I summarized this issue recently for some discussions with library users
of ISO 639-2, and here is what I said.
The question arose on the ISO 639 Joint Advisory Committee why there was a
discrepancy between the information on the ISO 639-2 web site concerning
the status of the two-character code for Serbo-Croatian. At the first
meeting of the JAC in 2000 it was decided to deprecate the two-character
code for Serbo-Croatian, since the group approved the addition of Bosnian,
so that the 3 separate languages (Serbian, Croatian and Bosnian) were all
defined. (Serbian and Croatian were already defined as part of the
balloted ISO 639-2 standard). Thus there would be no need for
Serbo-Croatian. However, in the revision of the two-character code list,
which was published as ISO 639-1 in 2002, Serbo Croatian was listed with
the code "sh". It isn't clear if that was a mistake or intentional.
In MARC and in ISO 639-2, there are separate codes for Serbian (scc) and
Croatian (scr); after the publication of the ISO list, the captions in
MARC for those codes were changed from "Serbo-Croatian (Cyrillic)" and
"Serbo-Croatian (Roman)" although the codes stayed the same.
Currently work is proceeding on a code list for ALL languages (ISO
639-3) without the need for a certain body of literature. The question
came up as part of that effort and a code "hbs" has been included for
Serbo-Croatian. It was requested that a code for Serbo-Croatian be added
to ISO 639-2.
Defining this in 639-2 would cause some confusion. Serbo-Croatian would be
like a collective code, but all of its parts are also defined. It also
would be confusing since libraries used to use the current Serbian and
Croatian codes for Serbo-Croatian in its 2 different scripts.
This situation would be similar to the case of Norwegian, where there is a
general code for Norwegian but 2 separate codes for varieties (Norwegian
Bokmal and Nynorsk) were later defined. In the MARC community we have said
that LC would not apply the new codes for Bokmal and Nynorsk, although
they are listed in the documentation.
There is the possibility of applying such a code (hbs) only to the
non-standard, non-literary varieties. For the MARC list we would probably
call it "Serbo-Croatian (Other)" to distinguish it from the other defined
On Wed, 22 Jun 2005, John Clews wrote:
> I agree with Peter on both his points quoted below.
> On the specific "hbs" code for ISO 639-2, I would also like to see a
> summary of the discussions about the previous USMARC (and UKMARC) code,
> included in the round up of this. The way that they were split into "scr"
> and "scc" was quite well put, from memory, and quite a logical thing to do
> at the time, even though Serbo-Croatian still exists as an entity.
> Rebecca or Milicent: any chance that you could send me a summary of that?
> I seem to have archived it out of sight as it were.
> Many thanks in advance
> > There are a couple of items I've mentioned recently that were discussed
> > for a while but which need some action:
> > - Serbo-Croatian: I suggested that we add "hbs" to 639-2 (it's in the
> > draft table for 639-3), but at the least we need to take action to
> > clarify the status of "sh" in 639-1.
> > - "no linguistic content": I suggested that we add this; I'd like to see
> > us move toward a vote on it.
> > Thanks.
> > Peter Constable
> > GIFT | GPTS | MICROSOFT