LSE PhD Student Session: Nicolas Cote & David Kinney

Nicolas Cote: “Liberalism and Social Choice”

Abstract: One of the core commitments of liberalism is to the protection of individuals against intolerable intrusions upon their liberty. It enshrines these protections through rights, which safeguard us from coercion. Sen’s celebrated “Impossibility of a Paretian Liberal” result proved that welfarism, as expressed by the Pareto principle, is inconsistent with even the most minimal concern for liberal rights. However, though Sen is acknowledged to have shown the existence of a conflict between liberalism and welfarism, his conception of liberal rights generates paradoxes. I present a novel approach to the representation ofrights in the language of social choice, one which avoids the paradoxes of Sen’s original view, while allowing us to express a whole range of liberal commitments. Most notably, my framework allows us to state precisely how liberal rights constrain welfarism, and even to define meaningful measures of illiberalism, allowing us to rank social states with respect to how deeply illiberal they are.

David Kinney: “Pragmatic Causal Feature Learning”

Abstract: In series of recent papers, Chalupka et al. (2015, 2016a, 2016b, 2017) define a procedure for coarsening the measure spaces of fine-grained variables in causal models. This procedure is said to “learn” the salient macro-level features of a target causal system. Chalupka et al. prove that their coarsening procedure requires a small number of interventions on the causal variable, relative to the total number of values of the fine-grained variables being coarsened. Their procedure implicitly assumes that one should produce as coarse-grained a representation of any pair of cause and effect variables as possible while preserving all mutual information shared between the two variables; this assumption echoes work by Woodward (2010) and Griffiths et al. (2016). In a forthcoming paper, I put forward a decision-theoretic account of causal coarsening that rivals these information-theoretic accounts. It would be a bad result for my view if it could not also claim the positive upshots of Chalupka et al.’s results. In this talk, I will argue that applying Chalupka et al.’s framework to my account of coarsening yields similarly positive results, and even demonstrates some advantages of my proposal.