Re: Turkish "script"? (was: New contribution)

From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Sat May 01 2004 - 19:52:24 CST


From: "Peter Kirk" <peterkirk@qaya.org>
> Someone mentioned "Turkish script". This would be rather like trying to
> define a Turkish script which is similar to Arabic until the 1920's and
> then suddenly became similar to Latin script!

Is uppose that this was a confusion of terms. In Turkey or for linguists, the
alphabet is not called "Latin" or "Roman", but "Altaic".

Linguistically, the Altaic alphabet is distinct from the Latin alphabet, but in
Unicode they are unified in the same "script".
Unification in Unicode allows better interoperability of data and easier
interchanges. A common Unicode script does not necessarily mean that this is the
same alphabet.

Note however that Turkish, the language, is not in the Altaic group of languages
according to some sources, although other Altaic languages share the same
alphabet (for example Azeri). So some prefer the term "Turkic" to designate this
alphabet. Still this alphabet is a clear subset of the Unicode "Latin" script.

The same thing could be said about the Celtic alphabet or the Baltic one, which
are also unified with the Latin one in the same script.

The concept of "scripts" in Unicode is not strictly the same as for linguists,
due to unification (unification occured long before Unicode, within legacy
charsets because of processing problems, so that the same charset could be used
to write both Turkish and English for example).

If we were strict, a "script" would be a precise instance resulting of a
combination of a language and a writing style. If so, there would be several
Arabic scripts, and even several Latin scripts (like Roman, Gothic, Italic,
Suetterlin, handwritten cursive, etc...). This could reach a situation where a
specific calligraphic style woudl become a separate script. An unification is
then necessary to help keeping together in the same code, several styles that
are mutually intelligible, instead of creating a new script for each possible
graphic variant.



This archive was generated by hypermail 2.1.5 : Fri May 07 2004 - 18:45:25 CDT