CHAT: barbarisms

On Sun, 13 May 2001, John Cowan wrote:

> dirk elzinga scripsit:
>
> > Is this really Henninian pronuncation of Greek, or simply the
> > principles governing English stress applied to names of Greek
> > origin?
>
> If that were so, I'd expect /&akrow'powlIs/, not /@k'rAp@lIs/,
> wouldn't you?
No, I don't think so. In words built on the Greek _polis_
(policy, politic, etc), the {o} is consistently given its
checked pronunciation [A], indicating a short underlying vowel
for this root. Stress in a large portion of the English lexicon
falls on the penultimate mora, given final syllable
extrametricality (i.e., final syllables don't count). This means
that the underlying phonological representation of orthographic
{-polis} is /'-pA<lIs>/. The final syllable is extrametrical,
and doesn't count for stress assignment. /A/ is short, and hence
monomoraic, therefore stress falls on the syllable immediately
preceding. So we have Me'tro-polis, Minne'a-polis,
India'na-polis, etc. Stress will never fall on the /A/.

If the {o} were given its free value indicating a "long" vowel,
then I would indeed expect stress to fall on it--as it does in
Minne'sota, Ari'zona, etc, since in these cases the "long" {o}
(= [ow]) is bimoraic. Given final syllable extrametricality, the
stress would then fall on the first mora of "long" {o}.

Dirk

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

"The strong craving for a simple formula
has been the undoing of linguists."               - Edward Sapir