Sorry that I'm a couple of weeks behind you all on this thread.

John Clews wrote on 03/14/2003 04:57:53 AM:

> This also points to distinct roles for both ISO 639-2 and ISO 639-3.
> It would also enable Rebecca to be able to send a reply to the
> enquirer stating that Valencian would be handled in ISO 639-3, still
> under development, but that at present it would not be included in
> ISO 639-2, or a similar straightforward reply.

Can you please clarify how you envision this being handled in ISO 639-3?
That standard is intended to cover individual languages, and I have not
seen any indication that Valencian is to be considered a distinct language
on whatever basis.

It seems fairly obvious to me that the problem involved here is not one of
needing new entries or new category types. Rather, it is purely and simply
an issue of documentation. I have long contended that attempting to
document the denotation of a language identifier by listing only English
and French language names is inadequate. Including indigenous language
names is an improvement, but only slightly. Language names alone don't
fully indicate what the intended denotation is. Clearly, we want the
denotation of ca / cat to be understood as including Valencian (and, I
think Balear -- certainly Ethnologue considers these to be variants of a
single language). I suggest that all we need to do is to have the code list
include annotations that explains this. (Note for comparison that the
Unicode standard includes annotations for characters to explain what they
are and, sometimes, what they are not with the purpose of making the
intended interpretation clear.) This would be similar to the MARC Code List
in providing more information, but different from it in that these dialects
would not be given their own entry and their own identifier.

- Peter

Peter Constable

Non-Roman Script Initiative, SIL International
7500 W. Camp Wisdom Rd., Dallas, TX 75236, USA
Tel: +1 972 708 7485