Re: Optimum number of symbols

i disagree with most arguments in this thread and i really hope i get more
time this week-end to elaborate:

imagine a young child learning how to write his own mother tongue. he can
anticipate a word in a text from the context and the first or few letters or
signs of a following word: "i'm tir(e)d, i'm go(i)ng t(o) b(e)d". when i read
japanese i often remember what a kanji is read because i guess the word which it
is part of from context.
now, this is worth only for people who can speak the language they learn to
write. imagine you learn english as a second language and read: "i'm tird, i'm
ging t bd" while you barely know "tired" and "bed". guessing pronunciation is
even worse with logographic and ideographic systems, which makes them much more
difficult than full alphabets. and learning to add a few extra alphabetical
signs to the ones needed to pronounce a word is way easier than guessing signs
needed to pronounce and recognize a word.

could the non-native people on this list who learned to read books or
handwrite--i mean not only read kanas, remember kanjis or typewrite on a word
processor, but read aloud or handwrit even simple texts--in japanese or chinese
or hebrew or arabic or khmer, abjads, abugadis or ideograms tell us how easier
that was from doing the same with alphabetical spanish or classical greek? i
argue that reversely japanese and khmer children have little problem learning to
read foreign languages written in the latin alphabet. my private informant also
told me that learning how to write spanish and english for her hebrew pupils
looks way easier than learning how to write and read hebrew for american
children. learning the respective languages is another thing--and other people
on the list will certainly contend or develop this.

regarding syllabaries, i've read here what i wrote a few days ago: kanas are
easy to read but it takes way more time to write them fluently than to write a
plain alphabet. maybe because learning 31 signs is easier than learning 50 or
100 signs (math reason: imagine learning a syllabary of 200 signs). or maybe
because a greek pi can be equated with a latin /p/  while kanas can't be broken
down (cultural reason). or because |ne| looks like |wa| which looks like |re|,
|me| looks like |nu| and |sa| mirrors |chi|. or i'm the only stupid guy on this
list who had a hard time learning how to write kanas. anyway, learning a
syllabary is learning more signs than learning an alphabet for the same number
of phonems and reading a CV syllable alphabetically is getting two clues to know
how it reads. the learnability of a linear alphabetic system seems to me better
compared to any other system.

this thread discusses what writing system is optimal for a language to be
learned to write and read by people already speaking it.
i argue that the writing system optimal for a language to be learned to write
and read by people who don't speak it already is a full, linear alphabet,
however phonemically tinkering this alphabet may sound.

David Durand wrote:
>>>
The point is that the criteria for optimality are so divergent,
across cultures and languages, that even objective facts are
susceptible of differing interpretation.
<<<
Mathias
http://takatunu.free.fr/tunugram.htm

Reply