Oct 13 JDN 2460597

Introduction

What freedom do we have to choose some actions over others, and how are we responsible for what we do? Without some kind of freedom and responsibility, morality becomes meaningless—what does it matter what we ought to do if what we will do is completely inevitable? Morality becomes a trivial exercise, trying to imagine fanciful worlds in which things were not only other than they are, but other than they ever could be.

Many people think that science and morality are incompatible precisely because science requires determinism—the causal unity of the universe, wherein all effects have causes and all systems obey conservation laws. This seems to limit our capacity for freedom, since all our actions are determined by physical causes, and could (in principle) be predicted far in advance from the state of the universe around us. In fact, quantum mechanics isn’t necessarily deterministic (though in my preferred version, the Bohm interpretation, it is), but a small amount of randomness at the level of atoms and molecules doesn’t seem to add much in the way of human freedom.

The fear is that determinism undermines human agency; if we are part of a closed causal system, how can we be free to make our own choices? In fact, this is a mistake. Determinism isn’t the right question to be asking at all. There are really four possibilities to consider:

• Acausalism: Actions are uncaused but inevitable; everything is ultimately random and meaningless.
• Libertarianism: Actions are uncaused and free; we are the masters of our own destiny, independent of the laws of nature.
• Fatalism: Actions are caused and inevitable; the universe is a clockwork machine of which we are components.
• Compatibilism: Actions are caused but free; we are rational participants in the universe’s causal mechanism.

Acausalism

Hardly anyone holds to acausalism, but it is a logically coherent position. Perhaps the universe is ultimately random, meaningless—our actions are done neither by the laws of nature nor by our own wills, but simply by the random flutterings of molecular motion. In such a universe, we are not ultimately responsible for our actions, but nor can we stop ourselves from pretending that we are, for everything we think, say, and do is determined only by the roll of the dice. This is a hopeless, terrifying approach to reality, and it would drive one to suicide but for the fact that if it is true, suicide, just like everything else, must ultimately be decided by chance.

Libertarianism

Most people, if asked—including evolutionary biologists—seem to believe something like libertarianism. (This is metaphysical libertarianism, the claim that free will is real and intrinsically uncaused; it is not to be confused with political Libertarianism.) As human beings we have an intuitive sense that we are not like the rest of the universe. Leaves fall, but people climb; everything decays, but we construct. If this is right, then morality is unproblematic: Moral rules apply to agents with this sort of deep free will, and not to other things.

But libertarian free will runs into serious metaphysical problems. If I am infected by a virus, do I choose to become sick? If I am left without food, do I choose to starve? If I am hit by a car, do I choose to be injured? Anyone can see that this is not the case: No one chooses these things—they happen, as a result of the laws of nature—physics, chemistry, biology.

Yet, so much of our lives is determined by these kinds of events: How can Stephen Hawking be said to have chosen life as a physicist and not a basketball player when he spent his whole adult life crippled by amytropic lateral sclerosis? He could not possibly have been a professional basketball player, no matter how badly he might have desired to be. Perhaps he could have been an artist or a philosopher—but still, his options were severely limited by his biology.

Indeed, it is worse than this, for we do not choose our parents, our culture, our genes; yet all of these things strongly influence who we are. I have myopia and migraines not because I wanted to, not because I did something to cause it to happen, but because I was born this way—and while myopia isn’t a serious problem with eyeglasses, migraines have adversely affected my life in many ways, and while treatment has helped me enormously, a full cure remains elusive. Culture influences us even more: It is entirely beyond my control that I speak English and live in an upper-middle-class American family; though I’m fairly happy with this result, I was never given a choice in the matter. All of these things have influenced what schools I’ve attended, what friends I’ve made, even what ideas I have considered. My brain itself is a physical system bound to the determinism of the universe. Therefore, in what sense can anything I do be considered free?

