This was brought up before, but I don't know that it was closed "closed" exactly. As I said earlier today, though, using the autonyms directly creates any number of problems, one of which could even be the claim that doing so makes these words more culturally biased than the international standard.
I disagree. If the Finns call their language 'suomen', I place more weight on that than on what other people call it.
But some Finns, namely the Swedish-speaking ones, call the Finnish language "finska." It seems like, from what you're say below, that we should call Finnish both finska and suomi. But my question is, why? What does that get Lojban speakers or learners?
What about groups that don't have names for their languages?
In those cases, an exonym is fine. Although if they have a name for the group, maybe use that?
But then you've reintroduced cultural bias: languages which have speakers who name their own language get to name it in Lojban; if you don't name your language, you're out of luck. Plus, in most of these groups, their language is "the people's language" because the name of their group is "the people." (Birdwalk: this is portrayed with surprising fidelity in _Avatar_.)
Do we keep the "ki-" prefixes, along with their variants, for all of the Bantu languages? If so, then why not keep the "lingua" with "lingua latina"?
I would dispense with those as derivative, but again, that's not the point I'm making.
What about if the language name was borrowed from a Bantu language, and so speakers don't realize that the ki- prefix is a prefix, and treat it as part of the word?
I realize this isn't your point, and to some extent I'm thinking out loud; it just seems like any sort of autonym-based system is going to have a ton of problems and inconsistencies and difficulties in coming up with the lexical items for different languages, as opposed to something based on or derived from the ISO. If there is going to be that degree of inconsistency, it seems like there should be some benefit from doing so.
So I overstated my case; clearly we can't use autonyms for all languages. Perhaps there is a general opinion that it is desirable to use the same method for devising words for all languages. I would disagree with that & argue that it's better to use autonyms where you can (the majority? the vast majority?), and exonyms otherwise. But I'm not here to stand in the way of progress.
Well, see above; the only way I can see using autonyms sometimes but not others would introduce a bunch of cultural bias, which was one of my main goals in doing this. (Also, it depends what you mean by "majority:" if we're going with only major world languages, the exonym thing would be an issue, but in terms of actual numbers of languages, disregarding their distribution or number of speakers, I'd bet the groups with no language name would outnumber the other, if only slightly.)
Is there, actually, an argument for including the ISO code for the language family? Because I still can't wrap my brain around that. And I'd like to. (And that is my main point, which I failed to make clear earlier.)
A couple reasons. If you buy the idea that the ISO language code should be the basis, then you have a problem (if you don't, then you don't): they're far too short to be fu'ivla. There are other ways to make them longer, of course (doubling, using the autonym, etc.) but using the language family is, I think, cool, for a number of reasons: it incorporates information about the languages into their names, which is great from a linguistic perspective. Plus, English does something similar, where many language names end in -ese or -ian. I've been thinking of the ISO language family codes as a sort of suffix which says, not only, "Hi, I'm a language!" like -ese does in English, but also says something about where the language comes from. Which, I will repeat, I think is cool :) I'm not opposed to other ideas for lengthening the ISO language codes, I just haven't seen any mentioned on the list yet.
Chris