Re: tonal languages

On Wed, Jan 08, 2003 at 01:12:37PM +0800, Florian Rivoal wrote:
[snip]
> >But coming back to what I *am* aware of... some examples of tone sandhi in
> >my Hokkien dialect include the following: (sandhi'd words marked with '*')
> [snip]
>
> one of the thing which amazed me with the shanghainese, is that quite
> often, the tone resulting of the sandhi was not par of the tones listed
> as "shanghainese tones".
Unsurprising. The same probably also happens in Hokkien, it's just that
I'm not conscious of it. Tones are not fixed, rigid features; the
"official" tones are often just what they would be in isolation. When in
combination, they tend to mold and merge with each other, leading to tone
sandhi and "intermediate" tones which don't always correspond to one of
the "official" tones.

> A total count of all the possible variants(only on the material i have
> read up to now, so maybe there are more) would bring the theorical 5 to
> a total amount of 9, or even 12 if you count the glotal-stoped tones as
> distinct one.
Well, for the longest time I've considered Hokkien as a 4-tone language
myself; but apparently some linguists think otherwise, and decide to
combine long/short into the definition of tone as well.

> But i have always seen shanghaienese refered as a 5 tone language. Does
> this mean every body is wrong, and it is not a 5-tone one, or does this
> mean it allows off-tones tonal variations?
Probably the latter. I think it's analogous to phones vs. phonemes.
Perhaps you can think of "tonemes"? :-) The point is, tones aren't as
fixed as some people might imagine. They are more like "tonemes" that
could correspond to one or more actual tones. (And as a native speaker of
a tonal language, I can say that mentally, I regard sandhi'd tones the
same as the un-sandhi'd versions---they just come out different,
spontaneously, when in certain contexts. I am also slightly aware,
however, of specific tone sandhi's in certain contexts that are done for
the sake of euphony.)

> an interesting thing to note is that expect someone with linguisicts
> background, shanghainese people will tell you their language has NO
> tone.
Unsurprising. There is this whole (mis)conception that Mandarin is the
"most authentic" (even though that's not really true), Chinese
dialect/language. And therefore, it is supposedly the most aesthetically
pleasing in many ways, including tonal shapes. Other dialects/languages
with tones that are perceived to be "flatter" than Mandarin are often
considered "atonal". (Atonal in the sense of "discordant" rather than
"tone-less". Discordant relative to the "official" Mandarin, of course.)
This doesn't mean that said languages are really tone-less; it's just a
matter of perception.

[snip]
> It ain't so simple. not all tonal language do so. I believe that
> cantonese and vietnamese ( and maybe other, but i don't know) have some
> constraints on music making, such as: you can only put a rising tone on
> a rising melody, for exemple. of course the melody is not exactly the
> initial tone, but it preserves its spirit. rising is rising, falling is
> falling. But it is true that mandarin does not folow this rule at all,
> and do not respect the tones for music.
True. Actually, now that you mention it, I do remember someone mentioning
that melodies written by Cantonese speakers often roughly corresponds to
the tonal shape of the lyrics. But this doesn't apply when putting lyrics
to an existing non-Cantonese tune, though.


T

--
"Uhh, I'm still not here." -- KD, while "away" on ICQ.

Reply