Published online by Cambridge University Press: 01 January 2022
I present three reasons why philosophers of science should be more concerned about violations of causal faithfulness (CF). In complex evolved systems, mechanisms for maintaining equilibrium states are highly likely to violate CF. Even when such systems do not precisely violate CF, they may nevertheless generate precisely the same problems for inferring causal structure from probabilistic relationships in data as do genuine CF violations. Thus, potential CF violations are particularly germane to experimental science when we rely on probabilistic information to uncover causal structures since we cannot then use those structures to predict the right experiments to ‘catch out’ hidden causal relationships.
Many thanks to audiences at PSA 2012, University of Minnesota, and CaEits 2011 for helpful questions and feedback. Special thanks to Endre Begby and Kathleen Creel for ongoing discussion and feedback on drafts and to Creel for presenting this article on my behalf at PSA 2012.