Freedom As Determination

                                                Freedom As Determination



John goes into a café for lunch. He looks over the menu, carefully considering each option (John is a fastidious eater): he reviews the possible alternatives and weighs up which will please him the most, the price of each, what is most healthy, and what sort of figure he will cutting eating it. After assiduous deliberation he makes his selection, feeling happy that he has made the best choice in the circumstances. He places his order, waits in anticipation, and then proceeds to consume the dish he ordered. He is pleased with the outcome.

            This would appear to be a paradigm case of a free decision: John has considered a range of possible alternatives, reviewed them for desirability, and then selected according to his deepest culinary wishes. It was entirely up to him what he chose; there was no compulsion or coercion from without or within. And yet it has seemed to many philosophers that there is a deep problem here: John’s action was not free, contrary to appearances. For his action was determined—by the laws of physics and the initial conditions and by his overall psychological state. Given this determination John could not have acted otherwise—he would have done the same thing in identical circumstances. Thus human action is not free, and paradigm cases of freedom are no more genuinely free than any other occurrence of nature.

            Some respond by denying determinism, hoping thereby to make room for freedom: either there is indeterminism in physical nature or in the mind. Others contend that on closer examination freedom and determinism are not incompatible, once we understand what freedom actually involves (lack of constraint from outside or inside). But it is agreed on all hands that freedom and determinism are prima facie at odds with each other; it takes work to reconcile them, often ingenious work. The compatibilist is under suspicion of changing the subject in order to save the phenomenon. What I intend argue is that freedom actually entails determinism—that free will makes no sense without determinism. The two are not just compatible but deeply connected: it is a conceptual truth that freedom requires determinism of the most robust kind. Only in a deterministic world is freedom possible. To put it differently, freedom of choice logically requires causal and nomological necessitation by antecedent states of the universe. Specifically, freedom requires strong psychological necessitation: we can be free only if we are bound in our choices by our prior psychological states. In Hume’s terminology, liberty logically depends upon necessity.  [1] This is not merely compatibilism but what I will call determinationism—the “determinationist theory” of free will (it could equally be called necessitationism).  [2]

            Before I lay out this theory I want to consider how freedom might be abrogated in a case like John’s. It is certainly abrogated if someone puts a gun to John’s head and orders him to choose (sic) the salami sandwich: that would be a clear case of external constraint, preventing him from ordering what he most desires (fish and chips). But there are other ways in which his ability to act freely might be thwarted: for instance, he might suffer from a brain complaint that prevents him from envisaging the range of alternatives—he suffers a kind of modal blindness. Whenever he tries to review the available alternatives he finds that nothing comes to mind; he cannot mentally scan through the possibilities, finding himself stuck on a single option. He has lost his sense of alternative futures. Or it may be that he can envisage a range of alternatives but has lost the ability to select one—he suffers from selection impotence. Whenever he tries to select a particular option his mind goes haywire, either selecting none or weirdly skipping to another option than the one he deems best. Or he can select among envisaged alternatives but his action is wildly out of sync with his selection: whenever he tries to order the fish and chips the words “Can I have a salami sandwich?” pop out of his mouth. In all these cases he has lost his ability to act freely, or it has been drastically depleted. We might think of these aberrations as internal constraints interfering with the normal deliberative process; they are freedom-destroyers. They destroy John’s ability to choose and obtain what he most desires from an array of alternatives.

            Let me now state my thesis as baldly as possible. Freedom entails that the agent falls under the law: “People act in accordance with their wishes, unless prevented from doing so”. I intend this law to be analogous to the law of inertia: “Bodies maintain their state of motion, unless imposed upon by an outside force”. That is, it is a law of nature—deterministic and nomologically necessary–that governs how choices are made. I will call this law the “Law of Strongest Desire”, or “LSD”.  [3] The thesis, then, is that free will entails LSD: it is not possible to be free unless you fall under LSD– that is, unless your actions are necessitated by your wishes (desires, inclinations, likes, whims, practical judgments, pro-attitudes, or whatever motivates). That is what freedom is—determination by what motivates. Choice is “wish-fulfillment”. This kind of determinism is not just compatible with freedom; it is what freedom requires and consists in. Free choice just is selecting from among a range of alternatives in accordance with natural law. It is a law of nature that people act according to their strongest desire, and free will depends on the obtaining of that law.

