https://altaica.ru/LIBRARY/KOREAN/franc ... tation.pdf ), proto-Japanese had /i e a o u ə/, and proto-Korean had /i e a o u ɨ ə/ (which is unchanged from proto-Korean-Japanese).

I thought Korean-Japanese was not considered plausible?

Yeah, I don’t believe it is. I don’t think we should be worrying about it at this stage of development of the ID.

(Also, apologies for not working on this. I should get back to it sometime soon.)

Man in Space wrote: ↑Wed Jan 08, 2025 4:42 pm

jcb wrote: ↑Wed Jan 08, 2025 2:40 pm

bradrn wrote: ↑Sat Oct 14, 2023 8:19 am

An interesting discussion, but could we perhaps have it outside the ID subforum please?

(Unless you can transcribe a paper on the topic, that is!)

According to Francis-Ratte ( https://altaica.ru/LIBRARY/KOREAN/franc ... tation.pdf ), proto-Japanese had /i e a o u ə/, and proto-Korean had /i e a o u ɨ ə/ (which is unchanged from proto-Korean-Japanese).

I thought Korean-Japanese was not considered plausible?

I think that Proto-Korean-Japanese has (rightfully) gotten a bad rap because of the poor methodology (mass comparison and statistics) used by Ruhlen, Starostin, and others for "Altaic" (which includes Korean and Japanese), but Francis-Ratte excludes their reconstructions and starts anew, using only the comparative method and Korean and Japanese data.

Also, I am not qualified to say whether the Korean and Japanese languages are related, but I find that Francis-Ratte's interpretation of the history, archeology, and genetics of Korea and Japan makes far more sense than Vovin's and other detractors'.