A reference class problem for causal explanation?

Katie Steele (ANU)

About the talk: Much ink has been spilled over the value judgments associated with ‘inductive risk’, i.e., the risk one assumes when determining whether there is adequate evidence to assert some (scientific) claim. Comparatively little emphasis is given to other potential ‘risks’ involved in science communication, specifically, when providing causal explanations of phenomena. This talk seeks to rectify this imbalance by focussing on the choices associated with selecting particular causes as most significant for an effect. I argue that value, and not just epistemic, judgments are needed to resolve what is effectively a reference class problem associated with causal selection.

About the speaker:Katie Steele studied mathematics at the University of Queensland (BSc 1999). She later pursued an MA in philosophy at the University of Queensland (MA 2003), followed by a PhD in philosophy at the University of Sydney (PhD 2007). Katie was subsequently a Postdoctoral Research Fellow in the Australian Centre of Excellence for Risk Analysis at the University of Sydney. She then took up a position in the Department of Philosophy, Logic and Scientific Method at the London School of Economics and remained there from 2008 to 2016, when she joined the School of Philosophy at the ANU. Her research interests include rational choice and attitude change, agency and accountability, ethics and risk, scientific inference and explanation, evidence-based policy, and climate change policy.