Lawful Humean explanations are not circular

Abstract

A long-standing charge of circularity against regularity accounts of laws has recently seen a surge of renewed interest. The difficulty is that we appeal to laws to explain their worldly instances, but if these laws are descriptions of regularities in the instances then they are explained by those very instances. By the transitivity of explanation, we reach an absurd conclusion: instances of the laws explain themselves. While drawing a distinction between metaphysical and scientific explanations merely modifies the challenge rather than resolving it, I argue that it does point us towards an attractive solution. According to Humeanism, the most prominent form of the regularity view, laws capture information about important patterns in the phenomena. By invoking laws in scientific explanations, Humeans are showing how a given explanandum is subsumed into a more general pattern. Doing so both undermines a principle of transitivity that plays a crucial role in the circularity argument and draws out a central feature of the Humean approach to scientific explanation.

Transitivity and Humeanism about Laws

“…Earlier, Dretske also contested the view that mere generalisations could have any explanatory power over their instances: "Subsuming an instance under a universal generalization has exactly as much explanatory power as deriving Q from P & Q. None" (1977: 26).6 See Lange (2013),Hicks and van Elswyk (2015),Marshall (2015),Miller (2015),Roski (2018),Shumener (2017),Marshall (2015),Dorst (2018),Emery (2019),Bhogal (2020),Hicks (2020),Kovacs (2020) andDuguid (2021).7 However, not everybody endorses that view-according toEmery (2019), it is the other way round: laws ground their instances.…”