Re: more English orthography

Nik Tailor wrote:

>Marcus Smith wrote:
>> In fact, the people I tested were consistent in responding with
>> [h{lVkIti]
>
>Well, [V] and [@] are allophones of each other, at least in my dialect,
>[V] used in stressed syllables, and [@] in unstressed.
But [@] is an allophone of all the lax vowels, so I don't see what this is
supposed to prove.  Also, in my dialect there are times when unstressed [V]
resists changing to [@].  My first phonology professor said that wasn't
unusual.

>Also, I have a (probably nonphonemic) distinction between "really short
>schwa" as in the third syllable of dilapidated (the schwa seems to be
>less a phoneme and more a transition between sounds - it lasts just as
>long as it takes for the tongue to move up to the roof of the mouth,
>starting when the /p/ opens) and "regular schwa" as in "kappa", in which
>both the /&/ and the /@/ seem to take the same length to pronounce.
Interesting.  I've never heard of this before.

>> Since they are willing to provide a form
>> for these when there could not possibly be any stored in their head, then I
>> think we have to be suspicious of any pronunciation of an uncommon word.
>
>Ah, but what if instead of *telling* them the word, you'd shown them the
>word and asked them to pronounce it?  I suspect you'd find a reasonably
>consistent guessing there.
Of course.  But this would be guidance from orthography, not some underlying
representation.  This doesn't tell us anything useful, because people often
let
the orthography "incorrectly" influence their pronunciation.  There are many
people who pronounce "often" with [t].  Historically the [t] was lost, but
under the influence of spelling, the [t] reappeared.  So I don't think it is
fair to say that the pronunciation offered under the influence of orthographic
representations is any more correct than from purely auditory stimulus.

That basic point still stands, that I don't see how we could find an
underlying
pronunciation for the schwa in "comma."  It may be best to consider it a
phoneme.  I just don't know how to test that.

Marcus

  The use of [V] makes sense, when forced to
>choose a transformation from [@] into another vowel, they'd pick the
>closest one, namely [V], which is, in fact, an allophone of [@] in many
>(most? all?) dialects.