            There are two questions about this thesis: is LSD necessary for freedom, and is it sufficient for freedom? The reason it is necessary is that in cases where it breaks down the agent is not free. Suppose John plumps for his fish and chips after prolonged and careful reflection, coming to the conclusion that this would be the best possible lunch in the circumstances; but then a sudden brain spasm leads him to choose the salami sandwich, which ranks low on his preference list. He may or may not notice the switch, depending on the nature of the brain spasm; but in any case he chooses what he does not most desire. Surely this is not a case of free choice: it is like the gun to the head scenario. He most wants fish and chips but he doesn’t choose in accordance with that wish; he has been subjected to a freedom-destroyer, pushing him in the direction of the unwanted salami sandwich. Or suppose he suffers from a strange psychological quirk whereby his judgments of desirability are regularly subverted according to the rule, “Choose your third-ranked option every time”. He always chooses the third item on his list of desirables without changing his evaluations. He thus acts against his wishes in a predictable way, violating LSD. Again, he is not acting freely: he lacks what an ordinary free agent possesses, viz. the ability to act in accordance with what one most strongly desires. Imagine if his pathology consists in choosing the opposite of what he most desires—choosing, say, discomfort over comfort, and suffering the consequences. We would rightly regard this as a strange compulsion, inimical to freedom, and suggest that John consult a freedom doctor. Therefore falling under LSD is a necessary condition for freedom.

            Is it sufficient? Intuitively, it is—any normal person would agree that this is what freedom comes down to. But the fretting philosopher is apt to sense difficulty: he will ask how LSD is compatible with the principle that a free agent could always have acted otherwise. And it is true that on this view the agent cannot act otherwise given that she is subject to the law of strongest desire—she will always of necessity act in the way that fits what is most strongly desired. She has no alternative given her motivational structure: her desires (etc) determine (uniquely so) her actions. But does the ability to act otherwise really contradict that kind of determinism? No, because it can simply mean, “She could have acted differently had she had different desires (or a different ranking of desires)”. Of course she could have acted otherwise—just as John went into the café the day before and in fact ordered a salami sandwich, because that’s what he felt like having then. There is no logical requirement that agents can act otherwise given their total psychological state—whoever said that that is what freedom involves? It is a bizarre scenario that is being imagined: an agent has exactly the same psychological makeup at two different times but acts differently, intentionally so. What reason could there be for acting thus differently? What went through the agent’s mind that justified the difference of action (and choice)? How would the agent explain herself?

            At this point Buridan’s ass is likely to make an entrance: suspended between two piles of equally delicious hay, it nevertheless picks one pile over the other. But surely that could happen only if the animal had a sudden, if subtle, hankering after (say) the pile on the left—the way it reflected the light at that moment, a general tendency to favor the left, etc. If the two alternatives really were exactly psychologically equivalent, the animal would not be able to make a choice: it would have to manufacture a distinction in order to be triggered into action. But this is not the kind of situation in which we find ourselves when deliberating; we generally find it easy to appreciate distinctions among the options before us, even if it is just a matter of momentarily fancying one thing over another.

