Re: THEORY: Ray on ambisyllabicity

Jesse S. Bangs:
> > How would you account for the distribution of [?]?
> >    better   [bE?@]
> >    winter   [wIn?@]
> >    untidy  *[Vn?AIdi]
> >    today   *[?@dEI]
> >    monitor  [mQnI?@]
>
> Woah, woah, woah!  There are some serious dialectical problems here, which
> may be contributing to the confusion.
I don't think there's confusion. Just rational disagreement.

> In my dialect (Central American) *none* of these have [?].
Of course not. Nobody in N. America AFAIK has [?] for /t/ except
adjacent to nasals. But so what? If we're enquiring about whether there is
evidence for ambisyllabicity in English, positive evidence from a single
dialect will suffice. There's no subsidiary claim that it will generalize
to all dialects of English, let alone to all languages.

> The initial /t/ is [th]
> (aspirated) while the medial ones here are voiced flaps.  This argues for
> /t/ being included in the onset of the following syllabe, since, for
> example, the same thing can happen across acknowledged word boundaries:
>
> Go to the store.
> [g@u d@ D@ stOR]
The environments for tapping in N. American & Australian seem to be
a subset of the environments for English glottaling, the subset being
defined by an additional requirement that the /t/ be intervocalic. It's
not obvious to me that the best way to state the intervocalic requirement
is in terms of syllable structure, but at any rate the data from tapping
dialects is less supportive to an ambisyllabicity analysis.

> Based on your other remarks I'm inclined to beleive that the sounds in
> question are indeed ambisyllabic, but this particular argument doesn't
> hold up.
I'm not sure I grasp your reasoning here. The argument doesn't hold up
for Cockney because of GenAm data? Or the argument doesn't generalize to
all dialects of English? (which is true, but where does that leave Cockney?)

--And.