Causal Necessity

The general sentiment in contemporary metaphysics is that there is a very strong case to be had against the necessity of causal connections (‘CASE’, for short). Indeed, even causal realists tend to accept that causation— although a real and substantial connection—falls short of being necessary (Schrenk 2009; Mumford & Anjum 2011,2018). To my mind CASE consists largely of a collection of straw man arguments that only address a conception of causation—which I will call ‘relational realism’—that to my knowledge has never been put forward in defence of the necessity of causal connections. Indeed, I don’t know of any arguments that address the standard view, as I have presented it. In this chapter I will first show that the main arguments now comprising CASE cut no ice against the standard view, and then argue that the argument developed by Mumford & Anjum, which does prima facie appear to address a view much like my own, ultimately does not work.

I will not attempt to address every argument ever put forward against causal necessity, but enough to cover the main bulk of CASE. As a consequence, I can’t claim to completely vindicate causal necessity, but only reveal how overestimated the strength of the arguments against it have been. The main bulk of CASE consists of five main objections: (i) Hume’s discussion of causation (1739, 1748), which I have already addressed at length in Chapter 2, (ii) Anscombe’s criticism of a logically/conceptually necessary connection, (iii) Russell’s (1912) problem of action at a temporal distance, (iv) the family of counterarguments sometimes denoted collectively as the problem of interference and prevention, which has been used to criticise both the kind of conditional account of causation that J.S. Mill (1843) defended and more recent powers-based conceptions of causation (Mumford & Anjum 2011), and (v) the objection that even if everything in the actual world happens in accordance with causal laws or in virtue of dispositional essences, causation would still fail to be necessary because the laws of nature and/or the dispositional essences are themselves contingent (Armstrong 1983; Lange 2004).

I have already discussed Hume’s argument in Chapter 2, arguing that it really only argues that there is no logical/conceptual necessary connection between impressions and/or ideas. The question therefore is whether anyone has argued that this is what causal necessity is, which is what (ii) is all about. I won’t address Hume’s own argument any further. I will also ignore (v) because its rebuttal requires a more serious discussion of the possible worlds metaphysics than is possible here. I will however go some way to criticise its applicability to explain causation in Chapter 9, but that discussion is not really geared to the problem of causal necessity, but the more general question of whether possible worlds metaphysics can offer any account at all of causation. That I don’t discuss (v) has no serious impact on the discussion of the other objections, since they don’t address views that claim validity beyond the actual world.