Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site talcott.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!harpo!decvax!genrad!panda!talcott!tmb
From: tmb@talcott.UUCP (Thomas M. Breuel)
Newsgroups: net.nlang
Subject: more diacritical marks...
Message-ID: <487@talcott.UUCP>
Date: Wed, 14-Aug-85 05:23:56 EDT
Article-I.D.: talcott.487
Posted: Wed Aug 14 05:23:56 1985
Date-Received: Sun, 18-Aug-85 05:15:42 EDT
Organization: Harvard University
Lines: 44

[in reply to a letter by sommar@enea.UUCP, in which he argues that
the problems connected with national letters and diacritical marks
are problems with computers rather than of the language]

Aside from that you learned in school that 'oA', '"A' and '"O'
*are* letters, what is the necessity for their existence? Why
do they need a separate slot in the dictionary? If they are
anything like the German umlaute, you could easily replace
them with 'Oa', 'Ae', and 'Oe', for example (if those letter
combinations are not used in your language).

Let me give you an example of what I would consider a valid
*reason* for keeping a national character set: the Japanese
very seriously considered Romanisation of their written language.
They ultimately decided against it. Japanese is too rich in
homophones (due to borrowing words from Chinese and dropping
the tones) to represent it adequately in writing with a
purely phonetic system. The advantage of having an internationally
'compatible' writing system did not outweigh the advantage
of being able to distinguish homophones in the written language.

If you can't come up with a very good linguistic reason for
keeping your specific national characters, I think you
should re-consider your position: most computers happen to
be made in America, most typewriters do not have *your* national
character set, programming languages use those codes that
you are using for national characters for punctuation, and
most people neither know nor care about your special way
of arranging words in a dictionary or how to write your
national characters. The full 'English' alphabet happens to be
known to all users of the Roman writing system, and it happens
to be the common subset of characters on typewriters and
computers. And it is perfectly usable for the phonetic representation
of languages as rich in sounds as English, German, or Chinese.
I sincerely doubt that *your* language is phonetically so much
more complex than these that you could not represent it easily
(and in fact without any serious changes to your current use
of the Roman writing system) with 26 letters and a handful
of letter-combinations (like the German 'ch', 'sch', &c),
and I am not aware of any Germanic language so rich
in homophones that the introduction of special characters
to distiguish them is warranted.

					Thomas.