What do /a/ and /O/ merge to?

[Beverly Flanigan]

On environments:  J.C. Wells, in _Accents of English 3_ (non-British Isles,
here AmEng),  cites K & McD on the different merger realizations, of
course, and also Wetmore on western PA and Hankey on eastern OH.  Quoting
Bailey (1973), he suggests the change toward merger as affecting first __tV
words, such as 'naughty', then those with following alveolars ('caught',
'dawn'--though some distinguish the nasal set as later, I believe), then
those with following velars ('hawk').  He notes Allen's (1976) claim that U
of Minnesota students even then tended to have only /a/ except before /r/
and claims that this goes at least as far west as Utah and even California
(my two Utahan colleagues both have Don/dawn merger at /a/, and Matt
suggests that Nebraska does too).  Wells says that some words are still
highly variable though, like 'fog', 'water', and 'wash' (see pp. 473-476),
but they aren't down here.  He says nothing about following /l/, but in
southern Ohio the merger to 'turned script a' (intermediate between /a/ and
/O/) occurs in words like 'collar' and 'caller' too--as well as in 'color',
where the schwa (or wedge) has lowered and backed.  So these are worth
checking out in Indiana!

At 10:11 PM 3/4/2003 -0500, you wrote:
>Of course we shouldn't, and in my question I indicated two different
>results, one with /a/, or something close to it, and one with /O/, or
>something close to it.  The Phonological Atlas of North America indicates
>several other possibilities.  The question is who does what with the merger.
>I could go on to ask in what ways the merger is conditioned.  Some of my
>students have /O/ before /l/ and /a/ elsewhere.  Others have /O/ everywhere,
>and still others have /a/ everywhere.
>
>But these different results should have some discernable social and/or
>geographical distribution.  I'll have to find Beverly's article.  I suspect
>she provides some of this answer.
>
>Herb
>
>-----Original Message-----
>From: American Dialect Society [mailto:ADS-L at LISTSERV.UGA.EDU]On Behalf
>Of Mark A Mandel
>Sent: Tuesday, March 04, 2003 8:17 PM
>To: ADS-L at LISTSERV.UGA.EDU
>Subject: Re: What do /a/ and /O/ merge to?
>
>
>I don't think we should expect this question to have a uniform answer.
>The whole point of phonemic merger is loss of distinction. It seems
>perfectly reasonable that different communities could realize a single
>phoneme (in the same context) in different ways -- in fact, once you
>abstract from the issue of merger, it's all over the place.
>
>-- Mark A. Mandel