Latvian extlang subtags

Doug Ewell doug at
Sun Jan 24 00:58:25 CET 2010

Kent Karlsson <kent dot karlsson14 at comhem dot se> wrote:

>> People talk about "Chinese" when they might say "Mandarin" or 
>> "Cantonese" if pressed for specifics.  Likewise, so I have been told, 
>> people talk about "Latvian" when it turns out they might really mean 
>> either Standard Latvian or Latgalian.  If this is the criterion by 
>> which "zh" was assigned extlangs, then that same criterion should be 
>> applied to Latvian,
> If that was the case, *all* macrolangauges would have *each* of their 
> encompassed languages as extlangs. And that is not the case. Only a 
> small subset was selected.

You're right, I misspoke.  The criterion for that small subset of 
macrolanguages to be associated with extlangs, as stated in Section 
4.1.2, is that they:

(1) "have a specific dominant variety that is generally synonymous with 
the macrolanguage," or

(2) "have traditionally used their primary language subtag, possibly 
coupled with various region subtags or as part of a registered 
grandfathered tag, to indicate the language."

Those who have knowledge about Latvian and Latgalian seem to indicate 
that these conditions apply.

> And that was for purposes of the compromise only; not something to be 
> continued to be reanalysed.

Where is that stated?

Section 4.1.2 says: "To accommodate language tag forms used prior to the 
adoption of this document, language tags provide a special compatibility 
mechanism: the extended language subtag."  Notice that compatibility 
with existing *tagging practice* is what is described here, not existing 
*tags*.  According to the experts, both Latvian and Latgalian have 
likely been tagged 'lv', both before and after the adoption of RFC 5646. 
That existing pactice is what makes 'ltg' and 'lvs' candidates to be 

Unless I am missing something, Section 4.1.2 does not say that no newly 
created macrolanguages may have extlangs associated with them.

Section 3.4, item 10 does allow for the possibility that ISO 639-3 may 
add or remove "macrolanguage" status to an existing code element, so 
even if we in LTRU may have thought that was unlikely, it is explicitly 

>> Your fundamental objection to the extlang mechanism should not enter 
>> into this.
> I still see my objection as perfectly valid.

You are at liberty to hold the belief that the extlang mechanism should 
not have been created in the first place.  I object when that belief is 
used as the basis to argue that application of the mechanism should be 
stopped when new 'zh'-like, 'ar'-like scenarios present themselves.

Doug Ewell  |  Thornton, Colorado, USA  |
RFC 5645, 4645, UTN #14  |  ietf-languages @ ­

More information about the Ietf-languages mailing list