Like consciousness, free will has an aspect that seems outside the scope of scientific investigation almost by definition. Are “you” the author of your choices? Well, what exactly do we mean by “you”? No matter what sequence of physical causes, random events, or even supernatural interventions led up to your making a particular choice, a skeptic could always claim that the choice wasn’t really made by “you,” but only by an impersonating demon that looks and acts like you.
On the other hand, there’s another aspect of free will that’s perfectly within the scope of science. Namely, to what extent can your choices actually be predicted by an external entity constrained by the laws of physics? Obviously they’re at least partly predictable: advertisers, seducers, and demagogues have always known as much, and modern fMRI scans confirm it! But is there any limit to the accuracy of prediction? If there isn’t, then can at least the probabilities be predicted to arbitrary accuracy, as they can in (say) the case of radioactive decay? Or are they subject to Knightian uncertainty?
Of course, even supposing your choices were unpredictable in the strongest sense imaginable, a philosopher could always say that that’s just a practical problem, and still doesn’t mean that your choices are in any sense “free.” Well, fine—but let’s at least try to answer the “straightforward empirical question,” of whether your choices are predictable or they aren’t!
Now, it’s on this latter, predictability question that quantum mechanics might finally become relevant. Or it might not—we don’t know yet. But it’s at least conceivable that the No-Cloning Theorem puts some fundamental limit on how well you can learn the state of a chaotic dynamical system, like (say) a human brain, and that it might forbid you from making a second copy accurately enough to “instantiate a new copy of the person, with the same will” (whatever that means!). Again, I stress that I have no idea whether this is true or not; many people feel confident that the “classical, macroscopic, non-invasively measurable” degrees of freedom should already contain all the relevant information. If it were true, though, then as many others have pointed out, it could have all sorts of “applications” to resolving classic philosophical paradoxes involving brain-copying, by simply explaining what goes wrong when you try to set up the paradox.
Even more from the same source: