Abstract
Defenders of compatibilism occupy one of two camps: those who think that free will requires the ability to do otherwise, and those who deny this. Those compatibilists who think that free will requires the ability to do otherwise are interested in defending a reading of ‘can’ such that one can do otherwise even if determinism is true. By contrast, those compatibilists who think that free will does not require the ability to do otherwise tend to join incompatibilists in denying that determinism is compatible with the ability to do otherwise. The primary goal of this paper is to fight for the first camp and against the second camp. The secondary goal is to draw greater attention to a relatively overlooked aspect of the debate about alternative possibilities: that giving up on alternative possibilities might mean giving up on much of morality.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Frankfurt (1969).
Interest in compatibilist readings of ‘can’ are arguably on the upswing, as evidenced by those mentioned in footnote three, above. See also Haji (2012).
These principles and how they interact with determinism have also been explored by Copp (2008), Fischer (2003), Graham (2011a, 2011b) and Yaffe (1999), though they focus on the connections between OUGHT/REASON and moral responsibility. Haji (2012) discusses the connection between OUGHT/REASON and a variety of other normative properties and CAN, but is explicitly noncommittal about CAN.
It is compatible with these arguments that one can have reasons for what one actually does. Haji (2009) argues for the more ambitious claim that alternative possibilities are needed for any reasons at all, including for actions one is determined to do.
An incompatibilist can more easily accept that determinism threatens more than just moral responsibility, and so might allow that if determinism is true then the claims constituting FC are true. But even hard incompatibilist Pereboom (1995) denies FC, and insofar as he is committed to denying FC, he is just a much a target of the coming arguments as are compatibilists.
I have claimed that if no one ever has most normative reason to do otherwise then no one would be morally blameworthy for not having done otherwise. Many compatibilists endorse this view, e.g. Widerker (1991), Copp (1997, 2008), and Fischer and Ravizza (1998, 2003). I do not know whether Haji would endorse this view as well. In Haji (2012, 168), he denies that being blameworthy entails one has done something morally wrong, but this is distinct from denying that being blameworthy entails one has done something for which one had overall normative reason not to do (since, even on Haji’s view, the set of moral reasons an agent has are only a subset of all the normative reasons an agent has). However, so long as Haji rejects the claims constituting FC, which he does, the coming arguments apply to him. Further, one need not accept the principle as grounds for rejecting FC; the principle that ‘If no one ever has most normative reason to do otherwise then no one would be morally blameworthy for not having done otherwise’ is neither a premise in the arguments that have as their conclusions the constituents of FC, nor is it a premise in the arguments to come.
A reviewer for this journal has objected to this line on two fronts. First, the reviewer thinks it objectionable that I allow for cases of moral blameworthiness when no moral wrong has been done. I do not see a problem here. Moral blameworthiness need not be directed at morally wrong actions; a lazy agent who never does anything with his life except watch reality TV is a candidate for the sort of blame we call “moral blame,” and while he may have breached prudential obligations, or simply failed to do what he has most prudential reason to do, he has not breached any moral obligations. In fact, there can also be a case in which one is morally blameworthy even though one has acted on a moral reason. Our lazy agent finally decides to get a job and is standing in the unemployment line. While there, a member of a charitable organization asks if he would help, right now, for an hour or so, and the agent agrees. This is meant to be a case in which he has some moral reason to help but overwhelming non-moral (prudential) normative reason to stay in line and continue on his path to making something of himself (and we can stipulate that helping in this case is not a step down that path). I think this agent is morally blameworthy; we could say to him, “Why did you do that?! It was nice of you…but c’mon! You were finally making progress in sorting out your life and you abandoned it on a whim! A morally admirable whim, sure, but a whim nonetheless that you should have resisted!” And he would be more blameworthy were this the fifth time it has happened.
Second, the reviewer has pointed out that one might think moral blameworthiness is not appropriate when one fails to do what one has most normative reason to do, but rather only when one fails to do what one believes one has most normative reason to do. If that were true, the reviewer explains, then a compatibilist might deny OUGHT/REASON without any qualms, since blameworthiness would be a function of what the agent believes she has most normative reason to do, not of what the agent has most normative reason to do. This is, as the reviewer noted, highly controversial, and I think it is false. Failing to do what one believes one has most normative reason to do is a case of weakness of will, and it would be very surprising to think that the only thing one is ever morally blameworthy for is having a weak will. Standardly, one is blameworthy for breaking that promise, injuring that innocent person, failing to manifest sufficient respect for her, and so on. Similarly, when one has done something wrong, it is not one’s believing it would be the breaking of a promise or one’s believing it will injure the innocent person that makes it wrong, but rather its being the breaking of a promise, the injuring of an innocent person, etc. that makes it wrong [see Dancy (2000), especially chapter 3, for persuasive arguments for views along these lines]. A view according to which blameworthiness attaches to a disconnect between beliefs about the weight of normative reasons and action, rather than a disconnect between what the agent’s actual normative reasons are and what the agent does, strikes me as deeply implausible (though I do not pretend to have dealt fully with that issue here). I say more about this in the discussion of coherence and veridical deliberation, below.
