Re: THEORY: Sandhi

From: "Cheng Zhong Su" <suchengzhong@...>
> As for hard to learn, it will depend
> on what we want. If we want knowing more in life time,
> we has to detect more information in every single oral
> actions.
Er, not actually.  An example is that lang "Ehmayghee chah" where all the
numbers were single syllables with the same consonant and different vowels.
1234567890 - gay ghee guy gaw gow gah geh gih ga(t) goh.  He used no phonetic
representation so it's not easy to tell what sounds exactly he meant, but it's
clear that such a system would be useless in an area with imperfect transmission
quality (such as, say, over some cell phones...) because *the information is too
tightly condensed*.  If you miss a bit, the whole word is lost or misunderstood.

(Dunamy suffers from this too, actually: 345 are /f@/ /fo/ /fAv\/)

An optimal language will have a balanced signal-to-noise ratio....

> After all, when we used to the system,
> it want be a hard job, some tone language children
> even don't know what is a 'tone'.
Actually, it's irrelevant whether they know what a tone is.  (Do
English-speaking children generally know what 'stress' is?)  What matters is
that they can _produce_ tones, whether they know what they are doing or not.
(Grammar still exists, even if it isn't studied!)  And if someone whose mother
tongue is toneless wants to learn a second language that happens to be a tonal
language, then they _will_ have to *both* know what a tone is *and* learn to
produce them.

    *Muke!

Replies