I’ve spent a few more weeks reading about free will and the varieties of compatibilism and incompatibilism. And—much to the regret of some of my readers, I suppose—I haven’t changed my mind. I still don’t think that we can make real “choices” at any given moment; I feel that all of our choices are predetermined by the laws of physics and chemistry, and I think that all the attempts to save the notion of free will via philosophical “compatibilism” are unconvincing.
And my feeling that the common notion of free will—that at any given time, if the past history of an individual and all of her molecules were replicated, she would always choose the same way—was confirmed by discussions I had with three scientist colleagues. None of these colleagues had thought much about the problem of free will, but all of them, when pressed, thought of “free will” in the way I’ve characterized it. Further, all of them raised the similar objections to my claim that we have no free will in that sense: Wouldn’t that lead to nihilism? What about moral responsibility? But can’t people be persuaded to act in a certain way?, etc. This is an anecdotal and small sample, but it’s a sample of smart scientists, and all of them initially conceived of free will as the ability to make decisions independent of the laws of physics.
Before I talk a tiny bit about compatibilism, let me present this video, which shows a functional magnetic resonance imaging (fMRI) experiment like that used in the famous work of Bode et al. (see reference below for a free download), showing that one can predict the outcome of a decision up to seven seconds before the subject is conscious of having made a decision. The YouTube description says this:
In this clip, Marcus Du Sautoy (Professor of Mathematics at the University of Oxford and current Simonyi Professor for the Public Understanding of Science) participates in an experiment conducted by John-Dylan Haynes (Professor at the Bernstein Center for Computational Neuroscience Berlin) that attempts to find the neurological basis for decision making.
Hayes was one of the authors of the Bode et al. paper cited below.
It’s a complicated set-up, but the results and explanation are cool, even if you think they have no bearing on free will:
Now I’m perfectly aware that the “predictability” of the results is not perfect: it seems to be around 60%, better than random prediction but nevertheless statistically significant. I think, though, that as our ability to image and understand the brain improves, the predictability of which decision the subject will make will improve. After all, fMRI is rather crude, based as it is on blood flow to certain areas of the brain. And I know some will object that even if the decision was “predictable” up to seven (and probably ten) seconds in advance, it still could have been a decision, but an unconscious one.
I maintain that if a decision is unconscious—if it takes place in your head well before you’re aware of it—then that is not free will, which involves conscious decisions. After all, every “decision” has to be reflected somehow in brain activity that is correlated with an action, so we’d expect to see predictable pre-conscious brain activity if there were no free will. For those who say that seven seconds isn’t long enough, would you deny free will if I could tell you what flavor of ice cream you’d choose while you were on the way to the store, knew what flavors were on offer, but said you didn’t yet know what you wanted?
Now the version of free will I’ve adumbrated is contracausal free will, and it’s clear that I’m an incompatibilist—I believe that our actions and “decisions” are solely the results of the laws of physics and chemistry, and that such decisions are in principle incompatible with my definition of free will. But I think that nearly all smart philosophers and scientists agree with me on at least one point: our decisions are basically deterministic (perhaps tempered with a bit of quantum indeterminacy, which can’t be part of free will) and are the result of physical laws. Few people believe in mental/physical dualism thse days.
What people differ about is whether determinism removes our notion of free will. And so they concoct “compatibilist” definitions of free will—ones that make free will compatible with physical determinism. I have not found one of their arguments remotely convincing, for I adhere to the same notion of free will as most folks do, and am unwilling to change it to conform to some philosopher’s attempt at reconciliation. To me, free will means “I could have decided otherwise,” and if we can’t do that, then we don’t have free will. We have something else, and I wish that philosophers would use another term if they’re compatibilists.
I’ve read about compatibilism because someone asked me about the philosophical arguments for it. I’ve only found four or so that stirred me even remotely, but, as I said, none were convincing:
- Free will is shown when people’s decisions are seen to respond to reasoned argument. That’s not convincing for two reasons: reasoned argument is still an environmental influence which can impinge on the brain to affect people’s decisions. Second, whether or not someone is responsive to reasoned argument is itself determined by the laws of physics.
- Free will is shown when someone’s “decision” is compatible with their backgrounds, temperament, habits, and personality. This isn’t acceptable because it doesn’t show that someone is making a free choice—only a choice that’s consistent with decisions and actions they’ve evinced before. It doesn’t show that they could have chosen otherwise, either.
- Maybe you can’t decide freely to do something, but you can decide freely not to do something. This is the version of free will suggested by Benjamin Libet, who did the first experiment showing predictability of “choice” by brain imaging. Dismayed at the implications of his result, he suggested the idea of “free won’t.” That’s bogus, however, because you don’t have any choice whether to veto a contemplated action, either. (The icing on the cake is that “vetoing” takes place in precisely the same brain regions as “choosing.”)
- Free will represent the “choices” made by rational, contemplative beings whose faculties have evolved to weigh many factors before making a decision. This subsumes a number of ideas suggested by different philosophers, including Dan Dennett. I don’t find them convincing because to me they just show that our brains are complicated computers made out of meat, evolved to weigh lots of inputs before giving an output. But computers that spit out a single output—a choice—after absorbing many inputs are still computers, and we don’t think that computers programmed to respond to complicated inputs have “free will.” Does a chess-playing computer have free will? If you think so, then go tell it to the philosophers.
I still think that compatibilism represents a sort of kneejerk philosophical response to the fact that nearly everyone finds totally unpalatable the idea that we are automatons whose actions are completely determined by the laws of physics. And, as Harris says in his upcoming (and excellent) small book, Free Will, all versions of compatibilism essentially boil down to one pithy description:
“A puppet is free so long as he loves his strings.”
Bode, S., A. H. He, C. S. Soon, R. Trampel, R. Turner, and J.-D. Haynes. 2011. Tracking the unconscious generation of free decisions using uItra-high field fMRI. PLoS ONE 6(6): e21612. doi:10.1371/journal.pone.0021612