Fatalism

This reasoning leads quickly to fatalism, the notion that because everything we do is controlled by laws of nature, nothing we do is free, and we cannot rightly be held responsible for any of our actions. If this is true, then we still can’t stop ourselves from acting the way we do. People who murder will murder, people who punish murderers will punish murderers—it’s all inevitable. There may be slightly more hope in fatalism than acausalism, since it suggests that everything we do is done in some sense for a purpose, if not any purpose we would recognize or understand. Still, the thought that death and suffering, larceny and rape, starvation and genocide, are in all instances inevitable—this is the sort of idea that will keep a thoughtful person awake at night.

By way of reconciling determinism with libertarian free will, some thinkers (such as Michael Shermer) have suggested that free will is a “useful fiction”.

But the very concept of anything being useful depends upon at least a minimal degree of free will—the ability to choose actions based upon their usefulness. A fiction can only be useful if beliefs affect actions. If there even is such a thing as a “useful fiction” (I’m quite dubious of the notion), free will is certainly not an example, for in order for anything to ever be useful we must have at least some degree of free will. The best one could say under fatalism would be something like “some people happen to believe in free will and can’t change that”; but that doesn’t make free will true, it just makes many people incorrigibly wrong.

Yet the inference to fatalism is not, itself, inevitable; it doesn’t follow from the fact that much or even most of what we do is beyond our control that all we do is beyond our control. Indeed, it makes intuitive sense to say that we are in control of certain things—what we eat, what we say, how we move our bodies. We feel at least that we are in control of these things, and we can operate quite effectively on this presumption.

On the other hand, different levels of analysis yield different results. At the level of the brain, at the level of biochemistry, and especially at the level of quantum physics, there is little difference between what we choose to do and what merely happens to us. In a powerful enough microscope, being hit by a car and punching someone in the face look the same: It’s all protons and electrons interacting by exchanging photons.

Compatibilism

But free will is not inherently opposed to causality. In order to exercise free will, we must act not from chance, but from character; someone whose actions are random is not choosing freely, and conversely someone can freely choose to be completely predictable. It can be rational to choose some degree of randomness, but it cannot be rational to choose total randomness. As John Baer convincingly argues, at least some degree of causal determinacy is necessary for free will—hence, libertarianism is not viable, and a lack of determinism would lead only to acausalism. In the face of this knowledge, compatibilism is the obvious choice.

One thing that humans do that only a few other things do—some animals, perhaps computers if we’re generous—is reason; we consider alternatives and select the one we consider best. When water flows down a hill, it never imagines doing otherwise. When asteroids collide, they don’t consider other options. Yet we humans behave quite differently; we consider possibilities, reflect on our desires, seek to choose the best option. This process we call volition, and it is central to our experience of choice and freedom.

Another thing we do that other things don’t—except animals again, but definitely not computers this time—is feel emotion; we love and hurt, feel joy and sorrow. It is our emotions that motivate our actions, give them purpose. Water flowing downhill not only doesn’t choose to do so, it doesn’t care whether it does so. Sometimes things happen to us that we do not choose, but we always care.

This is what I mean when I say “free will”: experiences, beliefs, and actions are part of the same causal system. What we are affects what we think, what we think affects what we do. What we do affects what we are, and the system feeds back into itself. From this realization I can make sense of claims that people are good and bad, that acts are right and wrong; and without it I don’t think we could make sense of anything at all.

It’s not that we have some magical soul that lives outside our bodies; we are our bodies. Our brains are our souls. (I call this the Basic Fact of Cognitive Science: We are our brains.) Nor is it that neuron firings somehow “make” our thoughts and feelings as some kind of extra bonus; the patterns of neuron firings and the information that they process are our thoughts and feelings. Free will isn’t some mystical dualism; it is a direct consequence of the fact that we have capacities for conscious volition. Yes, our actions can be ultimately explained by the patterns in our brains. Of course they can! The patterns in our brains comprise our personalities, our beliefs, our memories, our desires.

Yes, the software of human consciousness is implemented on the hardware of the human brain. Why should we have expected something different? Whatever stuff makes consciousness, it is still stuff, and it obeys the laws that stuff obeys. We can imagine that we might be made of invisible fairy dust, but if that were so, then invisible fairy dust would need to be a real phenomenon and hence obey physical laws like the conservation of energy. Cognition is not opposed to physics; it is a subset of physics. Just as a computer obeys Turing’s laws if you program it but also Newton’s laws if you throw it, so humans are both mental and physical beings.