            So it is not required for freedom that one be free of one’s own mind, i.e. free to act independently of what is desired (valued, judged best all things considered). Freedom is freedom to act dependently on one’s desires. Thus LSD is sufficient for freedom: there is not some further ingredient to the concept that requires a free action to be one that transcends desire. Suppose an agent regularly flouted his own strongest desires: does that make him or her freer than an agent who always acts in accordance with his desires? Hardly. Freedom does not involve the ability to act against what one desires—except in the trivial and irrelevant sense that freedom often involves the ability to resist certain of one’s desires. If a person judges it morally best to do such and such, contrary to what he desires prudentially, he is not violating LSD, since in this case his moral wishes outweigh his prudential wishes. People do indeed sometimes act from their moral wishes, but that is just one more example of the desire-choice hook-up. What doesn’t happen is that people act from no wish at all, or quite contrary to their wishes–to do so would be to cease to act freely, as in cases of brain spasm. There is no merit in that, and little sense: what would it be, from the inside, knowingly to choose according to what one does not most desire? Does one say to oneself, “I know that I don’t really want to do A all things considered, but would much rather do B, and yet I will deliberately and self-consciously do A”? Even supposing such a state of mind to be possible, why would having it make one freer than someone who simply acted in accordance with what he most desired (or judged best)? Why should ignoring one’s deepest desires enhance one’s freedom, even if it were possible?

            We can all agree that freedom requires the ability to ignore or suppress or regulate some desires, so that we are not pressed into action by our appetitive promptings or baser inclinations; but that is a far cry from the idea that freedom requires total liberation from all desire, even the most elevated, rational, and refined. Does God’s freedom require him to act independently of what he judges best, as if to prove that he is really free? No, God always acts according to LSD: he makes his actions conform to his divine wishes; he doesn’t disconnect the two. Could God have acted otherwise? Well, he could have had different plans for the universe and hence different wishes, but it is absurd to suppose that he could have done differently given his plans and wishes (despite his omnipotence). Are we to think that his actions sometimes mysteriously detach themselves from his omnisciently considered judgments about what he most wishes for his creation?  Could it be, for example, that what he wants from us is virtue but that he sometimes acts so as to promote vice, even though he has no desire that we be vicious?

            Thus LSD is necessary and sufficient for freedom, which means that freedom entails determinism (psychological determinism). It doesn’t entail determinism all the way down—it would be odd if freedom proved that the quantum world is deterministic. What it entails is that desire necessitates choice—to act freely is to act in accordance with one’s desires (considered judgments of desirability). And there is a further point: we know that this is so when we act. That is, we rely on an assumption of determinism when we deliberate—we rely on lawful psychological connections. Most obviously, we are aware of the lawful connection between desires, choices, and actions: we know that desires produce choices that produce actions. If these causal connections were to break down, we would lose the ability to act freely, because there would be no point in deliberating if we could not depend on such lawful causal connections. What point is there in deciding what to do if your decisions are powerless to lead to action of the desired type? What point would there be in deliberating if a review of alternatives could not lead to a mental act of selection? No, we assume that there is a lawful causal web here that justifies predicting what will happen when we deliberate; if that web became fragmented and dissolved, we would be bereft of the ability to act freely, i.e. in accordance with our wishes. There would be no point in John deliberating about his precious lunch if nothing he did mentally reliably led to certain outcomes, ultimately eating what he wants most. Reading the menu has to lead reliably to having an array of options in mind, which has to lead reliably to a considered survey of these options and their pros and cons, which has to lead reliably to a justified selection, which has to lead reliably to placing the right order, which has to lead reliably to getting the lunch he most desires. All this is lawful and causal, so deliberation presupposes determinism at this level; introducing a random and unpredictable element into the process would only destroy or diminish freedom, not generate it.

            This is what I meant by saying that freedom and determinism are not only compatible but conceptually connected. The concept of freedom analytically entails determinism of the kind enshrined in LSD, which is no less deterministic than other laws of nature, such as the law of inertia–hence the determinationist theory of freedom. Our free actions are governed by a law of nature that is as rigid as any law. It is the same with animals: they too obey the law of strongest desire—they act in accordance with what they desire most at the moment of action. They do not demonstrate their freedom by mysteriously flouting their desires, detaching their actions from what they want; they simply follow nature in acting according to their wishes. They could not act otherwise given their wishes, though they could act otherwise with different wishes; but that in no whit detracts from their freedom. Their will is entirely free, as is our will and the will of God—free to enact what they most desire (unless interfered with). Determination by desire does not rob us of free will; it constitutes free will. We should count ourselves lucky that our actions are determined by our desires, or else we would flail in a sea of randomness, or languish in a prison of external constraint. Our desires don’t confine us in determining our actions; they allow us to be free agents constrained by nothing alien and undesired.

