Re: USAGE: syllables

On Mon, 2 Feb 2004, Alexandre Lang wrote:

> On Sun, 1 Feb 2004 22:37:46 -0600, "Nik Taylor" <yonjuuni@...>
> said:
> > Alexandre Lang wrote:
> > > Maybe breaks between words, for example?
> >
> > Those are often marked by simple spaces.  And there is the use of # to
> > mark word-boundaries where necessary.
>
> Is there any phonetical difference between [.] and [#] though, besides
> length?
I'm not sure that there's necessarily a difference of length. Syllable
breaks, as far as I know, are mostly used to imply which allophone of a
phone would be used. For instance, in a language like (some versions of)
English, a t has a different pronunciation at the start to at the end of a
syllable, and different again when there's an [s] before it in an onset,
so you might want to point out the differences between /as.ta/, /a.sta/
and /ast.a/ (I think these differences are all implied in English by
things like stress and word breaks, but another language might be
different). Typically one wouldn't denote a syllable break when implied by
stress markes.

Word breaks, on the other hand, have other differences in addition to
those of syllable breaks. For instance, in most dialects of English tmk,
only one syllable in a word can have primary stress, or in French
according to Christophe* the last syllable of every word (which mightn't
correspond to a written word) has the stress.

There's probably other differences, too. Of course, in a purely mono-
syllabic language, there'd be no difference between word and syllable
breaks. The exact meanings is probably language-dependent.

* I doubt he's unique in this, I'm just citing my source.

> > Well, except that a language that nasalized a vowel when following a
> > nasal consonant would also do it when preceding a nasal consonant, so
> > you'd simply describe the rule as "nasalize vowels when a nasal
> > consonant exists in the same syllable".  As I understand it, the
> > combination onset + nucleus simply doesn't pattern distinctly from the
> > whole unit "syllable", so that there's no need to create a special term.
>
> So a vowel followed by a nasal always becomes nasalized? Why is it not
> possible to build allophones only depending on the onset?
It is. And IIUC, in French, /an/ isn't nasalised whereas /a~n/ is. And I'm
quite capable of saynig [in] with no nasalisation if I want. And you can
build allophones only depending on the onset. The degree of
diphthongisation in my /bi\u\/ is greater than the degree in either
/di\u\/ or /i\u\b/. Or /r/ is unvoiced in my English when after an
aspirated stop, but not after an unaspirated stop.

--
Tristan