Ur-axioms in Libet experiments.

I’d been wondering for a while why the Libet experiments would make a difference to free will arguments when I fell across Sam Harris arguing that any experimental outcome would have been consistent with the denial of free choice (see 6:40 here). The case against freedom can be made from the belief that the brain suffices to initiate behavior, which seems to be an ur-axiom that Libet himself is assuming. Even if Libet hadn’t found evidence of the brain preparing to move before the choice to move, and even if fMRI’s didn’t allow us to predict choices slightly better than chance, the choice itself is just another thing that happens to us. We simply find ourselves choosing or desiring something, and would find ourselves desiring something else if we had a brain tumor or different genes. What’s the difference between desiring something because brain structure demands it and desiring it because a mad scientist has found a way to manipulate us into desiring it, there being no freedom in either case?

True, cognitive science is assuming that the brain is in charge, but the hypothesis hasn’t crashed and burned yet, so the success of the cog-sci is evidence for the truth of the assumption, right? Well, no, not by the standard of proof that science holds itself to. We’d also need to see how successful the opposite hypothesis would be and then compare the outcomes. But this gives us a logical problem: if we think MN is necessary for science and want to prove it scientifically, we’d set up a science with MN and a science without one, which is to assume from the outset that our hypothesis is wrong. The problem generalizes to an incompleteness theorem: any attempt to prove what is necessary for science cannot be done scientifically since it requires begging the question. If you want to have the logical rigor of proving your case with control-experimental groups, you will need at least one foundational premise that cannot be proven in this way.

Harris recognizes that his argument against freedom goes through irrespective of experimental findings, but he’s resting his case on ur-axioms that ground science as such, namely (a) Every action is nothing but initial conditions and formal rules, and (b) all initial conditions (or at least all the ones after the Big Bang) arise from actions. Sean Carroll makes the same assumptions in his arguments against the existence of human souls.

What’s fascinating about (a) and (b) is that they are proposed as explanations of action which can’t account of the initiation or principle or arche of any action, since they give no account of how one could move from not acting to doing so. An initial condition is indifferent to action or stasis, and an abstract rule has no power to shove particles around. In fact, Harris and Carroll are both working from an ur-axiom that natural processes are not initiated* or, what is the same thing, any initiation of a natural process comes from outside nature. Alas for their arguments, this is the sort of belief that both allows for and is much more compatible with the existence of God and the freedom of intellectual substances. MN is better grounded by ~N than N.

Notice that this ur-axiom leaves everything in science in place. There is no need to touch or modify the Dirac equation to allow for some sort of ridiculous souleons or freewilleons or self fields to shove matter around and create the initial conditions of freely chosen brain states. Energy as such is posterior to all initiation of action and so has no role to play in its initiation. It stays as conserved as it ever was, and follows all the rules we have discovered it to follow, only it does so now as a secondary cause, the way that torque explains the motion of the wheels in your car. You need nothing other than energy to cause a natural motion just as** you need nothing other than torque to turn a wheel. It suffices to explain why it’s rotating, doesn’t it? You can even talk about torque initiating the motion, and there is a perfectly acceptable sense in which it does. For all that, this doesn’t mean that it’s torque pushing pistons, and in this sense it does not initiate the motion.


*One could speak of an “initiation” that was arbitrary or for the same of experimental convenience, of course, but this is by definition not a feature of the natural world.

** the “just as” indicates being analogous or comparable, not being identical.

%d bloggers like this: