This Henry Farrell Substack post about AI and modern social science argues that we need to reorient much of our work toward the kind of “Big Structures, Large Processes and Huge Comparisons” that Charles Tilly advocated we study. Yet this is precisely not what a lot of political science (and more generally social science) research does these days:
“the social sciences now lay a much greater emphasis on identifying causal relationships, which tends to push away from the kinds of big secular shifts that he suggests we ought focus on.”
As a result, Farrell argues that “we don’t just need well funded and politically and economically independent social sciences to understand the consequences of AI, but also to reorient the social sciences so that they are again well suited to deal with big structures and large processes. We are in a world where it is urgently important to understand change at scale, since it is just such change that we are going through.2 The social sciences need both solid funding, and a lot of internal work to reorient themselves towards the questions that matter.”
This reminded me of a recent PS article by Guy Grossman, William Dinneen, and Carolina Torreblanca with tons of fascinating data on the nature of books and articles published in Political Science between 2003-2023. Their bleak summary:
“Our analyses show that publication volume has tripled during that time, driven by new entrants and increasing collaboration. The field has become increasingly quantitative but remains substantively diverse, even though individual papers have become more topically focused. Younger scholars publish more articles but fewer books and book chapters, signaling strategic adaptation to institutional rewards. Although novel research attracts, on average, more citations over time, it does not consistently translate into higher journal placement, encouraging safer, incremental publication strategies. Together, these patterns depict a discipline that is larger and more productive than ever yet is increasingly organized around measurable outputs, rather than intellectual risk-taking or theoretical innovation [emphasis added]”
If Farrell is right, this suggest that doing work on “the questions that matter” may be a real struggle for the discipline, which is increasingly accustomed to making its judgments largely a matter of mechanistically counting lines on a CV rather than reading and evaluating the content of those CV lines. I’m not a political science pessimist – there is lots of great work being done in the discipline of various kinds – but I do think the trends Grossman et al. point to are deeply worrisome. This is especially true in a context of crises of higher education that focuses PhD students and junior scholars (understandably) on pumping out as much as humanly possible as quickly as possible in a desperate bid to a land some career security.
Paul Staniland 