            There may be a residual anxiety: if desires necessitate and cause actions, don’t they compel us to act, and isn’t that contrary to the sense we have of our freedom? We don’t feel compulsion from desire, but shouldn’t we if determinism holds? First, we should not model the causality here on clinical cases of compulsion, such as drug addiction or OCD, which are (to some degree) freedom-destroyers. Second, we could follow Hume in claiming that this kind of causation, like all causation, is not perceptible by the mind, so the fact that we don’t feel it doesn’t show it isn’t there. Third, we don’t feel compulsion in other cases of mental causation either, though it is assuredly present, such as the causation of belief by perception, or imagination by recollection, or emotion by thought. These causal connections are lawful and involve causal power, and hence are “compelled”, but we don’t feel this compulsion, as we may feel the compulsion of drug addiction. Our desires necessitate our choices in virtue of LSD, but we don’t need to experience that necessitation—and such an experience is not possible if we follow Hume on causation. Perhaps the lack of an experience of necessity gives us an illusion of causal independence, but we must actually be subject to causality in the recesses of the mind. We are subject to necessitating psychological law even if it doesn’t strike us that we are—and we have to be if we are to function in such a way as to permit freedom. And clearly people are subject to all sorts of psychological laws that they may not suspect subsume them—as that they act from ambition or self-interest or vanity (this is a point of Hume’s). Being subject to LSD is not the same as consciously realizing that it necessarily governs one’s actions at all moments (again consider animals).

            Here is a further point (also made by Hume): we don’t tend to think our freedom is compromised by psychological laws that only hold generally and for the most part, so why should we be disturbed by laws that hold without exception? Being subject to psychological law is not ipso facto an abrogation or diminution of freedom: we don’t feel less than fully free because we are generally motivated by ambition or self-interest or vanity (or indeed by good will and public spiritedness). It is not that we would have our freedom increased by having no lawful tendencies at all, by being (unlike the rest of nature) entirely anomalous. So why should being very lawful imperil our freedom? Freedom has nothing to do with escaping natural law; but it has everything to do with obeying the right kind of law to enable freedom, i.e. the law of strongest desire. This law is what is constitutive of free will, so it can hardly be such as to undermine free will. If I rigidly obeyed the law of always acting rationally and morally, that would have no tendency to show that I was not as free as someone less unwavering. Similarly, if I rigidly obeyed LSD that would have no tendency to show that I was less free than someone who suffered intermittent breakdowns of that law (by brain spasms, etc)–quite the contrary. As Hume says, liberty is not opposed to natural necessity but to external constraint (or internal constraint). To lack freedom is to have our desires imprisoned, not for our desires to imprison us—for how could that conceivably be? Desires are not like external agents (or internal agents) dictating that we should act contrary to them. Freedom is determination by our desires (considered judgments of desirability).

            To defend free will is not to defend moral responsibility. For us to be praiseworthy or blameworthy would involve our desires arising from a process involving the exercise of our free will; but this is seldom if ever the case. Someone subject to a vicious and corrupting childhood, who forms corresponding bad desires, will naturally act in accordance with them, thus acting freely; but we cannot blame such an individual for those desires (she had no say in their formation), and hence we cannot blame her for her actions. When this person performs an immoral act she is acting freely from her own desires—she is not externally (or internally) constrained—but she is not responsible for the action in the moral sense, since she cannot be blamed for the formation of her bad desires (the same goes for good desires, of course). Freedom of action does not imply freedom of desire. Thus the person may not be free though her actions are (since they follow from her desires). The actions of a criminal may be free but not blameworthy.

