Social science all shook up

The Times Higher Education recently published an article titled Do the social sciences need a shake up? which is a response of sorts to Nicholas Christakis’ NY Times op-ed, Let’s shake up the social sciences. The central thesis of both is that the social sciences have stagnated, largely because of disciplinary silos, and would better serve society if reorganized. Here are my thoughts as a recent transplant from the biophysical to social sciences.

On the Christakis piece:

Some reorganization of social science to create departments (or some university structure) of cohesive, innovative areas such as network science and neuroeconomics would almost certainly catalyze progress. As the THE authors put it, “The truly serious issue is whether immovable department names are a worrying signal of immovability of thought. Sadly, we suspect that, in the UK at least, they are.” That has been my experience as well. Siloization is a huge problem in academia and we seem not to know how to fix it, but the old boundaries feel like British-drawn lines in Middle East sand.

However, the reorganization seems to have begun, if modestly and perhaps in the margins. I haven’t tracked it formally, but it sure seems like there are more job announcements every year for environmental science departments, computational social science centers, and the like.

Undergraduates absolutely should have the equivalent of physical science labs. Further, curricula need to be reformed. Statistics and coding are huge parts of modern social science research. Those skills aren’t essential to do quality research, but imagine two applicants to your graduate program, both environmental studies majors, one with a minor in stats, math, or CS and the other without. ‘nuf said. And of course those skills open myriad doors outside academia.

What’s more, I think there’s sizable group of smart, motivated undergraduates that could be recruited to social and social-environmental science programs if they were seen as more “hard core”. I studied chemistry as an undergrad not out of any particular interest or career direction but because it was challenging and engaging and there was a sense that it was leading toward good jobs. Social science is nearly universally intriguing, but the smart kids over in engineering think – with some justification – that their talent and tuition would be wasted in a social science program. In my Department of Environmental Science and Policy there are dual environmental science and environmental management programs. This kind of multi-tracking seems good. Why not have computational social-environmental science major that draws from CS, statistics, environmental science, and political science?

On the THE piece:

That the American Political Science Review hasn’t published a single article about climate change is staggering. I don’t know why this is, but I’m shocked and appalled.

“Simon Bastow and Jane Tinkler … argue that the social sciences are increasingly important in the study of human-influenced systems, such as the planet’s climate, and that we need substantially greater interaction and integration between science and social science subjects.” I’m not sure. Interaction, certainly; integration, carefully. Biophysical solutions without attention to the social side of the equation are doomed. But, expertise is extremely difficult to obtain; one essentially cannot be an expert on both the biophysical and social science. And given the increasing quantitative and computational demands on social scientists, asking for more than passing competence in biophysical subjects might mean compromising disciplinary expertise. Of course we need some competency on the other side of the divide, but mostly to bring formidable expertise to bare on both sides of the equation, we need incentive for collaboration, respect for other disciplines, and common language. It is astonishing how different vocabulary can be in different departments for the same subject.

On social science articles being too long, I mostly agree. To some extent, I think the move to quantitative investigations helps with this. Verbal arguments take up a lot of space and often substitute for mathematical models in the social sciences. Sometimes this is necessary, but when equations can substitute for words, they should. From McElreath and Boyd’s Mathematical Models of Social Evolution, “[F]ormal (mathematical) models are also important because they give us the ability to clearly communicate what we mean. The looseness of verbal models sometimes allows scholars to argue for years about their implications. Such wrangling is much rarer in mathematical disciplines because you have to precisely specify how every piece of the theory relates to every other piece.” So more quantitative arguments could mean shorter articles and more precise arguments.

One caveat is that in the biophysical sciences it is much easier to define the components of the system under investigation: “protein kinase A” can be defined in a sentence without difficulty; on the other hand, “bridging social capital” – while no less real or important – is a much tougher thing to define and constrain. For this reason, social science may always need to be more verbose, especially as frameworks are in development.

Finally, I think we have to keep in mind the role that funding plays in the desired transformation. The fewer resources that are available, the more academics have to retreat to safe plays to keep their career prospects alive. Both in quantity and in organization, funding agencies could do a lot to bring about the shake up we’re looking for.