Process Tracing, Mechanisms and Causation

Alexander George, Andrew Bennett and others have defended case studies in political science under the method George first termed ‘process tracing’. The systematic methods they recommend provide the best guide to doing case study research. George and his followers have demonstrated through their own substantive research how much we can learn from careful historical case studies or process tracing. However, their theoretical justification of process tracing in terms of its ability to pin down mechanism and causal processes is less assured than the substantive work itself. I will critique their account of causation, their account of the relationship between generalizations and mechanisms. I will argue that despite their systematic methods case studies still suffer from the specification problem. I then provide a rather simpler and more straightforward methodological defence of historical narratives.