Appeal to ISO 639 RA in support of a 639-3 code for Elfdalian

Michael Everson everson at
Wed Apr 20 11:56:14 CEST 2016

New draft. New title. If anyone has comments, please provide edits. There’s also a long list of literature and other resources in a file that can go with it. 

This is an appeal by the group responsible for the IETF language subtags to the ISO 639 RA to reconsider and revert their earlier decision and to assign an ISO 639-3 language code to Elfdalian.

The group responsible for the IETF language subtags is deeply concerned about the reasons given by the ISO 639 RA for rejection of the Elfdalian language. There is no doubt that its linguistic features are unique in the continuum of North Germanic languages. The reasons supporting rejection are weak and invoke mainly arguments from outside of the field of linguistics. These arguments, originating from one particular Swedish governmental agency, are successfully rebutted in the other contributions, all of which support the addition of Elfdalian to ISO 639-3. The rebuttal documents (given in provide convincing evidence that Elfdalian is not merely a dialect, but rather is as distinct from Swedish as Norwegian is (indeed, as distinct from Swedish as Icelandic is); those have separate language codes. No reasoned responses to those documents and their evidence was provided by the ISO 639 RA.

We understand that the Swedish government considers it a dialect of Swedish, but that seems to be a political decision, not a linguistic one. Certainly, Elfdalian is influenced by the national language, Swedish, just as Frisian is influenced by Dutch. But its grammar and phonology are clearly distinct (vowels are not lengthened in open syllables, medial /ð/ and /ɣ/ and Old Norse nasal vowels are retained, and four cases (nominative, accusative, genitive, dative) are preserved at least in definite nouns. The theoretical categorization about “Abstandsprache" vs “Aufbausprache” given in Kristine Zach’s MA thesis are not convincing given the actual linguistic data. Elfdalian has archaic features as well as innovative features, and is unique, being closer to Icelandic and Faroese in many ways than it is to Swedish.

We ask the ISO 639 RA to consider this. Elfdalian is a unique and independent language from Swedish as can easily be demonstrated by its phonology and morphology. The request for a language tag for Elfdalian was not made in order to establish it or give it status vis à vis any political process in Sweden. The request for a language tag for Elfdalian was to enable texts written in Elfdalian to be tagged as such. 

It appears to us that the RA has evaluated the sociolinguistic status of Elfdalian in the territory of its speakers, rather than the linguistic characteristics of the language itself. If such a criterion were commonly applied, many minority languages might not have language tags at all. We believe that what the RA ought to have done is to recognize the fact that Elfdalian *is* a language, that it is not identical to or a dialect of Swedish, and to assign it a three-letter code for the identification of its name. 

Our aim is to persuade the RA to reverse its decision and issue a code element. We would like the RA to recognize that it is “linguistic identity” which matters for Elfdalian, to the extent that such identity supersedes any political or diplomatic considerations for that language. 

There is a legitimate user need for a language subtag for Elfdalian. Up until now, these have always been supplied by ISO 639-3. However, there is the possibility in BCP47 for registration of a language sub tag that is not based on ISO 639-3, and it has been suggested that that process be initiated. We do not wish to do this because:

(a) There is the possibility of conflict or redundancy if the RA later approves a code element. 

(b) Some processes are incompatible with 5- to 8-letter language subtags, which would not be beneficial to Elfdalian data.

(c) While it is an option for BPC 47, it would decouple BCP 47 from ISO 639-3 in a way which neither we nor the ISO 639 RA may want.

But we have no recourse but to consider it, if the situation does not change. It would be far better for all users of BCP 47 and ISO 639-3 if this step were not taken, but that could depend on speedy action by the ISO 639 RA and the intent and needs of the requester. Thus we write to the ISO 639 RA in appeal, and strongly request that the language code be added to ISO 639 within two months of the receipt of this mail, to avoid this situation.  

Michael Everson (BPC 47 Language Subtag Reviewer)
Martin Dürst
Mats Blakstad

More information about the Ietf-languages mailing list