Causal counterfactuals are not interventionist counterfactuals

Tyrus Fisher

pp. 4935-4957

in: Matteo Colombo, Raoul Gervais, Jan Sprenger (eds), Objectivity in science, Synthese 194 (12), 2017.

Abstract

In this paper I present a limitation to what may be called strictly-interventionistic causal-model semantic theories for subjunctive conditionals. And I offer a line of response to Briggs’ (Philos Stud 160:139–166, 2012) counterexample to Modus Ponens—given within a strictly-interventionistic framework—for the subjunctive conditional. The paper also contains some discussion of backtracking counterfactuals and backtracking interpretations. The limitation inherent to strict interventionism is brought out via a class of counterexamples. A causal-model semantics is strictly interventionistic just in case the procedure it gives for evaluating a subjunctive conditional requires making the values of the variables implicated in the antecedent independent from the values of the parents of these antecedent variables. Most causal-model semantic theories that have gained attention are strictly interventionistic.