The following is cross-posted from Dead Voles. It originated in a message to participants in the Socnet and Pajek email lists. It speaks to the issue raised by Michael Alexeevich Popov in his blog post Two Anthropologies, but suggests a less apocalyptic view of our current predicament.
There are, it seems to me, three distinct approaches to social network analysis.
1. The Mathematicians
2. The Social Scientists
3. The Anthropologists and HIstorians
The mathematicians, who now include physicists and computer scientists, develop the algorithms and complex modeling techniques on which we all depend. For the rest of us, the critical question is not so much how do they work as what do they do for us.
The social scientists are driven by a vision in which interactions between network vertices result in network parameters that can then be treated as independent variables in causal analysis. Sampling and statistical inference are critical issues here.
What, then, of the anthropologists and historians? My prototype is a presentation by Dr. Lothar Krempel and a group of German historians I heard at my first Sunbelt conference at St. Pete Beach, Florida, in 2008. The problem was to map the social networks generated by the correspondence related to the Newton-Leibniz controversy over who had invented the calculus. The focus of the presentation was neither a new algorithm nor an attempt at causal explanation. Here SNA was used to enrich our understanding of an already heavily studied bit of intellectual history.
I was struck by this presentation because my own project is an experiment in using SNA in a similar way, to trace the history and enrich understanding of the creators who are members of teams whose ads have won awards in one of Japan's major advertising contests. As someone who has worked in and around the industry for nearly three decades and knows many of these people personally, I bring an ethnographic perspective to the project. I am also working in a heavily documented field; books by and about Japanese creators, a lively and active trade press that has published hundreds of pages each month for, in some cases, more than a half century, a wealth of government and other statistics—the historical context is there for the reading. It just takes time. What SNA brings to the table are tools that let me quickly and persuasively demonstrate connections among, just in my current database, over 4000 winning ads and over 8000 creators who were members of the teams that created them.
My preferred tool is Pajek. Exploratory Social Network Analysis with Pajek gave this newbie to the field a quick and effective introduction to the basic concepts of SNA, along with instruction on how to use the software. I have nothing against other packages; just don't have the time, energy, or compelling motivation to switch to something else with another learning curve.
When I think of where I am with Pajek now, I recall the piano lessons that I took as a child. I have learned my scales and a few basic chords and am now beginning to explore new techniques. I do not aspire to be a composer (Mathematician). Nor do I want to confine myself to a particular modernist style (Social Science). But a certain level of competence, a bit of improvisation, adding a new twist to my understanding of an industry in which I have spent a large part of my life. That seems doable.