What Counterfactuals Can Be Tested
This addresses a foundational challenge in scientific discourse for researchers and practitioners dealing with causal inference and hypothetical reasoning.
The paper tackles the problem of testing counterfactual statements, which are hypothetical and often incompatible with observed data, by providing a complete characterization of testable counterfactuals and procedures to infer their probabilities from physical experiments.
Counterfactual statements, e.g., "my headache would be gone had I taken an aspirin" are central to scientific discourse, and are formally interpreted as statements derived from "alternative worlds". However, since they invoke hypothetical states of affairs, often incompatible with what is actually known or observed, testing counterfactuals is fraught with conceptual and practical difficulties. In this paper, we provide a complete characterization of "testable counterfactuals," namely, counterfactual statements whose probabilities can be inferred from physical experiments. We provide complete procedures for discerning whether a given counterfactual is testable and, if so, expressing its probability in terms of experimental data.