While case studies are often used in political science research, the discipline has yet to explicitly address the challenges and benefits of studying specific historical events. This is according to SIS professor Joseph Torigian, who recently authored an article published in Global Studies Quarterly advocating his case for event-specific research. We spoke with him to ask some questions about his article and the study of such events.

You also can read the full original article, “A New Case for the Study of Individual Events in Political Science,” with access provided by American University Library.

What do you define as an “individual event” in political science, and what role does case study research currently play in the field of political science?

In the article, by “individual event,” I mean a specific, concrete historical moment in time whose outcome is interesting for a social scientific reason. Case studies are a popular approach in political science, and they have been used to great effect, but, in my article, I argue that we can still think more rigorously about what individual events can actually teach us. Case studies should not really be used as a crutch or supplement to “prove” a mechanism allegedly identified in a causal regression.

When doing small-n to medium-n research, the primary value of the cases is not the variation in outcomes but whether the case studies themselves are convincing—you either get the cases right or wrong, so the onus should be on within-case evidence. Political scientists also use case studies for process-tracing, but in the article, I point out that we have reasons to think that reifying a causal chain may not be the best or only approach to individual events (although I draw upon many of the insights “process-tracers” have written about over the last few years).

What are some of the methodological challenges or constraints in studying individual events in political science?

If the primary “value-add” of your research is a deep dive into a specific event, then you will not be able to answer questions like the size of an average causal effect on a population. You are not trying to identify a constellation of variables that tends to have a certain outcome. Doing research into specific events is just as rigorous as any other kind of social science, but it can also be messy. Since the world files its information so miscellaneously, no topic can be meaningfully researched by going to just one archive, writing up the findings, and moving on. Often, the evidence is ambiguous, and conclusions may change as more information becomes available.

Despite these challenges, why should political scientists study individual events, and what can event-specific research tell us?

First, although a focus on one or two individual events does not allow us to make broad sweeping claims about the homogenous effects of a variable, we can still learn a great deal. A very rigorous investigation into a single event can help us identify the structures that had a gravitational pull on the outcome, as well as the mechanisms. They help us get a sense of how a particular political environment works and make other similar cases more “legible.” Second, it’s at specific moments of political contestation when politics are at their most visible. Third, people outside of academia tend to view important historical moments as symbols representing a particular view of how politics works, and professionals in academia should provide their own accounts of these crucial moments in time. And finally, even though this kind of research often faces evidentiary constraints, if the topic is important enough—and there is enough new evidence—we should still do our best to provide continuous updates on what we know.

You explain in the article what assumptions are most useful when undertaking this type of research method. What are these assumptions that researchers should consider when studying individual events?

First, with regards to generalizability, we should prioritize the discovery of driving forces or structures, not covering laws. Second, as for research design, we no longer need to select targets of inquiry based on the “regularist” or “frequentist” tradition. Third, when we think about evidence, instead of variables, we should seek to find “diagnostic evidence” about what is driving an event within a case. Fourth, we need to manage the tension between driving forces and contingency by accepting that contingency is relative and accepting that some outcomes are more “determined” than others. Fifth, to achieve rigor and falsifiability, we should ask specific questions that can be answered by the available evidence and which have clear social scientific implications.

You used the study of nuclear crises, like the Berlin 1958-1962 nuclear crisis and the Cuban missile crisis of 1962, as examples of event-specific research in your article. What other individual events in political science do you think should be studied in the future?

In my own research, I have focused on elite power struggles in China and the Soviet Union. Political scientists who study authoritarian regimes, especially Leninist ones, have often emphasized that institutions matter, especially in China and the Soviet Union during the eras of Deng Xiaoping and Nikita Khrushchev. But if you look at newly available archival material and the research done by top historians in those countries, you see that those previous accounts need to be updated. Succession politics after Mao and Stalin were about historical legacies and grudges, manipulating ambiguous rules, and courting the military and political police, not competing policy platforms in a semi-democratic environment. Understanding those two moments in time not only helps us understand key moments in Chinese and Russian history, but also shapes our views about the nature of authoritarianism.