[This was written for the Ayn Rand Institute's essay contest for graduate students, 1995. --mh]

The Objectivist Theory of Free Will

by Michael Huemer

Imagine we are at a murder trial. Randy Smith is accused of killing his Aunt Millie. The defense admits that on the night of the murder, Smith had an argument with his Aunt, that he took a pistol out of his jacket and shot her. She died of the gunshot wound. Smith knew that the gun was loaded, that Millie was directly in front of it, and that he was pulling the trigger. He was not insane at the time, there were no abnormal chemicals in his brain, and he was not acting in self-defense. He killed her knowingly, intentionally, and unjustifiably.

Nevertheless, Smith maintains, he cannot be held responsible for his action, because, in the strongest sense, he could not help it. It was, he says, physically impossible for him to avoid shooting his Aunt. He argues:

Physics teaches us that all physical changes transpire in accordance with the laws of nature. Now my firing of the gun, along with my aunt's ensuing death, were physical events. So, if the dictates of science are to be accepted, these events were ultimately the outcome of events occurring in (say) 2 million B.C., together with the laws of nature. But it is not up to me what went on 2 million years ago. And it is not up to me what the laws of nature are either. Therefore, the consequences of these things, including my present actions, are not up to me either.(1)

Is this argument valid? If it is, parallel reasoning also applies to every human action, whether for good or ill. If so, then literally no one can control anything.

In the past two centuries of philosophy, three responses to this issue have predominated. First, there are the 'hard determinists', who agree that since determinism is true, free will is an illusion. These include, notably, Baruch Spinoza, who argues that there is no such thing as free will because "all things have been predetermined by God, not from his free will or absolute pleasure, but from the absolute nature of God"(2); and Arthur Schopenhauer, who argues more secularly directly from the law of identity:

[E]very thing-in-being must be something, must have a definite nature. It cannot exist and yet be nothing, it cannot be something like the ens metaphysicum, that is, a thing which simply is and no more than is, without any definitions and properties, and consequently, without a definite way of acting which flows from them. . . . But all this is just as true of man and his will as of all other beings in nature. . . . Freedom of the will, when carefully analyzed, means an existence without an essence, which means that something is and at the same time is nothing, which in turn means is not, and consequently is a self-contradiction.(3)

Those who have rejected free will have generally done so because they assumed, as Schopenhauer demonstrates, that free will requires indeterminism -- i.e., that man act without cause -- and have considered this idea irrational.

Their position is hard to believe, however. Throughout his life, every normal person deliberates, imputes responsibility, and recommends or proscribes courses of action. In each of these activities, he presupposes that alternatives are available to himself and others -- for it would make no sense to deliberate over what one had no choice about, or to recommend to a person what he either could not do or could not help doing, etc. So a second school of philosophers have held that, since man certainly does have freedom -- that is, he often has multiple alternatives available to him -- indeterminism must be true. Mankind, according to this view, constitute the sole exception to the law of causality.(4)

This view has its own problems. If my actions are not caused, then it seems they must be mere random, inexplicable happenings. A third line of philosophers, urging that this cannot be the meaning of freedom, have claimed that free will is, after all, compatible with determinism -- perhaps even entails determinism. They have sought to explain how this is possible by means of definitions of "freedom" that remove the apparent conflict. Usually they say that a person is 'free' if there are no external obstacles hindering him; if he is able to do what he wills; or if, had he tried to act differently, he would have succeeded -- none of these definitions imply the man's action is uncaused. Locke, for example, urges "that freedom consists in the dependence of the existence, or not existence of any action, upon our volition of it"(5), while Hobbes writes, "LIBERTY, or FREEDOM, signifieth, properly, the absence of opposition; by opposition, I mean external impediments to motion," later concluding, "Liberty, and necessity are consistent."(6) Thus, the answer to the argument I put forth at the outset would be that Smith was free not to shoot his Aunt, because, if he had tried not to shoot her, then he wouldn't have shot her. This latter fact remains the case even if determinism is true.

But this last position grants us freedom only in a Pickwickian sense. It allows that Smith could have done otherwise than he did if he had tried to, while granting that, actually, he could not have tried to, and that, just as Smith maintains, given his circumstances only one course of action was possible.

