The Library of Congress received an official request to change the ISO
639-2 bibliographic codes for Serbian and Croatian to the terminology
codes, i.e. srp and hrv. Of course, this is not the first time this has
happened, but this time it is a letter that was delivered by the director
of the National Library of Serbia and signed by both him and the director
of the National Library of Croatia as well as both country's
standards institutes. Arguments are the usual. I only have a printed copy
of the letter and could scan it if necessary. But I will summarize.
They start by saying that scr was based on "Serbo-Croatian-Roman" and scc
on "Serbo-Croatian-Cyrillic" and that these have not changed since 1998.
Excerpted from the letter:
"Main reasons for our request are:
1. The official language in the Republic of Croatia is CROATIAN-HRVATSKI
in the vernacular.
The official language in the Republic of Serbia is SERBIAN-SRPSKI in the
2. The language names and the 2-letter codes for Croatian (hr) and for
Serbian (sr) were already included in the first edition of the
international standard for codes for the representation of names of
languages ISO 639:1988.
3. We propose hrv as the 3-letter code for Croatian and srp for Serbian
for bibliographic and terminological purposes (they have already been
defined for terminological purposes in ISO 639-2-- the usage of one code
for both purposes would be much simpler and unambiguous).
4. Our proposal is based on the fact that according to the ISO 639-2, the
first criteria for selecting the language code is the preference of
country using the language (Clause 4.1).
5. The language codes have also to be derived from the language name--
CROATIAN or HRVATSKI and SERBIAN or SRPSKI (Clause 4.1).
6. The language code hrv for Croatian would not be in contradiction with
the codes for the representation of names of countries (ISO
7. The code hrv has already been used in library and information community
in Croatia (e.g. the online bibliographic database of the National and
University Library in Zagreb)."
They continue to say that it will improve their databases and enable
effective exchange of records between their countries and catalogs.
My initial response (before the visit) was that they could use the 639-2/T
codes by agreement between their countries (which they are doing anyway),
but that our large bibliographic databases would be adversely affected by
changing these codes.
That of course wasn't an adequate response.
We have discussed the situation with people here at LC, with OCLC and a
few research libraries that have large Cyrillic collections. OCLC has
73,000 records with the code scc and 115,000 with scr. Since there are
tools to convert these and it is a one-to-one match, we and those with
whom we have consulted felt that we could honor their request.
The question of course is precedent that we would set. How many other
countries may ask for the same? In this case we would retire the 639-2/B
code in favor of the T code, which of course is easier to rationalize
than introducing a totally new code (for instance, if the Spanish
countries asked for "esp").
We do say in the standard that considerations include usage in
bibliographic databases and the codes are only to be changed for
I don't believe these people will ever leave us alone or accept any of our
arguments, so in this case, I am inclined to give in. But certainly the
JAC should weigh in here.
We need to make a decision very soon-- the request came in the beginning
of April and we are starting to get some pressure.