¹ Versions of the Alternative Possibilities Argument represent only the most obvious type of objection to compatibilism. For other objections, see, for example, McKenna, Michael and Coates, D. Justin, ‘Compatibilism’, in Zalta, E., (ed.) The Stanford Encyclopedia of Philosophy, 2015Google Scholar, URL =<https://plato.stanford.edu/archives/win2016/entries/compatibilism/>, §2; Vihvelin, Kadri, ‘Arguments for Incompatibilism’, in Zalta, E., (ed.) The Stanford Encyclopedia of Philosophy, 2011Google Scholar, URL = <https://plato.stanford.edu/archives/fall2015/entries/incompatibilism-arguments/>.

² The first of these is a version of a limerick that, according to the Oxford Dictionary of Quotations, was originally composed by Maurice E. Hare. The second limerick is my own contribution to the genre.

³ For the explanation of why I now capitalize the initial letters of the words ‘Determinism’ and ‘Indeterminism’, see the penultimate paragraph of this section.

⁴ For similar characterizations, see Mele, Alfred, Free Will and Luck (New York: Oxford University Press, 2006), 17CrossRefGoogle Scholar; Vihvelin, ‘Arguments for Incompatibilism’, §1; also McKenna and Coates, ‘Compatibilism’, §1.

⁵ Mele, Free Will and Luck, 3. The characterization quoted by Mele is from Inwagen, Peter van, An Essay on Free Will (Oxford: Clarendon Press, 1983), 3Google Scholar.

⁶ ‘Global’ in the sense that it applies to the world as a whole. For parity, I use ‘Determinism’ (with capital ‘D’) for the (global) thesis that I have defined in this section.

⁷ See, for example: McKenna, Michael and Pereboom, Derk, Free Will: A Contemporary Introduction (New York and London: Routledge, 2016), 23–24Google Scholar; Kane, Robert, A Contemporary Introduction to Free Will (Oxford & New York: Oxford University Press, 2005), 8–10Google Scholar; Watson, Gary (ed.) Free Will (Oxford: Oxford University Press, 2^nd edition, 2003), Editor's Introduction, 9–10Google Scholar; Honderich, Ted, How Free Are You? The Determinism Problem (Oxford & New York: Oxford University Press, 1993), 66, 79Google Scholar.

⁸ McKenna and Pereboom, Free Will, 17 (italics in the original).

⁹ Cf. Mele, Alfred, ‘Libertarianism, Compatibilism, and Luck’, Journal of Ethics 19 (2015), 1–21; 14–15CrossRefGoogle Scholar.

¹⁰ Of course, there are those who think that free will is impossible, regardless of whether Determinism is true. But these theorists (‘impossibilists’) should still not think that these types of indeterminism are freedom-prohibiting in a sense of taking away a freedom that would or might otherwise obtain. For the term ‘impossibilism’, and a discussion of some versions of this view, see Vihvelin, ‘Arguments for Incompatibilism’.

¹¹ My argument in this section obviously raises the question why any compatibilists have ever been one-way compatibilists. Yet the literature is replete with references to classical compatibilists who are attributed this position (allegedly holding that free will is not only compatible with, but also requires, Determinism). I cannot pursue this intriguing question here. But one possibility is that at least some of these theorists intended to hold, not that free will requires that the world as a whole be deterministic, but the more restricted thesis that free actions must be deterministically caused.

¹² Cf. Mele, Free Will and Luck, 58. Principle (L) is related to what Robert Kane has called ‘the Indeterministic Condition’ that typical libertarians insist on as a requirement for free action (Kane, Contemporary Introduction, 38).

¹³ For the same reason, I have deliberately used ‘does something other than A’, rather than ‘does not do A’ in (U). It should be noted, though, that ‘doing something other than A’, where doing A is making a decision to do C, does not entail ‘making a decision to do not-C’. It could, for example, include deliberately (or at least voluntarily) postponing making a decision about whether to do C. (For emphasis on this point, see Steward, Helen, ‘The Truth in Compatibilism and the Truth of Libertarianism’, Philosophical Explorations 12 (2009), 167–179CrossRefGoogle Scholar; also Steward, Helen, A Metaphysics for Freedom (Oxford: Oxford University Press, 2012), 154–156CrossRefGoogle Scholar. But see also note 28 below.)

