In Defense of Compatibilism: A Response to Edwards and Coyne
William Edwards frames the quarrel as an argument between thinkers who believe in free will
I was excited to read William Edwards’s article “The Academic Quarrel Over Determinism.” As a philosophy professor, I teach my students about that quarrel every semester. As I read the article, however, I was disturbed by what was missing from Edwards’s account. Jerry Coyne’s rejoinder “Why We Shouldn’t Bet on Having Fee Will” only deepened my frustration.
Edwards vs. Coyne
William Edwards frames the quarrel as an argument between thinkers who believe in free will and those who believe we live in an entirely deterministic universe. He argues that it feels like we’re in control of our actions in a way that makes us morally responsible for those actions but that this feeling is “dismissed as an illusion by serious, contemporary neuroscientists.” Rather than bowing to the verdict of the neuroscientists, though, Edwards recommends that we see the question of whether free will exists as “the Pascal’s Wager of the twenty-first century.” There’s “too much about the universe that we don’t understand” for us to be confident that the neuroscientists are right, so it’s appropriate for us to decide what to believe on the basis of factors other than evidence. Along the lines of Pascal’s original argument for belief in the existence of God, Edwards says we have much to gain and nothing to lose by understanding ourselves to be free and morally responsible beings. As such, we should believe that determinism is false.
In his response, Coyne makes several good points. He understates the potential downsides of abandoning our belief in moral responsibility, but he’s on firmer ground when he argues that, although we have good reason to wish that something is true, this isn’t a good reason to believe that it is true. Moreover, we may not be any more able to force ourselves to believe something that we know defies the available evidence “than a tiger could force itself to turn vegetarian.”
Coyne does not, however, question Edwards’s assumption that a kind of human agency and determinism are incompatible. In fact, he explicitly affirms it. Thus, both articles exclude the possibility that, as I’ve argued elsewhere, we don’t have to choose between the benefits of believing in human will and moral responsibility on the one hand, and forming our beliefs on the basis of evidence on the other. That’s a shame because—in academic philosophy, at least—the question of whether agency and determinism are compatible is what the quarrel is about.
Jerry A. Coyne
Hume and Later Compatibilists
David Hume was exactly the kind of Enlightenment philosopher who regarded belief in empirically baseless but attractive claims the way that a tiger might regard a plate of tofu. Hume argued against belief in Heaven and Hell, for example, at a time when that sort of thing could land a person in serious trouble. Nevertheless, he argued that we have free will.
Coyne and Edwards both seem to think that the kind of “free will” we’d need to be held morally responsible for our actions is “contra-causal free will.” In other words, if we’re truly free, then our decisions can override a physical chain of cause and effect. Hume didn’t think we have this kind of freedom. In Chapter Eight of his Enquiry Concerning Human Understanding (“Of Liberty and Necessity”), he argued that, even when we don’t know everything about what led Person P to decide to perform Action A instead of Action B, we shouldn’t jump from that premise to the conclusion that there isn’t a coherent cause-and-effect explanation. Hume compared human psychology to clockwork. When a clock stops working, a peasant might only be able to report that “sometimes they stop working,” whereas a trained clockmaker would be able to find the tiny grain of sand that got caught in the gears. Nevertheless, Hume thought, as long as P picked A because he preferred A rather than because, for example, an assassin threatened to shoot him if he picked B, P chose A of his own free will. In other words, Hume thought that the freedom we need to be responsible for our actions is freedom from coercion.
This was an early version of compatibilism, and it’s vulnerable to some obvious objections. What about a kleptomaniac who steals a bauble to satisfy his compulsion, or a heroin addict feeding an addiction? More fancifully, what if a mad scientist were to implant a chip in P’s brain that caused him to have an overwhelming desire to do A? In each of these scenarios, coercion isn’t present, but P doesn’t seem to be in control of his actions in a way that makes him responsible for them—or, at the very least, he seems to have diminished control and hence diminished responsibility.
Many post-Hume compatibilists would say that in order to be responsible, P must be free not just from coercion but from other factors such as addiction, compulsion, and manipulation by brain implant. Nevertheless, they do not claim that P must be placed entirely beyond the influence of genetic and environmental factors to be responsible.
Responsibility and Punishment
The compatibilist conceptions of free will that have been discussed so far might strike some readers as silly and uninteresting redefinitions of free will. Coyne has certainly suggested as much elsewhere. According to this critique, “humans have free will if you define ‘free will’ in a compatiblist way” might be equivalent to saying, “God exists if you define the sum total of all material things that exist as ‘God.’” Any atheist worth his salt would respond to this by saying that in that sense God exists, but who cares? This simply isn’t the “God” whose existence is disputed by atheists.
This objection misses the point. The argument about whether to use the phrase “free will” to describe (a) freedom from coercion, (b) some more sophisticated compatibilist definition of freedom, or (c) contra-causal freedom might indeed be a matter of different thinkers talking past one another. Even the semantic argument about which one of these accounts best fits what ordinary speakers mean when they use the term is a matter of relatively limited interest. The question of which of these kinds of freedom is necessary for moral responsibility, however, is an extremely interesting question.
To see why it matters whether people are ever morally responsible for their actions, think for a moment about criminal punishment. The idea that at least some criminals deserve to be punished only seems to make sense if those criminals are morally responsible for their crimes.
The philosopher Derk Pereboom, a skeptic about free will, has argued that we can justify punishing violent criminals even if we don’t believe them to be responsible for their crimes. He uses the analogy of quarantining plague victims. It’s important to isolate them from the general population for the sake of public safety even though they don’t “deserve” to be locked up.
