Re: Phonetic scripts and diphthongs ...

Andreas wrote:

<<But the same applies to affricates, and
single letters for affricates are all over the place; English supplies 'j'
for
/dZ/ (a case where IPAoid phonemic representation is _farther_ from the
1-to-1
ideal than the much-maligned English orthography). Anyone's got any thoughts
as
to the possible causes of this apparent asymmetry?>>

The IPA representation is the way it is because affricates are sequences of
two different sounds.   Psychologically, though, affricates are pretty much
always thought of as single "letters", no matter how they're represented
(e.g., "ch").   John Ohala (the former phonetics professor at Berkeley)
proved
this during one of our classes.   He chose a random student (native English
speaker) and gave them a series of words.   After each word, they had to
say whether it was in the set, or whether it wasn't.   The professor would
then respond "correct" or "incorrect".   So, he started out with "couch", the
student said "in thet set", and he said "incorrect", and it went on this way,
and pretty soon the whole class got the idea that the set was whatever began
with a complex coda.   Once it was clear to everyone, and once the student
obviously knew what the set was, he said the word "charm", and the student
said "not in the set".   That was my natural reaction, too, even though I
should've
known better.

So maybe if a phonemic alphabet is the desired alphabet, the Americanist
transcription system, or something similar, would be better.   For Turkish,
for
example, the Americanist system would probably work better, because the
affricates in Turkish function exactly stops in every respect.   If a strict
phonetic
alphabet is desired, though, it should be two graphemes, because there're two
sounds there.   In my opinion, of course.

-David
*******************************************************************
"sunly eleSkarez ygralleryf ydZZixelje je ox2mejze."
"No eternal reward will forgive us now for wasting the dawn."

-Jim Morrison

http://dedalvs.free.fr/

Reply