This paper explores the epistemological and methodological foundations of process tracing. Recent attempts to understand qualitative research through the lens of mainstream statistical work and the experimental template have been largely misguided. Nor does it seem fruitful to understand process tracing solely on the basis of causal mechanisms, as many quantitative researchers also regard mechanismic reasoning as central to their work. Instead, this paper will contribute to an emerging literature that approaches the methodology of process tracing from the inferential logic it employs. I argue that the notion of the ‘trace’ is more than a metaphor in this respect. The concept has been explored in philosophy and history, most notably by the ‘microhistorian’ Carlo Ginzburg. I show that ‘tracing’ entails a particular use of evidence and logic of causal inference in line with what Ginzburg called the ‘conjectural paradigm’ of science. This model has not received the attention it deserves in political science, despite some interlinkages with recent advances in the literature on qualitative methods (e.g. the ‘folk-Bayesian’ interpretation of process tracing). The paper ends with a discussion of the limitations of the ‘conjectural model’ of qualitative research.
