Anthropology Co-Citation Graph

Recently Kieran Healy posted a link on Twitter to a co-citation graph he’d made to try to understand what philosophers “have been talking about for the last two decades?” He also posted a nice poster he made from this data [PDF]. I reposted these and mentioned that it would be great to have something similar for anthropology. The internet being the wonderful place that it is, I shortly had my wish, courtesy of Jonathan Goodwin.

anthropology co-citation graph

This chart isn’t as clean as Kieran’s – and probably has too much data (four journals going back to 1973), but Jonathan has helpfully provided instructions for how he did it in case anyone is interested in pursuing it further. I’d love to be able to create separate charts for each of the various sub-disciplines in anthropology, but that might be harder to do since they often appear in the same journals. Still, hopefully some interesting insights can be gleaned from this kind of data. If you are able to do anything with this, let us know in the comments!

UPDATE: Jonathan made a new, lower-density, chart for just 1998-to-the-present.

UPDATE: And a new one, with a chronological slider.

6 thoughts on “Anthropology Co-Citation Graph

  1. There is too much data in this version for it to be very clean, though there are probably tweaks I could make to the link distance, ‘gravity,’ and line thickness to make it display better. Chrome is much snappier with it than Safari on my computer, so you could also try that it if it’s too slow. I believe that Healy’s graphs didn’t go as far back in time as these do, which is another route I could go. A cleaner graph of only American Anthropologist is available here:

    One thing I could implement is another slider for date of publication, but it would require more coding than I have time for at the moment. Also, theoretically, the community-detection algorithm that colors the nodes should be able to identify sub-fields in a general journal.

  2. Jonathon. Hat’s off to your programming skills. If I were using Pajek to generate the static graph I would use the Kamada Kawai spring algorithm and the separate components setting to break up the central hairball a bit.

  3. Jonathan,

    It is hard to predict what various changes will do to the data – but I think your suggestions would probably help: restricting to the last 20 years and reducing the gravity. It’s worth a try anyway!

    What I really hope is that someone with more experience working with this kind of data will be inspired by your work…

  4. Alan. Certainly. All content on this site is licensed under a Creative Commons Attribution-NonCommercial License. But please do come back and leave a comment if anyone does anything interesting with this data!

Comments are closed.