3D power maps and mountain top castles: Bangime as a Linguistic Island

The life of a linguist can be mystical and magical; this is one of those times for sure. I’m currently attending the Workshop on Language Islands, hosted by three incredible members of Goethe University’s Institute for African Studies, Marie Ngom, Rainer Vossen, Ulrike Zoch, and set in a former castle atop a spectacular mountain range!


Certainly there was no better setting to test out the long-standing German tradition of classifying languages spoken in areas detached from their homelands as islands to see if the same analysis works in the African context. The range of experts gathered here are more than happy to be together in one turret-like round room to discuss the origins and implications of complex data gathered in remote regions in Africa.

Since Rachel Watson and I had been discussing ideas for visualizing complex comparative data for some time now, I tried my hand at Excel’s new add-in, 3D maps (replacing previous Power Maps).  I uploaded the videos I created into YouTube and put them in Prezi. Of course when you try out something new, there’s bound to be glitches, but nevertheless I think the presentation was a fun way to visual multifaceted data.


Certainly not everything in this presentation will make sense to a viewer without a discussant but I do want to share it at least for the purpose of non linear data visualization. In particular, I wanted a way to represent the multiple levels of factors which impact Bangime. The 3D map highlights the relative population density of each of the ten or so neighboring languages which, in sheer numbers, completely eclipse the 1500 or so Bangande population. The fact that Bangime speakers retain their own non-borrowed lexical items in an ocean of widely used terms is surprising to most, but it seems to be a common feature among the potential cases for language islands presented here. Clearly there is more than one model of language contact.


Further, though we don’t often pay literal credence to oral histories, the game “telephone” illustrates the hazards of orally transmitted data, the numerous lexical correspondences found among core vocab Bangime words such add body parts, with far flung Mande and Dogon languages suggest a potential truth to the Dogon-Mande origin myth. In other words, where Bangime does have lexical items reminiscent of other area languages, these are often unexpectedly found among the least likely of words to be borrowed in the language, and then, when from Dogon or Mande, not from the immediately neighboring Dogon languages nor from the Bozo dialect spoken in the adjacent valley. The clear correspondences that Bangime body parts have with Dogon and Mande languages, or those with some Dogon languages and Mande languages, but then not with other Dogon languages (an East-West split has been proposed to have occurred between the Dogon languages – see references in the presentation) currently spoken outside the range of daily contact certainly suggests that a) these core items were not borrowed, but are cognates, and (even if they were borrowings) b) the time at which these items first came into the lexicon was when all the Dogon and Mande languages were spoken in geographic proximity, most likely at the time and location when Proto-Dogon/Mande was spoken.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s