Justifying Interventionist Assumptions
Michael Baumgartner, University of Bern and Center for Philosophy of Science

Abstract: One of the core qualities of interventionist methodologies of causal reasoning is widely taken to be their ability to disambiguate causal inferences that other methodologies leave ambiguous. Essentially, such disambiguation is possible within the interventionist framework because it adopts stronger causal assumptions about analyzed causal structures than other frameworks. Justifying these stronger assumptions is not normally seen to be one of the cardinal requirements for establishing the soundness of interventionist causal inferences. Advocates of interventionism often (explicitly or implicitly) consider these stronger assumptions to be reasonably justified by their plain power to resolve ambiguities. In this talk, I shall cast doubts on this pragmatic acceptance of interventionist assumptions. Notwithstanding their indisputable inferential strength, these assumptions require further justification, for ambiguous causal inferences could often be disambiguated in radically different ways by means of different assumptions with equal inferential strength. Further justification of interventionist assumptions within the interventionist framework, however, tends to trigger an infinite regress.