Re: THEORY: Ray on ambisyllabicity

On Sun, 8 Oct 2000, And Rosta wrote:

> Ray:
>
> > Before I'd read Dirk's letter I'd have shown the Welsh as:
> >          s     s
> >         /|\   /|\
> >        o n c o n c
> >        | | | | | |
> >        h a p p 1 s
>
> If bottom tier is not taken to be representing timing units in addition
> to phonetic content then it may be that the difference is purely
> theory-specific. Dirk may know better.
He may, but then he's been looking at a lot of Shoshoni lately, which
has very different prosodic properties :-). It is a usual shorthand to
let phonetic symbols stand in for feature bundles and the "root" or
organizational node which gathers them together. I already replied
that I would indeed include a timing tier for this example.

> > [...]
> As for the point about [V] (= your [@]) and [I] in word-final position, I don't
> think it's valid (I mean: it's pertinent & well-taken, but turns out to be
> invalid), for two reasons. The main reason is that the generalization
> about lax vowels can be restated in one of various ways which remain arguments
> in favour of ambisyllabicity but rule out your counterexamples as irrelevant.
> For example:
>   I.  stressed syllables containing a lax vowel must be closed
>   II. stressed syllables must contain a minimum of two segments in the rhyme
Yes. The latter is crucial to Hammond's analysis. He couches it in
moraic theory but the idea is the same; a stressed syllable must be
minimally bimoraic. In his analysis, I follows from II: since stressed
syllables must be bimoraic, a stressed syllable containing a lax vowel
(which is monomoraic) must have a coda consonant and the mora which it
carries with it. So such a syllable will snatch a medial consonant
leaving the second syllable without an onset.

> [...]
The stuff just snipped concerns evidence from insular varieties of
English so I forbear comment, pleading ignorance of the (admittedly
fascinating) facts.

Dirk

--
Dirk Elzinga
dirk.elzinga@m.cc.utah.edu