I'm wondering whether anyone knows of any machine parsable database of the geographical distribution of natural languages (for example the geographical distribution of (native) speakers)?

I know WALS has some basic information (it assigns some kind of point of origin for each langauge), but this is not what I'm looking for (indeed, I would like to have such a database to cross reference with WALS).

Edit: As the confusion in the comment section suggest my wording was rather unfortunate, as it sounded like I had a specific problem in mind. I'm not looking for anything very specfic, but I'm wondering generally about what (machine parsable) databases containing data about geographical aspects of natural langauges there are. Two very reasonable examples suggested in the comments are databases outlining the geographical regions languages are spoken and (dually) databases for querying what languages are spoken in any geographical region. Even more specific could be something along the lines of a database for querying (native?) speaker density of larger languages (for small ones I guess WALS would usually be enough).

One thing I've thought of using such a database for is generating more informative maps from WALS, than those already supported by the web interface (having English be a point in England seems somewhat unintuitive).

  • If what WALS has is not useful enough for you, what then do you want? What do you want to ask of this database? "at position X, what languages are spoken?", "for language Y, what is the bounday of locations where it is spoken?"? Or some other kind of question? What purpose would you give such a database? Do you want just a more refined linguistic atlas than are in existing maps?
    – Mitch
    Oct 16, 2011 at 15:25
  • @Mitch I was generally interested in anything giving more geographic data than WALS. Both your suggestions would be very interesting, but I was generally curious about what databases linking geographical and linguistic data were available. Oct 16, 2011 at 16:04
  • @TiloWiklund You're still not telling us exactly which kinds of data you're looking for. All you're saying is that WALS doesn't meet your needs.
    – Alek Storm
    Oct 16, 2011 at 19:56
  • 1
    There's not much and parsing is a problem for all. The only thing that tries to deal with the whole world is Ethnologue. Merrit Ruhlen has compiled a document listing information for the world's languages, including location, but it's a big .pdf document here. Covering just endangered languages is: UNESCO Atlas of the World's Languages in Danger; and forthcoming (one day!) ELCat. Oct 16, 2011 at 23:19
  • As @GastonÜmlaut mentioned, the closest resource to what you're looking for would probably be the Ethnologue, which does list the names of geographical locations for native speakers of natural languages. But then you'd need to somehow link this to a separate database that could match up those names with a map. I'm guessing that Google Maps should be able to do this, though I'm unfamiliar with their API personally. Oct 17, 2011 at 1:56

1 Answer 1


Probably the best online source of information on the geographical distribution of natural languages would be Ethnologue. Ethnologue aims to catalogue all of the world's languages and provides speaker numbers, rough location, family affiliation and other information, along with maps for many areas. Conversely, for every country in the world it lists the main languages that are spoken there, along with an estimate of speaker numbers. It is not designed to be parsed, but it can be.

An alternative is the World Language Mapping System, but this is not free and uses a very dodgy language classification scheme (Ruhlen's).

Some other sources that deal with restricted subsets of the world's languages are: UNESCO Atlas of the World's Languages in Danger, AustLang (a complete catalogue of Australian languages), and the forthcoming ELCat.

  • 1
    Ethnologue is not particularly precise for where some languages are spoken, and trying to capture the difference between where a language is spoken and 'also spoken in' is not always clear. For example, that Mandarin is also spoken in the USA as well as China is probably not as important as the fact that English is 'also spoken in' the USA as well as the UK. Also, Ethnologue doesn't make clear if it's listing things by district, state or country. Still, it's the most compete source and should be relatively extractable.
    – LaurenG
    Oct 22, 2011 at 1:09
  • @LaurenG Not sure I follow what you say about '...district, state or country', Ethnologue lists languages by country and gives fairly good 'region' info for each lang. Oct 22, 2011 at 1:38
  • 1
    I quite agree with LaurenG, but I still think this is a quite reasonable answer, and nobody seems to know of any more precise databases. Oct 22, 2011 at 12:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.