Tracing Causal Mechanisms

Abstract

“This argument is too structural. It's under-determined and based on unrealistic assumptions. Moreover, it tells us little about how the world really works.” Among many scholars (the present author included), these are an oft-heard set of complaints. Consider two examples. The central thesis of the democratic peace literature—that democracies do not fight other democracies—is hailed as one of the few law-like propositions in international relations. Yet, as critics rightly stress, we know amazingly little about the mechanisms generating such peaceful relations (Rosato 2003:585–586, passim; Forum 2005; Hamberg 2005). And in Europe, scholars have for years debated the identity-shaping effects of European institutions. One claim is that bureaucrats “go native” in Brussels, adopting European values at the expense of national ones. But here, too, critics correctly note that we know virtually nothing about the processes and mechanisms underlying these potentially transformative dynamics (Checkel 2005a, 2005b). So, to paraphrase a former US president, “it's the process stupid.” To invoke process is synonymous, I argue, with an understanding of theories based on causal mechanisms. To study such mechanisms, we must use a method of process tracing. But this is not easy. Proponents of process tracing should be wary of losing sight of the big picture, be aware of the method's significant data requirements, and recognize epistemological assumptions inherent in its application. Consistent with a central theme of the present Forum, I argue that process tracers are well placed to move us beyond unproductive “either/or” meta-theoretical debates to empirical applications in which both agents and structures matter as well as are explained and understood through both positivist and post-positivist epistemological–methodological lenses.