Article idea: scholarly social network mapping

Several days ago I had an idea for an article: a study of humanities scholars' social and professional networks, using names mentioned in the acknowledgments sections of scholarly monographs. If you assembled enough data, I bet you could build a network graph, and if you had some way of representing people's areas of specialization, you could start to map the way various fields overlap and intersect. And then you could use that information to think about any number of library-oriented questions -- the nature of scholarly communication, the "invisible college" phenomenon, how to tap existing networks to do outreach, and so on. It would be a bit like doing a citation study, but more geared toward finding out the less obvious connections, i.e. who helped a book get written as opposed to who gets directly cited in footnotes.

My main concern is about the ethics of aggregating and publishing this kind of information, even though all of it would already be publicly available. Even though it's common knowledge, in theory, that X thanks Y in her acknowledgments, it's not quite the same to put together a big pool of data showing X's connections to Y (and Z, and W, and A and B) all in one place. Collegial relationships often shade over into personal friendships, and it feels a bit intrusive to map those connections. Could one anonymize the information, somehow, and just label everyone with letters of the alphabet or something else neutral, and only identify their research specialty? It would probably mean more work, but I'm still thinking it would make a really interesting project.

So, O blogosphere, is this a viable idea, do you think? And if it turns out to be, would anyone be interested in co-authoring?

Comments

[Valdis Krebs]

Public info does not need special permission to analyze/map. Here is a network map of social network scholars taken from public data available on who co-authored with whom in the journal "Social Networks". This data was given out to all participants at this year's INSNA [International Network for Social Network Analysis] conference -- Sunbelt 2008.

http://www.orgnet.com/SN.html

It is also easy to get other famous co-author data sets such as those of mathematician Paul Erdos.

[Patricia]

Fascinating--and a quite useful idea. You might want to take a look at Amanda French's post on networks, titled "Networks & Literary Influence," inspired by her reading of "Six Degrees: The Science of a Connected Age," by Duncan Watts. See:

http://amandafrench.net/2008/08/12/networks-literary-influence/

It occurs to me that, despite the many acknowledgments one typically sees in the forewords of scholarly monographs by humanists, scholars in the humanities still shy away from collaboration when it comes to their research and writing (and, sometimes, their teaching). Your idea would be a step in the right direction toward changing this situation for the better.

[Patricia]

Ah, I see from Amanda F's comments that you already know of her post, and she knows of yours. Perhaps a co-authoring opp is in the offing!

[k8]

I say go for it. I do know someone in my primary field who researched how a few primary ideas in the field could be traced back to the fact that several important people all happened to be in the same program or a cognate program at a particular university.

[Amanda]

Ah, yes, I'd forgotten about Erdos numbers. And I like the co-authoring angle!

[Valdis Krebs]

Here is a network map of Erdos' collaborators -- these nodes all have an Erdos number of 1 and the links show which collaborators co-wrote papers together. Erdos' node is not shown to make the map clearer [his node is not shown because it is understood he is connected to everyone on the map].

http://www.orgnet.com/Erdos.html

[Amanda French]

Honestly, I don't see any ethical problem with this -- and it's interesting! Would you include dedicatees, or only acknowledgees?

I especially remember going through a number of scholarly works from the 50s and 60s and being struck by the number of wives thanked "for typing the manuscript." Man.

One thing you'd run into, I'm sure, is the number of acknowledgees who aren't scholars. Also, I'd think it'd be important to establish some way of limiting the scope up front, at least at first. One way to do it would be to start with a central figure (like Erdos) & trace outward.

I'd offer to co-author, but honestly I'm pretty booked with writing commitments right now. Someday we'll definitely collaborate, though!

[Amanda]

I seem to recall reading an acknowledgments section that commented on that whole "thanks to my wife for typing the manuscript" phenomenon (and ended with "and I did the typing myself," which was kind of nice).

I've been thinking about both of those problems -- non-scholar acknowledgees, and limiting scope. I think I'd have to keep it within a set time frame -- just books published in the last five ten years, say -- but working outward makes sense. And I'd have to figure out some way to identify the acknowledgees. Sometimes it's easy to tell when people are thanking their siblings or parents or next-door neighbors or cats, but sometimes it's harder. I considered using the MLA directory, except not everyone is a member.

Pity you're busy at the moment, but then again, so am I, so who knows when this project might actually get off the ground?