1 As I shall suggest below, the likelihood of success that ordinary ability requires varies with certain characteristics of the agent and of the action to be performed. It will be important to remember that what the required likelihood is in any particular instance is determined ultimately by the ordinary concept of ability we actually use, not by what philosophers say.

2 I assume that overdetermination is absent. I discuss the bearing of overdetermination on free action elsewhere. See my doctoral dissertation Freedom and Criticism: An Account of Free Action (Princeton University 1984), Ch. 4.

3 This characterization raises questions which there is no point in trying to answer here: for instance, how often must one possess an ordinary ability in order to have the corresponding standing ability? What are the constraints, if any, on the periods of time which attributions of standing ability implicitly invoke? What is the point of having this concept as well as that of ordinary ability?

4 Harry Frankfurt shares this interpretation of the traditional approach to the problem of freedom. He writes, ‘a dominant role in nearly all recent inquiries into the freewill problem has been played by a principle which I shall call “the principle of alternate possibilities.” This principle states that a person is morally responsible for what he has done only if he could have done otherwise.’ ‘Practically no one … seems inclined to deny or even to question that the principle of alternate possibilities (construed in some way or other) is true.’ See ‘Alternate Possibilities and Moral Responsibility,’ Journal of Philosophy 66 (1969), 829.

Frankfurt, Wright Neely, and Gary Watson (among others) have tried to abandon the orthodox approach by suggesting that relations between higher-order wants or value judgments and actions are what is crucial for free action, not an agent's ability to do otherwise. See Frankfurt's ‘Freedom of the Will and the Concept of a Person,’ Journal of Philosophy 68 (1971), 5-20; Neely's ‘Freedom and Desire,’ Philosophical Review 83 (1974), 32-54; and Watson's ‘Free Agency,’ Journal of Philosophy 72 (1975), 205-20.

5 An Essay Concerning Human Understanding, ed. Peter H. Nidditch (Oxford: Oxford University Press 1975), 237

6 An Inquiry Concerning Human Understanding, ed. Charles Hendel (Indianapolis, IN: Bobbs-Merrill 1955), 104

7 Ethics (London: Oxford University Press 1966), 105-6

8 ‘Freedom and Action,’ reprinted in Keith Lehrer, ed., Freedom and Determinism (New York, NY: Random House 1966), 12

9 ‘The Incompatibility of Free Will and Determinism,’ Philosophical Studies 27 (1975), 188

10 Since I am particularly interested in ordinary ability, where I use ‘ability’ I will mean ‘ordinary ability,’ except where context clearly indicates otherwise.

11 I note below that cases of psychological compulsion illustrate that having the ordinary ability to do otherwise is not enough to make one a free agent.

12 This does not presuppose that a causal theory of action-explanation is correct. To explain my movements my decision must be sufficient for their occurrence, given various background conditions. But this does not necessarily mean that my decision causes those movements to occur, even if such an account turns out to be correct.

13 I am not claiming that proponents of these accounts saw themselves as offering theses about ordinary ability. The modern debate between compatibilists and incompatibilists has been plagued with confusion about what sort of ability is at issue. I am only trying to assess familiar compatibilist and incompatibilist positions in terms of my understanding of what sort of ability is relevant to free action.

14 See Moore, 110.

15 Remember that one may intend to throw a bull's eye and subsequently succeed in throwing one without intentionally hitting the bull's eye and without having the ordinary ability to hit bull's eyes. Myra has the ordinary ability to throw bull's eyes because she throws them intentionally, not simply by luck.

16 I offer an explanation for these variations in the occasional failure phenomenon in section IV.

17 My attack on conditional analyses follows lines similar to one of J.L. Austin's objections to certain of those accounts, though my objection was not inspired by his. See ‘Ifs and Cans,’ reprinted in Bernard Berofsky, ed., Free Will and Determinism (New York, NY: Harper & Row 1966), 315, 318-9.

18 See the following for representative presentations of this objection: Chisholm, 15-16; C.A. Campbell, ‘Is “Freewill” a Pseudo-Problem?’ reprinted in Berofsky, 124-5; C. D. Broad, ‘Determinism, Indeterminism and Libertarianism,’ reprinted in Berofsky, 141. For a compatibilist who accepts this objection see Donald Davidson, ‘Freedom to Act,’ reprinted in his Essays on Actions and Events (Oxford: Oxford University Press 1980), 68.

19 I will be neither surprised nor worried by disagreement over how and where to draw the boundary between conditions constituting the opportunity to do something and conditions constituting the ability to do it. As far as I can tell, nothing in my argument here depends on that issue. (Cf. note 22 below.)

20 As much of the remaining discussion will focus on opportunity, one possible confusion can be avoided by distinguishing a categorical from a hypothetical sense of opportunity. If certain conditions present one person with the opportunity to do something in the categorical sense, then the presence of those conditions would afford any person with the categorical opportunity to perform that action. Conditions of categorical opportunity are not essentially indexed to the abilities of particular persons. Conditions of hypothetical opportunity are. Due to differences in ability, two persons may be in qualitatively identical circumstances though one has the hypothetical opportunity to perform an action which the other does not have the hypothetical opportunity to perform. So, for instance, persons usually have the categorical opportunity to open a good lock only if they are given the key to it. A professional crook, however, may only need a bobby pin to open it. The crook can have the opportunity in the hypothetical sense to open the lock without having the categorical opportunity to open it. I take opportunity in the categorical sense here, though my remarks about it can without difficulty be adapted to the hypothetical notion of opportunity as well.

21 Incompatibilists have no grounds for categorically denying this. Bad aim is not suspect as a causal agent, as choice and other acts or states of will are sometimes alleged to be.

22 I have not given a noncircular account of either ordinary ability or opportunity. One of the central features of ability that I have emphasized makes mention of opportunity; and now we see that one requirement on conditions of opportunity refers to ability. This does not especially worry me as I have not set out to present a complete account of either ability or opportunity. In fact I am not sure what significance such an account could have.

23 It is possible to read Austin as embracing this position. See Austin, 308, fn. 9.

24 I refer again to the innovations Frankfurt, Neely, and Watson have presented. See also my Freedom and Criticism.

25 I would like to thank T.M. Scanion, Bas van Fraassen, anonymous referees for the Canadian journal of Philosophy, and numerous members of discussion groups at Princeton University, Franklin and Marshall College, and Virginia Polytechnic Institute and State University for their helpful comments on earlier versions of this paper.