While showdowns over methodology are nothing new in the social sciences (and, indeed, were far more intense in the 1990s than they are today), they continue to provoke impassioned debate between scholars. From disputes between qualitative scholars and quantitative scholars to arguments between different paradigms, questions concerning methodology and analytical frameworks are wide-ranging and seemingly irresolvable. As in most areas of life, I tend to take a rather moderate position on the question of research methods. Nearly every theoretical model and research method provides certain unique advantages, but no single technique is sufficiently powerful to render other methodological approaches obsolete. Thus, I support a research style that incorporates elements of all forms of research, as I feel that only this “wide net” approach is sufficient to capture all the salient variables and trends needed to develop rigorous and useful explanations for complex and often opaque events.
Unsurprisingly, I am frustrated by scholars who exclusively rely on one method of research, as this invariably leads to deficient results that are incredibly easy to criticize and dismiss. Researchers that rely exclusively on qualitative studies can generate deep, precise explanations of certain events, for example, but they aren’t able to analyze enough data points to find meaningful trends across a large period of time. Quantitative scholars can generate data sets that cover centuries, but this numerical approximation inevitably leads to a degradation in the fidelity of the study. Because of this, cases can be coded incorrectly, and unique and novel events tend to be ignored or simply homogenized into the broader group of data.
I have a particular frustration with scholars who rely solely on abstract theory, though, as their work strikes me as both parochial and of relatively little value. To be clear, I think theory development is a valuable exercise, and I also believe it is inevitable simply because the world is too complex to understand without simplified frameworks designed to emphasize certain variables that play an outsized role in shaping events. My frustration arises, however, from scholars deliberately pigeonholing themselves into one theoretical paradigm or another and ignoring empirical examples that refute their theoretical framework or, even worse, twisting historical events so violently as to draw conclusions that no historian of the period would ever advance. Unfortunately, this dynamic exists in all of the social sciences in which I have any degree of formal training (and I’m sure it exists in many of the disciplines that I haven’t studied). Not only does this approach to scholarship necessarily force polarizing debates between academics, but it also leads to fundamentally flawed analyses. For example, in their book The Behavioral Origins of War, D. Scott Bennett and Allan C. Stam find that nearly every contemporary IR theory regarding conflict initiation highlights statistically significant variables that help elucidate why wars occur. Paradoxically, though, every major theory attempts to downplay many of these variables precisely because they are generated from other theories that directly conflict with one another.
The problem isn’t the theories themselves, it’s scholars inability to use theories correctly. Theories are just a simplification of the world that allows us to examine events through a certain lens that highlights certain aspects while downplaying others. This is useful when one seeks to tease out causal linkages between certain variables, but it is actively deleterious to knowledge production when it suppresses other variables simply because they aren’t emphasized in the theory. Good scholars should be able to draw from a range of explanatory models to generate the most accurate understanding of empirical events. Those who remain trapped in meaningless and counterproductive paradigm wars do a disservice to the disciplines they serve and to human knowledge production more broadly.
Theoreticians ensconced in their preferred models respond to criticisms like mine by contending that good theories must be parsimonious because overly detailed theories defeat the purpose of having theoretical models in the first place: They gunk up the pristine frameworks designed to simplify the world sufficiently so as to allow for academic study. The problem with this retort is that it conflates theories with actual explanations of historical events. Theories should be simple, but explanations of nuanced and complex occurrences should not. Obviously having parsimonious theories is valuable, but when one then refuses to incorporate a number of theories into his or her explanation of complicated events, one develops a deficient explanation of phenomena. In other words, if one is going to use parsimonious, over-simplified theories, one must compensate by drawing from a number of them to explain necessarily complex events. Yes, this requires extra work, but it is the only way to honestly and accurately explain the linkages between causal variables and their impact on the world.