Philosophy

A Causal Theory of Confirmation for Bayesians. 2024, Synthese [PDF] [Slides] 

• Theme: Confirmation 

• Abstract: This paper proposes a new Bayesian confirmation theory, according to which confirmational relations are causal relations between credences. The idea is that proposition E is evidentially relevant to proposition H relative to a credence distribution cr just in case cr(E) is a cause of cr(H), which is understood from an interventionist perspective as intervening on cr(E) would make a difference to the value of cr(H). E confirms H means that under an intervention on cr(E), cr(H) and cr(E) would covary in the same direction; disconfirmation means that they would covary in the opposite direction. I argue that this causal theory explains how orthodox Bayesian theory succeeds in non-extreme credences and fails in extreme credences, the latter known as “the old evidence problem”. Furthermore, the causal theory provides a solution to the old evidence problem that avoids serious problems faced by some other solutions. Ultimately, the causal theory of confirmation subsumes orthodox Bayesian theory as a special-case application to non-extreme credences.

Simpson's Paradox Beyond Confounding. 2024, European Journal for Philosophy of Science, with Zili Dong & Weixin Cai [PDF]

• Theme: Representation

• Abstract: Simpson's paradox (SP) is a statistical phenomenon where the association between two variables reverses, disappears, or emerges, after conditioning on a third variable. It has been proposed (by, e.g., Judea Pearl) that SP should be analyzed using the framework of graphical causal models (i.e., causal DAGs) in which SP is diagnosed as a symptom of confounding bias. This paper contends that this confounding-based analysis cannot fully capture SP: there are cases of SP that cannot be explained away in terms of confounding. Previous works have argued that some cases of SP do not require causal analysis at all. Despite being a logically valid counterexample, we argue that this type of cases poses only a limited challenge to Pearl’s analysis of SP. In our view, a more powerful challenge to Pearl comes from cases of SP that do require causal analysis but can arise without confounding. We demonstrate with examples that accidental associations due to genetic drift, the use of inappropriate aggregate variables as causes, and interactions between units (i.e., inter-unit causation) can all give rise to SP of this type. The discussion is also extended to the amalgamation paradox (of which SP is a special form) which can occur due to the use of non-collapsible association measures, in the absence of confounding.

A Means-End Approach to the Reference Class Problem (under review)

• • • • Theme: Confirmation

      • Synopsis: Describes a modern solution to the reference class problem, based on an analysis of the main statistical and causal inference methods of selecting variables and models. 

      • Full draft available upon request. 

Probabilistic Causation: An Interventionist Trilemma (under review)

• Theme: Explanation

• Synopsis: Describes a tension between interventionists' commitment to counterfactuals and the use of objective probabilities, explores various solutions, and finds them all lacking.

• Full draft available upon request.

Interventionism and the Challenge of Counterfactual Skepticism (in progress) [Draft]

• • • • Theme: Explanation

      • Abstract: What is the relationship between the counterfactual theory of interventionism and the position of counterfactual skepticism? I argue that the skeptical challenge reveals a deep inadequacy in the interventionist framework in probabilistic contexts. This argument is based on an analysis of the argument from chance and a demonstration of the incompatibility of contextualist replies and interventionist theses.

      • Any comments or suggestions are welcome.

A Functional Inquiry into Probabilistic Causation  (in progress) [Draft]

• Theme: Explanation 

• Abstract: What function, if any, could probabilistic causal concepts serve, beyond the various deterministic ones? I argue that their functions are exclusively pragmatic: to supply reasons and to justify actions under uncertainty. This argument is based on the finding that probabilistic causal concepts are required for navigating practical tasks (e.g., promoting a desired state), but seldom appear in scientific analysis of causality, either qualitative or quantitative.

• Any comments or suggestions are welcome.

Reinforcing the Causal Foundation of Evolution (in progress) [Project Proposal]

• Theme: Representation 

• Abstract: What is the precise causal structure of genetic drift in evolutionary theory? I answer this question by presenting formal causal models underlying two types of genetic drift—gametic sampling and demographic stochasticity. The formal results are then used to resolve controversies around genetic drift.

• Any comments or suggestions are welcome.

Inference to Independent Manipulability (in progress) [Abstract]

• Theme: Representation

• Abstract: I argue that factor analysis—one of the most common methods for constructing social variables—can be interpreted as a sophisticated, two-step, abductive inference to independent manipulability, thus compatible with the interventionist philosophy.

• Any comments or suggestions are welcome.