Free will has recently come up in other threads, and as such I thought I'd start a thread about it. What is free will? In this thread I will be talking about the particularly philosophical topic that has been debated for millenia--literally--and about which, in the words of the Stanford Encyclopedia of philosophy, "just about every major philosopher has had something to say."
WALL OF TEXT AHOY
1: The Positions:
Speaking of the Stanford Enyclopedia of Philosophy, I can think of no better source to give the general overview:
“Free Will” is a philosophical term of art for a particular sort of capacity of rational agents to choose a course of action from among various alternatives... Most philosophers suppose that the concept of free will is very closely connected to the concept of moral responsibility. Acting with free will, on such views, is just to satisfy the metaphysical requirement on being responsible for one's action... But the significance of free will is not exhausted by its connection to moral responsibility. Free will also appears to be a condition on desert for one's accomplishments (why sustained effort and creative work are praiseworthy); on the autonomy and dignity of persons; and on the value we accord to love and friendship.
The main perceived threats to our freedom of will are various alleged determinisms: physical/causal; psychological; biological; theological. For each variety of determinism, there are philosophers who (i) deny its reality, either because of the existence of free will or on independent grounds; (ii) accept its reality but argue for its compatibility with free will; or (iii) accept its reality and deny its compatibility with free will.
Those (i)-(iii) listed there give us the trilemma mentioned in this thread's title: they represent the standard responses to the problem of free will.
, in this context, has nothing to do with the politics. Instead, it expresses the attitude that determinism is false, and because determinism is false it is possible for our wills to be free.
, by contrast, is the attitude that determinism is true and that it is therefor impossible for our wills to be free.
, finally, is the attitude that determinism is true (or that it might as well be), but nonetheless our wills can be free. Spoiler alert: this one is right.
2: The Problem
We said about that libertarianism, determinism, and compatibilism all are responses to a problem, the problem of free will. But we still haven't actually said what that problem is! So here is a fast and loose version: according to our ordinary concept of ourselves and the world, we are only responsible for things that are under our control. It can be my fault that I did not mail a letter on time if I waited too long to leave for the post office, but it cannot be my fault that I did not mail a letter on time if there was an earthquake which destroyed the post office. Lack of control implies lack of responsibility, and this is true even in cases when there is no obvious intervening force from the outside: for instance, it can be my fault that I haven't picked up groceries if it is because I watched a movie instead, but it cannot be my fault that I haven't picked up groceries if I am an agoraphobe who is psychologically incapable of leaving the house.
But, once we accept this principle, then the various determinisms mentioned above threaten to make all our action look like it is of the second sort: out of our control, and hence, out of our responsibility. For instance, consider causal determinism. If the total state of the world as it was in 1000 AD settled how it would be for every moment onwards, then it follows that the total state of the world at 1000 AD must have settled whether my letter would arrive at the post office on time or not. But then how can it be my responsibility whether the letter is on time? It is certainly not my responsibility that the world be a certain way in 1000 AD, for that is obviously out of my control. And if I am not responsible for the state of the world at 1000 AD then I also cannot be responsible for anything that necessarily follows from it being that way. And so I cannot be responsible for the fate of my letter.
Now, someone might object: look, our best science tells us that determinism is not actually true. Quantum effects are indeterministic, and hence, the state of the world at 1000 AD did not settle how it would be for every moment onwards. But it's important to note that physical/causal determinism is just one of the determinisms that we can take to be threatening. We can take “higher level” determinisms (psychological, biological, social) to present exactly the same problem. For suppose it were the case that some set of biological, social, and/or psychological facts from the day before you were born settled whether your letter would arrive at the post office on time: for instance, once we settled the facts of your father's nasty temper and your mother's chronic laziness, along with your household income, neighborhood, and local public school, it necessarily followed that your letter would not get there on time. These things are just as obviously out of your control as the state of the world at 1000 AD, and so all the same consequences would follow.
So what we really must worry about is not the determinism of particles, or families, or middle-sized objects, but rather, something like the determinism of the total order of nature. Another way to put it is: what we must worry about is, when considering our actions, whether we can find anything that actually comes from us. This sentiment finds clear expression in Thomas Nagel's “The View from Nowhere:”
From the inside, when we act, alternative possibilities seem to lie open before us: to turn right or left, to order this dish or that, to vote for one candidate or the other—and one of the possibilities is made actual by what we do... but from an external perspective, things look different. That perspective takes in not only the circumstances of action as they present themselves to the agent, but also the conditions and influences lying behind the action, including the complete nature of the agent himself. While we cannot full occupy this perspective toward ourselves while acting, it seems possible that many of the alternatives that appear to lie open when viewed from an internal perspective would seem closed from this outer point of view, if we could take it up. And even if some of them are left open, given a complete specification of the condition of the agent and the circumstances of action, it is not clear how this would leave anything further for the agent to contribute to the outcome—anything that he could contribute as a source, rather than merely as the scene of the outcome... from an external perspective, then, the agent and everything about him seems swallowed up by the circumstances of the action; nothing of him is left to intervene in those circumstances.
Hopefully by now it should be at the very least clear what the problem is supposed to be. There is something threatening which Libertarians and Determinists both agree would rule out the possibility of free will. They differ just insofar as to whether they accept some form of the above physical/social/psychological/total order of nature/etc. determinism to be true, and hence accept that there can be no free will, or reject said determinism as false, and thus allow for such a thing as free will.
I really could go on and on here by offering yet more various descriptions of this problem, or trying to nail down a truly precise statement of exactly what is at stake. But I don't think that would be productive. For my ultimate conclusion is that the problem itself is deeply confused.
3: Libertarianism and Determinism
Why do I think the problem of free will is confused? The best way to approach this issue is to start by pointing out that the problem really isn't the problem of determinism per se
. It is the problem of efficacy—of seeing ourselves as the efficacious, and hence, responsible, authors of what we do—and that is a problem to which determinism and indeterminism are only incidental.
Consider this: suppose that there was a radium atom in my head, and that were it to decay it would result in my delivering my letter to the post office on time. If this were the case then it would be indeterminate whether my letter got to the post office on time, but that indeterminism doesn't appear to make me any more responsible for the result. By positing a radium atom in charge we have made me slave to an indeterminate process, but a slave all the same; it would be ridiculous to hold a person responsible for whether a radium atom decayed. Similarly, suppose that all my attitudes, beliefs, desires, and etc. are indeterminately caused by the world around me. And, in turn, they indeterminately cause certain actions to issue from me. The only thing that this changes from our previous deterministic picture is whether the forces that are acting through me are reliable. It doesn't explain how there could be a force that is specifically from
me. Indeterminism in and of itself is incidental to efficacy, and efficacy is what we are actually after.
But what of efficacy? If we conceive of it as a contribution of our own, one that is completely independent of the (in)deterministic forces operating through us, then it is an incoherent aspiration. Consider:
[T]o really be free we would have to act from a standpoint completely outside ourselves, choosing everything about ourselves, including our principles of choice—creating ourselves from nothing, so to speak.
This is self-contradictory: in order to do anything we must already be something. However much material we incorporate from the external view into the grounds of action and choice, this same external view assures us that we remain parts of the world and products, determined or not, of its history.
I think these considerations are sufficient to show that the libertarian is confused; she is reaching for something that she can never grasp. But where does that leave us? It might make us into determinists—not, as the name implies, because we think the world is actually deterministic, or that it would even be relevant if it were. But rather, because we agree with the libertarian about what is required for efficacy, but, unlike the libertarian, we can see that it is impossible. And hence we give up on the idea of free will altogether.
But this is at least prima facie
not an attractive position. Because, after all, it requires us to repudiate all sorts of everyday notions we have about our own agency and ability to interact with the world. If there is a slightly different notion of efficacy, one which would avoid committing us to a massive error theory about the folk concept, then that would count as a good reason to employ it instead. Frank Jackson is eloquent on this point in From Metaphysics to Ethics
4: Identity Theory and Efficacy
[M]y pre-analytic conception of free action is one that clashes with determinism. I find compelling Peter Van Inwagen's argument that because the past is outside our control, and any action fully determined by something outside our control is not free, determinism is inconsistent with free will. And so do many. Even the most dedicated compatibilists identify it as the argument they need to rebut. What compatibilists arguments show, or so it seems to me, is not that free action as understood by the folk is compatible with determinism, but that free action on a conception near enough to the folk's to be regarded as a natural extension of it, and which does the theoretical job we folk give the concept of free action in adjudicating questions of moral responsibility and punishment, and in governing our attitudes to the actions of those around us, is compatible with determinism. There is, accordingly, an extent to which the compatibilist is changing the subject, but it is a strictly limited sense. For compatibilists do, it seems to me, show, first, that the folk concept of free action involves a potentially unstable attempt to find a middle way between the random and the determined, second, that the folk conception is nowhere instantiated, and, third, that the compatibilist substitute does all we legitimately require of the concept of free action. If is hard to see how we could better motivate a limited change of subject.