Now the Objectivist theory of free will maintains the freedom of the will together with the law of causality, without resorting to the sort of redefinition of "freedom" that Locke, Hobbes, and others have found necessary. Objectivism affirms our freedom in the strong sense, in which "A course of thought or action is 'free,' if it is selected from two or more courses possible under the circumstances."(7) Unlike the previous views, however, the Objectivist theory does not assume this means our free actions are uncaused. Rather, when one performs a free action, the action is caused (generally, by one's values and factual beliefs), but other actions are still possible, because it is up to one which possible causes are operative in oneself.

Consider again the case of Smith, the murderer. The three traditional theories of freedom concur that, in order for Smith to have multiple courses of action possible to him given his circumstances, his action would have to be uncaused -- they agree, that is, that causality implies determinism. But according to Objectivism, Smith's action was caused (say) by his anger at his aunt; however, it remained possible for Smith not to shoot his aunt, because he could have not been so controlled by his anger. Many other people who get angry do not kill anybody; they have cultivated rational character traits, which they act on. Smith, too, could have chosen to act from more reasonable motives, and then he would not have killed Millie. For example, he could have thought about the consequences of shooting Millie, realized that these would be harmful, and acted accordingly. Thus, he could have chosen different causes for his actions.(8)

But why did Smith act on the motives he did, instead of more rational ones? Was there also some cause of this choice? At this point, it becomes evident that there must be either an infinite regress of choices, or else some primary choice, which gives rise to the others -- a choice about which one cannot ask further, why the person made that choice. Here Objectivism's second unique feature appears, that of identifying the primary choice as the choice to focus one's consciousness. Whereas previous writers have generally focused on man's physical actions as the locus of freedom, Rand recognizes that such external actions presuppose some previous mental activity. In order to act, one must first identify possible courses of action, how to carry them out, and what one wants to achieve. Hence, physical action presupposes awareness. Rand calls the state in which one's mind is alert and prepared to acquire information, "focus". Focus is a precondition on awareness. Furthermore, she recognizes that this state is achieved only through specific mental effort. Hence, the primary choice, without which other choices are impossible, must be the choice to focus one's consciousness. Without such a choice, one would be unaware of the possibilities of action.(9) This makes the Objectivist theory of free will uniquely epistemological, in the sense that it identifies the volitional nature of awareness as the source of man's free will.

Finally, Objectivism makes two observations about the validation of the theory of free will. Rand does not attempt to give a positive proof that our wills are free. Rather, in the first place, she observes that the fact of free will is available to introspection. Each of us can observe that he can focus his consciousness, or relax it. We can pay attention, or not. It would be out of place to ask for a proof of this fact, in the same way that it would be out of place to ask for a proof that trees exist, if you are standing in front of one, looking at it -- not because the fact is unknowable, but because it is known directly, rather than needing to be derived from something else.

Second, Rand argues that it is not possible consistently to deny that one has free will. Every human choice and every evaluation presupposes it. One cannot deliberate about something, unless one thinks it is within one's power to do it or not do it; one also can not say that something 'should' or 'shouldn't' be done, unless it is possible for it to be done or not be done. Consequently, if one is deliberating about whether to believe in free will or not, then one is already committed to its existence. Nor can the determinist tell us that we should accept determinism. Nor can he claim that he is advocating determinism because it is true -- since on his view, he is advocating determinism only because some blind factors beyond his control force him to advocate it. Thus, the determinist's position appears to devolve into incoherence, as soon as he tries to assert it.(10) This is not, strictly, a proof of the freedom of the will, however.(11) What it shows is that, in order to argue about free will (even to deny it), one has to already implicitly know that one has it; therefore, one must have learned it by some means other than argument -- in particular, Rand holds, one learns it by direct observation.


1. Cf. Peter van Inwagen's argument in An Essay on Free Will p. 222.

2. Ethics, appendix. See also his proposition 32.

3. From his "Essay on the Freedom of the Will".

4. See, for example, Duns Scotus, Thomas Reid, and Immanuel Kant for this view.

5. Essay Concerning Human Understanding, Book II, chapter xxi, section 27; italics Locke's.

6. Leviathan, chapter XXI.

7. Peikoff, Leonard. Objectivism, the Philosophy of Ayn Rand (hereafter: OPAR) (New York: Penguin, 1991), p. 55.

8. OPAR, pp. 65-7.

9. OPAR, pp. 55-60.

10. OPAR, pp. 69-72.

11. Aristotle calls this kind of argument "negative demonstration", as opposed to demonstration proper. See Metaphysics IV.4, where he argues that a person cannot deny the law of non-contradiction without implying its truth.