Re: tonal languages

>Kou scripsit:
>
>>  Voiced *anything* (b, d, g, v, z, l, m, n, ng, ny, voiced h, voiced
>>  pinyin j, voiced pinyin x) is going to be be either yang qu or yang ru
>>  in Shanghainese, both "rising tones", though yin ping syllables starting
>  > with l, m, or n, seem to retain their yin pinginess.
>
>Well, on a very quick inspection I think you are right, but there are also
>definitely yin ping vs. yang qu contrasts for y- initials, e.g. y@<rnd> = you3
>= 'exist' is yang qu, whereas the first syllable of y@ngwe = yin1wei2 =
>'because' is yin ping.  Sorry I don't have a proper minimal pair.
According to my two sources, these are not strictly contrasts.
"Exist" starts with voiced h (looks like an upside down cursive y in
IPA, hence yang qu. Like l, m, and n, which I mentioned above,
Mandarin syllables starting in w or y keep their yin pinginess with
their Shanghainese equivalents. To reiterate, Mandarin yin ping w's
stay yin ping, while other tones pick up a voiced initial (v, m, ng,
or voiced h) in Shanghainese, hence yang qu; Mandarin yin ping y's
stay yin ping, while other tones pick up a voiced initial (ny or
voiced h), hence yang qu. Yin qu syllables, of which there are scant
few with "w" or "y" also stay yin qu (at least at first blush). It's
late Friday, I'm tired, and am explaining this backwards, I think. I
shouldn't be using Mandarin as my starting point, but I hope you get
what I'm trying to say.

>Still, if we neglect this, we can say that yanginess goes with voiced
>initials, and ru-iness definitely goes with glottal stop, so the only live
>tone contrast is that between ping and qu, which explains why you can find
>pages that say Shanghainese has only two tones, or even no tones at all,
>but only a stress accent (presumably ping=stressed, qu=unstressed).
Interesting.

>I like this comment from Ramsey, which makes it clear that the tone changes
>are really phrasal tones rather than tone sandhi as such:
[snip the familiar quote]

I, too, found that paragraph very informative.

Kou