More precisely, the argument to come immediately in the next section only requires that a compatibilist think one of the claims constituting FC is false.
Strictly speaking, this is not the logical contradiction of CAN. It does, however, demonstrate the principled compatibility of determinism with the ability to do otherwise, which is just what the compatibilist needs.
Premise (5) is of the form P → ~(P → ~Q), from which we can derive premise (6), P → Q.
For the claim that it is unfair to blame someone for failing to do something he cannot do, see Wallace (1994, 154–194), and for a recent defense of the claim that one can be blameworthy despite acting out of compulsion, see Graham (2011a, b). For the thought that reasons must be capable of being pieces of advice, see Williams (1989), and for an argument against this claim see Schroeder (2007).
I. I assume that reasons are true propositions, and that true propositions are facts. These are not uncontroversial claims, but these metaphysical disputes do not affect the arguments to come.
II. For other defenses of OUGHT and REASON, see Vrnas (2007) and Streumer (2007). Schroeder (2007), 53 explicitly endorses REASON, apparently for the same reasons he takes OUGHT to be true, as does Haji (2012).
Streumer (2007) offers an argument similar to Flap and Fly, calling his argument “The Argument from Crazy Reasons.” While Streumer and I are both in the business of arguing that REASON is true, I am unconvinced by the argument from crazy reasons, or rather, am unconvinced that a denier of Reason should be convinced by his arguments. A denier of Reason may, in the face of his argument, deny the intuition that the reasons he offers are in fact crazy, insisting that one can have a reason to do what it is impossible to do, and in some cases that reason can be the strongest of all one’s reasons. What Streumer does not realize, in my estimation, is that a denier of Reason may say in such cases that one has most reason to do what one cannot do, in which case one should do what one has second-most reason to do, and if one cannot do what one has most second-most reason to do, one should do what one has third-most reason to do, and so on. Such a reply completely evades Streumer’s argument from crazy reasons. My intuitions align with Streumer’s, of course, but if we are to take the denier of Reason and her intuitions seriously, as I attempt below, then we must offer arguments that are not mere re-articulations of Reason-affirmers’ intuitions.
A reviewer has suggested that explanations for why the fact that ‘flapping and flying would result in escape’ does not constitute a reason might differ based on whether one is an existence internalist or externalist about practical reasons. Existence internalism is the view that a necessary condition for F to be a reason for A to ϕ is that ϕ-ing is somehow or other related to some motivational fact of A, e.g. that it would serve (one of) A’s desires, that A could be motivated to ϕ after (rational) deliberation, etc. The suggestion is then that an internalist would offer an explanation for why Agent lacks a reason to flap and fly that an existence externalist would not, viz. that A is not motivated to ϕ and cannot be gotten to be motivated to ϕ after rational deliberation. But I think this is misguided for three reasons. First, while an existence internalist claims that one explanation for why a particular agent lacks a particular reason is that acting on that reason is not appropriately related to A’s motivations, she need not claim it is always the explanation. An existence internalist may affirm REASON and OUGHT (as existence internalist Schroeder (2007) does), and so allow that even if the necessary condition for having a reason to ϕ as specified by the existence internalist is met, there is another necessary condition that might not be met that would explain why an agent lacks a reason. Second, it is not clear that ‘A cannot be motivated ϕ’ is a distinct explanation for why A lacks a reason to ϕ from ‘A cannot ϕ’. It seems the former is simply a further specification of the latter, viz. that is why A cannot ϕ. Lastly, we can simply stipulate in the case of Agent that Agent has a desire to ϕ. An account of desire as, roughly, a disposition to action, is perhaps the most widely accepted account of desire. On that view, A’s having a desire to ϕ just is A’s being disposed to ϕ in such and such circumstances, even if those circumstances never obtain. Thus, whether one has a desire is a function of what we say about certain counterfactual conditions. I can desire to eat cake even though no cake is before me since, were some cake before me, I would eat it (all else being equal). In the case of Agent, we can simply stipulate that Agent has a desire to flap and fly away. That is because, were certain counterfactual conditions to obtain, Agent would flap and fly away. (We may imagine in some counterfactual condition that God offers Agent the temporary ability to flap and fly, and stipulate that Agent would take God up on that offer). But since, ex hypothesi, Agent has a desire to flap and fly, her not being motivated to flap and fly cannot be the explanation for why flapping and flying does not count as a reason; the distinction between existence internalism/externalism is thus rendered irrelevant.
