For some reason that I don’t understand, I feel compelled to reiterate why I don’t believe in free will—at least in the common notion of free will. Perhaps it’s the combination of my genes and my environment, particularly the environmental stimulus provided by Massimo Pigliucci in a new post at Rationally Speaking, ”Jerry Coyne on free will.” Needless to say, Massimo takes me to task again, especially for my anti-free-will article in USA Today, “Why you don’t really have free will.“
I’d like to reiterate and then counter Pigliucci’s beef with my arguments. Here are his main points and my counterarguments.
1. My concept of free will is empirically untestable. As he says:
Before we continue, however, let’s hear Jerry’s definition of free will: “I mean it [free will] simply as the way most people think of it: When faced with two or more alternatives, it’s your ability to freely and consciously choose one, either on the spot or after some deliberation.” He continues: “A practical test of free will would be this: If you were put in the same position twice — if the tape of your life could be rewound to the exact moment when you made a decision, with every circumstance leading up to that moment the same and all the molecules in the universe aligned in the same way — you could have chosen differently.”
As Jerry knows, and immediately admits in the paragraph following this quote, such a test is anything but practical. In fact, it cannot be carried out, ever. Which is why I contend that Jerry and others who push the idea that free will (and consciousness, and moral responsibility) is “an illusion” are mistaken when they think they are doing so on the basis of science.
He argues, then, that my claim that we don’t have free will of this sort is not empirical but metaphysical.
I grant that we can’t rewind the tape of life, but that indirect evidence of two sorts suggests that my argument against free will is correct. First—and to me this is the decisive one—our brains are made of molecules. Molecules (and the neurons they make up) must obey the laws of physics. Our “decisions” are made by brains. Therefore our brains must obey the laws of physics. Absent any quantum indeterminacy, then (and that cannot constitute any portion of free will), what our brains do is determined and predicted by the laws of physics. Ergo, in any situation we could not have chosen otherwise: we cannot decide freely which of several alternatives we choose.
Suppose that an alien observes coin tosses, and sees that sometimes a coin comes up heads, and sometimes tails. The alien supposes that the coin is alive, and can “decide” which way to fall. The alien cannot exactly repeat each coin toss to see whether, given identical conditions, the coin can land differently. We can never repeat those initial conditions (though I’m told some magicians can bias the outcome by how they flip the coin). But we can conclude, without having to do the experiment, that the way the coin lands is absolutely conditioned by the laws of physics—by how it’s tossed, the air currents obtaining at the time, the height of the hand, and the nature of the surface on which the coin lands. We don’t have to do the experiment to conclude that.
Our brains are like that coin, except they’re far more complicated and made of meat instead of metal. But they still obey the same laws that govern coin tosses. Why, then, do our brains get to “choose” but a coin does not? Yes, we take in more inputs than does a tossed coin (but perhaps not many more), but so what? Physics still rules.
I argue that the onus for proof is on those people who claim that our decisions are free and not completely subject to the laws of physics. For these are the people who implicitly claim that our brains are free from physical law in such a way that allows us, at any given time, to freely decide among alternatives. It is not my responsibility to show that we have this sort of free will; everything we know about physics militates that.
Second, there are experiments like Libet’s (and better ones) showing that “decisions” can be predicted up to seven seconds before they’re made consciously. This time lag won’t always obtain, since some “decisions” are made more quickly, as when we decide how to hit a tennis ball; but if we can show that some decisions are made unconsciously, that militates against any conscious decisions, and to me conscious decision-making is essential for my form of free will to obtain (see below).
2. The unknowns of physics suggest room for free will. I note first that Massimo doesn’t define free will in his piece; he’s only arguing against my conception of it. At any rate, he argues:
Of course this conclusion depends on one’s concept of free will, and there are several on offer (more on this below). It also depends on entirely unargued for assumptions, including the following: causal closure (i.e., that the currently known laws of physics encompass the totality of causal relationships in the universe); a working concept of causality (one of the most thorny philosophical concepts ever); physical determinism (which appears to be contradicted by physics itself, particularly quantum mechanics); and the non-existence of true emergent properties (i.e., of emergent behavior that actually is qualitatively novel, and doesn’t simply appear to be so because of our epistemic limitations). I have opinions about all four of these points, but I don’t have a knockdown argument concerning any of them. The point is, neither does Jerry.
I’m just going by the currently known laws of physics, which appear to hold throughout the universe. As for determinism, yes, I do think there are true indeterminacies on the quantum level, but their influence on our behaviors and choices is dubious at best. And even if such indeterminacies did influence us, those influences would be random and not constitute any basis for free will.
