A 09:12 99-01-28 -0800, Peter_Constable@sil.org a écrit :
> I see the value of Rick's use of "alphabet" to exclude
> punctuation, etc. Writing system, then, would include
> everything used for the given language, including punctuation,
> etc. This makes sense in terms of Daniels typology since it
> does not address at all punctuation, etc. but only those
> elements that represent phonological units.
[Alain] Well, there are gray zones. For example in French, the letter "h"
is never ever pronounced (sometimes it is phonologically totally useless,
and only etymological, sometimes it has an effect on surrounding sounds
without being pronounced itself), and it almost has punctuation value (it
then is equivalent to a comma to indicate that no "liaison" should be made
between 2 words [*]). It is not an isolated case in French. Many letters
are not pronounced in this language (in fact so far there is only one case
of a letter always pronounced [still looking for counterexamples], the
letter "v"). So even talking about phonological units for an alphabet does
not cover the general case.
I would say "phonological" or "etymological" units.
Rio de Janeiro
* Example: "Des héros"
(pronounced "day-ay-row" to avoid "day-z-ay-row"
which would mean "zeroes" rather than "heroes")
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:44 EDT