            I have not said that freedom entails physical determinism, only psychological determinism. So psychological determinism is not incompatible with freedom: but what about physical determinism? There is some feeling that physical determinism is in tension with freedom, because all the movements of the body are uniquely predetermined by antecedent physical conditions, and freedom requires the ability to do otherwise. But this is confused: it doesn’t matter that all my movements are physically predetermined, so that my will has no power to change them to other types of movement; all that matters is that I have the power to move differently in the presence of different desires—which will require that the antecedent physical facts be different. Freedom does not require that someone just like me physically must be able to act differently from me—for that would imply that his action is disconnected from his desires (since he must have the same desires as me in virtue of the supervenience of the mental on the physical). The physical body is not like an outside agent interfering with my autonomy; it is simply the place where my desires naturally nestle. The fact that my body can no more depart from its causal history than a tree can is irrelevant to whether I am free: the question is whether I can act in accordance with my desires. There is an intuitive feeling that physical determinism conflicts with freedom, but there is no demonstration that the two are incompatible; and it is farfetched to suppose that our ordinary concept of freedom requires that physically identical agents should be able to act differently—that the future is not fixed by the past. That is a philosopher’s fancy not part of what our ordinary concept of freedom involves. The onus is on the philosopher to show us why there is any such entailment; and it would be strange if our ordinary concept implied such a surprising theoretical possibility. Rather, what our concept implies is something we know very well, namely that we act in accordance with our wishes; but that has nothing to do with breaches in physical determinism.

            I have argued that freedom entails necessitation and that necessitation does not undermine freedom. We thus cannot rule out freedom by establishing necessitation. Whether we are actually free is a further question, but I can see no reason why not and every reason to think we are. Freedom is acting in accordance with one’s strongest desire (all things considered), and this we do all the time; ergo we are free. My actions are free because I act as I wish to act. To be free is to be subject to law—the law of freedom.  [4]


Colin McGinn

  [1] David Hume, An Enquiry Concerning Human Understanding, Section VIII, “Of Liberty and Necessity”. It was reading Hume that led me to the proposals of this paper.

  [2] Ken Levy drew my attention to an excellent article by R. E. Hobart, published in Mind in 1934, entitled, “Free Will as Involving Determination and Inconceivable Without It”. This article is very much in the spirit of what I argue here (and others have said similar things over the years); if I add anything, it is perhaps a new way of formulating the same basic points.

  [3] There are all sorts of complexities about the notion of “strongest desire” that I don’t want to engage here. All I need for my purposes is the idea that all actions are preceded by some sort of motivational state that causes them to occur (or causes the agent to perform them) and which involves law and necessity. I am claiming that freedom implies determination by motivational states of some sort as opposed to lack of determination by anything in the agent. So, roughly: people do what they desire to do because of their desires—where these operative desires cause and determine. These desires can range from a simple desire for food to the most sophisticated moral evaluations. One’s strongest desire might fall into the latter category: one wants above all to act rightly, so one acts on that desire, thus acting freely. This is quite compatible with having strong appetitive desires that the moral desire overrides. The law of strongest desire is intended to include cases in which moral motivations outweigh other kinds. The question I am concerned with is whether necessitation by one’s motives (prudential, moral, appetitive) undermines one’s freedom.

  [4] Everything in nature is subject to law, even free decisions. So free decisions must fall under laws, and there is no reason the corresponding law should not take the form of LSD. The law of freedom is not something that detractsfrom freedom.

2 replies
  1. Javad
    Javad says:

    Dear Professor McGinn,

    I hope this message finds you well. I have been reading your writings and am curious to inquire about a potential distinction in meaning between “free will” and “freedom” as expressed in your work.

    Thank you for your time and consideration.



    • Colin McGinn
      Colin McGinn says:

      There isn’t much of a distinction, though “freedom” connotes something more political in many contexts. The phrase “free will” is a philosophical coinage and can be misleading.


Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.