Re: A few phonetics-related q's

Ray Brown wrote:
> But no one so far has noted the way /nt/ is commonly pronounced in modern
> British English, i.e. /n?/
>
> I leave it to my American cousins to explain the various US pronunciations
> of /nt/ in "twenty" (Colloquial Brit ['twEn?i])
In my experience we don't glottalize our t's in _that_ environment, but we
do in phrase-final, and pre/post-C: I can't [k_h&~?]; where's the cat?
[k_h&?]; hatrack ['h&?r&k]; button ['bV`n=], Clinton [klI~?n=]
>
> No doubt this is all very confusing, but I am reminded of Yuen Ren Chao's
> observation:
> "If /ni/ can change into /a/, then practically anything can change into
> anything...."
> BTW for the curious, /ni/ was archaic Chinese for "two". In the modern
> Yangzhou dialect it is /a/. The changes can be traced through other
> dialects; it appears to be:
> /ni/ --> /nz\i/ --> /z\i/ --> /z`1/ --> /r\=/ --> /@r\/ --> /ar\/ --> /a/
I must remember that. So far, the sound changes from Proto to Modern Gwr are
maybe too straightforward.  Cf. the word I mentioned the other day-- llaq
[d_la?] 'narrate' < *guláp (via glap, dlap, dla?) (Anyway, it ought to be
lloq, since *a > o/_p# but let that pass.)

And related gu@ng 'story' < gúlap via gul@, gu@l, gu@n, gu@ng.  It's all
much too simple :-((

Of course in related languages, many other outcomes are possible. Gwrs on
the other continent-- a more diverse lot-- generally kept to the CVCVC
structure and added to the skimpy proto-morphology.
-----------------------------------------
Unrelated message (free ride principle)--
In a volume of conference papers just published/purchased, there's a very
thorough article on Rotuman, entitled (haha) "Temathesis in Rotuman".
If anyone's interested, email me for the Biblio. data.