Do We Have Free Will?
The conundrum of free will and destiny has always kept me dangling.
—William Shatner
Since we have a physical component, most of us agree that some mental control over the body is necessary if the mind is at all real. Indeed, this ability is the most important of the mind’s functions, if it exists. And if the mind does not have such power, a person could not have true free will—we would not have the genuine power to choose how to act but would instead be completely controlled by nonconscious physical events in our body.1 But does the mind have causal powers, or is free will an illusion? How could freedom be even possible in a material being? Are what appear to be free acts really done by the body without our control and determined solely by a chain of physical and biological forces operating in our body? If we have the personal control of free will, we have genuine choice and agency to at least some degree, and our choices would then have to be taken into account for any complete explanation of our actions. (The classical formulation of free will in terms of “could have done otherwise” ends up causing more problems than clarification.)
So do we have free will or not?
Naturalists who are also structural reductionists reduce the mind to the nonconscious brain, and eliminationists eliminate the mind—either way, free will is eliminated altogether. Thus, mental phenomena are seen as at best powerless epiphenomena. An immaterial mind, even if it existed, could have no causal power. To use Thomas Huxley’s image, a mind could no more cause actions than a train’s steam whistle could cause the train to move. To accept more would be to deny that there can be a complete physical description of physical events. Moreover, a sufficient physical cause precludes the need for any other cause. Physical phenomena are the actual causes at work when we mistakenly think the mind is causing an event. Moreover, free will could not possibly arise even in principle in a deterministic world. Thus, we have no more free will than any other animal, including those animals that we think have consciousness. If determinism is true for all events in our universe, then there is no free will or random events: from the beginning of time (if time has a beginning) it was fixed that I would be typing this sentence right now on this planet and that you would be reading it whenever it is that you read it, and nothing could change these events from occurring. As Sean Carroll puts it, given the quantum state of elementary particles of a person and the environment, the future is fixed by the laws of physics—any true free will would violate everything we know about the laws of nature. If there are some random events at the quantum level or above, the history of the universe would be different, but there is no reason to believe that such randomness could somehow create free will.
To reductionists, only the brain acts, with its nonconscious actions based on only our accumulated past experiences, our abilities, and our environment—no mental decisions are involved. Those experiences fix what occurs next through the determinism of events governing nature. Even if our brain, acting like a computer, can distinguish different reasons and follow what the chain of past experiences dictates to be the best, no free decisions occur. The brain does not “decide”—there is no agent or choice but only mechanical operations. All events are really physical, and all physical events are determined completely by previous and concurrent physical events. The appearances of personal agency and free will is no more real than the appearance that a thermostat freely adjusts the room temperature. Only because we do not know what will happen next do our actions appear to be freely chosen—our brain does not (and perhaps cannot) predict the next outcome of the complex phenomena operating in us. But we are not the author of our actions: nonconscious matter is in control. A string of inanimate events in the brain that we think we initiate by free will are caused merely by neurological events. Actions happen to us, and the feeling of a conscious “will” is only an illusion generated by the brain thereby creating in us a sense of ownership of our actions. (Why evolution created the illusion of ownership is not clear since actions in a deterministic world would occur the same way without a sense of agency or ownership.) It is only because we, unlike everything else in the world, are conscious that we have this illusion.
Since Thomas Hobbes, “compatibilists” within the reductionist camp have tried to put a smiley face on determinism: as long as one is not restrained by outside forces or an inner compulsion caused by mental disorder, there are no compelled actions, and one is thus autonomous and free. Compatibilists think this the only “free will” we have. One is morally responsible for one’s actions as long as one acts only from one’s inner desires that are not caused by a brain disorder. So too, habits or dispositions conditioned by previous actions may determine an action.
Compatibilism is the most popular position in philosophy today, but many philosophers object that there is nothing “free” about our actions in compatibilist “free will.” Compatibilists have formulated several defenses to try to make the position convincing, but none are very strong: what they mean by “free will” simply is not what is normally meant—we are still left with a purely deterministic chain of physical events. For compatibilists, we are still only a biochemical puppet, and when Sam Harris says “a puppet is free as long as he loves his strings,” opponents would respond that he is only playing with words. How does the unconstrained or unhindered operation of a deterministic brain form “free will” in any sense? And if all acts are determined, it is not clear why the absence of any internal and external constraints makes a relevant difference—it is just another type of determination. It may make a significant difference legally but not philosophically: a person’s actions are still totally fixed by nonconscious events—“unconstrained” only means that normal bodily mechanisms determine our actions, not that any free will is involved.