The problem with this line of thought is that it doesn’t seem to be capable of making sense of some of the most important modern ideas about how criminal justice systems should work. Consider Blackstone’s maxim that it is better for 10 guilty persons to go free than for one innocent person to be sent to prison. This is the moral calculation built into the “innocent until proven guilty” standard. But would we apply a parallel standard of evidence to quarantining plague victims? It’s hard to imagine a public health official reasoning that it’s better for 10 plague-ridden individuals to go around spreading the disease than for one uninfected person to be placed in quarantine. The idea that criminal punishment isn’t like this—that it is categorically wrong to punish the innocent—seems to rely upon a moral distinction that disappears when we say that no one is responsible for their actions.
Frankfurt and Fischer
Even if we (a) accept that we live in a deterministic universe and (b) think that it’s implausible that no one is ever responsible for their actions, it is still possible to reject compatibilism. After all, if it’s literally impossible for individual P to take option B, given all the previous links in the relevant chains of cause and effect, it might seem obviously wrong to say that P “was free to” do B.
This line of thought has led compatibilist philosophers like Harry Frankfurt and John Martin Fischer to slice up their conceptual categories in a more cautious way. Frankfurt and Fischer are willing to concede the semantic issue and grant that if determinism is true, we can’t have free will understood as the freedom to do otherwise. They argue, however, that in a completely deterministic world, we’re still free to exercise the kind of control necessary to be morally responsible.
In a famous paper, Frankfurt defends this distinction with a hypothetical example. His original example has spawned a whole literature in which different philosophers have used similar examples—generally called “Frankfurt cases”—to debate his ideas. One of my favorite of these “Frankfurt cases,” suggested by Joshua Spencer, a philosopher who agrees with Frankfurt and Fischer’s version of compatibilism, takes the classic Grandfather Paradox for time travelers and turns it on its head.
Suppose that Martin is a time traveler; he travels in a machine that can transport him to various points in earth’s history. During one of his many trips, Martin rescued a man from plummeting to his death. Let us suppose the man rescued was a high-wire walker who was working without a net in a very desolate area with no one nearby. Martin arrived in his time machine just in time to see the man fall from his wire, head first toward the ground. Luckily, Martin quickly found a button labeled “Emergency Safety Net Release.” Martin pressed the button and an emergency safety net was deployed across the field underneath the high-wire walker. The walker landed safely in the net and walked away from the situation unscathed. If Martin had not pushed the button and released the emergency safety net, then the high-wire walker would have fallen to his death.
As it turns out, this high-wire walker was Martin’s grandfather at a very young age. At the time of his fall, Martin’s high-wire walking grandfather had not yet met Martin’s grandmother and he had not done anything that would have preserved his gametes for posterity. So, since Martin’s grandfather would have died without any progeny, it is clear that (1) if Martin had not released the emergency safety net, then Martin would never have existed. But, of course, it’s not possible for Martin to do anything if he never exists. In particular, (2) it’s not possible for Martin to do otherwise than release the emergency safety net if he never exists.
Despite the fact that Martin wasn’t free to do otherwise, Spencer thinks, Martin seems to deserve our praise for saving this man (especially since he didn’t know the high-wire walker was his grandfather). Martin would not deserve praise had he depressed the button by stepping on it without realising, even though the outcome would be the same. The difference is that, in the scenario Spencer described, Martin was in control—even though he wasn’t free to do otherwise.
John Martin Fischer explains this distinction as the difference between “regulative control” and “guidance control.” He asks us to imagine two scenarios. In the first, you’re guiding a car that’s functioning properly. You turn the steering wheel to the right and the car turns right. Had you turned it to the left, the car would have turned left. In the second scenario, you’re turning right in exactly the same way. Nothing about the cause-and-effect relationship between the way you turn the wheel and the way the car moves is different in this second scenario. However, this time, the car is malfunctioning such that it can’t turn left. Perhaps the steering wheel would jam if you tried to turn left. In the first scenario (but not the second), you’re exercising “regulative control” over the direction of the car. In both scenarios, though, you’re exercising “guidance control.”
Fischer’s idea is that, even in a deterministic world where we aren’t free to do otherwise, we have “guidance control” over our actions just so long as we’re appropriately ‘reasons-responsive’—that is to say, we understand reasons for and against possible courses of action and we’re at least somewhat moved by such reasons. Building on Fischer’s idea, Ryan Lake argues in his doctoral dissertation, No Fate But What We Make: A Defense of the Compatibility of Freedom and Causal Determinism, that the reasons that move us must be our reasons in order for us to be exercising the guidance control necessary for moral responsibility.
To get a better sense of Lake’s idea here, think about property rights. If I find a precious stone on a beach that no one has laid claim to, it’s fine for me to take possession of it. If I find the same kind of stone in a friend’s pocket, it would be wrong of me to take it without asking. In both cases, the stone comes from outside of me. I didn’t make it. The difference is that in the first case, I’m the only person with a claim to it. Similarly, if the mindless operation of natural laws playing out through a complex combination of genetic and environmental factors results in my weighing the pros and cons of a possible course of action in a certain way and coming to a particular decision, the reasons that motivated me are my reasons. If a mad scientist’s chip in my head brings about the same result, they are not. In both cases, my decision is determined by factors outside of me, but only in the first case do I have a unique “claim” on my reasoning.
Conclusion
These are obviously complex ideas about which there’s room for reasonable people to disagree. I don’t claim that anything I’ve written here constitutes a decisive defence of compatibilism. But I hope to have shown that compatibilism is at least plausible and that it deserves to be taken into consideration. We can’t have a complete conversation about whether to reject determinism in favor of betting on the existence of free will (as Edwards wants us to) or reject the existence of free will in the name of determinism (as Coyne wants us to) without at least seriously considering the possibility that this if a false dilemma.