,I do not mean to imply that a practical deliberator must take into account all her reasons in a given instance of practical deliberation. As a reviewer for this journal has said, “I have a chronic reason to take an expensive trip around the world to see spectacular sights that I have not yet seen, and yet it hardly seems to be the case that I am in error in not taking this fact into account in my daily deliberations.” This is surely right. What we should add is that a practical deliberator should take into account those reasons that are relevant to the context at hand. That is, a practical deliberator is in error not merely if she fails to perceive all the reasons she has at a given moment, but those reasons that are relevant to the context. In deciding what to do for vacation one would do well as a practical deliberator to see that there is reason to go see spectacular sights, but this is not relevant in the context of trying to escape from a pursuer. Similarly, if one is trying to escape from a pursuer, then the reasons relevant to escape should be taken into account. In fact, one fails as a practical deliberator when one considers reasons that are irrelevant to the context or task at hand.
Schroeder (2007) offers an example in which one likes surprise parties and ‘that there is a surprise party in the other room’ is a reason for one to go in the room, but one cannot act for that reason since, according to Schroeder, that would undo it as a reason. One might apply the same reasoning to whether the deliberator is in error for not taking the reason into consideration. But, even if that is successful (see footnote 22 for more on this), that only shows there are special cases in which a deliberator is not in error, not that there is no systematic connection between the goals of a practical deliberator as such and recognizing the reasons one has. Further, if Schroeder’s argument is successful, we can reformulate the premise by adding the qualification, “…so long as acting for that reason that one does not take into consideration would not undermine what makes that fact a reason.”.
This is a line of thought on which a reviewer at this journal has helpfully pushed me.
It might be thought that Fischer should claim that Agent’s deliberations are not only coherence legitimate the first time around, but also veridical legitimate. This move, however, is simply question-begging, since what is at issue is whether the fact Agent takes into consideration is really a reason or if Agent is just mistaken about this. Since facts that are considered reasons in deliberations that are veridical legitimate are, by the definition of ‘veridical legitimate’, actually reasons, it would be question-begging to declare that Agent’s deliberations are veridical legitimate from the start.
A reviewer has suggested that the source of the inability matters, such that inability can disable a fact from being a reason in one case (e.g. in Flap and Fly) but may not in other cases. But if inability disables in some cases and not in others, then an explanation is needed as to why sometimes inability disables and sometimes it doesn’t. Absent such an explanation, we have strong reason (sufficient, in my view) for concluding that the source of an inability does not matter to REASON/OUGHT. So one who thinks the source of an inability matters faces an argumentative dilemma: either point out the flaw with Flap and Fly or offer a plausible explanation as to why the source of an inability makes a difference to REASON/OUGHT.
A reviewer has helpfully pointed out that what I say here is incompatible with premise vi* of “No Goods” in Section II, and so I think “No Goods” is unsound. I have included “No Goods,” though, because many philosophers accept vi* (including both those who endorse a desire-based conception of value, e.g. Schroeder (2007), and a non-desire-based conception of value, e.g. Scanlon’s (1998) buck-passing account of value), and showing that those who endorse vi* face a problem I have articulated is, I think, philosophically worthwhile. Further, it is worth noting that the rejection of “No Goods” does not undermine the core argument of the paper, since the core argument only requires one of the arguments in Section II to be sound, not all of them.
Much of the same reasoning can be applied to Schroeder’s surprise party case (see footnote 18, above). That there is a surprise party in the other room shows that it would be good if one were to walk in there unaware that there is a party, but it does not follow that one has a reason to go in there.
A reviewer has pointed out that it follows from my arguments that in a Frankfurt Type Case, Jones is not morally responsible for shooting Smith since, on my view, Black makes it the case that Jones cannot not shoot Smith, REASON/OUGHT are true, and one cannot be morally responsible for an act unless one has reason/ought to do otherwise. Thus, I think FTCs do not successfully demonstrate that the principle of alternative possibilities is false. However, if my arguments are sound, this is of no consequence as regards the goal of the compatibilists. For the compatibilists goal is to vindicate moral responsibility, and FTCs are advanced in an attempt to show that moral responsibility is compatible with determinism even though determinism is not compatible with alternative possibilities. But if determinism is compatible with alternative possibilities, as I have argued, then we do not need FTCs, or any other argument, to show the falsity of PAP, since the falsity of PAP is not needed to vindicate moral responsibility.
