Name That Language!

[Richard W]

I'll make a wild guess at Ahamb (from Vanuatu). With one exception, the initial voiced stops are prenasalised, which looks a Papuan or Melanesian feature. The one exception is 'br-', which might be conceivably be a bilabial trill.

[Nortaneous]

Is it a Slavic language?

No.

Is it spoken in Southeast Asia?

Using Wikipedia's definition of "Southeast Asia", no.

(Cambodia, Laos, Myanmar, Thailand, Vietnam, Malaysia, Indonesia except Western New Guinea, the Philippines, and various small islands)

I'll make a wild guess at Ahamb (from Vanuatu). With one exception, the initial voiced stops are prenasalised, which looks a Papuan or Melanesian feature. The one exception is 'br-', which might be conceivably be a bilabial trill.

No.

[Pabappa]

well given the way you had to carefully define SE Asia, Im going to guess that it is in the wedge area, and that this is yet another Papuan language.

Am I right?

I cant even begin to guess which family, but thisll help get us going.

[Ares Land]

Is it Sino-Tibetan?

[Creyeditor]

Is it spoken in Asia at all?

[Nortaneous]

well given the way you had to carefully define SE Asia, Im going to guess that it is in the wedge area, and that this is yet another Papuan language.

No. (Carefully defining things is good practice anyway, for increased clarity.)

Is it Sino-Tibetan?

Yes.

Is it spoken in Asia at all?

Yes.

[Karch]

Is it some weird Qiangic language spoken in eastern Tibet?

[Nortaneous]

Is it some weird Qiangic language spoken in eastern Tibet?

As far as I can tell, it isn't spoken in the province of Tibet.

[bradrn]

Is it some weird Qiangic language spoken in eastern Tibet?

As far as I can tell, it isn't spoken in the province of Tibet.

That is some exceptionally precise wording there, especially since most Qiangic languages are spoken outside Tibet province! So: is it Qiangic at all?

[Nortaneous]

Is it some weird Qiangic language spoken in eastern Tibet?

As far as I can tell, it isn't spoken in the province of Tibet.

That is some exceptionally precise wording there, especially since most Qiangic languages are spoken outside Tibet province! So: is it Qiangic at all?

It's classified as Qiangic. (The precise wording here is just because I'm not convinced that Qiangic is a clade.)

[Karch]

Is it rGyalrongic?

[Nortaneous]

Is it rGyalrongic?

Yes.

[Karch]

Is it a rGyalrong language?

[Nortaneous]

Is it a rGyalrong language?

there are a single-digit number of Rgyalrongic languages, but yes

[Karch]

Is it Situ?

[Nortaneous]

Is it Situ?

yes (specifically bTsanlha)

[Zju]

Offtopic, but it kinda annoys me that the 'Name That Language!' game sort of boils down to tree search. The pattern of guessing is almost always the following:
1) Is it spoken in this continent or this giant continental area?
2) Is it from this or that language family?
3) Is it from this or that language subfamily?
4) Is it this or that or maybe that language?

It might be just me, but the game would be more engaging if only the fourth type of questions were allowed, and maybe the third, if the subfamily is altogether small and/or scarcely attested. No narrowing down bit by bit, as if following an algorithm.
It would make people explore linguistic diversity, typology and distribution.

[Richard W]

Offtopic, but it kinda annoys me that the 'Name That Language!' game sort of boils down to tree search. The pattern of guessing is almost always the following:
1) Is it spoken in this continent or this giant continental area?
2) Is it from this or that language family?
3) Is it from this or that language subfamily?
4) Is it this or that or maybe that language?

It might be just me, but the game would be more engaging if only the fourth type of questions were allowed, and maybe the third, if the subfamily is altogether small and/or scarcely attested. No narrowing down bit by bit, as if following an algorithm.
It would make people explore linguistic diversity, typology and distribution.

The problem is that the languages are often too obscure. Perhaps you should allow macrolanguage guesses. We recently had a Brahmaputran language so obscure that it doesn't seem to have a standard name, and its subfamily definitely doesn't (or alternatively, has several names). Even the 20 questions approach failed on that one. I also have doubts as to the reality of the orthography - what I found made me wonder if it was restricted to a single project and in competition with other orthographies. (I tried to find where the type of spelling was used, and instantly disqualified myself by googling one word and finding the sample text.)

Your post would be on-topic in the game's thread.

[Moose-tache]

Another possibility would be limiting people to two questions: one narrowing question and one final guess. Then if nobody guesses it, whoever is closest (according to the arbitrary whims of the person who chose the language) wins. Who-is-closest would probably be more fun than "let's just keep guessing until the most determined person finally guesses right."

[bradrn]

Your post would be on-topic in the game's thread.

Yes, I agree. Could we get this discussion moved to that thread please? (I don’t regularly visit this thread so only just saw this discussion now…)

Another possibility would be limiting people to two questions: one narrowing question and one final guess. Then if nobody guesses it, whoever is closest (according to the arbitrary whims of the person who chose the language) wins. Who-is-closest would probably be more fun than "let's just keep guessing until the most determined person finally guesses right."

I like this idea! (Except for the ‘arbitrary whim’ bit… I’d replace that with something a bit less fuzzy. Something like ‘tree distance according to Glottolog if possible, else use geographical distance’ could work well.)

EDIT: I was interested to see how well that suggested ranking system would work in practise, so I decided to try an example. Let’s say I post a Nuer text, and get the following attempted solutions back:

A: Luo
B: Yimas
C: Tagbu
D: Nandi
E: Nilotic
F: Oto–Manguean
G: East Sudanic

(Yes, I think that people should be able to suggest language families as possible solutions.)

And, for reference, here’s Glottolog’s Nilotic family tree:

More: show

nilotic-nuer.png (25.61 KiB) Viewed 9987 times

Finally, ‘tree distance’ is a bit of a vague term, so let me define the distance of language B from language A as ‘the number of generations from A to the closest common ancestor of A and B’.

Now that the preliminaries are over, let’s have a look at how those guesses fare:

• A’s guess of Luo is in the same family, and is a distance of 3 away from the correct guess of Nuer. (Last common ancestor is Western Nilotic, 3 generations away from Nuer.)
• B’s guess of Yimas is not in the same family, so fall back on geographic distance: ~12700 km.
• C’s guess of Tagbu is again in a different family, so fall back on geographic distance: ~550 km.
• D’s guess of Nandi is in the same family, and is a distance of 4 away from the correct guess of Nuer. (Last common ancestor is Nilotic, 4 generations away from Nuer.)
• E’s guess of Nilotic is also a distance of 4 away from the correct guess of Nuer (the last common ancestor of Nuer and Nilotic being of course Nilotic).
• F’s guess of Oto–Manguean is troublesome: this is a different family, so the scoring method above requires us to fall back on using the distance from Nuer to Oto–Manguean — an impossible task, given that Oto–Manguean is a language family.
• G’s guess of East Sudanic is even worse: this is the family usually assumed to contain Nuer, but Glottolog doesn’t recognise it, making it difficult to unambiguously determine what the distance is from Nuer to East Sudanic.

So the ranking ends up as being A < D = E < C < B, with F and G’s guesses being troublesome. Intuitively, the winner would seem to be A here, and the ranking agrees. So this does seem to be a fairly good ranking system, albeit not quite as good as I’d like.