Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.duke.edu!news-feed-1.peachnet.edu!insosf1.infonet.net!internet.spss.com!markrose
From: markrose@spss.com (Mark Rosenfelder)
Subject: Re: Russian vowel  bI
Message-ID: <D6Gz6n.FrI@spss.com>
Sender: news@spss.com
Organization: SPSS Inc
References: <3kcq7c$167k@news.ccit.arizona.edu> <D68CGy.8F3@eskimo.com> <3lf6uq$crf@news.ycc.yale.edu> <D6A86x.4tI@eskimo.com>
Date: Mon, 3 Apr 1995 17:27:57 GMT
Lines: 15

In article <D6A86x.4tI@eskimo.com>, Richard Wojcik <rickw@eskimo.com> wrote:
>It isn't natural at all for allophones of the same phoneme to be used in a
>contrastive way.  If you have two phonetically distinct pronunciations of
>letters, then you are using different phonemes to contrast them.  I can
>think of no other language in which the letters of the alphabet have names
>that violate the phonology of the language.

According to Geoffrey Sampson, the Latin alphabet may once have had such
a letter: H.  As the [h] sound disappeared from Latin (he speculates), 
speakers may have made increasingly desperate attempts to produce an [h]
when naming the letter (still needed in digraphs): [ahha], [axxa], then
[akka].  This last, of course, conformed once again to the phonology of
the language, and developed normally into e.g. French /aS/ (cf. _vacca_ ->
_vache_).  But in earlier stages of this process the letter H would have had
a name containing a sound that wasn't a phoneme of the language.