As for emergent properties, those too must obey the laws of physics, unless you hypothesize an “emergent property of free will” that is somehow physically unconnected with lower-level processes. Yes, there are emergent properties that cannot be predicted from knowing about their constituents (the wetness of water may be one), but that wetness still must conform to the laws of physics obeyed by its constituent molecules. The properties of water do not thereby become free from the laws of physics.
3. Experiments like Libet’s and more recent ones like that of Soon et al. (2008) Fried et al. (2011, reference below) are irrelevant to free will. They show only that our decisions might not be conscious ones.
Not even Libet himself took his experiments to show that people don’t make conscious decisions, in part because reporting awareness of an urge (in this case, of pushing a button) hardly qualifies as a conscious decision. The latter is the kind of reflective deliberation that Jerry and I engaged in while composing our respective essays, and it is simply not measured by Libet-type experiments. Indeed, it is not surprising at all that we make all sorts of unconscious decisions before we become aware of them, as any baseball batter, or anyone catching a falling object on the fly, will readily testify.
My argument is simple here: “decisions” made unconsciously don’t buttress my concept of free will, for they could simply reflect the generation of outputs from inputs in the same way that a computer generates outputs from inputs. The common conception of free will—the one I’m addressing—requires making a conscious decision. If you choose vanilla rather than strawberry at an ice cream store because your neurons and taste buds have determined that in advance, in what sense is the decision “free”?
I find it strange when some people argue that “we” are not making decisions if our subconscious is operating, since presumably we all agree that our subconscious is just as defining of “us” as conscious thinking is. Accordingly, “my brain made me do it” is hardly a defense that will fly in a court of law except, and for good reasons, in pathological cases such as behaviors resulting from brain damage.)
Well, I contend that every act of malfeasance results from “brain damage”—at least the same kind of neurological determinism that may have caused Charles Whitman, who had a brain tumor, to murder people at the University of Texas. And I’ve never argued that “my brain made me do it” is a valid defense against criminal conviction. Even if we don’t choose our acts of criminality, there are good reasons to incarcerate criminals, including protection of society and as an example to deter others. Such examples are, of course, environmental influences that can affect the workings of our brains, and hence our future actions.
4. I argue, without evidence, that free will is an evolved property of brains. Massimo argues:
But just for the sake of argument let us suspend judgment on all of this and ask Jerry the obvious question: why do we have such a pervasive “illusion” to begin with? Apparently, he knew this was coming, and answered thus in the USA Today article: “where do these illusions of both will and ‘free’ will come from? We’re not sure. I suspect that they’re the products of natural selection, perhaps because our ancestors wouldn’t thrive in small, harmonious groups — the conditions under which we evolved — if they didn’t feel responsible for their actions.”
As far as I can tell there is no empirical evidence whatsoever to support such speculation. To the contrary, we know of plenty of social animal species that seem to thrive very well indeed without requiring the illusion of free will to keep them in line. Certainly social insects don’t need to be fooled that way, and it is hard to imagine even species of social mammals, including most primates, needing to engage in deliberate reasoning before deciding how to behave toward fellow group members.
Note that I said, “We’re not sure.” I really have no idea why we have the illusion of agency, and was just speculating that it may be an adaptation. But it might not be—it could be an epiphenomenon of having complex brains. By the way, our brains are far more complex than those of social insects, and we process many more inputs than those of, say, ants. A big ant could not function as a human being. And we have no idea whether animals engage in deliberate reasoning, though this morning’s kitteh post suggest that cats can, and I certainly think that primates can.
But it doesn’t matter. I have no idea about why we have the illusion of free agency, nor am I deeply invested in an evolutionary, much less an adaptive, answer.
5. I ignore other philosophers’ concepts of free will.
In the above comment Jerry also ignores that philosophers have been debating various concepts (not definitions, because they are not ex-cathedra pronouncements) of free will for a long time. Competing approaches to free will have been put forth, among others, by Plato, Aristotle, Descartes, Hume, and more recently Daniel Dennett and Harry Frankfurt, to name but a few. It is a profound mischaracterization of the history of philosophy to present various takes on free will as being simply reactive to the latest scientific discoveries.
