Re: tlug: Re: Japanese input

[Manuel Chakravarty <chak@example.com>]

I think, Karl-Max, you are oversimplifying matters.

Karl-Max Wagner <karlmax@example.com> wrote,

> > Even if their speech was mutually
> > incomprehensible, educated men throughout the region could communicate 
> > through writing. They still can, to an extent. Give a Japanese person
> > a copy of Renmin Ribao, or a Chinese a copy of Asahi Shimbun, and
> > chances are they can at least get a general idea what
> > it's about. On a
>
> Yes. I use that, too. Well, I also can read most european
> languages that way. No problem. No Kanji required. 

Kanji provide much more accurate meaning in this situation.
I saw a Chinese friend who doesn't speak Japanese using
Kanji to communicate in Japan.  That was much better than I
could do with Spanish or Italian (although I am fluent in
German and English and had Latin in school).

> > If anything should be eliminated it's kana -- especially
> > katakana. That would have the beneficial side-effect of improving
> > Japanese students' English pronunciation -- since for the first time,
> > schoolchildren would have to deal with the actual sounds of English
> > words :-)
> Brutally speaking, the best would be to do what the French did
> in Vietnam or what the Turks did by themselves or what the
> Slovenes did some time ago: throw out their proprietary writing
> system and switch to the alphabet. Due to the multilingual
> setting in Europe, the alphabet is relatively good at
> representing different languages.

Sorry, but the alphabet plainly sucks for Japanese.  It
clutters a sentence up with symbolic noise, because of the
restricted set of phonemes used in Japanese.  The sound
system of Japanese and the alphabet are just incompatible.

Actually, I have heard several times now that people
starting to learn Japanese using Romanji instead of kana and
kanji have much more trouble with the correct pronunciation
and the structure within words (especially Chinese
compounds).  

So, if there is a choice, it is Kana, but definitely not
romanji.  But in the end, kanji are also quite helpful.  My
Japanese is very bad, but still, I am more comfortable with
a moderately kanjified text than with a pure kana text.  I
find, learning compounds without kanji quite difficult, due
to the many homophones.  Tell me what `koushi' means.

> > inherent agglutinating nature of the Japanese language looks to make the
> > mixed use of kana and kanji work very well, and this may be the
> > background of the fact that the Japanese language has successfully
> > expanded its vocabularies by applying Chinese words (in Chinese kanji
> > characters) especially to abstract concepts.
>
> Sure ? How many foreign words do you think we adopted in German?
> We have them from all directions: Latin, Greek, Italian, French,
> English, Japanese, Chinese, Russian etc. etc.

Come on, now you are starting to get ridiculous.  Anybody
who knows both German and Japanese will just laugh about
that comparison.  You can easily speak quite good German
while using foreign words rather seldom, but you can't even
go to a supermarket in Japan without being drowned in words
of Chinese origin.

> > 	As I'm not a linguistic expert, I can not jump to the conclusion that
> > we can not eliminate kanji. But practically speaking, not using kanji
> > characters will bring problems to Japanese, as listed below:
> > 
> > 	1. Conflict words: different words sharing one and the same sound.
>
> We have lots of them in German, too. Not much of a
> problem.

Now, you are really joking!  Compared to Japanese, German
basically has no homophones.

> > 	2. Confusion between elements of a sentence: it will be hard to
> > distinguish independent words and auxiliary words. This will diminish
> > readability, anyway.
>
> This is a problem in pretty much all languages. It is minor,
> however.

You can't know very much about Japanese...

In information theory, there is the notion of the `hamming
distance'.  That is, roughly speaking, the number of bits
that you can alter on average in a meaningful word before
getting another meaningful word.  The hamming distance of
Japanese is astonishingly small.  You can compensate for
that in speaking, in writing it is much more difficult.

And finally, I think computers shouldn't dictate which
languages and writing system we are using.  Instead, we
should dictate the languages used by computers.  (Being a
researcher in programming languages, this is a most natural
viewpoint for me ;-)

Manuel

--------------------------------------------------------------
Next TLUG Meeting: 13 June Sat, Tokyo Station Yaesu gate 12:30
Featuring Stone and Turnbull on .rpm and .deb packages
Next Nomikai: 17 July, 19:30 Tengu TokyoEkiMae 03-3275-3691
After June 13, the next meeting is 8 August at Tokyo Station
--------------------------------------------------------------
Sponsor: PHT, makers of TurboLinux http://www.pht.co.jp