In fact, the intuitive psychology of free will is among the most powerfully and precisely predictive scientific theories ever devised, right alongside Darwinian evolution and quantum physics.

Consider the following experiment, conducted about twenty years ago. In November of 2006, I planned a road trip with several of my friends from our home in Ann Arbor to the Secular Student Alliance conference in Boston that was coming in April 2007. Months in advance, we researched hotels, we registered for the conference, we planned out how much we would need to spend. When the time came, we gathered in my car and drove the 1300 kilometers to the conference. Now, stop and think for a moment: How did I know, in November 2006, that in April 2007, on a particular date and time, E.O. Wilson would be in a particular room and so would I? Because that’s what the schedule said. Consider for a moment these two extremely complicated extended bodies in space, each interacting with thousands of other such bodies continuously; no physicist could possibly have gathered enough data to predict six months in advance that the two bodies would each travel hundreds of kilometers over the Earth’s surface in order to meet within 10 meters of one another, remain there for roughly an hour, and then split apart and henceforth remain hundreds of kilometers apart. Yet our simple intuitive psychology could, and did, make just that prediction correctly. Of course in the face of incomplete data, no theory is perfect, and the prediction could have been wrong. Indeed because Boston is exceedingly difficult to navigate (we got lost), the prediction that I and Steven Pinker would be in the same room at the same time the previous evening turned out not to be accurate. But even this is something that intuitive psychology could have taken into account better than any other scientific theory we have. Neither quantum physics nor stoichiometric chemistry nor evolutionary biology could have predicted that we’d get lost, nor recommend that if we ever return to Boston we should bring a smartphone with a GPS uplink; yet intuitive psychology can.

Moreover, intuitive psychology explicitly depends upon rational volition. If you had thought that I didn’t want to go to the conference, or that I was mistaken about the conference’s location, then you would have predicted that I would not occupy that spatial location at that time; and had these indeed been the case, that prediction would have been completely accurate. And yet, these predictions insist upon such entities as desires (wanting to go) and beliefs (being mistaken) that eliminativists, behaviorists, and epiphenomenalists have been insisting for years are pseudoscientific. Quite the opposite is the case: Eliminativism, behaviorism, and epiphenomenalism are pseudosciences.

Understanding the constituent parts of a process does not make the process an illusion. Rain did not stop falling when we developed mathematical models of meteorology. Fire did not stop being hot when we formalized statistical dynamics. Thunder did not stop being loud when we explained the wave properties of sound. Advances in computer technology have now helped us realize how real information processing can occur in systems made of physical parts that obey physical laws; it isn’t too great a stretch to think that human minds operate on similar principles. Just as the pattern of electrical firings in my computer really is Windows, the pattern of electrochemical firings in my brain really is my consciousness.

There is a kind of naive theology called “God of the gaps”; it rests upon the notion that whenever a phenomenon cannot be explained by science, this leaves room for God as an explanation. This theology is widely rejected by philosophers, because it implies that whenever science advances, religion must retreat. Libertarianism and fatalism rest upon the presumption of something quite similar, what I would call “free will of the gaps”. As cognitive science advances, we will discover more and more about the causation of human mental states; if this is enough to make us doubt free will, then “free will” was just another name for ignorance of cognitive science. I defend a much deeper sense of free will than this, one that is not at all threatened by scientific advancement.

Yes, our actions are caused—caused by what we think about the world! We are responsible for what we do not because it lacks causation, but because it has causation, specifically causation in our own beliefs, desires, and intentions. These beliefs, desires, and intentions are themselves implemented upon physical hardware, and we don’t fully understand how this implementation operates; but nonetheless the hardware is real and the phenomena are real, at least as real as such things as rocks, rivers, clouds, trees, dogs, and televisions, all of which are also complex functional ensembles of many smaller, simpler parts.

