Potential outcomes, counterfactuals, and structural modelling. Causal approaches in the social sciences

This paper examines the potential outcome model developed by Rubin and its counterfactual underpinnings as developed by Lewis. Though a ma jor contribution of Rubin’s potential outcome model has been to stress the importance of the design stage, we recall the main methodological and epistemological flaws of this approach. We argue that the study of causes and effects does not necessarily require counterfactuals, once a structural modelling framework, as the one developed here, is adopted. Our approach emphasises and spells out the role of background knowledge, marginal conditional decomposition, and of stability for providing a causal explanation of a given phenomenon.