First caught wind of this at Roderick Long’s blog.
There’s been some discussion here and there about a recent study announced by the Max Planck Society on human consciousness. To quote from the press release I’ve linked:
In the study, participants could freely decide if they wanted to press a button with their left or right hand. They were free to make this decision whenever they wanted, but had to remember at which time they felt they had made up their mind. The aim of the experiment was to find out what happens in the brain in the period just before the person felt the decision was made. The researchers found that it was possible to predict from brain signals which option participants would take already seven seconds before they consciously made their decision. Normally researchers look at what happens when the decision is made, but not at what happens several seconds before. The fact that decisions can be predicted so long before they are made is a astonishing finding.
This unprecedented prediction of a free decision was made possible by sophisticated computer programs that were trained to recognize typical brain activity patterns preceding each of the two choices. Micropatterns of activity in the frontopolar cortex were predictive of the choices even before participants knew which option they were going to choose. The decision could not be predicted perfectly, but prediction was clearly above chance. This suggests that the decision is unconsciously prepared ahead of time but the final decision might still be reversible.
This is being heralded as a confirmation of the results of the famous experiments on people’s subjective experience of choice done by Benjamin Libet, which were widely claimed (though not by Libet) as proof that free will was illusory, due to the fact that electrical activity in the motor cortex of the brain (the “readiness potential”) appeared before the subject was conscious of choosing to act. I want to talk about Libet a little, because this study appears to have the same limitations, in terms of what conclusions we can draw about its implications.
First, however, there’s an elementary point of logic that needs to pointed out, which is this: Even accepting for the sake of argument that this research shows all actions are actually decided at the unconscious level and merely rationalized by the conscious mind afterwards, which I’m going to question in a minute, that would not prove determinism. As commenter Laura J. points out in the comments of Long’s post, the claim that this would prove determinism hinges on the unstated assumption that the unconscious processes running in the background of my conscious mind are not really “me,” that the self is only the fully conscious mind, and that unconscious influence on or control of it is an enslaving outside force, rather than an equally true part of my self. If you reject this assumption, the determinist would also need to prove that the unconscious mind is fully determined by something before he could claim to have shown that I- that is, the complete “I”- have no free will.
It should also be noted that, at best, this would refute only incompatiblist free will, and has nothing to say about compatibilism versus incompatibilist determinism. I don’t consider that distinction very meaningful, frankly- I’ve never heard an explanation of compatibilism that didn’t boil down to determinism with some of the terminology of free will thrown in- but many people would consider it important.
So, Libet. The first problem with a determinist interpretation of Libet’s work was, as Benjamin Libet pointed out, that his subjects would occasionally show the usual electrical activity in the motor center, the “readiness potential,” and then choose not to act. So it's a jump to assume that the electrical buildup demonstrates that the act is truly predetermined, and not merely an indication of a strong disposition to act. The press release for the new Planck study says that the accuracy of their predictions was imperfect but “clearly above chance,” with the precise percentage unspecified. The fact that modern scientists can make predictions of people’s movements seven seconds in advance with above-chance accuracy in this situation is extremely cool, but it’s pretty poor as a knockdown argument for determinism. A skilled bookie can predict the winner of a sporting event at a rate well above chance, but that hardly demonstrates that the competition is rigged. (Unless it’s boxing, of course.) It merely shows that the outcome of the game was affected by conditions in place before the game began. Unless, as some people do, you redefine “free will” to mean that choices are just random and completely unaffected by conditions in the physical world, such as electrical activity in the agent’s own brain, this doesn’t demonstrate much philosophically, however intriguing it might be as science.
There are some other problems, which wouldn’t go away even if the prediction in the experiment was never wrong. The actions taken by Libet's subjects were consciously preplanned. In Libet’s work, the subjects watched a timer, and were instructed to choose a random moment to hit a button, then report the time they perceived themselves willing to hit the button. The problem there is that the intention "hit the button" was, in an important sense, already consciously formed and chosen before the clock had even started. The subject had already decided to hit the button, already knew they were going to hit the button, and the only intention not yet consciously formed was the exact moment. In that sort of case, it's hardly surprising that the motor center was lighting up before the actual final decision, regardless of whether the subject actually had free will or not.
Another issue is that the action being taken- hitting the button- is completely random and meaningless, and the decision of what moment to hit the button is thus completely arbitrary. Thus, it is precisely the sort of thing that I would expect to be decided at an unconscious level, since there’s absolutely no reason for the conscious mind to care about the particular moment the button is hit, and thus no reason to deliberate about it at the conscious level, except perhaps to ratify a conclusion already reached unconsciously. We simply don’t know if the results are applicable to all experiences of choice, including those we have more reason to consciously deliberate on, and we have been given no reason to think that they are.
Based on the way the experiment is described, both problems seem to be present here as well. The press release says, “In the study, participants could freely decide if they wanted to press a button with their left or right hand. They were free to make this decision whenever they wanted, but had to remember at which time they felt they had made up their mind.” So, again, the subject had already decided to hit the button, and he knew that, barring some sudden freak occurrence that forced him to flee the room, he would carry out that decision. So, once again, the brain gearing itself up to act before the final moment of decision is to be expected with or without free will. The only new wrinkle appears to be having a choice of two buttons. The fact that they could predict which would be chosen is neat, but not terribly decisive, since it's not at all hard to imagine a free-willed person starting with a strong predisposition for one button over another without consciously realizing it. As in the choice of timing, the choice of buttons is completely meaningless and thus gives the conscious mind no reason to bother to make the decision itself, though it might will the actual execution of the decision.
Theses sorts of experiments are limited by the limitations of human focus. Since the subject has to concentrate on precisely monitoring and reporting his own mental state, he can’t really engage in any acts that aren’t tainted by a lengthy gap between resolving to act and actually acting, since a situation where he had to make a quick decision in response to something unexpected would occupy too much of his attention. Likewise, he can’t make any decisions that he actually has a reason to care about and consciously think through, while simultaneously closely monitoring and reporting the precise moment he became aware of his choice. If genetic engineering really takes off, we should see if we can crank out some posthumans with huge heads and extra brain lobes who can maintain multiple trains of thought at once. Short of that, the philosophical implications of this are limited.
For maximum effect, you should now reread this post, but this time do so aloud by shrieking out the entire thing in your best Geddy Lee impression. Actually, henceforth, I want everybody to read everything I write that way.