John Cowan cowan at
Thu Jan 25 01:11:18 CET 2007

Addison Phillips scripsit:

> In other words, if you make the Description 'Valencia' (the place), it 
> will be more accurate. Then "ca-valencia" would mean roughly "Valencian 
> as spoken in Valencia". Just because its a subtag of type "variant" 
> doesn't mean that the defined meaning isn't regional or geographic in 
> nature.

The story I've been able to pull together seems to be this:

0) Catalan, like a lot of languages, has a bunch of dialects and a
standard variety which is a dialectal compromise; it's closest to the
Central dialect spoken in and near Barcelona.

1) There is definitely a special variety of Catalan spoken in Valencia,
and it's plausible to call it Valencian.  It has sub-varieties.

2) The Valencian Academy of Language (Acadèmia Valenciana de la Llengua)
says Valencian is a dialect of Catalan, but is standardized separately
from Standard Catalan.  The standardized varieties are different but
are fully mutually intelligible.

3) Some members of the Academy think the names "Valencian" and "Catalan"
are synonyms.  (So does ISO 639-2.)

4) Some people, not members of the Academy, think Valencian is different
enough from other Catalan varieties to count as a separate language.

So I suggest the Description be "Varieties spoken in Valencia, including
Standard Valencian."

A general point:  Descriptions are going to be more verbose for variant
subtags than for other kinds of subtags, which are in the end defined
by encyclopedic descriptions at Ethnologue, Language List, Unicode,
the U.N. Statistics Division, and elsewhere.  When we invent a variant
subtag, we must explain what we mean, because we have nothing to fall
back on except offline sources like dictionaries and grammars.

John Cowan    cowan at
Half the lies they tell about me are true.
        -- Tallulah Bankhead, American actress

More information about the Ietf-languages mailing list