Conclusion

Libertarianism is largely discredited; we don’t have the mystical sort of free will that allows us to act outside of causal laws. But this doesn’t mean that we must accept fatalism; compatibilism is the answer. We have discovered many surprising things about cognitive science, and we will surely need to discover many more; but the fundamental truth of rational volition remains untarnished.

We know, to a high degree of certainty, that human beings are capable of volitional action. I contend that this is all the freedom we need—perhaps even all we could ever have. When a comet collides with Jupiter, and we ask “Why?”, the only sensible answer involves happenstance and laws of physics. When a leaf falls from a tree, and we ask “Why?”, we can do better, talking about evolutionary adaptations in the phylogenetic history of trees. But when a human being robs a bank, starts a war, feeds a child, or writes a book, and we ask “Why?”, we can move away from simple causes and talk about reasons—desires, intentions, beliefs; reasons, unlike mere causes, can make more or less sense, be more or less justified.

Psychological and neurological experiments have shown that volition is more complicated than we usually think—it can be strongly affected by situational factors, and it has more to do with inhibiting and selecting actions than with generating them, what Sukhvinder Obhi and Patrick Haggard call “not free will but free won’t”; yet still we have volitional control over many of our actions, and hence responsibility for them. In simple tasks, there is brain activity that predicts our behavior several seconds before we actually consciously experience the decision—but this is hardly surprising, since the brain needs to use processing power to actually generate a decision. Deliberation requires processing, not all of which can be conscious. It’s a little surprising that the activity can predict the decision in advance of the conscious experience of volition, but it can’t predict the decision perfectly, even in very simple tasks. (And in true real-life tasks, like choosing a college or a spouse, it basically can’t predict at all.) This shows that the conscious volition is doing something—perhaps inhibiting undesired behaviors or selecting desired ones. No compatibilist needs to be committed to the claim that subconscious urges have nothing to do with our decisions—since at least Freud that kind of free will has been clearly discredited.

Indeed, evolutionary psychology would be hard-pressed to explain an illusion of free will that isn’t free will. It simply doesn’t make sense for conscious volition to evolve unless it does something that affects our behavior in some way. Illusions are a waste of brain matter, which in turn is a waste of metabolic energy. (The idea that we would want to have free will in order to feel like life is worth living is profoundly silly: If our beliefs didn’t affect our behavior, our survival would be unrelated to whether we thought life was worth living!) You can make excuses and say that conscious experience is just an epiphenomenon upon neurological processes—an effect but not a cause—but there is no such thing as an “epiphenomenon” in physics as we know it. The smoke of a flame can smother that flame; the sound of a train is a sonic pressure wave that shakes the metal of the track. Anything that moves has energy, and energy is conserved. Epiphenomenalism would require new laws of physics, by which consciousness can be created ex nihilo, a new entity that requires no energy to make and “just happens” whenever certain matter is arranged in the right way.

Windows is not an “epiphenomenon” upon the electrons running through my computer’s processor core; the functional arrangement of those electrons is Windows—it implements Windows. I don’t see why we can’t say the same thing about my consciousness—that it is a software implementation by the computational hardware of my brain. Epiphenomenalists will often insist that they are being tough-minded scientists accepting the difficult facts while the rest of us are being silly and mystical; but they are talking about mysterious new physics and I’m talking about software-hardware interaction—so really, who is being mystical here?

In the future it may be possible to predict people’s behavior relatively accurately based on their brain activity—but so what? This only goes to show that the brain is the source of our decisions, which is precisely what compatibilism says. One can easily predict that rain will fall from clouds of a certain composition; but rain still falls from clouds. The fact that I can sometimes predict your behavior doesn’t make your behavior any less volitional; it only makes me a better psychologist (and for that matter a more functional human being). Moreover, detailed predictions of long-term behaviors will probably always remain impossible, due to the deep computational complexity involved. (If it were simple to predict who you’d marry, why would your brain expend so much effort working on the problem?)

For all these reasons, I say: Yes, we do have free will.