OK, I admit it: I have to work on writing better titles for my posts. “Network Theory and the Social Sciences” sounds a bit dry. And the last time I tried to write about network theory it didn’t exactly produce a riveting post. But Will Wilkinson and James Fowler’s diavlog on Bloggingheads makes the topic not only accessible but entertaining.
The topic of the talk is Fowler’s book Connected: The Surprising Power of our Social Networks and How They Shape Our Lives (co-authored with Nicholas Christakis). The book and the talk are about real world social networks, not online social networking, and pretty much all of the ground they cover is extremely interesting.
Early on they discuss the value of network theory to the social sciences. To quote Wilkinson:
One of the things that I liked a lot about the book was the idea that thinking about networks and the way they affect us offers a kind of ‘third way’ between a couple of extremes in the social sciences. So, on the one hand, the extreme atomistic individualism of homo economicus. On the other extreme there’s a kind of radical holism of certain kinds of sociologists or anthropologists.
As Fowler puts it:
The last few centuries in social science have been sort of like a tennis match where we’ve been going back and forth between the individual and the group… Physicists came on the scene in the late 20th century… and made the whole scientific community realize that to understand complexity you needed to move beyond the individual but not all the way up to the group. You needed to focus on the relationships between groups and how those relationships structure interactions and how they cause things to flow from one node to another in a network. And for human beings what that means is that what we should be thinking about is these social contagions: things like political behavior, things like emotions, things like love and sex and your health behavior
Here’s a four-minute clip of the section I’m quoting but the whole thing is well worth a look. Fowler’s work shows how ideas and behavior ripple through social networks such that you not only influence your friends but your friends’ friends. Rather than trying to summarize and risk misrepresenting the research, I suggest you check out the diavlog or the book for yourself.
At one point in the discussion Wilkinson takes issue with the application of the language of “contagion” to emotions or eating habits, especially given the lack of a clear transmission mechanism.
How is this different from just the fact that people pick up norms and habits from their immediate social milieu? Is that the same thing as ‘catching’ something? …So why isn’t this just a story about people adopting norms?
Whether the language is appropriate is an interesting question, but what I really want to focus on is the question “Why isn’t this just a story about people adopting norms?” And what I’d suggest (though as an amateur I suppose I should insist rather than suggest so people will believe me) is that the difference has less to do with the phenomena being examined and more to do with the methods being used in the examination.
In other words, the question isn’t whether my eating habits are the result of me adopting norms or the result of social contagion. It’s whether or not network science offers a useful framework for explaining and predicting patterns of human behavior.
I suspect Fowler would agree, based on his response, which focuses on the utility of network science to the study of culture:
I think this goes back to the discussion we were having about this change that we’re trying to get people to see in the sciences… I really think that what network science is ultimately going to help us to do is to help us think scientifically about the evolution of culture.
And shortly thereafter Wilkinson captures it more directly:
Just saying that people tend to download norms from the ambient culture really does leave it very vague where they’re getting their norms from or where they’re getting their ideas about what’s acceptable. And so I like the idea that the network gives you something more specific to look at.
To conclude: Fowler’s story is different than a story about norms not because he’s looking at different phenomena, but because he’s applying a different scientific framework to those phenomena. This might seem obvious but there’s one last point that, I think, is worth making in light of Will’s comment about network theory as a “third way” in social science.
The “third way” between atomistic individualism and group identity is extraordinarily appealing. None of us wants to feel that we are only a member of a group, that we lack all individuality. Most of us also recognize and accept that, at least to some extent, our lives are influenced by the complexity of circumstance.
And so the “third way” of the network seems appealing, even true. What we must strive to remember is that it is no more “true” than any other research method. It is simply a tool, to be embraced or discarded based not on whether we find its level of analysis intuitively appealing, but on the extent to which it allows us to explain and predict human behavior.
UPDATE: Here’s a fun article from The Washington Post on how loneliness is contagious within a social network.