Re: Phonetics vs. Phonemics (was: apparently bizarre 'A's)

On 2/24/06, Andreas Johansson <andjo@...> wrote:
> Quoting John Vertical <johnvertical@...>:
>
>  > >As to phonetic symbology, you're right that I overstated the
> > >precision.  Each symbol covers a spectrum of similar sounds.  But the
> > >difference between phonetics and phonemics is that in the latter case
> > >the sounds represented by a single symbol are identified by their
> > >equivalence within a given language, and need not even be phonetically
> > >similar at all.
> >
> > >Mark J. Reed < markjreed@...>
> >
> > That reminds me - what's the most different allophones of a single phoneme
> > you know of (either qualitively or quantitively)?
>
> An Argentinian I met some years ago appeared to have [h] and [C] in free
> variation as the realization of /x/ before front vowels. Can anyone familiar
> with Argentinian Spanish say whether that's normal down there, or just was some
> idiosyncrasy of his?
>
> I also seem to recall hearing of a language where /k/ had a [G] allophone
> between vowels, which, while easy to imaging diachronically, is a pretty stark
> difference synchronically.
Perhaps you recall reading one of my earlier posts on Shoshoni; it has
just this kind of  allophony. The voiceless stops /p, t, ts, k, k_w/
(/ts/ is a stop phonologically in Shoshoni) all have voiced continuant
allophones intervocalically. Thus:

/papi/ [paBi] 'older brother'
/ata/ [a4a] 'mother's brother; man's sisters' child'
/motson/ [mozo] 'facial hair; beard'
/kaku/ [kaGu] 'mother's father; daugher's son'
/j1k_wi/ [j1G_wi] 'say something:SG.SUBJ'

This is a productive phonological rule that applies across morpheme
boundaries as well. For example, the sentence 'I saw the horse'

/n1 puNkui puikka/

is realized

[n1 BuNgu Buikka_0] (Obviously a couple of other things are going on as well.)

Dirk
--
Gmail Warning: Watch the reply-to!

Reply