The Counterfactual NESS Definition of Causation

Abstract

Beckers & Vennekens recently proposed a definition of actual causation that is based on certain plausible principles, thereby allowing the debate on causation to shift away from its heavy focus on examples towards a more systematic analysis. This paper contributes to that analysis in two ways. First, I show that their definition is in fact a formalization of Wright’s famous NESS definition of causation combined with a counterfactual difference-making condition. This means that their definition integrates two highly influential approaches to causation that are claimed to stand in opposition to each other. Second, I modify their definition to offer a substantial improvement: I weaken their difference-making condition in such a way that it avoids their problematic analysis of cases of preemption. The resulting Counterfactual NESS definition of causation forms a natural compromise between counterfactual approaches and the NESS approach.

Action Languages Based Actual Causality in Decision Making Contexts

“…This test states that 'an act was a cause of an injury if and only if, but for the act, the injury would not have occurred' [36]. 'In the context of structural equations, this flawed account can be described as equating causation with counterfactual dependence' [6]. Given that overdetermination cases are not just hypothetical and rare cases (cases of pollution, suicide, economic loss, .…”

“…This assumption states that an event C causes a second event E if and only if, both events occur, and that, given an intervention allowing to fix the occurrence of a certain set of other events in the context-without being constrained to respect the physical coherence of the world-there is a context where if the first event had not occurred, the second would not have occurred either. This assumption is described by Beckers [6] using SEF notation as:…”

“…Interventionism-that Beckers' CNESS [6] and Beckers and Vennekens' BV [7] definitions reject-introduces non factual elements to the causal inquiry which appear problematic even for the author [19]: if I fix BH [Billy hits] to zero here, I am sort of violating the way the world works. [...] I am contemplating counterfactuals are inconsistent with the equations but I seem to need to do that in order to get things to work out right.…”

“…The actual causation definition we propose is an action languages suitable formalisation of Wright's NESS test. Even if accepted by influential counterfactual theories of causation authors as embodying our basic intuition of causationsuch as Pearl [6]-criticism of the use of logic as formalism has prevented the popularisation of this test. What is argued is the inadequacy of logical sufficiency and logical necessity to formalise these intuitions.…”

“…What is argued is the inadequacy of logical sufficiency and logical necessity to formalise these intuitions. Recent works have shown that rejecting the formalism is not a reason to reject the idea behind it by successfully formalising the NESS test in causal calculus [9] and in the structural equations framework [6]. It is conceivable to work on a way of compiling existing action languages problems and translating them into SEF.…”