Theiling Online    Sitemap    Conlang Mailing List HQ   

Re: Optimum number of symbols

From:julien eychenne <eychenne.j@...>
Date:Monday, May 20, 2002, 6:28
Tim May writes:

 >> On another point, I would in a conlang want to reject a phonemic
script
 >> because I reject the very notion of the phoneme.
 >>
 >>  And
 > Like others, I'd be interested in some elaboration of this statement.

So am I : it is very hard to reject the phoneme as a linguistic unit,
even in languages with syllabic writing systems. One can't forget that a
language is spoken before it is written, and there are people who have
no idea of how the language is written. That doesn't prevent them from
having a mental representation of the language, and this representation
is not only syllabic (on evenly phonemic). I think we should not confuse
:
  - phonological skills : syllable and phoneme are *both* adequate
levels of analyse, and the mind has a representation at each level of
the language. Here is an exemple where you can hardly do without the
notion of phoneme : il you pay attention to your own speech (I did, and
it works :-P) you will notice "mismappings" such as : "he bakes the
took" instead of "he takes the book". This is a clear evidence of the
phoneme being a unit, separated from the pure skeletal slots, because
two unit were mismapped. It would be crazy to say that it's a feature or
syllable shift. The syllable is another psychophonological unit, and in
my example it is certainly relevant, as the mapping occurs in the onset
of the syllable ;)

  - graphical skills : the mind may have (but not necessarily) a
graphical representation of the language, rather syllabic or rather
phonemic or whatever else.

Now, if you reject the phoneme as a monolithic
"Sound-Pattern-of-English"-like feature matrix, with no internal
structure, I just agree with you :) Autosegmental analyses gave clear
evidences (at least they convinced me) for separate tiers for almost
every feature. Moreover, new notions in phonology such as skeleton allow
to have deeper and "healthier" (though not perfect) representation of
the sound structure of language. But even if the notion has been deeply
modified and criticized, it is really hard to completely give it up.

Julien