The social sciences are insular. Arguably more so than the humanities or the natural sciences. Consider the map below from Jim Moody which looks at PhD training at Duke University and tries to capture “how faculty training crosses PhD programs [and]…how faculty publications cross multiple scholarly fields.” The social sciences are islands while the natural sciences and humanities tend to be part of denser networks.
But, I increasingly have the sense that the social sciences are converging both methodologically and substantively, not to mention in learning from and co-authoring with each other. I spent the summer reading some survey books on Psychology (Bloom’s Psych), Sociology (Collins’s Four Traditions in Sociology and Sociological Insight), and Anthropology (Henrich’s Secret of Our Success) and even one from my own discipline (Ansell’s Why Politics Fails) that seemed to suggest even more connections. Are the social sciences merging? Is there ultimately just one social science?
To start to answer these questions, I first thought about Political Science’s trade balance with other disciplines, which I wrote about here. The short story is that we tended to be importers from others, but I could identify a few examples of exports.
Then I went looking for overlaps or intersections between social sciences, which is my subject here. I tried to identify whether there were institutionalized subfields of each social science (as opposed to extraordinary individuals) that address key questions or topics in another social science.1
The table below lays out some of the intersections (apologies to those that I missed).2 Surprisingly to me, I was able to fill in most of the blanks even with a generous interpretation of social science that includes History and Philosophy.
That said, not all of these overlaps indicated genuine engagement. Political Science I think does pretty well. Political scientists tend to be genuinely aware of and responsive to developments in Economics, Philosophy (if you count political theorists), Psychology, and Sociology when their subject matter coincides. This might be due to our lack of a clear identity or the fact that many of our engagements are wholesale imports of other fields (see my earlier post). Similarly, sociologists are responsive to their peers in Political Science and Economics (even if they tend to be reflexively critical of the economists).
On the other hand, I don’t sense as much give or take between anthropologists and others. Part of this is presumably that their subject matter – frequently pre-industrial and extinct societies – doesn’t overlap much with the other social sciences who were born in trying to understand modern life. (Anthropology is also divided into four disparate subfields which make comparisons harder.) Historians might land here as well though their goals tend to be different from social scientists – they are less interested in developing theories or testing hypotheses – and so mutual exchange may be limited.
Economics represents another mode of interaction. It may study the same material as others (it is often seen as imperialistic), but it tends to reinvent those other fields rather than engage with them. Thus, when economists do psychology, it is as behavioral economics, and when they do sociology, it as network analysis.3 I sense that psychologists are somewhat similar in that they will cover ground close to Economics or Sociology, but on their own terms. A related kind of standoffishness can be seen in Philosophy which takes other field as objects of study rather than interlocutors – for example, in the philosophy of social science. Philosophy also addresses normative issues more than positive ones.
A couple of questions arise from this? Are there any empty boxes where interesting dialogues could be started? Are there examples of good and bad overlaps?
One absence that struck me was the isolation of Anthropology. Though it might be the most fundamental social science – the study of human culture – my sense is that there is relatively little exchange between Anthropology and the other disciplines. The implications of culture and cultural difference should arguably be central in the study the economy, politics, and society (and in the past they were central concerns and helped to motivate the growth of social science), but I sense that we have neglected them. This may have political roots. On the one hand, there is a hesitancy to compare cultures because previous attempts were in the service of colonialism and racism or because of a liberal belief that individual choice trumps culture. On the other, there is the tendency of some social sciences like psychology to develop supposedly universal theories based on WEIRD (Western, Educated, Industrialized, Rich, and Democratic) subjects. (I put together some data on this here.) I’m not sure of the reasons for Anthropology’s isolation, but there are attempts to restore culture to a central place in the social sciences. I am thinking of the work of Joe Henrich and his collaborators, which I wrote about here.
