Revisiting McKay and Johnson's counterexample to (β)

ABSTRACT

In debates concerning the consequence argument, it has long been claimed that [McKay, T. J., and D. Johnson. 1996. “A Reconsideration of an Argument Against Compatibilism.” Philosophical Topics 24 (2): 113–122] demonstrated the invalidity of rule (β). Here, I argue that their result is not as robust as we might like to think. First, I argue that McKay and Johnson's counterexample is successful if one adopts a certain interpretation of ‘no choice about’ and if one is willing to deny the conditional excluded middle principle. In order to make this point I demonstrate that (β) is valid on Stalnaker's theory of counterfactuals. This result is important and should not be neglected, I argue, because there is a particular line of objection to the revised formulations of the consequence argument that does not succeed against the original version.

KEYWORDS:

• Incompatibilism
• the consequence argument
• counterfactuals
• laws of nature

Acknowledgments

I would like to express gratitude to two anonymous reviewers from this journal who provided invaluable feedback on this paper. I am grateful to the reviewers from another journal who have made a number of criticisms to the original manuscript resulting in a much improved article. I am also grateful to Fabio Lampert, Julio de Rizzo, Nancy Cartwright, Matthew Tugby, Carolina Sartorio and Helen Beebee.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 As we now know, $P_0$ needs to be a proposition about the distant past (Campbell 2007). I am grateful to an anonymous referee for drawing my attention to this.

2 Perhaps the only exception is Blum (2000). Even so, he argues for the paradoxical conclusion that ‘N ought to be, and yet ought to fail to be, agglomerative’ (Blum 2000, 286).

3 Of course, Stalnaker (1981) would demur. On his approach might-counterfactuals are analysed as an epistemic possibility claim at wide scope to a subjunctive conditional, such as ${\diamond }_e(P\text{□}\to Q)$, where ${\diamond }_e$ stands for epistemic possibility. Lewis objects to this view (Lewis 1973, 80), but see DeRose (1994) for an intricate response.

4 Another proposal is that ♦ has wide scope, so that the logical form of the might-counterfactual is $\mathrm{♢}(P\text{□}\to Q)$. This is the view of Stalnaker (1981) and DeRose (1991), but since they both accept the conditional excluded middle, the consequence argument could be ran with N on their approach.

5 The dominant view as to ‘might’ is that it expresses some sort of epistemic possibility, something (not quite but) along these lines: ‘might P’ is true (as uttered by x) iff P is compatible with what x knows (Williams 2010, 655); De Rose's view (1991) is that ‘might P’ is only true when we do not know nor could in relevant ways find out anything incompatible with P. Interestingly, if anything along these lines is correct, it does not matter what the scope of ♦ is, for ML and M$P_0$ will be too difficult for the incompatibilist to defend. For instance, ML could be objected by simply pointing out that $\sim L$ is consistent with everything we know or that the conditional ‘If I were to do otherwise, $\sim L$’ is consistent with everything we know. This seems quite plausible because surely the incompatibilist will not want to be committed to the claim that ‘might $\sim L$’ is false according to the epistemic ‘might’.

6 Vihvelin (2013) is another influential compatibilist who denies NL. But she denies NL because she thinks – incorrectly in my view – that Lewis' theory of counterfactuals is enough to deny NL, regardless of whether or not we accept a Humean view of the laws (Vihvelin 2013, 163). I agree that if Lewis' theory of counterfactuals is correct, then NL is false. But his theory clearly hinges on a Humean view of the laws. One distinctive feature of the governing theories is that the laws are counterfactually resilient, that is, the laws of nature would still be true in counterfactual suppositions logically consistent with them. This ‘Nomic Preservation’ principle or something like it is accepted by mainstream governing views of the laws and is inconsistent with the denial of NL. See, for instance, Lange (Goodman 1983; Carroll 1994; Lange 2000; Maudlin 2007; Roberts 2008; Lange 2009).

7 Carroll (1994) offers another version of that argument, which is interestingly responded by Beebee (2000).

8 See, for instance, Dorst (forthcoming) with respect to the predictive BSA and (Demarest 2017) with respect to the potency BSA.

9 What is more, Tugby (2016) interestingly argues that it is far from clear how dispositional essentialists can explain that our world is regular rather than chaotic from moment to moment. An ideal knower could know a great deal of regularities up to now, but then if our world starts to be chaotic, the axioms of our best system will be altogether different from those regularities.

10 I regret that I do not know whether systematising views can or should account for the counterfactual resilience of the broad past; plausibly there is no broad past on systematising views. Should the broad past be fixed in the impoverished worlds scenario? If it is a law that all massive particles attract each other in that world, and the broad past is fixed, then the counterfactual ‘if there were two particles, they would attract each other’ is a counterpossible; but then it would also be true that ‘if there were two particles, they would not attract each other’. But it does not seem Demarest would count them as counterpossibles. Perhaps what is really a threat to free will on a view such as Demarest is not the deterministic laws, but the causal powers of the fundamental properties.

11 That is, provided we assume the counterfactual sufficiency interpretation of the N operator, which may be properly challenged. See, for instance, Lampert and Merlussi (2021a, 2021b).

12 Moreover, as an anonymous reviewer pointed out, we are also able to preserve the similarities between the Mind argument and the consequence argument. For example, if we interpret the relevant counterfactuals in terms of Stalnaker's theory, we could run the Mind argument with the N operator. This might be useful if someone wants to defend the view that free will is metaphysically impossible (because Np would be incompatible with determinism and indeterminism) or perhaps van Inwagen's mysterianism. My worry, however, is that the counterfactual interpretation of N may fail to properly capture the control condition related to free will. See, for instance, Lampert and Merlussi (2021a) for arguments against the counterfactual sufficiency interpretation of N. In fact, the arguments from Lampert and Merlussi, if successful, would affect any common formulation of the consequence argument using either of $\left(\alpha \right)$, $\left(\text{β}\right)$, or $(\text{β}-2)$, regardless of whether counterfactuals are interpreted with a Lewisian or an alternative, Stalnakerian semantics.

Additional information

Funding

Pedro Merlussi's research is funded by Sao Paulo Research Foundation – FAPESP – [grant number 2017/20532-8], with additional support from the LATAM Bridges in the Epistemology of Religion [grant number 61253] funded by the John Templeton Foundation.

Notes on contributors

Pedro Merlussi

Pedro Merlussi is a postdoctoral researcher at the Centre for Logic, Epistemology and the History of Science of the University of Campinas, Brazil. His work focuses on the metaphysical problem of free will and determinism, laws of nature and counterfactuals.