In opposition are the “libertarians” of the philosophical (not political) kind who claim that mental states have at least some causal power over the body. This permits the possibility of true free will: we can choose our course of action—the course is not fixed by physical events. Structural antireductionists can offer an explanation for this: nature has psychological structures that are real and on par with the physical and biological structures that set up minds with causal powers, personal choice, and control. Because mental phenomena have their own causal consequences, they are a nonnegotiable feature of reality and not reducible to products of the physical or biological structures. Making a choice is different from a physical action: it has an “inside” that mechanical actions do not. Mental phenomena are dependent on the physical (both matter and the physical level of organization) as the base-conditions for their appearance, but they are equally real in their own right and can exercise causal power over the body.
To structural antireductionists, the mind causally enters into events (as discussed in chapter 12). This is no more problematic than any other level of causation—it is simply another level of normal structure, and thus of ordinary causation, that nature has produced. Prior to the appearance of beings with self-awareness or a self-reflective mind in the evolution of the cosmos, a determinism may have prevailed (although uncaused events on the subatomic level would present an issue). But once the base-conditions for mental phenomena are fully assembled, such a mind arises, and psychological-level structures operate in implementing our beliefs and values, emotions, intentions, and so forth into actions. Indeed, how free will could evolve in a world that is deterministic before the appearance of self-conscious life if consciousness is not the result of a separate level of organization is not clear.
Thus, the antireductionists’ account accepts the autonomy and agency of a person, genuine choice, and mental causation as components of the overall causal system affecting our actions. Collectively, this permits the possibility of free will. If so, our behavior results from intentionality and agency, and these cannot be accounted for solely by the causal role played by physical forces. We then have at least some degree of choice in our actions, and a complete explanation of our actions would require reference to our choices as causes, not just to the physiological conditions of the event. Under this approach, there also is no causal overdetermination since physical events are not the complete causal account of the course of events. A complete account of the course of brain activity requires a role for beliefs and decisions, even if events are closed on a neural level.2 The mysteries surrounding mental causation are merely a subset of the general mystery of how any level of organization operates in nature. (But that the mental level is involved and is not determined by the physical level does not guarantee that there is free will: a more encompassing determinism including determination of events in the mind may still prevail. Mental structures must enable a freedom of choice for free will to prevail.)
To address whether we have free will or are determined, we must first distinguish causation from determinism—free will requires the former but is incompatible with the latter. If our actions were totally chaotic, we would not have the control permitting acts of will, but causal order permits control by enabling us to predict the outcome of our actions: if actions X and consequences Y in the past routinely occurred together, then we can confidently predict that if we do X, then Y will probably follow. Determinism, however, goes beyond causation: it entails a fixed chain of causes: A (along with some surrounding conditions) causes B, B in turn causes C, C in turn causes D, and so forth. For reductionists, all the causes are physical. In determinism, all events and properties are completely fixed by past causes: if we know all the causes and conditions, once a chain of events starts we can predict its determined end. Thus, causation is only about isolated lawful conditionals, while determinism is about the antecedents of those conditionals and how they lead to a fixed chain of consequences from an initial action. (Calling an isolated action “determined” is slippery: it can mean simply “caused” without loading in all the metaphysics of determinism.) With genuine free will, one can control some of the antecedents in causal chains, picking the antecedents as we go along. Thus, we can affirm the lawfulness of actions and consequences necessary for choice and affirm that every human event has a cause and yet still reject determinism.
Unfortunately, philosophers usually do not distinguish causation and determinism. Instead, they equate the two and thus see the rejection of determinism as having only one alternative: chaos—a randomness of physically uncaused events, like that theorized to be happening on the subatomic level of organization. But with the distinction of causal order from determinism, one can reject determinism and still affirm the order of causation: the alternative to determinism is not necessarily indeterminism but may be the middle ground of causation that permits predictive control. Thus, with causation the resulting actions of a human agent are not random but selected by our decisions and carried out by voluntary actions.
