Explanations of and in Worlds

Abstract Scientific explanations should have a character that matches their subject matter: if a fact is not apt for causal explanation, it should not be explained causally. In this chapter, the author first reviews some cases where causal explanation has been extended beyond its proper limits, identifying a pattern of error termed the problem of causal overreach. The discussion then relates these considerations to the debate over causal locality in Everettian (many-worlds) quantum mechanics (EQM), in the context of three metaphysical distinctions central to recent discussions of locality: event causation vs. worldshaping causation, local branching vs. global branching, and overlapping worlds vs. diverging worlds. The author argues that worlds in EQM are not properly regarded as themselves subject to causal explanation; instead, causation should be understood as a wholly in-world matter. This approach avoids any nonlocal causal explanation; quantum entanglement instead supports a new category of noncausal worldshaping explanations. The chapter concludes by considering the implications of worldbound causation for principles of causal inference, in particular for the Causal Markov condition, and recommends the use of quantum causal models as a foundation for in-world causal inference.