Re: R: Moraic codas [was Re: 'Yemls Morphology]

Dirk:
> Oops.
>
> On Thu, 12 Jul 2001, dirk elzinga wrote:
>
> > No need to invoke the classical foot typology. What it looks
> > like to me is that you have a system in which coda consonants
> > are not moraic, and that the stress falls on the penultimate
> > mora.
>
> I neglected the example /.ka'si.ku.la./, in which the final
> syllable is in fact not counted (just like you said). Sooo ...
>
> It seems that since coda consonants are not moraic, a final
> closed syllable only counts for one mora. But this mora is *not*
> ignored for stress in your first example, /.ir'man.den./, or in
> the examples with a final diphthong or long vowel /.am.ni'pai./
> and /.in'tu:t./. However, the final mora *is* ignored in
> /.ka'si.ku.la./. Right now, I can't think of a single set of
> rules which would cover both kinds of cases.
>
> On Fri, 13 Jul 2001, Roger Mills wrote:
> >
> > Rule 1:  a diphthong/long V in the ultima is stressed, otherwise--
> > Rule 2:  a heavy penult (-CVC-) is stressed, otherwise--
> > Rule 3:  stress the antepenult (no matter whether heavy or light)
>
> This is as good a description as I can see, and demonstrates
> that syllables show a ranking of "stressability":
>
>         CVV > CVC ( > CV)
>
> This is essentially the same hierarchy which linguists have
> proposed for heavy syllables. If a language counts closed
> syllables as heavy (bimoraic), it will also count long vowels as
> heavy.
Luca:
> GENERAL RULE: final syllables are not moraic (thus they can never get
> stressed), unless they contain a long vowel or a diphtong.
>
> The right-most foot gets stressed:
>
> IRMANDEN /ir'manden/
> KASIKULA /ka'sikula/
>
> Final syllables containing a long vowel / a diphthong are always
> stressed:
>
> AMNIPAI /amni'pai/
> INTUUT /in'tu:t/
>
> Would it be better expressing the general rule in these terms: the last
> foot is stressed (considering the abovementioned foots to be respectively a
> trochee, a dactylus, a hiambus and a spondee?).
&
> Rule 1:  a diphthong/long V in the ultima is stressed, otherwise--
> Rule 2:  a heavy penult (-CVC- or -CVV-, where VV marks a long vowel) is
> stressed, otherwise--
> Rule 3:  stress the antepenult (no matter whether heavy or light)
Here's my suggested formulation:

Stress the rightmost bimoraic syllable or the antepenult mora, whichever is
the rightmost.

* If 'antepenult' means antepenult mora:
1. Codas are moraic; hence IRmaden, irMANden, *IRmanden -- /man/ is bimoraic.
2. Word-final Cs are not moraic: hence IRmaden, *irMAden.

(2) could be explained by treating final Cs as onsets. A test would be
what happens if final CC clusters are allowed. We would predict that the
final syllable of CC-final words would attract stress because it is bimoraic,
e.g. irmaDENT.

If, though, the stress were IRmadent or irMAdent, then the story would have
to change. Can we have more data please?

--And.

Reply