I think history is absent from exchange in significant ways as well. Though some fields mine history for cases to study and as a testing ground for theories (see here), not many take seriously the idea that history is really a foreign country and people may have had different psychologies and motivations in the past. This is somewhat ironic as the social sciences originated in an attempt to explain the changes that modernity and the industrial revolution brought about. Muthukrishna, Henrich and Slingerland’s piece “Psychology as a Historical Science” is a fascinating attempt to restore a historical sense to the social sciences who tend to look for universal laws rather than historical change. (Here is my political science take on this.) Perhaps this will ultimately become Asimov’s psychohistory, a science that inspired Paul Krugman among others.4
Another outlier is philosophy or normative theory in general. A recent trend in empirical social science is to foreground justice or social justice as the goal of our work. I tend to be a stronger believer in the fact/value divide and try to bracket values in my work, but if normative issues are to play a role in social science, then there needs to be much better grounding in philosophy and ethics than has heretofore been the case. There are few easy answers to the question of what is just and any easy privileging of a particular approach to justice is bound to mislead. Some fields do provide their own training here (welfare economics or political theory or perhaps moral psychology), but the issues are capacious and aren’t solved by simple references to justice, social or otherwise.
Finally, I noticed a large gap between Psychology and Sociology that could be filled. Yes, Psychology has a thriving field of social psychology (though it has been decimated by the replication crisis), but it is mostly distinct from Sociology, not least in methods. Meanwhile, I don’t sense that Psychology (with a large or small p) plays much of a role in Sociology, perhaps because of Sociology’s fundamental insight that society does not reduce to individuals or because of its very different methodologies.5 (This may be why Political Science is more open to Psychology.) Yet, a unified social science would probably stand on a reconciliation of psychology and sociology.
Are there some positive examples of cross-field interactions? From my point of view, the topic of political economy has seen a lot of productive work that comes at the intersection of Economics, Political Science, Sociology, and even History (Gerschenkron, Pomerantz, and so on) and Anthropology (Polanyi). The study of democracy is another area that has been enriched by contributions from economists and sociologists and more recently anthropology and psychology (Henrich, Haidt, Stasavage’s use of anthropological data). The study of normative political thought in two separate departments (Political Science and Philosophy) has generated differences in approaches but also dialogues that are interestingly discussed by Jacob Levy.
This suggests that disciplinarity and insularity has benefits that might not be replicated if there were a single discipline of social science. Each brings a different wissenschaft to their subject matter. This raises an interesting question of industrial organization/organizational sociology – how should the disciplines be organized – that deserves more attention. I’ve often been skeptical of institutional attempts to push disciplines together in the name of interdisciplinarity. Perhaps the key is how it is done.
In his memoir Misbehaving, Richard Thaler writes about how he helped to found the relatively successful field of behavioral economics. His first attempt was to bring together a group of superstar scholars from different fields – psychologists along with economists - to lay out a template and come up with general principles. This did not work. As he describes it, “academics don’t want to talk about research in the abstract – they want to see actual scientific results. But if scientists from one field start presenting their research findings in the manner that the colleagues in their field expect, the scientists from other disciplines are soon overwhelmed by technical details they do not understand or bored by theoretical exercises they find pointless.”
The more successful tactic was to hold a two-week intensive summer school for graduate students as no university was teaching a graduate course in behavioral economics. This attracted extremely talented students who went on to become leaders in the field.
I’m curious if these results on interdisciplinary collaboration are more general. Part of my skepticism of interdisciplinary initiatives in the social science is that disciplines tend to be defined by methods (wissenschaft) more than common subject matter interest. The political scientist/economist Scott Gehlbach seems to say something similar here in conversation with the historian Tracy Dennison. They note that in a meeting of social scientists and historians each side was very skeptical of the other’s evidence – historians saw quantitative evidence as suffering from selection bias – after all, official data was collected by governments for political reasons – while political scientists were skeptical of archival textual evidence as being subject to cherry-picking by scholars.
Nevertheless, I am hopeful that more convergence is possible and that one day we will produce a Hari Seldon.6 At the least, reading up on other fields has made me a more generous, and hopefully better, thinker.
I also sidestep works that explicitly try to explain a particular field using methods from another field – eg, Downs’s An Economic Theory of Democracy.
A better analysis – perhaps for later – would look at citation networks and co-authorships as Moody does here.
See the political science roots of Piketty’s work on voting (the Twitter reaction from political scientists was far more negative). Or note how Akerlof wrote a book discovering the idea that identity shapes economic interactions.
I don’t view history as innocent here either. Historians’ aloofness from social science can lead to bad history as in much of the work on the history of capitalism which could be helped by some interaction with economics.
Healy notes that a frequent comment at ASA meetings is “Why is this sociology?”
One scholar who seems to be aiming for this role is Peter Turchin, but I am still unsure what to make of his work.