This is great!
It is along the lines, which I was thinking of and it proves that we need
not re-invent the wheel.

(1) There are/will be available
- language symbols (for spoken or written language - problem: how to cope
with script variants or change of the writing system? -->script code [can
all cases be covered?])
- country symbols (to 'qualify' language variants - problem: how detailed
can/shall dialects be covered? -->sub-national entities [can all cases be
- script code (to 'qualify' differences of writing system for same language)
- sign language symbol (--> sgn - question: are there more problems IN
REALITY, if "sgn" is combined with language symbols or if "sgn" is combined
with country/region symbols?)

In this connection it is important to decide IN PRACTICE whether to put
"sgn" into the list of language symbols or into the list of script symbols
or ???
(2) ISO 639-1/RA - assisted by the JAC - has to find out, whether an alpha-2
symbol for sign language is needed or not.

-----Ursprüngliche Nachricht-----
Von: ISO 639 Joint Advisory Committee [mailto:[log in to unmask]]Im Auftrag
von Rebecca S. Guenther
Gesendet am: Montag, 07. Februar 2000 13:41
An: [log in to unmask]
Betreff: Re: (iso639.122) AW: Sign language codes

The Library of Congress considered the introduction of a code for sign
languages into the MARC Code List for Languages several years ago.  This
is the language code list upon which ISO 639-2 is based.  We worked
closely with experts from Galludet University in determining which types
of sign languages to represent with separate codes.  In the end, the
decision was to define one code (sgn) for sign languages, because defining
separate ones for different country's signed languages would result in
hundreds of additional language codes. Because the purpose of this
language code list is for bibliographic control, and it is often used in
electronic catalog records using standard cataloging rules, the Library of
Congress provided a document with guidelines for indicating sign

The above document refers to specific fields in the MARC record used to
include language information about a resource.  Although these may not be
useful for those not using MARC, the point is that the particular type of
sign language may be indicated in the record in addition to using the sgn

Note that in the MARC Code List for Languages, references are included. We
have included references from particular types of sign languages to the
general sign language code . So far we have only included those for which
we have received material, but the list of individual sign languages for
which sgn is used as a collective code could certainly grow.

Note that the reason "sgn" was not included in ISO 639-2 is that the ISO
list was essentially frozen after the 1993 Berlin meeting of the ISO 639-2
Joint Working Group (except for resolution of some items from the
balloting process at various stages). The sgn code was approved for use in
MARC in late 1995.

^^  Rebecca S. Guenther                                   ^^
^^  Senior Networking and Standards Specialist            ^^
^^  Network Development and MARC Standards Office         ^^
^^  1st and Independence Ave. SE                          ^^
^^  Library of Congress                                   ^^
^^  Washington, DC 20540-4402                             ^^
^^  (202) 707-5092 (voice)    (202) 707-0115 (FAX)        ^^
^^  [log in to unmask]                                          ^^
^^                                                        ^^

On Sun, 6 Feb 2000, Michael Everson wrote:

> Ar 13:27 +0000 2000-02-06, scríobh Christian Galinski:
> >Dear All,
> >as there is nothing like a unified sign language, but many national sign
> >languages (plus an international one - possibly with variations), I would
> >like to repeat an earlier suggestion to formally handle this question
> >'script code'.
> As editor of ISO 15924 (Codes for the representation of the names of
> scripts) I have no idea why Christian thinks this makes sense. I think he
> is mistaken about what Sign Languages are.
> It is true that Sign Languages tend to be national and regional in nature.
> This is related to questions of educational systems and mobility.
> Interestingly, there are a few Sign Languages which can be shown to be
> "etymologically" related. Irish Sign Language and American Sign Language,
> for instance have a few features in common with French Sign Language, and
> this can be traced to the influence of French clergymen who set up Deaf
> schools.
> In general, however, even languages like these are not mutually
> intelligible without study. To my knowledge, there is no "international
> sign language", although many people outside of America who have studied
> Gallaudet University have learned American Sign Language.
> >In fact sign language(s) is/are often considered as language(s) of their
> >by applied linguists. Looking at their 'representation' by means of
> >handsigns etc. they are of course (a) semiotic system(s) like different
> >scripts.
> It is incorrect (or rather, outdated) to say that "Sign Languages are
> considered to be languages". This has been long since proven and all
> reputable linguists accept and understand that these are true languages
> with their own grammar and syntax.
> There _is_ another kind of Sign Language, which takes symbols from
> "natural" Sign Language and uses them according to the grammar and syntax
> of a spoken language. But here it must be understood that there is a
> difference between "French Sign Language" and "Signed French".
> >In case of the existence let us say of an English variant of sign
> >eg in the US one must think of a way of coding
> >name of language + country code + sign language symbol
> >which is quite parallel to the usage of the script code such as
> >name of language (eg Chinese) + name of country (eg Singapore) + script
> >variant (eg simplified characters or complex characters).
> I don't follow this logic at all. Please read in detail the request for
> code "sgn" from the Irish National Body and the Deaf Action Committee at
> You will
> there a scheme for unambiguously identifying sign languages by region and
> _also_ for identifying the signed variants of spoken languages. The
> "sgn" is proposed as a new code to describe gestural languages. Country
> codes from ISO 3166-1 are used except where that is not precise enough, in
> which case country subdivision codes from ISO 3166-2 are used (cf. the
> situation for Canada, where sgn-US is used to identify American Sign
> Language, sgn-CA-NU to identify Eskimo Sign Language, sgn-CA-QC to
> French Canadian Sign Language, and sgn-CA-NS to identify Nova Scotian Sign
> Language).
> Note sgn-VA Monastic Sign Language and sgn-US-SD Plains Sign Talk, which
> show that Sign Languages are not only used by the Deaf.
> Remember, ISO 639 exists to help people identify what language a book or
> web document is written in or what languages were spoken (or signed!) at a
> conference. Because there are as noted in the proposal document, there are
> numerous ways of writing or notating Sign Languages: SignWriting is
> to be the most practical; HamNoSys and Stokoe are used by many academics.
> All of these are potential codes for ISO 15924, because these are _writing
> systems_, but the Sign Languages themselves are _languages_ and therefore
> need ISO 639 codes. We believe that the ISO 3166 extensions are most
> appropriate for differentiating the different Sign Languages because there
> are (at least) as many Deaf Sign Languages as there are countries (plus
> signed versions of spoken languages) and a single generic code "sgn" is
> good economy.
> Best regards,
> Michael Everson  **  Everson Gunn Teoranta  **
> 15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
> Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
> 27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire