The incoherence of free will
What does it mean to have free will?
Posted Nov 19, 2009
I recently re-read a classic piece by J.L. Mackie (April 1955), entitled “Evil and Omnipotence,” a stupendous philosophical essay about why theologians like Richard Swinburne are forced by their belief in an omnipotent, omnibenevelont and omnipowerful god into incredible and rather painful feats of mental gymnastics. One of Mackie’s minor points in the essay is that the so-called “free will defense” for the existence of evil in the world is problematic because the concept of free will itself is incoherent. Although, sometimes accusations of incoherence are thrown around a bit too easily in philosophy, I think this one has the potential to stick. (Mackie goes on with a devastating critique of the free will defense, a critique that remains effective even if the core concept should in fact prove to be coherent.)
Philosophically speaking, I still think that the best treatment of free will is the one given by Dan Dennett in his Elbow Room, which is a delightful book to read in its own right. Nonetheless, one may wonder whether the concept that emerges from Dennett’s analysis is in fact what most people would recognize as “free will.”
Of course, both words making up the term have the potential to be problematic, since it is not necessarily clear what we might mean by “will.” However, for the purposes of this discussion I will simply say that the will — insofar as human beings are concerned — is whatever set of motivations (and underlying neurological mechanisms) are behind the fact that we do certain things rather than others or, indeed, that we do anything at all. (Indeed, patients affected by severe damage to their amygdalas, for instance, seem to loose the will to do anything, likely because they've lost any emotional attachment to themselves and to things in the world: just like David Hume famously predicted, without emotions “It is not contrary to reason to prefer the destruction of the whole world to the scratching of my finger.”)
Moreover, I do not see a problem in, for instance, the Aristotelian concept of “akrasia,” or weakness of the will. Some people find it contradictory, because if I end up doing something out of my own volition — like eating a piece of chocolate cake — I cannot simultaneously claim that I did this “against my will,” because I knew that eating chocolate cake isn’t healthy. However, any human being who has struggled with food, sex, and other desires can make perfect sense of the idea of a weak will that makes you act against your own best interest even when you know perfectly well where such interest lies.
Anyway, back to the “free” part of free will. The obvious question is: free from what? That’s where coherence quickly becomes a problem. Unless you are a dualist — a thankfully dying breed among philosophers — you can’t possibly mean free from causal interactions with matter/energy, i.e. independent of the laws and materials of the universe. The will, whatever it is and however we like to conceptualize it, is grounded in the biological activity of our neurons. And last time I checked our neurons are made of matter, exchange energy (in the form of electrical currents and chemical reactions), and are subject to the laws of physics. So if that’s what you mean by “free,” it’s a no starter.
The next popular argument for a truly free will invokes quantum mechanics (the last refuge of those who prefer to keep things as mysterious as possible). Quantum events, it is argued, may have some effects that “bubble up” to the semi-macroscopic level of chemical interactions and electrical pulses in the brain. Since quantum mechanics is the only realm within which it does appear to make sense to talk about truly uncaused events, voilà!, we have (quantistic) free will. But even assuming that quantum events do “bubble up” in that way (it is far from a certain thing), what we gain under that scenario is random will, which seems to be an oxymoron (after all, “willing” something means to wish or direct events in a particular — most certainly not random — way). So that’s out as well.
It now begins to look like our prospects for a coherent sense of free will are dim indeed. If it ain’t random-quantistic or independent from causal interactions with the rest of the world, in what sense is it “free”? But if the will is not free, are we then not simply lumbering robots at the mercy of a mechanical, uncaring universe? (Or, worse yet, puppets in some god’s hands?) This conclusion strikes most people as intuitively deeply unsatisfactory. Moreover, wouldn’t that mean that human behavior would be predictable, at least in principle, if reductionist/mechanistic science became sufficiently advanced? That also strikes many as clearly off the mark.
One possible response is that, frankly, if the conclusions of a rational analysis go against your deepest held intuitions, so much the worse for your deepest held intuitions. But of course we also know that there are in fact non-deterministic physical systems (the time of decay of an individual atom, for instance), and we even know of perfectly deterministic systems whose behavior is for all effective purposes impossible to predict (chaotic, i.e. highly non-linear systems whose status at any given point in time is highly sensitive to initial conditions). So having a will that is causally connected to the rest of the physical world does not imply that our behavior is rigid or predictable.
Still, does that mean that we are in fact lumbering robots, whose illusion of being free is a combination of our ignorance of the causal web within which we are embedded and of our limited ability to compute our own future status? I think the best answer here comes from research in the cognitive sciences, which increasingly points to (at least) two levels of decision making in the brain: on the one hand, we now know that our subconscious makes a lot of decisions before we are consciously aware of them (think of those experiments showing the time-delay in electrical potential between when a muscle is being activated to perform a given action and when the subject becomes aware of having made the decision to perform that action, for instance). On the other hand, consciousness still seems to be a bit more than just a “rationalizing” process, taking on instead the role of high-level filter, or moderator, of unconscious brain processing (e.g., we can still stop an ongoing action if our conscious attention becomes focused on it).
What all of this seems to suggest is that the undeniable feeling of “free will” that we have is actually the result of our conscious awareness of the fact that we make decisions, and that we could have — given other internal (i.e., genetic, developmental) and external (i.e., environmental, cultural) circumstances — decided otherwise in any given instance. That’s what Dennett called a type of free will that is “worth having,” and I consider it good enough for this particular non-dualist, non-mystically inclined human being.