Re: USAGE: English [N] (was: mu for [N])

Quoting Ray Brown <ray.brown@...>:

> On Saturday, January 22, 2005, at 07:24 , Philip Newton wrote:
>
> > On Fri, 21 Jan 2005 18:48:33 +0000, Ray Brown <ray.brown@...>
> > wrote:
> >> not all varieties of English have phonemic /N/;
> >
> > Eh? There are varieties where "bang" and "ban" are homophonic?
>
> Where did I say that!!
>
> In those dialects _bang_ is pronounced [baNg], which in (those dialects)
> is *phonemically* /bang/. Nor is this a spelling pronunciation; it is just
> the older pronunciation which has given way to [baN] or [b&N] in most
> modern dialects, including mine.
I've met speakers who unvoice that final stop, giving thinks like [sINg_0]. I
suppose this is the origin of the use of spellings like "sumfink" in written
representations of non-standard speech.

> Yes, that's how I say it also. The phonemic status of [N] in English is
> one of those things that phonologists like to argue about (see above). It
> has, for example, been pointed out more than once that [N] and [h] are in
> complementary distribution (at least in RP), so should that not mean,
> according to the phoneme theory, they are positional variants of the same
> phoneme?
It would seem to give raise to problems with words like "behave" vs such like
"singer", unless one were to posit a /./ phoneme. Or perhaps one could argue
that [N] is the realization of an ambisyllabic /h/?

                                                         Andreas