Con-reasons and the causal theory of action

Abstract

A con-reason is a reason which plays a role in motivating and explaining an agent's behaviour, but which the agent takes to count against the course of action taken. Most accounts of motivating reasons in the philosophy of action do not allow such things to exist. In this essay, I pursue two aims. First, I argue that, whatever metaphysical story we tell about the relation between motivating reasons and action, con-reasons need to be acknowledged, as they play an explanatory role not played by pro-reasons (the reason the agent takes to count in favour of the action taken). Second, I respond to an argument recently developed by David-Hillel Ruben to the effect that a causal theory of action – still known as ‘the standard story’ – cannot account for con-reasons. His argument attempts to show that a fundamental principle of the causal theory cannot be reconciled with the role con-reasons play in a certain kind of imagined case. I first argue that a causal theorist is not, in fact, committed to the problematic principle; this argument has an added benefit, since the principle has been taken by many to show that the causal theory generates a puzzle about the possibility of weak-willed action. I then argue that a causal theorist has good reason to reject the possibility of Ruben's imagined cases. If successful, my arguments make clearer the commitments of the causal theory and show that it can accommodate con-reasons in the way I think they ought to be accommodated.

Keywords:

• motivating reasons
• causal theory of action
• David-Hillel Ruben
• con-reasons
• akrasia

Acknowledgements

Special thanks to David-Hillel Ruben for helpful comments on earlier drafts, discussion and encouragement. Thanks to Sergio Tenenbaum and an anonymous reviewer for Philosophical Explorations for helpful comments on earlier drafts. Research for this paper was funded by the Social Sciences and Humanities Research Council, Canada.

Notes on contributor

Jonathan D. Payton is a PhD student at the University of Toronto. Between 2013 and 2016, he will hold a Joseph Armand Bombardier Doctoral Scholarship from the Social Science and Humanities Research Council, Canada, which will fund a dissertation on the metaphysics of negative actions.

Notes

1. I have distinguished the two kinds of reason in such a way that a normative reason can still count as a motivating one. If I actually remember that I will be visiting my spouse's parents for the next few days, then nothing I have said rules out that this fact can be a motivating reason. I have thus drawn the distinction so as to be neutral between those who think the categories are mutually exclusive (Smith 1994, chapters 4 & 5) and those who allow normative reasons to be motivating as well (Dancy 2000).

2. I should note one qualification about the causal theory of action. For our purposes, I will focus on bodily actions, that is actions that consist of an agent's body moving in certain ways or being positioned in certain ways. For simplicity, I ignore the possibility of mental action. Similarly, I ignore a causal-volitionist view on which all actions are mental acts of deciding or willing, which cause bodily movements. Presumably, Ruben's arguments apply to mental actions just as well as to bodily ones.

3. See (Ruben 2009, 2010).

4. I assume that Ruben's arguments can be transposed so as to be more directly relevant to a causal theory on which rationalizing explanation is causal explanation, but reasons are not literally causes.

5. See, for instance, (Mele 1983, 2003) and (Bishop 1989).

6. Mele actually develops his notion of motivational strength to argue against certain elements of Davidson's view of practical reasoning and motivation as presented in “How is Weakness of Will Possible?” (Mele 1998, 29–31).

7. See (Bishop 1989), (Davidson 2001), and (Mele 1983, 2003).

8. It has been suggested to me that if we accept con-reasons, then it might be possible to act akratically by acting on a con-reason. That is, can I Φ on the grounds that p, when I take p to count against Φ-ing. Perhaps the obvious example of an agent Φ-ing for a reason they see as counting against Φ-ing is Milton's Satan: perhaps when he declares ‘Evil, be thou my good’, he does not see the evilness of Φ-ing as genuinely counting in favour of it, but he does nonetheless Φ on the grounds that doing so would be evil. If Satan really does preserve his judgment that Φ-ing would be evil, and that all things considered he should not Φ, then he satisfies the general conception of akrasia. On the other hand, it is hard to make sense of his declaration on these grounds; on the contrary, he seems to think that, all things considered, he should be evil rather than good. Unfortunately, I cannot deal with the possibility of this sort of akratic action here, at least not in the detail it deserves.

9. Ruben actually raises a second difficulty for the causalist in his essays: not only must a causalist find something for my con-reasons to cause, but she must also say that they cause the action they recommend, which is precisely what is ruled out in Type II cases. I will not deal with this difficulty here; since it is also motivated by (D) and the possibility of Type II cases, it requires no separate treatment.

10. See, for instance, Mele (1983, 345) and Mele (2003, 76–77).

11. For discussion, see Schueler (1995, 28–34).

12. Tenenbaum (2007, 1, n.2) accurately cites “How is Weakness of the Will Possible?” as evidence of how popular this thesis was at the time of its writing. It is important, when reading Davidson's essay, to remember just how widespread the view was in comparison with the causal theory of action, which was only beginning to see a revival in the wake of “Actions, Reasons and Causes.”

13. Ruben actually considers this objection, but his response seems to be not to answer it directly, but to shift the burden of his argument to cases in which the agent does not undergo any conscious, explicit deliberation at all (2010, 174).

14. Even this claim can be doubted. Even if both pile A and pile B are within the ass's visual field, you might think that in order to see that one is much better than the other, the ass must shift his attention from one to the other. If shifts of attention like this take time, then the ass will go for pile A more quickly in the counterfactual scenario where pile B is absent and no shifts of attention are necessary.