Changing definition of German (was: Re: ISO 639-3 releases list of 2009 changes)

[Kent Karlsson]

Den 2010-01-23 21.25, skrev "Leif Halvard Silli"
<xn--mlform-iua at xn--mlform-iua.no>:

> And don't forget the region subtag, as a tool for making the variety
> distinction. ;-)
> 
> It looks as if macrolanguages status is most often awarded to a code
> which cover two or more "language instances" that aren't divided by a
> border. That is: When there is no region tag to make the distinction.

I think you are mixing up the LSR with ISO 639-3. 639-3 supposedly only
awards distinct codes for different languages, whereas the variant
subtags (including region subtags) are supposed to distinguish (a
very small subset of) dialects or orthographies of a language.
The latter has no hope of ever becoming anywhere near complete using
the methods for allocating variants. IIUC, ISO 639-6 is an attempt atb
eing fairly complete w.r.t. dialects and orthographies in existence.

Granted, there is the ever-present difficulty of drawing the line
between language differences and dialect differences, so there is always
a judgement call. (Plus other difficulties that Randy alludes to.)

> This, again, in my view, hints that extlang subtags are very practical,
> as they can be used very much in a similar fashion to how region
> subtags are used. There are quite some examples of tagging that shows
> that taggers think this is logical, even when extlang subtags is not an
> option. ;-)
> 
> But I wonder: If 'de' means 'Standard German', then could it actually
> make sense to define a subtag which meant 'dialect of Berlin' and use
> 'de' as prefix for it? If the variant is not coverd by the 'de' then
> that doesn't make sense to me. I would almost claim that if 'de' can't
> _in theory_ become macrolanguage, then it can in theory also not become
> the preferred prefix of variant tag for a German dialect. (However, it
> could become the preferred prefix of a variant of Standard German.)

This is complicated by that there is a group of German languages,
a subset of "Germanic", but a superset of {(Standard) German, Swiss
German, Colognian} (to arbitrarily pick 3 out of 19 listed in Ethnologue).
See http://www.ethnologue.com/show_family.asp?subid=209-16.
It's unfortunate that the name "German" is used for nodes at different
levels in this (approximate) hierarchy.

    /kent k