I’ve read about the many ways philosophers have defined free will differently from me. And yes, if you change what you mean by “free will,” then you can find a way that we do have it. But I am addressing what I think is most people’s notion of free will, which I think is to some extent dualistic. Philosophers may have given up dualism, but my experience discussing this issue with others, including my biology colleagues, shows that almost without exception they have an unconscious dualism: that somehow we have some capacity to step inside our minds and influence their workings. To me, the only free will that matters is the ability at a single moment to choose freely between alternatives—that we could have done otherwise.
6. I argue that free will means the end of religion. Massimo:
Jerry claims that the death of free will spells the death of religion, although ironically he then mentions the Calvinist view of pre-determination. In fact, plenty of religious beliefs are compatible with lack of free will, so it seems like religion will survive even this assault (as befits an infinitely malleable tradition of made up stories).
I never argued this. What I said is that many important religious precepts depend on free will, and those will go away if free will is an illusion. Here’s what I said:
But there are two important ways that we must face the absence of free will. One is in religion. Many faiths make claims that depend on free choice: Evangelical Christians, for instance, believe that those who don’t freely choose Jesus as their savior will go to hell. If we have no free choice, then such religious tenets — and the existence of a disembodied “soul” — are undermined, and any post-mortem fates of the faithful are determined, Calvinistically, by circumstances over which they have no control.
I have no idea why Massimo thinks that this spells the end of religion. (By the way, an important argument in theodicy, that we have evil in the world because God gave us free will, will also go away.) Unless Massimo was tired when he wrote this, he’s failed to grasp my point. ”Many faiths” does not equal “all faiths.”
7. Why try to argue what we “should” do if we have no free will? In my piece I end by talking about the implications of realizing that we don’t have free will in the sense I define it. One is to give up the idea of punishment as retribution, another is to have more understanding for criminals, and not blame them for “making the wrong choice.” (I add here that perhaps we should not have so many regrets about our past, since we couldn’t have chosen otherwise.) Why would I make such prescriptions if I believe that all our actions are predetermined?
The answer is twofold. First, I make such arguments because I cannot do otherwise: this is what I have been conditioned to do by my genes and my personal history. But second, and more important, such arguments do affect people, and can influence their decisions. It’s a common misconception of those who argue against my views that those views ineluctably promote a kind of nihilism, in which we should do nothing. Well, we don’t have the choice to do nothing: we’re humans and we must act as though we have free will, even if we don’t. It is an all-powerful illusion—perhaps an evolved one—which conditions all our behavior.
But we shouldn’t feel that our behaviors are ineffectual. We can convince people to act in different ways through our words. Making people love us, or hate us, are simple examples. We are not powerless, though all of our behaviors—including whether we’re susceptible to the ministrations of others—are determined.
8. I am a “radical skeptic.” Pigluicci accuses me of this for several reasons:
In the end, skepticism about free will seems to me to be akin to radical skepticism about reality in general (the idea that all of reality is an illusion, or a computer simulation, or something along those lines): it denies what we all think is self-evident, it cannot be defeated logically (though it is not based on empirical evidence), and it is completely irrelevant to our lives.
But what is “self-evident” to people has been shown over and over again by science to be wrong! It is self evident to many that the “design” of animals and plants is evidence for a God. Free will may be one of these things. I think science can help us show logically that we don’t have free will in the way I define it, and if we don’t have it, then there are implications for our lives, some of which I mention in my USA Today piece.
I find it curious that Pigliucci says this at the end of his piece:
That said, we should then proceed by ignoring the radical skeptic in order to get back to the business of navigating reality, making willful decisions about our lives (including New Year’s resolutions, which actually succeed surprisingly often), and assign moral responsibility to our and other people’s actions.
How does he know that he can make “willful decisions”? And we really must rethink the idea of moral responsibility in a world in which there may not be free will. Perhaps we can dispense with the idea of moral responsibility along with the idea of free will! That may be radical, but it may also be sensible. We can replace the idea of “moral responsibility” with that of “actions inimical to others,” or something of the sort. The only things this would change would be our idea of punishment as retribution, as well as the underpinning of many religious beliefs.
Now I feel compelled to have some coffee.
Fried, I., R. Mukamel, and G. Kreiman. 2011. Internally generated preactivation of single neurons in human medial frontal cortex predicts volition. Neuron 69:548-62.
Libet, B. 1985. Unconscious cerebral initiative and the role of conscious will in voluntary action. Behav. Brain Sci. 8:529-566.
Soon, C. S., M. Brass, H. J. Heinze, and J. D. Haynes. 2008. Unconscious determinants of free decisions in the human brain. Nature Neuroscience 11:543-545.