Re: Unnecessary diacritics [was: Re: Contrastive distribution]

On Sun, 14 May 2006 16:03:40 +0000 (UTC), "Aaron J. Dinkin"
<dinkin@xxxxxxxxxxxxxxxxxxxx> said:

On Sun, 14 May 2006 13:07:03 GMT, Bob Cunningham <exw6sxq@xxxxxxxxxxxxx> wrote:

On Sun, 7 May 2006 14:54:05 +0000 (UTC), "Aaron J. Dinkin"
<dinkin@xxxxxxxxxxxxxxxxxxxx> said:

For instance: I say "cat" [k<h>&t], but someone from Cleveland would say
[k<h>e@t], and someone from Knoxville would say [k<h>&j@t].

Do you know of any English words beginning with a "k" sound
in which the [k] is not aspirated?

There are none; it's a phonological or phonetic rule of English.
(Studying for a phonetics exam, I am beginning to suffer from some
uncertainty about which kind of rule this in particular is.)

If not, I see no need for your "<h>"s. I like the practice of eschewing
diacritics with sounds that are pronounced no other way in
the language and in the position in the word being
discussed.

...So you prefer phonemic transcriptions?

God no! I abhor phonemic transcriptions.

All right, there is a difference (though a small one) between phonemic
and broadly phonetic notation.

There seems to be a huge difference. Phonetic
representation should correspond to how something is
pronounced. Phonemic, given the meaning of "phoneme", shows
no particular pronunciation, and can suggest in general a
different particular pronunciation for each reader. But the
perception of that particular pronunciation wouldn't be
being shared with others, so there would be no communication
of it among readers.

If I write /pak/, it can cover all vowel sounds that I
perceive to be in the same phoneme as [a], so it wouldn't
matter whether I wrote /pak/ or /p&k/, the only difference
being what I choose to call that phoneme, and I would
transcribe "pack" also as /pak/ or /p&k/. But if I write
[pak], my intent is to refer to the way I think Bostonians
pronounce "park", while if I write [p&k], the reference is
to the way I pronounce "pack" and I think speakers who
distinguish "pack" and "park" pronounce "pack".

Some of us in alt.usage.english had a very practical
experience with this difference a number of years ago. It
occurred to me that it would be interesting to exchange
postings written entirely in ASCII IPA. A few others also
thought it would be interesting, and we started doing it..

A British reader, Neil Coffey, responded to one of my
postings to say something like "I didn't realize you talked
like that". From the vowel symbols I was using, it was
apparent that my speech was greatly different from that of
other parts of America.

That had been my idea, to show each other how we talked.

But a poster from Boston insisted on using phonemic
transcription, with the result that the coverage of his
symbols would for the most part show no difference between,
for example, his speech and mine. That defeated the
purpose of the scheme, so the ASCII IPA postings soon ended.

I tried to persuade him to use phonetic transcription and to
use symbols that corresponded to his actual pronunciations.
He responded something like, "If I used phonetic
transcription, it would be full of diacritics". I wish I
had at that time seen Ladefoged's discussion of his American
speech transcription in the _Handbook of the International
Phonetic Association_. It would have agreed with my point
of view, and it would have been nice to have Ladefoged to
back me up.

If Ladefoged had used phonemic transcription in the
_Handbook_, his article wouldn't have shown usefully how the
California speech he was describing differed from American
speech in any other part of the country.

Incidentally, the demise of the ASCII IPA exchanges was
speeded to some extent by one poster who decided he had
defined a set of vowel pronunciations that were typical of
all varieties of American speech and who tried to persuade
the rest of us to conform to it.

I continue to conceive of no way in which phonemic
transcription can be useful to people who want to show how
they versus others pronounce things.

We might say that, in a sense, phonemic transcription would
pertain in one fell swoop to how they *and* others
pronounce things.

Using phonemic transcription is akin to saying things like,
"People in Boston pronounce 'park' the way people in Boston
pronounce 'park', while people in Califonria pronounce
'park' the way people in California pronounce 'park'".

But I chose a somewhat narrower
transcription than might have been strictly necessary in order to
emphasize the fact that I was writing phonetically, not phonemically.

.

Relevant Pages

  • Re: Pronunciation dictionaries?
    ... As you do with any adequate phonemic transcription keyed to phonetic ... Instead of exceptional words creatinɡ fresh rules, ... titles like 'Yet Another Proto-Austronesian Phoneme') ... can be divided into short and long vowels, but epsilon is not so easy to ...
    (sci.lang)
  • Re: Pronunciation dictionaries?
    ... phonemic variation as a modified-orthography transcription. ... (You must remember article titles like 'Yet Another Proto-Austronesian Phoneme') ... orthography phonetically misleading, ... can be divided into short and long vowels, but epsilon is not so easy to ...
    (sci.lang)
  • Re: some more Irish vowels
    ... Linguistics doesn't give the slightest good god damn about orthography. ... unambiguous to refer to the Spanish phoneme, ... because there are so few vowels. ... in the phonetics literature before. ...
    (sci.lang)
  • Re: Transcribing rhotics for ESL
    ... >> I'm sure Mxsmaniac knows this perfectly well. ... >> and the English phoneme involved are judged to be markedly different ... What is wrong with a phonemic transcription? ... the "r" in American English is a schwa with a hook on ...
    (sci.lang)
  • Re: some more Irish vowels
    ... Mary is a label for a certain quality of vowel, ... Pronunciation, and the right column shows a rhotic pronunciation ... phoneme in) a set of words, the last sentence would be nonsensical. ... it indicates phonemes but perhaps intends a phonetic transcription. ...
    (sci.lang)