This distinction also permits dismissing a standard refutation of free will. The argument goes: either determinism is true or it is not; if it is true, then all our actions are fixed by prior states, and thus there is no free will; if it is false, then there is only subatomic-like randomness, and so we have no way to control our actions and no way to guide their results through predictions, and thus once again there is no free will. In sum, there is either an unstoppable, uninterrupted chain of physical events and so no free will, or there is no predictable outcome and so no meaningful free choice can be exercised. But if we can predict the short-term outcome of our actions at least to a degree through our knowledge of recurring causal patterns and can also control our actions at least to a degree, then there is a third option—causal control without determinism—and free will is possible.3 In sum, free will does not entail randomness but requires control, and causal lawfulness without determinism provides such control.4
This distinction also renews the possibility of an “agent causation” in which a person can exercise free will in his or her actions without any determinism. Peter van Inwagen believes that free will remains a mystery even if there is agent causation because free will and determinism both obviously exist but are incompatible. But if causation and determinism are differentiated, this particular mystery disappears: there is in fact no determination of human actions, and we can cause actions in an ordered way. Every event still has a cause and is lawful, but mental causes can be in its chain of causes of human actions and so for our actions there is no determinism of inanimate causes. Causal order prevents randomness by giving persons the predictive control needed to exercise free will, and we then supply a cause in the chain of actions. But the question then is: How does the decision of how to act arise? Is there still a gap in the causal chain? Are the decisions determined by something other than physical events? Does some causeless event still occur? Does a thought how to choose just magically appear from our subconscious—just “pop into the mind”?5 And are subconscious events determined by physical events alone? This may remain a mystery, but libertarians accept that in some way a person enters the picture as a cause by freely choosing how to act.6
Thus, the stark contrast: under determinism our actions are a fixed chain of physical events, while under libertarianism we can control at least some causes by selection. Can we ascertain if the events in our brain are deterministic or permit causal control? Neuroscience may be a way. Benjamin Libet conducted experiments in the 1980s that showed that apparently freely chosen acts were in fact initiated by the brain a fraction of a second before the conscious decision to act occurred. These experiments have since been replicated and expanded by others. Determinists have jumped on these results as proving that free will does not exist and that in fact the conscious mind plays no role in the chain of our events—indeed, for them this is the last step necessary for science to remove all aspects of mind from science.
But neuroscientists have not been so quick to reach that conclusion. They have suggested other explanations for the results. For example: that the “readiness potential” (which occurs in the sensory motor-cortex) that these experiments actually measure is unrelated to the decision making (which occurs in the parietal lobe); that the readiness potential actually begins to build up before the choice has to be made in anticipation of having to make a choice when the participants in the experiments are told that they will have to make a decision, and thus it is unlikely to be related to the actual decision of choosing which way to act—that is, the urge to move is unrelated to the decision itself; that it simply takes more time for the conscious mind to register its actions in the brain; that the participants could not report the timing of their acts of will accurately; or that these results apply only to snap judgments rather than complex thought-out decisions that require reasoning, planning, and choosing and thus take much longer. They also note that the predictions are correct only about 60 percent of the time—which is clearly better than a 50/50 guess but not anywhere near certainty. Libet himself still believes in a “robust free will”: he affirms that the conscious mind has veto power over the unconscious originating events—the conscious free will not does initiate acts, but our conscious ability to veto has a control function. Nor, he claims, is there any evidence or even a proposed experimental design that definitively or convincingly demonstrates a physical determinism of human action. In fact, he thinks there is prima facie evidence that conscious mental processes can control some brain processes.
Thus, neuroscience to date has not decided the issue. And there does not appear to be any other empirical way to reach a resolution. We are once again left with metaphysics and mysteries. The determinists’ mystery is why we have the illusion of free will since it cannot be of evolutionary value—the illusion would cause us to believe that we could act freely, but if all actions are determined there is no point in the brain creating that illusion. The libertarians’ mystery is how mental action occurs. Libertarians can readily accept some points made by determinists: that there are a myriad of nonconscious events in the body in any human action; that all our actions are preceded by nonconscious causes and conditions, biases, and influences; that we are much more conditioned by our genetic and social background than most of us realize; and that subconscious processes dictate the options given to the conscious mind. But these points do not mean that our conscious mind cannot then exercise free will control over what is presented to it—they mean only that what options we have are more limited than we usually accept. That is, we cannot control the cards that are dealt us, but we have some freedom in how we play them.
Thus, current neuroscience does not refute free will, and if current theories in physics are correct, not all of reality is deterministic. Subatomic indeterminacy is currently considered ontic—a feature of reality and not merely the result of our cognitive limitations. So science has not answered the Big Question of free will yet, nor is it obvious that it will ever be able to answer it since a test for it is hard to devise. And there is a simple explanation for why there has not been any progress on the matter in philosophy: once again we apparently lack the cognitive apparatus to answer a vital question due to our physiological limitations in how we have evolved, just as we are incapable of answering the Big Question of consciousness.