In some ways, we can see a parallel issue of efficacy in the philosophy of mind. In the 19th century there was a popular view called Epiphenomenalism; on this view, the mental and the physical were separate and parallel. When an epiphenomenalist's finger was pricked, there would be a chemical pain signal that raced through his body, and there would also be a mental pain signal that raced through his mind; but it was never the case that the lines were crossed and anything in his mind actually caused anything in his body. The pain in his mind did not cause him to recoil: instead, the pain in his body caused him to recoil. The pain in his mind was just an extra on the side, a ghostly other that existed over and above the physical goings on. The reason I take this to be topical is because this view gained in popularity at the same time that science was unlocking more and more of the secrets of the biological human organism; as scientists and philosophers began to see a series of causes and forces in the objective order of the world—chemical transfers, neural firings, muscle contractions—that could explain a physical action, suddenly they could no longer see any room left over for the person themselves, in the form of their minds, to have caused that very same action. Suddenly, just as in the free will debate, the appearance of forces operating through a person seemed to negate any possibility of a force originating from them. There was nothing left for that concept to be.
One response to Epiphenomenalism, in philosophy of mind, is Identity theory. I am not going to go into the details, which are both nitty and gritty, but on the broad picture Identity theory claims that the mental just is the physical. Those two components that the epiphenomenalist held separate, the chemical signal in the finger and the pain in the mind, were actually both one and the same. On that picture there is thus no problem explaining how pain in the mind could do something like cause a person to recoil—the pain in the mind is the very same thing as the objective physical event in the finger and has all the same causal properties. One might say that when the identity theorist was confronted by a series of objective events and forces in the world that seemed to leave no room left over for the agent himself to exert an influence, the identity theorist responded by identifying some of those objective events and forces as
the agent himself. And then there was no problem with seeing the agent as part of the objective order of the world, or as seeing him as the cause of various events in that order.
What I am trying to argue is this: if I am correct that the threat to efficacy comes from an objective picture of events and forces in the world that seems to leave no room for the agent herself to be efficacious, then the way to answer that threat is to make room in that objective picture for the agent. And the way to make room for the agent in that objective order is not to shoehorn her in as an uncaused cause or a sui generis force, but rather, the way to make room for her is by identifying her as actually being the very same thing as some of the events and forces in that objective order. And once we have done that, there is no mystery left over about how the agent can be efficacious, for we have now located her among the objective order of causally efficacious events and forces.
5: The Scientific World View--A Last Word
One might object to the previous strategy on the following grounds: one might say “look, once you have the objective order of events and causes sans
agent, there is no need to put her back in there. You have the complete picture! The fact that we can have such a picture without an agent's participation is just evidence that agents don't exist. We should really be determinists after all.”
For my response here, Christine Korsgaard in “Self-Constitution” is my go-to girl:
[T]he scientific or mechanistic conception of the world... is the result of pressing our understanding of the world until the idea of an object, as a unified and independent being within the world, begins to look spurious. You might think that you're an object, indeed even an agent, but to a flea or a nit you are merely a rather nutritious and specific region of the environment, like a Pacific island. If the flea or nit could thing, it would think itself an object, perhaps even an agent, but to the cells in its body it is merely a rather nutritious and specific environment... and so on. Even we self-identifying self-conscious and supposedly self-maintaining substances fail to see how thoroughly embedded we are in an environment that supports us from the outside, how thoroughly our perceived internal unity and cohesion depend on what goes on around us. A chemical change, a rise in the temperature, a stray bullet, and the transient whirling vortex of forces that thought itself an immortal thing puffs away...
Are the [conceptions of the world that involve agents] then related to the Scientific World View as illusions to fact? If that were so, whose illusions would they be?
Our concept of ourselves as agents is primary in a way that doesn't even make sense to doubt, let alone to try to eliminate from the picture. After all, if there were no such thing as an agent, then how could anyone doubt anything, let alone the reality of agents? That one can even doubt it only proves one's doubts wrong.