Eric Christopherson wrote:

> Nik Taylor wrote:
> >
> > Eric Christopherson wrote:
> > > ACK! Yeah, I messed up those examples. I think the difference betwe=
en
> > > initial and final /k/ is not aspiration, but point of articulation.
> > > The /k/ in "kill" is more uvular than the one in "pack" or somethin=
g.
> >
> > /k/ in kill *is* aspirated.  The /k/ in pack, as Tom Wier has already
> > pointed out, is unreleased.  Actually, the /k/ in "kill" *is* pronoun=
ced
> > at a somewhat more forward position than "pack" or "code", it's
> > partially palatized.
>
> I disagree that final /k/ is unreleased, at least from the speech I
> hear every day. I pronounce "pack" [p&k_h] and "tap" [t&p_h], but
> "cat" [k&t7] ([7] representing no audible release). I'm inclined to
> think that [t7] is the same sound as in "cattle" [k&t7l=3D], but it
> sounds just like [d] and I have seen it represented as a tapped /r/
> before. Anyone know what the *real* story is? :)

Well, *most* dialects *do* have unrelease in those positions.  In your
case, it might actually be due to substrate influences from foreign langu=
ages
of generations ago.  In your general area of the country, it's actually b=
een
attested that a number of Germanisms have found their way into the speech=
 of
English speakers (like "Do you wanna come with?" exactly paralleling
German "Willst du mitkommen?"), or perhaps Scandinavianisms [note
to our Scandinavian readers out there:  do your languages feature unrelea=
se
in final position?].  For me, all stop consonants feature unrelease word-=
finally.

In any event, /t/ intervocalically in most American dialects is a voiced =
tap,
which in many languages, like Spanish, is an allophone of /r/.  In most
American dialects, this does sound like a [d] to some extent, but the fac=
t
that it's still considered an allophone of /t/, and not /d/, is shown by =
the
fact that people will actually say things like [stVt_hi:] for "study" (as=
 I
have personally heard) when trying to emphasize the fact (though it might
be said that this is just an example of hypercorrection), where the <d> i=
n
'study' is also a voiced tap.

(The whole situation is kinda weird to me, in fact.)


=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Tom Wier <[log in to unmask]>
ICQ#: 4315704   AIM: Deuterotom
Website: <http://www.angelfire.com/tx/eclectorium/>
"Cogito ergo sum, sed credo ergo ero."

"S=F4=F0 is gecy=FEed / =FE=E6t mihtig God manna
cynes / w=EAold w=EEde-ferh=F0."
 _Beowulf_, ll. 700-702
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D