Changing definition of German (was: Re: ISO 639-3 releases list of 2009 changes)

Doug Ewell doug at ewellic.org
Sun Jan 24 01:15:40 CET 2010


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

>> 'chs' was retired from ISO 639-3 as a decades-old error.  I haven't 
>> seen any evidence that 639-2 or 639-5 intends, or was requested, to 
>> pick this up as a collection.
>
> ISO 639-3 isn't decades-old yet...
>
> See
> http://www.sil.org/iso639-3/cr_files/639-3_ChangeRequests_2008_Sum&Add.pdf

Gosh, OK, if you insist.  From Change Request 2008-030:

"In the 13th edition of Ethnologue and in older linguistic literature on 
Native America 'Chumash' was listed by itself as a single language. 
Since the middle of the 20th century, however, it has been recognized 
that Chumash was a family of six mutually unintelligible languages. In 
the 14th edition of Ethnologue, the six constituent languages were 
added, but the entry 'Chumash' was not deleted, resulting in the 
retention of 'Chumash' as a constituent of itself. All specialist 
literature on the languages of California list 6 languages as 
constituents of the Chumash language family."

The error of identifying "Chumash" as a single language was cited as 
dating from "the middle of the 20th century."  That's decades.  The 
error was carried forward until quite recently in Ethnologue and, of 
course, in ISO 639-3.

I don't think this example necessarily supports the conclusion that ISO 
639-3/RA will happily reclassify individual language codes as collection 
codes at the drop of a hat, as implied.  One would hope they would 
correct information that is clearly and widely known to be incorrect.

> In the introductory "Summary of requested changes" it says:
>
> "1 retirement of a code element representing a language family whose 
> constituents were already present in code set. This code element will 
> be recommended for inclusion in ISO 639-5."
>
> I don't know if a request was actually sent to the ISO 639-2/639-5 RA. 
> The requests to that RA do not seem to be publicly recorded (as 
> opposed to the formal requests to the 639-3 RA).

The transparency of the 639-3/RA operations is quite refreshing compared 
to 639-2 and 639-5.

--
Doug Ewell  |  Thornton, Colorado, USA  |  http://www.ewellic.org
RFC 5645, 4645, UTN #14  |  ietf-languages @ http://is.gd/2kf0s ­



More information about the Ietf-languages mailing list