For more on this, see footnote 10, above.
References
Clarke, R. (2008). Dispositions, abilities to act, and free will: The new dispositionalism. Mind, 118, 323–351.
Copp, D. (1997). Defending the principle of alternate possibilities: Blameworthiness and moral responsibility. Noûs, 31(4), 441–456.
Copp, D. (2008). ‘Ought’ implies ‘can’ and the derivation of the principle of alternate possibilities. Analysis, 68(1), 67–75.
Dancy, J. (2000). Practical reality. Oxford: Oxford University Press.
Fara, M. (2008). Masked abilities and compatibilism. Mind, 117(468), 843–865.
Fischer, J. M. (1994). The metaphysics of free will: A study of control. Oxford: Blackwell.
Fischer, J. M. (2003). ‘Ought-implies-can’, causal determinism, and moral responsibility. Analysis, 63(3), 244–250.
Fischer, J. M., & Ravizza, M. (1998). Responsibility and control: A theory of moral responsibility. Cambridge: Cambridge University Press.
Frankfurt, H. (1969). Alternate possibilities and moral responsibility. Journal of Philosophy, 66: 829–39. (Reprinted in Fischer, Ed., 1986 and Frankfurt 1987).
Ginet, C. (1990). On action. Cambridge: Cambridge University Press.
Ginet, C. (1966). Might we have no choice? In K. Lehrer (Ed.), Freedom and determinism (pp. 87–104), New York: Random House.
Graham, P. (2011a). Fischer on Blameworthiness and ‘Ought’ implies ‘Can’. Social Theory and Practice, 37(1), 63–80.
Graham, P. (2011b). ‘Ought’ and ability. Philosophical Review, 120, 337–382.
Haji, I. (2009). Freedom and practical reason. Ethical Theory and Moral Practice: An International Forum, 12(2), 169–179.
Haji, I. (2010). Intrinsic value, alternative possibilities, and reason. Journal of Ethics, 14(2), 149–171.
Haji, I. (2012). Reason’s debt to freedom: Normative appraisals, reasons, and free will. Oxford: Oxford University Press.
Kane, R. (1996). The significance of free will. Oxford: Oxford University Press.
Nelkin, D. (2011). Making sense of freedom and responsibility. Oxford: Oxford University Press.
Pereboom, D. (1995). Determinism Al Dente. Nous, 29(1), 21–45.
Scanlon, T. (1998). What we owe to each other. Cambridge: Harvard University Press).
Schroeder, M. (2007). Slaves of the passions. Oxford: Oxford University Press.
Streumer, B. (2007). Reasons and impossibility. Philosophical Studies, 136, 351–384.
van Inwagen, P. (1983). An essay on free will. Oxford: Oxford University Press.
Vihvelin, K. (2004). Free will demystified: A dispositional account. Philosophical Topics, 32, 427–450.
Vrnas, P. (2007). I ought, therefore i can. Philosophical Studies, 136(2), 167–216.
Wallace, R. Jay. (1994). Responsibility and the moral sentiments. Cambridge: Harvard University Press.
Warfield, T. A. (2000). Causal determinism and human freedom are incompatible: A new argument for incompatibilism. Philosophical Perspectives, 14(s14), 167–180.
Whittle, A. (2010). Dispositional abilities. Philosopher’s Imprint, 10(12), 1–23.
Widerker, D. (1991). Frankfurt on ‘ought implies can’ and alternative possibilities. Analysis, 51(1991), 222–224.
Williams, B. (1989). Internal reasons and the obscurity of blame. Cambridge: Cambridge University Press. (Reprinted in Making sense of humanity, and other philosophical papers, pp. 35–45, 1995).
Wolf, S. (1990). Freedom with reason. Oxford: Oxford University Press.
Yaffe, G. (1999). ‘Ought’ implies ‘can’ and the principle of alternate possibilities. Analysis, 59(3), 218–222.
Acknowledgments
I am particularly grateful to David Palmer for his insightful feedback on a variety of drafts of this paper. I would also like to thank the reviewers for this journal, both of whom took the time to read the paper carefully and offer helpful comments and criticisms. More thanks are owed to those who gave feedback on early versions of this paper, including John Martin Fischer, David Copp, Gideon Yaffe, Ish Haji, Jacob Klein, Alex Grzankowksi, and Eric Vogelstein.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Blackman, R. Why Compatibilists Need Alternative Possibilities. Erkenn 81, 529–544 (2016). https://doi.org/10.1007/s10670-015-9753-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10670-015-9753-y