Given the state of science on this issue, should we remain agnostic about free will? One might conclude that considering what is at stake about our view of what we are that we ought to affirm the obvious despite its problems—that the mind has causal power—over a determinism of brain events. In addition, there is a clash of metaphysics based on conflicting intuitions (free will versus reality being deterministic), and the intuition about determinism has already been damaged by particle physics. Thus, common sense says to affirm a free will. (William James saw the question as a quarrel of unverifiable metaphysics, not science, and chose to reject the pessimism of determinism.) In light of our experiences, we can continue to believe in it even if we cannot explain it. We only have to give up the philosophical demand of an explanation of how free will would work before we decide to affirm it, which is only a consequence of the philosopher’s disease.
But there is an odd twist here: we cannot help but presume to have free will even if we did not want to. It is hard, if not impossible, to give up a sense of the mind’s control and simply let our body do whatever it was conditioned to date to do—we feel we are choosing and acting. (This leads pragmatists following C.S. Peirce to conclude that it makes no difference if we have free will or not since it does not affect our disposition to act.) To put the point ironically: we have no choice but to believe we have a choice. Even determinists admit that we must act as if free will is real. And fatalists still must seem to themselves to choose actions because they do not know in advance which actions are predetermined or the result of fate—if you fell off a boat into the ocean, you would still try to save yourself, no matter what you think about fate and determinism. And there is a further twist: how can one pretend not to be determined without having the actual mental causal power and free will to do that? It is hard to see how those who deny free will could convince themselves into pretending to have free will. It is one thing to feel controlled by physical causes and to consider free will as an illusion as one goes through life—like the illusion of a straw looking bent in a glass of water that we still see even though we know better—but how does one believe that we are puppets on a string and pretend to have free will? And again, if we have that ability to pretend, how can we not have genuine power? In this way, the very ability to pretend that we have free will becomes an argument for the existence of free will.
Thus, at worst we should be agnostic about whether there is genuine free will—it is an open empirical question that we simply cannot answer at present or perhaps ever. And those who remain agnostic about free will on philosophical grounds can go on acting exactly as they must in any case.
1.Lack of control over our actions would also raise the issue of whether we are morally responsible for our actions. Criminal punishment would be only a type of conditioning—a way to adjust the pool of experiences from which the brain derives our next action.
2.The events in the brain may still follow a strictly physical order of causation. (The analogy to the mind as software to the brain’s hardware was noted in the last chapter.) When we walk, our actions are constrained by the law of gravity—indeed, our actions must conform with all physical laws—but gravity does not determine where we walk. So too, if free will is genuine, the course of the neural activity is not determined by the physical laws that govern the activity of the neurons in the brain: the laws governing neural activity no more determine our choices than gravity and the physical laws governing our bodily movements determine what direction we choose to walk in. If free will is real, persons are free even if their brains are not—it is the person who makes decisions freely, not the brain. The brain would still affect the way we operate, but it does not explain all of how we think, desire, and choose.
3.If reality is organized into levels of causation, indeterminacy on the subatomic level is irrelevant to decisions on the everyday level. In any case, science has not shown that subatomic randomness affects a predictable causal order of the everyday world—billiard balls still behave like billiard balls despite what is going on at their quantum level. It also raises the possibility that causation is a power produced only on higher levels.
4.Another standard problem is that if there is an omnipotent god, his omnipotence and our free will are not compatible: if we have free will, then there is something even a god could not control and thus he is not all-powerful; conversely, if God has all the power, we have no control of events. God’s perfect knowledge would also be incompatible with our free will: if God knew from the beginning of time what I am going to have for breakfast tomorrow, do I have any free will in the matter now? If God knows now that I am going to have pancakes, that fact is now set and there is nothing I can do about it. One might respond that he knows only what I am going to freely choose tomorrow. But there is still a problem: one can only know what is true; thus, if God now knows what I am going to have, it must be so now, and thus I now have no freedom to do otherwise tomorrow morning. Some theologians try to get around these problems by making the ad hoc assumption that God somehow withdraws his omnipotence and omniscience in the case of human action.
5.An illustration of this occurred while I was writing this chapter. One Saturday morning I was planning on getting pizza for lunch. Then “out of the blue,” the idea of getting a falafel came to me. At that moment, I was not thinking about lunch or where to go for lunch—the thought “just came to me,” and no conscious decision making or act of will was involved. (I went for the falafel. The question is whether I had the free will to veto that impulse.)
6.Whether this requires a commitment to the metaphysical concept of personhood—a unified agent or center of action—or whether one can accept all the personal properties and capacities without such a commitment remains an issue.