Re: [Ietf-languages] Latin Sub tags

[Doug Ewell <[e-mail]>]

Hugh Paterson III wrote:

> Specifically though I am curious about the conceptual entities such as
> Early Latin, Classical Latin, and Medieval Latin. These three eras are
> well attested in Latin oriented scholarly literature. [...]
>
> In considering an application for these three varieties of Latin, the
> time-depth boundaries for these three "varieties" are contested,
> therefore I don't think that a variant tag relating to a time frame is
> especially neutral or clear.

If these three stages of Latin are in fact “well attested” (which they are), and their main distinguishing characteristic is temporal (as opposed, say, to other dimensions as described by Sebastian Drude), then any variant subtags *should* relate to time frames. Better that than inventing some bogus relationship.

The assumption, of course and as always, is that these varieties warrant separate tagging; that is, someone has a need to identify or search for content in each of these varieties.

> Some languages such as Greek have separate ISO639-3 codes for
> different time depths: e.g., [ell] and [grc]. My understanding is that
> this time based split was introduced to ISO639-3 via compatibility
> requirements with ISO639-2 and time depth indication via ISO639-3 is
> not a viable way to argue for the introduction of a "new language"
> code to ISO639-3.

Every language is different, and must be considered separately. The normal disagreements over what distinguishes a language from a dialect or other variety (don’t say it, John) become even less consistent when considering “old” and “middle” and “modern” varieties. Basically all linguists consider Old English, Middle English, and Modern English to be separate languages, but that might not be true for some other languages with “Old” varieties.

ISO 639-2 has little to do with this. Numerous code elements for “Old” and “Middle” languages have been added to 639-3 that were not distinguished in 639-2.

> Additionally, the ISO639-2 registrar could be approached to add these
> as "languages" -- there is sufficient publishing in these "languages"
> to meet the guideline ISO639-2 requirements. The result being that
> they would be de facto introduced to the ISO639-3 set of codes.
> However, I'm not at all sure that pathway really solves any long term
> goals for indication of time depths. (Is the ISO639-2 registrar on
> this list?)

I don’t think the focus on 639-2 is well placed. A code element does not have to be added to 639-2 before it can be added to 639-3; it’s the other way around. The only updates to 639-2 now are to stay aligned with 639-3.

> Type: variant
> Subtag: peano
> Description: Latino Sine Flexione

Latino sine flexione is an odd bird: an auxiliary language wholly constructed by modifying a natural language. It is a variety of Latin, but of a different sort from Early and Classical and Medieval Latin, which evolved organically over time. Both types of language variety (and more) can be described by variant subtags.

The correct process, if there is a question whether a “variety” is actually a separate language, is to propose it to ISO 639-3/RA *first*. If that proposal is rejected on the basis of being a dialect or other variety, then we can have that debate here, and in the vanishingly unlikely case we think we know better, we can register a 5- to 8-letter primary language subtag. The much more likely outcome is that we register a variant subtag (or none at all).

--
Doug Ewell, CC, ALB | Lakewood, CO, US | ewellic.org