Re: USAGE: Stress in English

Mark J. Reed wrote:

> On Wed, Feb 25, 2004 at 11:21:22AM -0500, Shreyas Sampat wrote:
> > There are several interacting regularities, but I don't think you can
> > argue convincingly that stress is phonemic.
>
> Sure I can.  Here's a minimal pair: /'pr=.mIt/ (n) a document entitling
> someone to perform some task.   /pr='mIt/ (v) to allow someone (to do
> something).
>
This is "phonemic" in the old-fashioned sense, viewing only the surface
manifestations.  Consider Shreyas' further statement (which also applies to
_ínvalid ~inválid_ cited by Teoh)---

>>The patterns are just slightly difficult for nouns and
verbs: unsuffixed verbs will count a final consonant as a syllable, and
parse that way, while nouns will count a final C as a coda of the final
syllable. That's why verb stresses are stable over the base, -ed forms,
and -ing forms.>>>

As best I recall, that's just one of the many, many, many factors that
Chomsky & Halle bring up in Sound Pattern of English.  Another crucial one
was whether a given word is [+native] or [-native] i.e. French/Latinate in
origin..
One could say that SPE was either a reductio ad absurdum or a triumph of
generative phonology--  and Lord knows, as a reading experience it's totally
coma-inducing-- nevertheless they do find a surprising amount of regularity
and predictability in English stress.

Reply