Tuesday, 17 September 2013

Maps - Language and Population Density

The maps in my post yesterday, showing the distribution of some language families in South America, got me thinking.  Most of the land coloured in as containing speakers of a particular language is uninhabited, and the languages themselves are spoken by relatively low numbers of speakers.  Moreover, there is often great diversity at a lower level.  As Michael Heckenberger tells us of the Upper Xingu, plenty of indigenous South Americans speak not only different languages, but even languages of different stocks.  The solid blocks of colour are completely inaccurate and give the bluntest, least nuanced understanding of the distributions.  I understand that many of them were produced by amateur cartographers/Wikipedians, but most maps in professional publications are no better.

So I was wondering if there was a way to simultaneously encode population density and language to give a more rounded picture of how things are on the ground.  There must be a way to do it, and then you'd only need accurate data to put on the chart.  Granted, South American languages families are poorly documented in every sense, including their present-day (let alone prehistoric!) distributions, and the numbers of speakers are low anyway and would probably barely show up on a map of the whole of South America, but...

I'm busy this evening, but if anyone out there knows of any attempt to do this, I'd really appreciate it if you could pop a note or link in the comments.  If I find anything myself in the next few days, I'll post something about it anyway.


  1. Well, there's this (though it's for race/ethnicity, not language): http://demographics.coopercenter.org/DotMap/index.html

  2. I've so far struggled to find anything similar for language, surprisingly. So maybe that's the closest we're going to get. I'm curious as to why this method isn't widely used for maps of language, though, as it seems to resolve many of the problems of displaying information about linguistically diverse regions.

  3. And for language, even:

  4. Hi there A.J.,
    You need to look for your solution in the framework of G.I.S. technologies (geographical Information Systems).

    Work has been done in the general area of interest you raise in this article. What appears to be a good product is supplied by World Geo datasets:


    They claim that as of version 1.6 of their software they now enable mapping also with population densities shown

    You may want to look in particular at the work that has been done by Dr. Steve Huffman in mapping the world's phyla/family of languages:

    By the way, I am not connected in any way with the above mentioned company.


You can post anonymously if you really want to, but I would appreciate it if you could provide some means of identifying who you are, if only for the purpose of knowing who has written what.