Re: 新たに発見された、木簡上の和歌について

Marc Adler wrote:
On May 29, 7:47 pm, Bart Mathias <math...@xxxxxxxxxx> wrote:


The most reasonable hypotheses have always seemed to me to be the ones

[snip]

Let me see if I got this right.

甲　　乙
i yi
e ye
o wo

No, it would be

yi i (If we pretend that "yi" could really mean something. At
least the 甲 version would be more like current "i.")

It seems that 甲 yi is a natural vowel, while 乙 i is a contraction of
things like ui and oi.

ye e

Neither of these go back to earlier single vowels, with 甲 ye usually a
contraction of ia and 乙 e a contraction of ai.

wo o (round o vs. non-round o?)

The 乙 o is the "real" one, possibly starting as mid *front* vowel (so
that earlier Japanese might have had an [e]-like vowel later replaced by
the new "e"s). 甲 wo is often obviously a contraction of things like au
and ua (cf. kazwof[e]- < kazu af[e]-), but given that it often
alternates with u, one might suppose that there was a mid back rounded
vowel to start with, and the contractions blended with it.

A few questions, in no particular order:

1) Where does [wi] (ゐ/ヰ) fit into this?

Basically just the wa-gyou i-retsu vowel. If there were any cases of 乙
i after w, the distinction with natural 甲 must have been lost.

2) Vovin bases his judgments on intonation, right? How is intonation
related?

I am currently unaware of how intonation would relate to this matter.

3) If the distinction seen in the manyogana stops being made in the
Heian period, why is it that we have e/ye, o/wo, and i/wi distinctions
into the Meiji Period? (Cf. Meiji ads for ヰスキー).

I don't see how there could be a Meiji e/ye distinction. How would it
be indicated? o/wo distinctions were a matter of writing, a (relatively
late) revival of the "historic" spelling that lasted through WWII. I
hadn't seen mention of a ヰスキー ad before, but if anyone actually read
it [wi] they must have been speakers of a [wi]-preserving dialect who
would also have said things like [dareka wiru?]. In the Edo period,
before the old spelling revival, the wi and i kanas were pretty much mixed.

4) What is your reasoning for arriving at the above hypothesis?

The key thing in my 1962 (? "1961" was probably wrong) paper was the
environments the distinctions were lost in. There are no kou/otsu
distinctions of front vowels ("i" and "e") after tongue-tip consonants
(t, d, s, z, r, y, n). A similar phenomenon occurred in many varieties
of English, where "dew" and "due" came to be pronounced like "do." In
such English one has /kyuw/ vs. /kuw/, /pyuw/ vs. /puw/, etc., but no
/-yuw/ vs. /-uw/ contrast after t, d, s, z, n (l belongs on the list,
too, but I for one have a "learned" pronunciation of "lieu" which does
have /ly-/, although I probably usually say "in loo of").

It looked and still looks to me like a physiologically conditioned
merger. For other consonants you can coarticulate a [j], but with these
it takes a quick tongue shift.

My arguments for the o's are similar--loss of labial vowel distinction
after labial consonant. In fact that may be where I started; I remember
being shocked by O(o)no Susumu's contention in _Joudaitokushukanadukai
no kenkyuu_* that the distinction was back rounded vs. front rounded
because that distinction would be hard to preserve after labial
consonants. Yeah, like French can hardly distinguish "peu" from "peau"!
Notice also that there are /w/ comes after most consonants in English,
but can you think of an English word that starts with pw-, mw-, or fw-
("bwana" makes me drop bw- from the list, but...)? Ono summarizes in
the book mentioned the work of several Japanese scholars that he didn't
notice was better than his.

Having learned Korean before Japanese, the Korean values of the vowels
in man'yougana also influenced me. Kourui o matches Korean /o/ or /u/
while otsurui o frequently matches Korean /e/ (or "eo"). Otsurui i
often comes out /ui/ (unrounded "u") in Korean.

5) What language did the Japanese and Koreans speak to each other
during the 5th-6th century Japanese occupation(s?) of Korea? Might
there be any Korean place names (for example) that would shed light on
Japanese phonology of this period?

Good question(s) for which no answers spring to what's left of my mind.

6) Aren't there a couple of Manyoshu poems that are nonsensical unless
read as Korean?
If this is true, couldn't the "manyogana transliteration" of the
Korean sounds shed any light on what the sounds actually were?

I suppose. I never got around to taking the opportunity to be convinced
by that Korean Manyoshu poem stuff.

Bart
__________________________
* Sorry about not being able to produce kokuji in this by other than
copy-and-paste. I messed up my Linux three weeks or so ago, trying to
upgrade Debian Sarge to Etch, and haven't quite recovered yet.


.

Relevant Pages

  • Re: (Wandered from) Re: English as a creole.
    ... Yale transcription of Korean, invented by Sam Martin. ... it by writing the round high back vowel as "wu" except after labials. ... The second Japanese vowel is either a recent ...
    (sci.lang)
  • Re: (Wandered from) Re: English as a creole.
    ... Yale transcription of Korean, invented by Sam Martin. ... It has the distinct advantage of avoiding ugly things like "eo" and "eu" for the vowels written that way in a popular system, but one pays for it by writing the round high back vowel as "wu" except after labials. ... When dealing with closed syllables those clever Japanese simply added a vowel. ...
    (sci.lang)
  • Re: (Wandered from) Re: English as a creole.
    ... it by writing the round high back vowel as "wu" except after labials. ... Korean, Tungusic and Mongolic have closed syllables, so anything like ... The second Japanese vowel is either a recent ... note that it's _absolutely natural_ for both languages to form ...
    (sci.lang)
  • Re: Tefu-tefu read as Choo-choo
    ... brief stage as diphthongs, which Japanese, and Korean, abhor. ... Japanese", to save "MJ" for "Middle/medieval J." ... (I also wish information on Chinese and Korean diachronics were easier to find on the Internet.) ... The best case for echo vowel would have it */tiara/, ...
    (sci.lang.japan)
  • Re: Arabic characters
    ... Every Arabic grammar will tell you that every ... Also things like shadda goes on top of the vowel sign, ... few days he was "the other John Lawler," but then he begain writing ... I'd have thought that for someone with an interest in writing systems as well a background in typography the distinction would have fairly leapt off the screen. ...
    (sci.lang)