¹⁴ As before, ‘action’ must be interpreted so as to exclude involuntary ‘actions’ such as dropping dead.

¹⁵ Bear in mind that Alan's decision, in my example in §3, might be an action that satisfies (a).

¹⁶ Recall the association between free will and responsibility in my initial characterization of free will (in §1).

¹⁷ The term ‘problem of luck’ is sometimes used in the literature to include what I call ‘the problem of control’ as well as what I call ‘the problem of luck’. For reasons that should become apparent, I prefer to distinguish the problems.

¹⁸ Inwagen, Peter van, Metaphysics (Oxford: Oxford University Press, 1993)Google Scholar.

¹⁹ Ibid., Ch. 11, 192–193 (italics in the original).

²⁰ Cf. van Inwagen, Metaphysics, Ch. 11, 193. Van Inwagen does not, however, use the word ‘control’ in his presentation of the argument, but speaks instead of whether the agent ‘has a choice about’ whether the outcome occurs.

²¹ Cf. van Inwagen, Metaphysics, Ch. 11, 193–194.

²² For discussion of agent causal libertarianism, see Watson, Gary (ed.) Free Will (2^nd edition, 2003)Google Scholar.

²³ Including van Inwagen (Metaphysics, 194). Contemporary philosophers who have attempted to defend libertarianism without appeal to agent causation include Robert Kane. See, for example, his ‘Responsibility, Luck, and Chance: Reflections on Free Will and Indeterminism’, Journal of Philosophy 96 (1999), 217–240Google Scholar.

²⁴ Mele, Free Will and Luck, 9.

²⁵ See Mele, Alfred, ‘Libertarianism, Luck, and Control’, Pacific Philosophical Quarterly 86 (2005), 395–407CrossRefGoogle Scholar, and Clarke, Randolph, ‘Agent Causation and the Problem of Luck’, Pacific Philosophical Quarterly 86 (2005), 408–421CrossRefGoogle Scholar.

²⁶ Concerning the absence of these kinds of randomness at the macroscopic level, see Fischer, John Martin and Ravizza, Mark, S. J., Responsibility and Control: A Theory of Moral Responsibility (Cambridge: Cambridge University Press, 1998), 253–254CrossRefGoogle Scholar. See also Honderich, How Free Are You?, 66.

²⁷ Ayer, A.J., ‘Freedom and Necessity’, in his Philosophical Essays (London: Macmillan, 1954)Google Scholar.

²⁸ If principle (L) can be satisfied merely by its being undetermined which token action of a given type the agent will do, this raises the question whether (L) is, after all, strong enough to capture the openness of alternative possible choices that the typical libertarian has in mind as a requirement of free action. Unfortunately, I cannot pursue this issue here. (See also note 13 above.)

²⁹ Beebee, Helen, Free Will: An Introduction (Palgrave Macmillan, 2013), 129CrossRefGoogle Scholar (italics in the original).

³⁰ Ibid. (italics in the original).

³¹ See, e.g. Kane, ‘Responsibility, Luck, and Chance’.

³² Kane, ‘Responsibility, Luck, and Chance’, 237 (italics in the original). Here I am assuming that the case of Jane exemplifies the class of ‘Self-Forming Actions’ that Kane regards as directly free actions, and whose occurrence is (according to his theory) necessary for any other actions to be free.

³³ Ibid. (italics in the original).

³⁴ For a forceful expression of scepticism about whether this condition is fulfilled, see Beebee, Free Will, 126–127.

³⁵ It has recently been argued that compatibilism faces a ‘problem of luck’ analogous to the traditional problem for libertarians, but which arises even if supposedly free actions are causally determined. (See Levy, Neil, Hard Luck: How Luck Undermines Free Will and Moral Responsibility (Oxford: Oxford University Press, 2011), Ch. 4CrossRefGoogle Scholar; de Calleja, Mirja Pérez, ‘Cross-world Luck at the Time of Decision is a Problem for Compatibilists as Well’, Philosophical Explorations 17 (2014), 112–125CrossRefGoogle Scholar, and, for discussion, Mele, Alfred, ‘Libertarianism, Compatibilism, and Luck’, Journal of Ethics 19 (2015), 1–21.)CrossRefGoogle Scholar It should be noted that this problem is distinct from the problems for compatibilists concerning luck and control that I have discussed here, which concern the implications for (two-way) compatibilism of the concession that free actions may not be causally determined.